Client factors associated with psychotherapeutic process and ...

Client factors associated with

psychotherapeutic process and outcome:

A concept analysis

A thesis submitted to the University of Manchester

for the degree of Doctorate in Counselling Psychology (D.Couns.Psych.)

in the Faculty of Humanities

2018

Caroline Vermes

School of Environment, Education and Development

2 of 271

Contents Abstract ......................................................................................................................................................... 6

Declaration .................................................................................................................................................... 7

Copyright ....................................................................................................................................................... 7

Acknowledgements ....................................................................................................................................... 8

Dedication ..................................................................................................................................................... 9

The Author .................................................................................................................................................... 9

CHAPTER 1: INTRODUCTION ....................................................................................................................... 10

1.1 Chapter overview .................................................................................................................................. 10

1.2 Study problem ....................................................................................................................................... 11

1.3 Why concept analysis ............................................................................................................................ 16

1.4 Study purpose and questions ............................................................................................................... 21

1.5 Key terms .............................................................................................................................................. 25

Concepts ................................................................................................................................................. 25

Psychotherapy......................................................................................................................................... 27

Causal syntax in psychotherapy .............................................................................................................. 29

Outcome ................................................................................................................................................. 34

1.6 Study structure ...................................................................................................................................... 37

CHAPTER 2: LITERATURE REVIEW ............................................................................................................... 41


2.2 Common factors theory ........................................................................................................................ 41

2.3 Client factors ......................................................................................................................................... 47

2.4 Therapeutic alliance .............................................................................................................................. 58

2.5 Change process theory.......................................................................................................................... 60

CHAPTER 3: METHODOLOGY ...................................................................................................................... 64


3.2 Critical realism and theoretical assumptions ........................................................................................ 65

3.3 Muted group epistemology .................................................................................................................. 69

3.4 Concept analysis design ........................................................................................................................ 73

3.5 Participant recruitment and data collection ......................................................................................... 80

3.6 Participant data extraction ................................................................................................................... 88

3.7 Participant data interpretation ............................................................................................................. 93

3 of 271

3.8 Research data collection ..................................................................................................................... 104

3.9 Research data extraction .................................................................................................................... 110

3.10 Research data interpretation ............................................................................................................ 110

3.11 Combining participant and research frameworks ............................................................................ 116

3.12 Rigour ................................................................................................................................................ 121

3.13 Ethics ................................................................................................................................................. 123

CHAPTER 4: RESULTS ................................................................................................................................. 127

4.1 Chapter overview ................................................................................................................................ 127

4.2 Helpful client factors ........................................................................................................................... 127

Stage A: Problem recognition ............................................................................................................... 129

Stage B: From other-focussed to self-focussed commitment to change ............................................. 130

Stage C: New social connections fuel change ....................................................................................... 136

Stage D: Reflective and exploratory learning ....................................................................................... 139

Summary of key findings: Helpful client factors ................................................................................... 146

4.3 Unhelpful client factors ....................................................................................................................... 147

Stage A Non-change: Problem-maintaining social environment .......................................................... 149

Stage B Non-change: No commitment to change ................................................................................ 150

Stage C Non-change: No new connections ........................................................................................... 153

Stage D Non-change: Resource-depletion cycle ................................................................................... 154

Summary of key findings: Unhelpful client factors ............................................................................... 158

CHAPTER 5: DISCUSSION ........................................................................................................................... 160

5.1 Chapter overview ................................................................................................................................ 160

5.2 Study question 1: Factors salient to participants ............................................................................... 161

5.3 Study question 2: Client factors in psychotherapy research .............................................................. 161

5.4 Study question 3: Key components of the reconstructed concept .................................................... 162

5.5 Study question 4: Concept utility for counselling psychology theory ................................................ 165

5.6 Domain ................................................................................................................................................ 178

5.7 Fecundity ............................................................................................................................................. 178

5.8 Differentiation ..................................................................................................................................... 179

5.9 Resonance ........................................................................................................................................... 179

5.10 Consistency ....................................................................................................................................... 184

5.11 Operationalisation ............................................................................................................................ 185

5.12 Recommendation for further study .................................................................................................. 185

4 of 271

5.13 Study limitations ............................................................................................................................... 187

CHAPTER 6: CONCLUSION ......................................................................................................................... 191

6.1 Contributions to knowledge ............................................................................................................... 191

6.2 Closing remarks: Results in relation to aims ....................................................................................... 194

REFERENCES .............................................................................................................................................. 196

APPENDICES .............................................................................................................................................. 226

Appendix 1: Database search screenshots ............................................................................................... 226

Appendix 2: Primary study inclusion criteria ............................................................................................ 228

Appendix 3: 49 studies extracted from 14 reviews/analyses ................................................................... 230

Appendix 4: Research data extraction spreadsheet ................................................................................. 231

Appendix 5: Quality control exclusion criteria .......................................................................................... 233

Appendix 6: Tests, variants and causal syntax chart for moderators ....................................................... 234

Appendix 7: Tests, variants and causal syntax chart for mediators ......................................................... 237

Appendix 8: Tests and causal syntax chart for predictors ........................................................................ 241

Appendix 9: Participant recruitment poster ............................................................................................. 243

Appendix 10: Participant information sheet ............................................................................................. 244

Appendix 11: Interview schedule.............................................................................................................. 246

Appendix 12: Participant consent form .................................................................................................... 247

Appendix 13: Transcript portion ............................................................................................................... 248

Appendix 14: Transcription considerations .............................................................................................. 250

Appendix 15: Interview summaries .......................................................................................................... 253

Appendix 16: Entire interview data spreadsheet ..................................................................................... 260

Appendix 17: Example ephemeral reflective notes page ......................................................................... 260

Appendix 18: Example of peer debrief meeting minutes ......................................................................... 261

Appendix 19: MIE Approval emails ........................................................................................................... 262

Appendix 20: Adverse event flow chart .................................................................................................... 263

Appendix 21: Client factors in Castonguay and Beutler (2006) ................................................................ 264

Appendix 22: Demographic information sheet......................................................................................... 265

Appendix 23: Definitions of causal relationships in research ................................................................... 266

Appendix 24: Included studies with social factors findings ...................................................................... 266

Appendix 25: Helpful elements density table ........................................................................................... 268

Appendix 26: Unhelpful elements density table....................................................................................... 270

5 of 271

Word count: 52,183

List of tables

Table 1: Participant demographic characteristics ....................................................................................... 83

List of figures

Figure 1: Thinking, behaving, experiencing, reconceptualising cycle ......................................................... 13

Figure 2: Referents, concepts and meanings .............................................................................................. 20

Figure 3: Role of concept analysis ............................................................................................................... 21

Figure 4: Causal influence flow diagram ..................................................................................................... 31

Figure 5: Incorrect causal assumption ........................................................................................................ 32

Figure 6: Simple causal assumption ............................................................................................................ 33

Figure 7: Smoking cessation flow diagram ................................................................................................. 34

Figure 8: Wampold's pie ............................................................................................................................. 46

Figure 9: Concept analysis method steps flow diagram ............................................................................. 80

Figure 10: Example notes of preliminary interview summary .................................................................... 89

Figure 11: Participant data extraction spreadsheet sample ....................................................................... 91

Figure 12: Hand notations for spreadsheet data cleansing ........................................................................ 92

Figure 13: Initial sketching for participant-informed concept hierarchy .................................................... 96

Figure 14: Intermediate sketch of participant informed hierarchy ............................................................ 97

Figure 15: Participant-informed concept framework ................................................................................. 98

Figure 16: Helpful super-factors and stages of change ............................................................................ 100

Figure 17: Unhelpful super-factors and stages of non-change ................................................................. 101

Figure 18: Categories of activity ............................................................................................................... 102

Figure 19: Categories of inactivity ............................................................................................................ 102

Figure 20: Elements of helpful activity categories 7 to 9 .......................................................................... 103

Figure 21: Unhelpful elements connoting non-change categories 7 to 9 ................................................ 104

Figure 22: Study selection flow chart........................................................................................................ 105

Figure 23: Manual interpretation of study data tags ................................................................................ 111

Figure 24: Labelled research data tags ..................................................................................................... 112

Figure 25: Portion of data analysis spreadsheet ....................................................................................... 113

Figure 26: Research-informed concept framework .................................................................................. 115

Figure 27: Reconstructed concept framework ......................................................................................... 119

Figure 28: Example of color-coding on reconstructed concept framework ............................................. 120

Figure 29: Helpful factors in the reconstructed concept framework ....................................................... 128

Figure 30: Unhelpful factors in the concept reconstruction framework .................................................. 148

Figure 31: Therapist relational skill causal influence ................................................................................ 167

Figure 32: Reconstructed helpful client factors causal influence ............................................................. 168

Figure 33: Reconstructed unhelpful client factors causal influence ......................................................... 169

file:///C:/Users/Admin/Dropbox/My%20Documents/caroline/D.Couns.Psych%20CURRENT/Year%203%20Assignments/Dissertation/All%20chapters%20in%201/Submitted/Caroline%20Vermes%20Client%20Factors%20Concept%20Analysis%20UoManchester%202018.docx%23_Toc536862193

file:///C:/Users/Admin/Dropbox/My%20Documents/caroline/D.Couns.Psych%20CURRENT/Year%203%20Assignments/Dissertation/All%20chapters%20in%201/Submitted/Caroline%20Vermes%20Client%20Factors%20Concept%20Analysis%20UoManchester%202018.docx%23_Toc536862207

6 of 271

Abstract

Background and objective

Common factors theory proposes that clients make significant contributions to psychosocial change

processes and outcomes in psychological therapy. However, the client factors concept contains such a

wide range of client activities, characteristics and extratherapeutic circumstances that without some form

of structural logic, it is difficult to understand how clients contribute to change processes, or how

counselling psychology theory might be shaped to work with the most potent client factors. Using first-

hand participant experience and published peer-reviewed psychotherapy outcomes research, this study

sets out to identify client factors associated with psychosocial change and non-change, and to organise

them into a conceptual framework.

Method

A deductive qualitative ‘hierarchy of abstraction’ concept analysis method is used to interpret and

organise data from nine participant interviews and 68 research studies. Graphic frameworks derived from

each data source are compared, then combined into a single reconstructed concept framework.

Findings

The resulting conceptual hierarchy model presents client factors associated with change and non-change

over four levels of abstraction. The detail of experience is located at the least abstract elemental and

categorical levels of change-related activity and inactivity; then abstracted to stages of change and non-

change; then further abstracted to two broad client factors. In this provisional model, the first of these

factors is client use of available social support resources, which inspires and reinforces the second factor,

which is reflective and experiential learning. Inadequate social support is implicated in the learning and

experiential stagnation associated with non-change. The adequacy of the reconstructed concept is

assessed against seven criteria; and by comparing its fit with two social learning theories.

Conclusion and implications

In view of the importance of reflective and experiential learning in this conceptual model, social learning

theory is proposed as relevant for informing future developments in counselling psychology theory and

research. In this model, the term ‘social learning factors’ more accurately points to the potentially major

constellation of wider client and extratherapeutic social contributions to therapy outcomes currently

indicated by ‘client factors’ in common factors theory.

Key words: client factors; concept analysis; concept adequacy; psychotherapy process; psychotherapy

outcome; common factors theory.

7 of 271

Declaration No portion of the work referred to in this thesis has been submitted in support of an

application for another degree or qualification of this or any other institute of learning.

Copyright (i) The author of this thesis (including any appendices and/or schedules to this

thesis) owns certain copyright or related rights in it (the “Copyright”) and she

has given the University of Manchester certain rights to use such Copyright,

including for administrative purposes.

(ii) Copies of this thesis, either in full or in extracts and whether in hard or

electronic copy, may be made only in accordance with the Copyright, Designs

and Patents Act 1988 (as amended) and regulations issued under it or, where

appropriate, in accordance with licensing agreements which the University

has from time to time. This page must form part of any such copies made.

(iii) The ownership of certain Copyright, patents, designs, trademarks and other

intellectual property (the “Intellectual Property”) and any reproductions of

copyright works in the thesis, for example, graphs and tables

(“Reproductions”), which may be described in this thesis, may not be owned

by the author and may be owned by third parties. Such Intellectual Property

and Reproductions cannot and must not be made available for use without

the prior written permission of the owner(s) of the relevant Intellectual

Property and/or Reproductions.

(iv) Further information on the conditions under which disclosure, publication and

commercialisation of this thesis, the Copyright and any Intellectual Property

and/or Reproductions described in it may take place is available in the

University IP Policy (see

http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=24420), in any

relevant thesis restriction declarations deposited in the University Library, the

University Library’s regulations (see

http://www.library.manchester.ac.uk/about/regulations/) and in the University’s

policy on Presentation of Theses.

http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=24420

http://www.library.manchester.ac.uk/about/regulations/

8 of 271

Acknowledgements Thank you to the nine participants who graciously contributed time and experience to

this research.

Thank you to my husband Mark for your unfailing encouragement. Through the happy

times and the difficult times, you are my security. I could not have completed this work

without you by my side. It is a privilege to share this life with you.

Thank you to my sons Ben and James for bringing me home time and again with stories

of intrepid learning journeys, which cannot be completed without our daemons: The

White Mare of the Light, Pantalaimon, Shadowmere, and Aslan to name a few.

Thank you to my colleagues at Oakwood Psychology Services for your tireless, kind

and enthusiastic support. In particular, Joanne Blezard, Julie Keir, Fiona Whelan, Sue

Haywood, Rachel Burgon, Molly Keir, Jane Oldfield, Ros Rheinberg, Dan Owens-

Cooper, Ellie Fretwell, Marie Mellor, Susan Murnane and David Fenton.

Thank you to members of the Chorlton Home Group for praying with me for endurance

to complete this project, which has rightly taken its place behind family and work

priorities.

Thank you to Dr Terry Hanley and Dr Laura Winter for supervising this project with

patience and wisdom.

Thank you to fellow students Didier Danillon and Nicky Toor. Our thesis meetings have

been reassuring and motivating. Special thanks to Lina Efthyvoulou for your solidarity

particularly through the toughest parts of the past four years.

Thank you to the composers and musicians whose performances have been uplifting

companions on this long learning journey. In particular, Brian Eno, David Sylvian,

Robert Fripp, Steve Reich, Bill Laswell, Jim White, Joe Frank, Laraaji, Nils Frahm,

Ólafur Arnalds, Klaus Schulze, Poppy Ackroyd, Sophie Hutchings, Kai Engel, Arvo Pärt,

Max Corbacho, Daniel Segerstad and Johannes Hedberg (Carbon Based Lifeforms),

Hidetoshi Kolzumi and Andre Scheffer (Connect.Ohm), Joe Acheson (Hidden

Orchestra); Markus Sieber (Aukai), Kieran Hebden (Four Tet), Simon Green (Bonobo)

9 of 271

and Skip McDonald (Little Axe). Super-thanks to John Schaefer (newsounds.org) for

pinning back my ears for the past thirty five years with your brotherly erudition. It is a

privilege to share a lifetime with you all.

And within all, thank you to God of love, the beating heart of generous creativity.

Dedication This research is dedicated to the memory of a participant.

The Author I earned my Master’s degree in Counseling Psychology (M.Ed) from Rutgers University,

New Jersey, USA in 1994. I earned another Master’s degree in Counselling Studies

(MA) from University of Manchester, UK in 2010. I have worked in full-time practice in a

range of mental health settings since 1994. Since 2006 I have worked in social sector

counselling service development and leadership. My academic research to date

includes this thesis; and my 2010 Master’s thesis on the cost-effectiveness of a guided

self-help intervention. I have published articles, guidance and textbook chapters in the

area of counselling psychology theory and practice including social entrepreneurship.

10 of 271

Flower in the crannied wall,

I pluck you out of the crannies,

I hold you here, root and all, in my hand,

Little flower – if I could understand

What you are, root and all, and all in all,

I should know what God and man is.

Alfred Tennyson, 1863

CHAPTER 1: INTRODUCTION

1.1 Chapter overview This chapter aims to orient the reader to the study. It is set out over five sections. In the

first I explain the problem this study addresses, which is the disorganised state of the

client factors concept within common factors theory. I justify the need for an evidence-

based conceptual framework that accommodates a wide range of known client factor

dimensions, states and behaviours over various levels of conceptual abstraction; that

accounts for the temporality of change processes; and that includes unhelpful as well as

helpful factors.

In the second section, to reduce possible doubt for the reader about why I chose a

concept analysis methodology for this study, I explain why I treat the term ‘client factors’

as a concept. I then explain why empirical concept analysis is a better fit than more

‘mainstream’ qualitative methodologies including thematic analysis, interpretive

phenomenological analysis and grounded theory. I also explain the tradition and context

for the reconstructive ‘ladder of abstraction’ concept analysis used here.

In the third section I detail the study’s two-fold purpose, which is to reconstruct the client

factors concept and then to consider the utility of the reconstructed concept to

counselling psychology theory. This section also includes the four research questions

that guide me in achieving these aims.

In the fourth section I define and discuss key terms including concepts; psychotherapy;

causal syntax; and outcome.

11 of 271

The chapter concludes in the fifth section with an overview of the study’s chapter layout

and content.

1.2 Study problem Sometimes, when reading textbooks and research articles that are relevant to my work,

I encounter an abstract technical term whose precise meaning, structure or function is

not fully explained. When I am particularly committed to the topic, I am troubled by an

under-defined term. It is likely to be derived from a detailed set of conceptually

interrelated phenomena, but I cannot extrapolate its significance if I don’t understand

what the concept encompasses, how its components hang together, or where its

mechanisms of action lie. For me, ‘client and extratherapeutic factors’ is such a term. I

became interested in it as an important but incoherent concept when, around the turn of

the century, I enthusiastically read The Heart and Soul of Change, Hubble, Duncan and

Miller’s (1999a; updated by Duncan, Miller, Wampold and Hubble in 2010) textbook

about the common factors that enable psychotherapy to achieve its results. Two key

messages in this book deeply influenced my subsequent thinking and work as a

psychotherapist. The first, from Tallman and Bohart (1999) is that the client is more

responsible than the therapist or therapeutic approach for making therapy work. The

second, from Hubble and Duncan and Miller’s (1999b) concluding chapter, is that the

practice of therapy should not be about classifying disease but about facilitating change.

Here they use the term ‘client and extratherapeutic factors’ to encompass client

characteristics that impact outcome; aspects of client activity within and beyond the

therapy relationship; and social and emotional resources that clients may use to

facilitate change. Overall, 18 disparate client and extratherapeutic factors can be

summarised from chapters contributed to this textbook by Asay and Lambert (1999),

Tallman and Bohart (1999) and Hubble, Duncan and Miller (1999b):

1. Duration and severity of client’s problem

2. Pre-treatment change

3. Stage of change

4. Motivation

5. Hopes and expectations

12 of 271

6. Helpful behaviours outside therapy

7. Nature, strength and quality of social support

8. Client’s theory of change

9. Self-expression and self-disclosure

10. Using self-help media

11. Participation and openness

12. Clarity of goals

13. Agency in pursuing goals

14. Active learning

15. Creative conceptualisation and re-conceptualisation

16. Reflexive self-understanding; gaining new perspective and realisations

17. Exposure and extinction

18. Reality testing, mastery and corrective experiences

In the textbook’s second edition (Duncan, Miller, Wampold and Hubble, 2010) Bohart

and Tallman (2010) add a further seven client factors gathered from literature review:

19. Suitability for therapy

20. Cooperation and interactive collaboration

21. Contribution to the therapeutic bond

22. Perception of therapy

23. Resilience and post-traumatic growth

24. Early change

25. Higher level of distress

Bohart and Tallman (2010) describe this known constellation of client and

extratherapeutic factors as a “potpourri of findings from unrelated research areas”

(p.94). A less charitable metaphor than a bowl of dried flowers and cinnamon sticks

could be applied. For me, the primary problem with this list is its disorganised

randomness. It includes behavioural and cognitive functions, active and passive states,

intra-therapeutic and extratherapeutic elements; intimations of different theories such as

‘stage of change;’ various degrees of abstraction from lived experience; temporally

determined aspects such as ‘pre-therapy change’ and ‘early change;’ and difficult-to-

13 of 271

measure categories such as ‘suitability for therapy’ (see for example Laaksonen,

Lindfors, Knekt and Aalberg, 2010 for discussion of a ‘suitability for therapy’ scale that

predicts outcome on the basis of participant personality function). This potpourri does

not tell me how change comes about, nor why some clients find change difficult to

achieve. In other words, it does not, by itself, offer a cohesive theory of ‘the active

client’, despite its appearance in a textbook dedicated to illuminating conceptual

mechanisms of client change. It requires some sort of analytic technique to render

theoretical coherence to its components.

To this end Tallman and Bohart (1999) briefly hypothesise (after Kolb, 1984) that client

change happens in cycles of thinking, behaviour and experience which lead to shifts in

problem and solution conceptualisation, which in turn lead to further thinking. To assist

reader comprehension I have adapted this change model into the graphic shown in

figure 1.

Figure 1: Thinking, behaving, experiencing, reconceptualising cycle

14 of 271

I find this conceptual model of client change inadequate for five reasons. Firstly, it is

over-simplified and does not show where the range of 25 client factors summarised

above might fit in, or how they make a difference. For example, the influence of social

support is missing. Secondly, it cannot distinguish, accommodate or organise the

various levels of abstraction inherent to the listed client factors. For example, ‘active

learning’ is a more abstract term (i.e. a general category of activity) than ‘exposure’ (i.e.

a behavioural technique associated with active learning). Thirdly, it does not suggest

where this cycle sits in wider temporal planes of change. For example, the client factor

‘stage of change’ implies that different tasks, frames of mind, states of awareness, or

activities are present at different times in a change process. Does this cycle happen

early or late in a change process? Or does it represent the entire change process?

Fourthly, the model does not account for why people struggle, or fail, to make desired

changes, which might be even more important for therapists to understand than how

people successfully change. Lastly, Tallman and Bohart state that there is “no specific

evidence” (p.113) to support their use of the thinking-behaving-experiencing cyclical

model in this way. Further work is required to render theoretical coherence to the

concept of client factors and its component parts.

Client factors as a concept

Before I explain why I chose concept analysis to help me construct a client factors

theoretical framework, I first explain why I believe the term is a concept.

In this study concepts are defined as abstract linguistic entities (for example, nouns)

that mediate between collections of interrelated embodied experiences, and the

meaning or interpretation that is made of them in mental representations such as

thoughts and ideas (Rundle, 1995). Concepts lend themselves to analysis and structural

mapping to assist with theorising and knowledge synthesis (Fauconnier and Turner,

1998; Knafl and Deatrick, 2000). As this study will show, it is possible to create a

structurally logical framework to map the meaning of the term ‘client factors’ through a

process of organising findings from quantitative research, thus determining it is a

concept (Laurence and Margolis, 1999).

15 of 271

The structural prototype that best describes the client factors concept is Laurence and

Margolis’ (1999) ‘containment model:’ an abstract concept that is a composite of

multiple other interrelated, less abstract concepts, as is the case with ‘client factors.’

Concepts as legitimate subjects for research enquiry

Much psychotherapy research over the past fifty years has been devoted to

understanding which therapeutic approaches and therapist attributes produce the best

outcomes. But this vast literature, according to Kazdin (2009), offers little explanation

about why psychotherapy works, or how client change comes about. Kazdin suggests

that “fresh approaches are needed in conceptualisation” of key change mechanisms

(p.419). Similarly, Beutler (1991) asserted that while a wide range of client variables are

theoretically important predictive constructs, they lack consensual or consistent

meaning, are poorly defined, and have weak conceptual linkages to psychotherapy

theories. These views indicate that client factors are worthy of improved

conceptualisation.

My starting point with this study is that I believe Bohart and colleagues are right: as

conceptual entities, client and extratherapeutic factors make more difference in therapy

outcomes than therapist or technique factors. By extension I also believe that the client

factors concept must point to a complex theoretical architecture of process mechanisms

in which most if not all of the factors identified above (and potentially others not listed)

must fit in a logically interrelated manner. I require an analytic method that enables me

to redress the five limitations with the thinking-behaving-experiencing-reconceptualising

model shown in figure 1. So, it must permit me to (a) consider the boundaries of the

concept and the associated rules by which member objects are included and excluded

(even if the rules are provisional and experimental); (b) distinguish and accommodate

various levels of conceptual abstraction amongst the member objects; (c) create a

graphic illustration of influential relationships between member objects in a way that

respects and highlights the temporal nature of human psychosocial change processes;

(d) include non-change factors as well as change factors and, if possible, illuminate

relationships between these; and (e) ensure that the framework is evidence-based.

16 of 271

1.3 Why concept analysis Rennie (2012) helpfully distinguishes three types of qualitative research and the

varieties of methods that suit each type. The first type, ‘experiential’ research,

conceptualises reported experience into structures or categories using methods such as

narrative analysis (McLeod and Balamoutsou, 2006), grounded theory (Strauss and

Corbin, 1998) and interpretive phenomenological analysis (Smith, Flowers and Larkin,

2009). The second type, ‘discursive’ research, is applied to the study of language use,

with methods such as conversation analysis (Sacks, 1972) and discourse analysis

(Parker, 1998). The third type, ‘experiential/discursive’ blends these two, and according

to Rennie is characterised by two methods, thematic analysis (Braun and Clarke, 2013)

and case studies (Stiles, 2003) because each has been applied to both the experiential

and discursive types of research.

This study lies in the experiential type, as its purpose is to conceptualise experience

reported by participants and published literature into a theoretically-informed structure.

In this study, if I had wanted simply to qualitatively describe, list or thematically organise

influential client factors that might be located in research literature or participant

experience I might have chosen thematic analysis, interpretive phenomenological

analysis (IPA) or grounded theory. Any of these methods would facilitate an evidence-

based analysis, but I will here explain why each approach is not the right fit for the aims

of this study, which is to create a tentative theoretically informed framework for the client

factors term.

Why not thematic analysis

Thematic analysis is unsuited to this study’s aims for several reasons. Rennie (2012)

notes that it is best suited to discursive and experiential/discursive - and by

extrapolation, not experiential - types of qualitative research. Although Rennie does not

explicitly explain why he makes this assertion, I wonder if it might be because thematic

analysis is not designed to create, develop or improve theory (University of Auckland,

no date). It tends to suit inductive research (Braun and Clarke, 2013) that creates

concepts from various forms of discourse analysis (Rennie, 2012). By contrast, my

theory-engaged study requires a deductive approach that starts with an existing

17 of 271

theoretical concept and critically examines the particulars of qualitative and quantitative

evidence for the theory. Furthermore, thematic analysis lacks strategies for

distinguishing different levels of conceptual abstraction or temporal zones, and thus

forces the researcher to work within a single plane of interpretation (University of

Auckland, no date). In short, while thematic analysis is useful for locating and analysing

themes and patterns in qualitative material, it is not designed to deal with abstract,

theory-bound concepts.

Why not IPA

Considering IPA next, Rennie (2012) suggests it is useful for experiential qualitative

research that creates structured conceptualisations. IPA permits two levels of thematic

abstraction - emergent and superordinate. But these functions cannot adequately

accommodate the multi-layered abstraction inherent in the client factors concept.

Moreover, IPA tends to focus on the individual participant (Larkin, Watts and Clifton,

2006) which is not suited to the aim here. On a personal note I would not choose IPA

because I do not agree with some terminology used by its proponents. For example I do

not believe that themes ‘emerge’ as foregone ontological entities waiting to be

discovered (see for example Smith and Osborn, 2007) but are epistemologically created

by the researcher. Additionally, I do not like the idea of the IPA researcher claiming the

power to give ‘voice’ to participants (see for example Larkin, Watts, and Clifton, 2006)

especially when participant contributions are sometimes heavily interpreted through the

IPA researcher’s frame of reference.

Why not grounded theory

Grounded theory at first seems an appropriate qualitative methodological choice. It

comprises a taxonomical, hierarchical analytical process, which identifies and organises

associated categories and properties of data under conceptual labels. It deals well with

analyses of social processes which can be translated to graphic models (Morrow and

Smith, 1995). It produces a discursive set of theoretical propositions that account for

causes and consequences related to the examined phenomena (Strauss and Corbin,

1990). It has been adapted for use in research which (as I explain further in Chapter 3,

is the case with this study) is rooted in critical realist ontology (e.g. Hoddy, 2018). It

18 of 271

includes criteria for judging analytic adequacy including fit, relevance, workability and

modifiability (Glaser, 1998). However, upon further consideration grounded theory also

proves inappropriate. Where grounded theory starts with raw data from which it seeks to

inductively draw up abstract concepts, in this case I am starting with an abstract

concept and require a methodology that deductively drills down, examines and maps its

structure and generative mechanisms using selectively chosen data. It is not possible

for me to ignore literature or theory in the topic area (as suggested by Glaser and

Strauss, 2000). Indeed, the starting point for the study is my knowledge of, and

reference to, an influential body of existing theory and research on client contributions to

therapy process and outcomes.

Why concept analysis

My research position is illustrated by the Tennyson poem that opens this chapter. The

mystery of the flower growing in the crack in the rock is like the concept of client factors

in the puzzle of what makes therapy work. If I can pull up and examine the root structure

of this concept I might achieve a better understanding of what clients do to bring about

psychosocial change, as a fundamental aspect of human adaptation and survival.

And so I arrive at concept analysis as the favoured qualitative methodology. Although in

Chapter 3 I describe and distinguish between different types of concept analysis and

related methodologies, and explain the ‘reconstructive hierarchy of abstraction’ method

(Sartori, 1984) chosen for this study, I here briefly orient the reader to concept analysis

methodology. For the avoidance of confusion, it is important that I first distinguish

‘concept analysis’ as an empirical research methodology (as used here) from ‘concept

analysis’ as a process of defining and clarifying key terms at the outset of a study (as

used, for example by Aveyard, Payne and Preston, 2016). Empirical concept analysis

research methodologies are traditionally employed in nursing scholarship (e.g. Rodgers

and Knafl, 2000); sociology (e.g. Fredericks and Miller, 1987); political science (e.g.

Collier and Gerring, 2009); cognitive science (e.g. Laurence and Margolis, 1999); and

gender studies (e.g. Goertz and Mazur, 2007). Each of these disciplines tends to prefer

its own philosophical and ontological schools of thought on the nature of concepts and

the purpose of concept analysis. This study employs a form of concept analysis derived

19 of 271

from the political science work of Sartori (1984) and Gerring (1997; 2012).

‘Reconstructive’ concept analysis aims to map relevant data units to a theoretically

informed hierarchically abstract definitional framework, and also aims to assess the

pragmatic adequacy of the result. Reconstructive concept analysis offers a critical

approach to understanding conceptual terms that have become arbitrarily fixed to

empirical findings, which I will argue later, may be the case with the term ‘client factors.’

As I discuss further in Chapter 3, counselling psychology has yet to engage with

concept analysis methods, even though our work is concerned with innumerable

relatively abstract concepts such as ‘emotional distress,’ ‘resilience,’ ‘collaboration,’

‘facilitation’ ‘goals,’ ‘family’ or ‘multiculturalism.’ Such concepts are ‘tools of the trade,’

but the scientist-practitioner’s ability to teach, practice, research and theorise effectively

using these tools depends on a precise, deep and clear understanding of their meaning

and referents, as well as the ability to measure them or assess their impact in people’s

lives. Achieving such clarity, in turn, depends on research methods that enable analysis

of conceptual structure and linkage to theory.

Empirical concept analysis

One starting point for understanding the role of concepts in language is to distinguish

‘referents,’ ‘concepts,’ and ‘meanings’ (Jackson, 1984). Referents are contextual, social

or personal characteristics, dispositions, processes, events and other phenomena that

are most directly visible in lived experience. Concepts are verbal denotations (linguistic

representations or symbols) for these phenomena. They are created though social

convention. Meanings are the individual and collective mental constructions predicated

by concepts. They are developed socially and are vulnerable to misunderstanding or

distortion if not adequately researched, taught or communicated. Figure 2 provides an

illustration of the relationship between referents, concepts and meanings.

20 of 271

Figure 2: Referents, concepts and meanings

In this diagram we can see that concepts are tools that bridge referents and their

meanings. By extension, concept analysis is a methodology that maintains conceptual

tools in an “orderly condition” (Jackson, 1984, p.206) by epistemologically mediating the

relationship that referents have with their meanings, and by enhancing contextual clarity

and definitional specificity, particularly as a preparatory step where a concept is

required to form part of a scientific investigation. Some types of concept analysis also

strive for ontological depth by pointing to possible mechanisms of action or the nature of

a process in order to advance existing theory (Risjord, 2009), which is the case in this

study. Figure 3 provides a basic illustration of the mediating role of concept analysis

research between concepts and theory development.

21 of 271

Figure 3: Role of concept analysis

1.4 Study purpose and questions Bohart has noted that “the role of the client as an active change agent who makes

significant contributions to the change process has been neglected in the

psychotherapy literature” (2004, p.119). He recommends, “There are now many reviews

of specific aspects of client functioning but none give a comprehensive picture of the

client in therapy….It seems more useful to investigate what these characteristics

mean....” (Bohart and Wade, 2013, p.221, 246).

This study takes up Bohart’s research recommendation. It has two aims. The first, as

previously noted, is to sort out the jumbled meaning of the ‘client and extratherapeutic

factors’ concept by creating a provisional theoretical model of how people participate (or

not) in processes of psychosocial change. The second is to consider the utility of the

concept for counselling psychology theory. I explain these aims further. then detail the

four research questions used to address them.

22 of 271

Aim 1: Analysis and reconstruction of the client factors concept

Clients operate in patchily mapped conceptual space in terms of their contribution to

psychotherapeutic process and outcome. One possible reason for this is that client

factors discussed in empirical or theoretical literature can sometimes be chosen or

described from a ‘therapy-centric’ position (Bickman and Salzer, 1996; Bohart, 2004).

Therapy-centrism refers to an assumption that the therapist (or therapeutic

approach/model, or therapeutic relationship), is the primary active agent in

psychotherapeutic work. Authors adopting a therapy-centric perspective tend to

acknowledge client influences on psychotherapeutic outcome only insofar as they

provide a raison d’etre for the therapist (e.g. functional deficits). Therapy-centrism tends

to position the client in a more passive or impaired role compared to that of the

therapist. Therapy-centrism also refers to professional blindness to these assumptions.

The first aim is therefore to bring a coherent structure to the client factors concept while

attempting to avoid compounding therapy-centrism. To achieve this, it is important to

be careful in choosing the epistemological frame and data sources for the analysis

(which I discuss in Chapter 3). But I acknowledge at the outset that as a therapist

myself I cannot eradicate my personal aim to serve therapy by trying to improve my

knowledge certain theoretical directions in psychotherapy literature.

Concept analyses traditionally examine the context, use, variants and/or definitions of

the concept in question, in data samples drawn from relevant disciplinary research (for

example Olsson, Bond, Burns, Wella-Brodrick and Sawyer’s 2003 study on adolescent

resilience; Burrows’ 1997 concept analysis of facilitation; and Ridner’s 2004 analysis of

psychological distress). In this case, because ‘client factors’ is a multidimensional

containment concept, the aim is to create an organising structure that adds definitional

specificity while respecting its contextual dimensionality.

Once the concept has been analysed, clarified and in some cases tentatively re-defined

through literature review, some researchers use case studies to demonstrate how

concept components work (for example Knafl and Deatrick’s 1990 concept analysis of

family styles for managing life with a chronically unwell child; and Brush, Kirk, Gultekin

and Baiardi’s 2011 concept analysis of overcoming). However the addition of case study

23 of 271

material does not use participant experience to define and structure the concept, only to

illustrate it once it has been derived from the literature.

Penrod (2007) observes that concept analyses derived solely from professional

literature can lack pragmatic utility because the results do not necessarily fit with

people’s experiences. I agree with Penrod’s position, in that some psychotherapy

research focuses too narrowly on professional activity to the exclusion of client activity,

and thus misses valuable information about the mechanics of change. I make an

assumption at the outset of this study that research based solely on diagnostic or

psychometric data, or on therapy-centric questions about therapy’s effectiveness, will

yield some useful information about client characteristics and activities that influence

processes of psychotherapeutic change and outcome. But because this body of

research tends not to explore the extratherapeutic resources and activities clients use to

make desired changes while in therapy, it does not provide a well-rounded picture of the

active client.

In this study I follow Penrod’s (2007) proposal to ‘advance’ literature-based concept

analysis by contrasting and combining the categorical framework derived from

published research data with a similarly-structured framework derived from participant

interview data. In my study, the participant framework is ‘dominant’ (in line with Morse

and Penrod, 1999) – in other words it forms the primary analysis against which the

literature conceptualisation is compared and contrasted (this will be explained further in

Chapter 3). During the process of comparison, the conceptual frameworks are held over

each other to determine which components are shared and which are unique to their

data source. To enhance the reliability of the results (Stavros and Westburg, 2009), the

two are then integrated in a form of data triangulation to advance a single concept

(Hupcey and Penrod, 2003; selective mapping of conceptual properties from different

source domains into a blended space is described further in Fauconnier and Turner,

1998). In the present study, this methodological triangulation approach is used, but

consistent with terminology adopted in Sartori’s (1984/2009) hierarchical concept

analysis methodology (explained further in Chapter 3), the process is termed ‘concept

reconstruction,’ This term will be used for the remainder of the study.

24 of 271

Aim 2: Consider the utility of a reconstructed client factors concept in

counselling psychology theory

An organising framework for understanding, structuring and synthesising the operational

contents of the client factors concept could offer a range of opportunities for the

advancement of counselling psychology theory. Counselling theories are intended to

guide the counsellor’s role, attitude, skills and techniques in the therapeutic encounter,

based on organising principles that represent interpretations of client problem

presentations (Drapela, 1990). Counselling psychology is aligned with humanistic and

integrative counselling theories with various problem-oriented principles. Without

wishing to over-simplify the differences between these theoretical approaches, nor to

trivialise their philosophical rigour, nor their helpful contribution to people’s lives, I

suggest that psychodynamic variants are devised around redressing problematic

attachment history; cognitive-behavioural variants around adjusting maladapted reactive

thoughts and actions; person-centred variants around releasing thwarted emotions;

narrative approaches around externalising and re-authoring the problem and so forth.

To my mind the organising principle of a counselling theory devised around a coherent

and comprehensive client factors concept would focus less on interpreting problems

and more on harnessing psychosocial processes and resources associated with

generative change. It might also inform the counsellor how to help counteract factors

associated with therapeutic failure.

Research questions

Four questions guide me in reaching these aims:

1. What do participants describe as salient factors in their efforts to make desired

psychosocial life-changes, and how are these best conceptualised? This data

informs the primary concept framework.

2. How are client factors represented in psychotherapy research? This data informs

the comparator framework.

3. What are the key helpful and unhelpful components of a reconstructed client

factors framework?

4. What is the utility of the reconstructed concept to counselling psychology theory?

25 of 271

The study’s first aim is addressed and answered by the first three study questions, and

in particular the third question which represents the combination of information from the

first two questions. Its second aim is addressed by the fourth question. The answers to

these questions are discussed in Chapter 5.

1.5 Key terms In this section I define and discuss five relevant terminological areas: Concepts;

psychotherapy; causal syntax in psychotherapy; and outcome.

Concepts Here I describe the role of concepts in mental health theory; the relationship between

concepts and theory; and some difficulties inherent in their abstract nature.

Concepts in mental health theory

Concepts facilitate communication within a knowledge community by linking a class of

referents to their meaning by one (or in the case of ‘client factors’, two) words. Concepts

in mental health care include, but are not limited to, practices, aims, traits, actions,

states of being, and social positions related to the roles of caregiver or care recipient. A

concept is best described in a brief statement of necessary and sufficient conditions.

However, the meanings of concepts in mental health care are rarely so definitionally or

ideologically ‘pure.’ Definitionally they overlap with related terms arising in practical use

in specific contexts over time. Ideologically they represent the language of professional

interests and then enter research analysis and theory as facts of life (Pilgrim, 2016,

2017).

Building blocks of knowledge? Or theoretical niches?

Concepts have been described by some authors as building blocks of theory in the

social sciences because of their inherent explanatory powers (e.g. Chinn and Kramer,

1994; Pintrich, Marx and Boyle, 1993). From this viewpoint, concept research extends

theory by bridging the gap between current theoretical understandings, and intellectual

goals in theory development (Rodgers, 2000b). On the other hand Paley (1996)

reverses the concept-theory equation, stating that theory creates concepts to occupy

26 of 271

meaning ‘niches.’ Paley asserts that “the only way to clarify a concept is to explicitly

adopt a theory that determines what its niche will be” (p.577).

In this study I take the ‘either/both’ ground by conceiving theory and concept

development as interdependent activities. Concepts are the ‘offspring’ of theories in that

theorists invent niche concepts to explain aspects of observed or imputed cause and

effect. For example, ‘client factors’ is an offspring of common factors theory (this will be

discussed further in Chapter 2). Once ‘born’ concepts exert formative influence on their

theoretical ‘parents.’ For example, if one theorises that client factors exert considerable

influence on outcome one can examine and adjust common factors theory in light of

new knowledge about client factors.

It is helpful to remember here that the purpose of theory, and associated concept

clarification, is the advancement of practice. Careful and thorough work on evidencing,

justifying and clarifying concepts makes them better tools with which to reflect on and

refine the theories and practices with which they are associated, but they must be

associated with a theory, even if not the originating theory, to make this work

meaningful.

The problematic nature of concepts

As useful as concepts are as communication devices and holders of meaning, I briefly

discuss here two difficulties inherent in their use and development in psychological

science: contextuality: how concepts are defined by power-interests; and obscuration of

connotation: how vagueness shrouds meaning.

Contextuality

Mental health concepts are created and shaped by the knowledge-contexts in which

they are used. The substantive content of scientific conceptual models, and what is

accepted as ‘core knowledge’ are influenced by sociological and historical factors

(Pintrich, Marx and Boyle, 1993). For reasons I shall explore further in Chapter 2,

common factors theory provides relatively little information about the client factors

concept, although it is potentially the most potent of all the common factors. Thin

descriptions of key concepts within a theory leave room for the representation of

27 of 271

incorrect information or the incorporation of “a false belief or two into [the] essence

placeholder for a concept” (Laurence and Margolis, 1999, p.48). As I will argue further in

Chapter 2 the belief, strongly held by some psychotherapy professionals, that the

therapeutic relationship directly ‘accounts for’ client therapeutic change (see for

example Patterson, Uhlin and Anderson, 2008), while unsubstantiated by research

evidence (Ardito and Rabellino, 2011), represents an example of therapy-centric

extrapolation of common factors theory.

Obscured connotation

Although the client factors concept appears to introduce client activity and

characteristics into a causal argument about the efficacy of psychotherapy, it also

linguistically obscures the magnitude and diversity of client influence on the process. As

abstractions, concepts lose the detail of their implications, particularly where there is

little consensus on their meaning even within their contexts of origin (concepts

described by Pilgrim and Bentall as “woolly,” 1999, p.271). The client factors concept is

such a case, where even in psychotherapy literature, the term does not

comprehensively or coherently point to its connotations, including sociological matters.

Shrouded concepts are uninformative and limit the theorising that can be done with

them. On the other hand, overly narrow or rigid conceptual definitions exclude

knowledge development from diverse sources.

Psychotherapy In this study the term psychotherapy encompasses the work of qualified and in-training

professionals including counsellors, psychologists, CBT therapists, occupational

therapists, social workers, psychiatrists and others who may not uniformly call their work

‘psychotherapy,’ in assisting individuals, families, couples or groups with a recognised

method of talk-therapy to improve mental health; to change problematic behaviours,

personality difficulties or relationship patterns; and/or to pursue personal growth: “self-

efficacy, competence, and mastery” (Luborsky, Crits-Christoph, Mintz and Auerbach,

1988, p. 165). Furthermore, psychotherapy is viewed a relationally-based joint activity

that depends considerably, if not primarily, on the quality of client involvement for its

effectiveness (Bohart and Tallman, 1999; Duncan, Miller and Sparks, 2004).

28 of 271

‘Client’ as a problematic term

‘Client’ connotes the passive role of a person being dealt with by social or medical

services. It is derived from the Latin ‘cliens’ which means “a person under the patronage

or protection of another;” and in ancient Rome referred to a “plebeian under the

protection of a patrician” (Oxford English Dictionary, 2017, no page number). The term

also, less obviously, connotes co-dependency: psychotherapists depend on clients to

make a living; and clients depend on psychotherapists to help with their emotional and

behavioural problems.

Use of the term ‘client’ lends patronising power to the person who uses it to describe

someone with whom they have a working relationship. Psychotherapists commonly use

the term ‘my client,’ as if the client is blessed by ‘belonging’ to the therapist.

‘Client’ also connotes participation in marketised care services in which the ‘customer’

has a choice of provider when they pay privately. But people who access publicly-

funded mental health services are rarely given much choice over type of therapy or

provider. In looking to alternative terms, ‘service user’ and ‘patient,’ are more commonly

associated with public sector services, but are even less apt descriptors for people who

partner with psychotherapists to achieve desired changes in life. McLaughlin (2009)

criticises the term ‘service user’ for prescribing a limited identity that points to the

person’s use of disability services and obscures their preferred, able statuses - which

may involve admirably resourceful and engaged ways of living. The term ‘patient’ is

derived from medicine and invites the sort of ‘pathologisation’ and ‘treatment’ of

individuals and groups of people with mental health difficulties from which some (but not

all) counselling psychologists seek to distance their practice (Woolfe, 2016).

There is no commonly used term for people who access psychotherapy relationships

that does not imply - and maintain - a power difference between the participants that

privileges the therapist as the agent most responsible for outcome. The absence of an

egalitarian term for therapy participants points to a “conceptual practice” (Harding, 2004,

p.30) which is complied with by therapists and clients alike. However, with regret, I have

chosen to retain the term ‘client’ in this study because ‘client factors’ is the subject of the

concept analysis. In Chapter 6 I address changing the concept title.

29 of 271

Where an author whose work I describe in some detail refers to ‘patients’ (e.g. Orlinsky,

2009) I retain this terminology in those passages.

Causal syntax in psychotherapy Bohart, Tallman, Byock and Mackrill (2011) argue against the use of experimental and

randomised controlled trials to judge whether, how, or which psychotherapy or client

factors ‘cause’ changes in client presenting problems. They contrast psychotherapy to

cancer treatment. Radiation might ‘cause’ cancer cells to die in a direct and linear

fashion. But psychotherapy is a complex interactive process in which the relationship

between inputs and outcome is “fundamentally probabilistic” (p.105). Bandura (1978)

also dispenses with unidirectional causal models of human behaviour, and instead

describes how people’s personal and cultural characteristics and social roles may

determine the interaction between environmental events, and internal cognitive

processes such as reflective thought, selective attention, planning and foresight, beliefs

and preferences, incentives, self-evaluation and self-efficacy. In Bandura’s ‘reciprocal

determinism’ model (1978), individuals influence environments and vice versa. Certainly

if one thinks about events in one’s daily life they are hardly separable from preceding

events, cognitive states, free will, chance, social context and so on. The idea that

something happens, or changes because of one entity’s sole influence on another does

not match phenomenological experience.

However, we cannot completely rid psychotherapy theory of an ontology of causality,

because without it, there would be no way of understanding how psychotherapy helps,

or indeed, why it should exist. In this study I suggest that the best we can do is point to

aspects of ‘causal influence’ and ‘causal probability’ in psychotherapy process and

outcome. To describe these we need causal syntax.

Gerring (2012) provides a clear explanation of causal syntax in social science which I

briefly summarise in the remainder of this section.

Independent and dependent variables

Dependent variables are the measured outcome of therapy (Kazdin, 2009).

Independent variables are causal inputs and generally take the form of client and

30 of 271

therapist cognitive, relational or creative activity. Making any causal argument, even

about probable influence, requires one to be able to say that a change in X

(independent variable, input or predictor) generates a change in Y (dependent variable,

outcome or response) relative to what Y would be without the change in X, given certain

background conditions and the scope of the empirical phenomena of interest. How X

changes Y may be obvious and straightforward; or it can occur over a pathway of

intermediary mechanisms. So, the distance between X and Y can be anything from

proximate (i.e. few if any intermediary factors) or distal (indirect effects with complex

and multi-staged mechanics). Intermediary generative mechanisms in the process of

change are known as mediators; while intervening variables and pre-existing conditions

that change the relationship between X and Y are known as moderators (moderators

and mediators are explained further in Appendices 6 and 7 respectively; and a summary

of definitions for mediators, moderators and predictors is available in Appendix 23).

Predictors of treatment outcome are conceptually similar to moderators but are

associated broadly with outcome across all treatment conditions. For example, the

client’s problem complexity or chronicity will have a direct bearing on outcome

regardless of the type of therapy provided (Beutler, Clarkin and Bongar, 2000; Beutler,

2016). Predictors are naturally occurring variables that can be measured and

statistically calculated in experimental research. They may include aspects of therapy,

client factors originating in or before therapy, or extratherapeutic factors.

Causal factors, as independent variables (X) be capable of change, and mediators

capable of activation, even if such dynamism is only a potentiality or cannot be directly

manipulated for example in experimental conditions. Gerring includes a range of related

terms for causal influencers, such as variables, conditions and factors (2012, p.202).

Furthermore, the outcome Y must be clearly specified and in some way measurable.

Gerring also asserts that important, or ambiguous background conditions ought to be

explicitly mentioned. Finally he argues that self-maintaining processes are not causal

because in these cases, X and Y are insufficiently distinguished. Figure 4 shows causal

influence in a flow diagram:

31 of 271

Figure 4: Causal influence flow diagram

In this causal syntax diagram, X (causal factors) lead to M (mechanisms of change)

which in turn leads to Y (changes in outcome). B (constant background factors) have an

antecedent and indirect effect on X and therefore an even less determinable effect on Y.

Incorrect causal assumptions

Beckwith, Dickinson and Kendall (2008) provide an illustrative example of how incorrect

causal inferences arise. An individual attends a smoking cessation class and

subsequently quits smoking (outcome Y). A causal inference could be made that the

causal factor (X) was the individual’s adoption of information and advice provided by the

class, as shown in Figure 5.

32 of 271

Figure 5: Incorrect causal assumption

In the Beckwith et.al. (2008) scenario, the causal influence had actually been the

participant’s partner’s insistence that he quit smoking. If the outcome Y (quit smoking)

happened irrespective of the advice provided in smoking cessation classes, or any other

facilitative factor, the causal assumption diagram would be as shown in figure 6.

33 of 271

Figure 6: Simple causal assumption

One can make further causal extrapolations beyond those made by Beckwith et.al.

(2008), if additional facilitative factors leading to the participant quitting smoking are

considered. For example, background factors include a government campaign to place

graphic health warnings on cigarette packaging. Additionally, alongside attending

classes which the participant found helpful, he obtained a nicotine replacement patch

from his GP which reduced his urge to smoke. He had managed to quit smoking a few

years ago and so had self-efficacy for quitting. The causal assumption diagram for this

scenario is shown in figure 7, where (M) mechanisms of change (or non-change) may

be mediators or moderators.

.

34 of 271

Figure 7: Smoking cessation flow diagram

Outcome In this study, outcome is defined as changes to client symptoms, diagnosis, behaviour,

attitude or psychosocial functioning as a result of having undertaken psychotherapy. In

distinguishing process from outcome, Orlinsky, Grawe and Parks (1994) propose that

process includes all in-session impacts, including the immediate effects on the client of

therapist interventions. Describing outcome, Orlinsky (2009) distinguishes between

outcomes and outputs. Outcomes relate to the aim of therapy (for example reduction in

presenting problems). Outputs are further-reaching consequences on clients’ life-

trajectories and social productivity.

Clients may experience a range of meaningful outcomes and outputs, in addition to the

ones they sought psychotherapeutic help to achieve, or different from those measured

by the therapist or researcher. For example Binder, Holgersen and Nielson, (2010) and

Castonguay et.al. (2010) found that clients deemed good outcome for mental health

psychotherapy to include new ways of relating to others, and improved self-confidence.

35 of 271

It is possible that some clients may only experience or become aware of certain

outcomes and outputs months or years after therapy is complete.

In the current study, where participants could discuss the ‘outcome’ of their choice,

some discussed clinical ‘outcomes’, while all described ‘outputs.’ Several also said the

core problem for which they sought therapy was not fully remitted, which is also an

‘outcome.’

Outcome is a particularly nebulous and unreliably formed concept in psychological

research. Here I briefly describe four problems with the outcome concept: Mis-

conceptualisation; reporting bias; unclear definition of improvement; and the non-linear

nature of change.

Mis-conceptualisation

Outcome mis-conceptualisation refers to an incorrect assumption that an observed

effect or change is an outcome of psychotherapy. It can be seen, for example, when an

aspect of therapist or client experience in therapy is described as an outcome but is

actually a process effect that dissipates soon after therapy ends (Nilsson, Svensson,

Sandell and Clinton, 2007). It may be possible to distinguish ‘lasting outcome’ from

process effects by gathering follow-up data some months after participants finish

therapy in outcome research (Hill and Lambert, 2004).

Another form of mis-conceptualisation is the erroneous conflation of client positive

feedback with therapeutic improvement. Studies have found no relationship between

client satisfaction and treatment efficacy (Stiles, Shapiro and Firth-Cozens, 1990;

Lunnen and Ogles, 1998; Pekarik and Guidry, 1999). Lastly, Kenniston, Boltax and

Almond (1971) explain how what is described as an outcome might actually be a

reflection of the quality of client participation in the social structure of therapy.

Reporting bias

Client change might be judged from the perspective of the researcher, the therapist, an

observer, or the client, but it has been found that therapist and client views of ‘success’

in therapy do not necessarily correspond (Cartwright and Roth, 1957). As Mintz (1972)

36 of 271

showed in a study on the determinants of global outcome, judging ‘successful’

improvement is a process particularly vulnerable to therapist bias. Mintz found that

therapists thought clients who started therapy relatively well and made minor changes

were more ‘successful’ than those who started with more troublesome problems and

nonetheless made unexpected gains. This suggests to me that ‘success’ might

sometimes be conflated by therapists with ‘easier therapy.’

Results of psychotherapy effectiveness studies can be negative or null; but such

findings are less likely than positive results to be reported in the research literature

(Lilienfeld, et.al. 2014). Likewise there is a risk in psychotherapy outcome research that

causal significance can be incorrectly attributed to therapy rather than to potentially

greater influences that take place outside the consulting room (Lilienfeld, et.al. 2014).

Unclear definition of improvement

Examples of studies included in this concept analysis that clearly indicate thresholds for

improvement are Dow et.al., (2007), whose marker of ‘improvement’ was ‘35%

reduction in Y-BOS questionnaire’; and Alonso, Muchon, Pifarre, Malaix-Cols and

Torres (2001) who indicated ‘absence of panic attacks over past 2 weeks.’ Other

studies such as Kleindienst et.al. (2011) indicate that a ‘statistically significant’ change

has been achieved. However, 48/68 (70.6%) of studies did not include a specific

threshold for defining improvement either in terms of changes in outcome measures

from start to end of therapy, or changes in problems reported by other means. Without

such markers it is unclear when outcomes are reported as ‘better’ or ‘worse,’ what the

relative value of this change is.

The non-linear nature of change

Hayes, Laurenceau, Feldman, Strauss and Cardaciotto (2007) discuss how outcome

research might accommodate the idiosyncratic nature of psychotherapeutic change.

Using dynamic systems theory (Thelen and Smith, 1996) they explain how human

cognitive and behavioural change processes tend to be triggered by critical fluctuations

or disturbances in a person’s life that are in turn set off by events that arise in highly

interdependent social systems. Such fluctuations might treated as ‘error’ or ‘noise’ in

37 of 271

traditional pre-and post- therapy designs, which, for convenience, conceptualise change

as gradual and linear. But these disturbances give vital clues to the transformative

impact of destabilisation and reorganisation in clients’ lives. Hayes et.al. (2007) suggest

that if we want to understand what mobilises client change we must study ‘transition

points’ rather than relying solely on time course outcome data. I consider these in the

concept reconstruction.

1.6 Study structure This doctoral study is presented over six standard chapters: Introduction; Literature

Review; Methodology; Results; Discussion; Conclusion. Here I briefly explain the writing

style in which this study is presented; and then provide an overview of each chapter.

Style

In terms of structure, all chapters start with an overview but for the sake of parsimony,

do not end with summaries.

In terms of research design, the job of clarifying what ‘client factors’ are, and how they

can be synthesised into a coherent framework requires a deductive, qualitative,

exploratory and graphically rendered methodology that considers the interrelated

functions of different types of client activities and resources within and beyond

psychotherapy processes. This research approach might be thought of as a counselling

psychology equivalent of a map maker surveying and sketching up a broad and

diversely featured landscape.

In terms of language use, I use first person voice in order to take responsibility and

ownership of the views expressed, while also acknowledging that my work rests on

important ideas rooted in my readings of influential authors in and around

psychotherapy research.

In terms of authorial stance, I examine the client factors concept and its uses in

psychotherapy literature in explicitly critical ways. A fundamental supposition (following

Bohart and Wade, 2013) is that client factors are considerably more potently associated

with process and outcome than most psychotherapy research accounts for. Sometimes

where client factors are identified in the literature they may be those that endorse the

38 of 271

therapist’s role as skilled helper (e.g. ‘preference for type of therapist,’ Beutler,

Harwood, Someah and Miller, 2016) and client’s role as functionally impaired (e.g.

‘resistance,’ Beutler, Harwood, Michelson, Song, and Holman, 2011). Incomplete and

biased conceptualisations of client activity in and beyond therapy limit therapists’ (and

clients’) capacity to perceive and criticise taken-for-granted assumptions about ‘what

works’ in therapy (Beck, 2013).

Consistent with this critical approach I do not bracket off my assumptions about the

topics addressed here as some qualitative researchers advise (e.g. Tufford and

Newman, 2010). I have instead attempted to transparently incorporate them into the text

as I go along, because my views infuse my curiosity about client factors with

investigative vigour. To suppress them would be a form of self-denigration, and a denial

of the socially experiential approach to knowledge-development used here (Fivush,

2000; Rennie, 2012). My assumptions are included to help the reader understand the

perspective from which I interpret the material handled here.

Study overview: Chapter 2: Literature review

The literature review develops my argument for the need for an organising conceptual

framework for client factors. It does this by first situating the client factors concept within

common factors theory, and then by scoping the diversity of client factor dimensions in

relevant literature. I highlight the disjointed nature of knowledge about the active client.

I then look at another common factor, therapeutic alliance, which has received more

research attention than client factors. I find it to be poorly conceptualised. I discuss

evidence that it may not directly influence outcome. I round out the chapter by

introducing change process theory but critiquing the conceptualisation of client factors

found in Orlinsky’s (1986/2009) process model.

Chapter 3: Methodology

The methodology chapter opens with an explanation of the critical realist ontology that

informs my handling of causality in this study. This is followed by a reflection on why

and how the study is shaped by a ‘muted group’ epistemology (Ardener, 1975;

Kramarae, 1981). My selection of a deductive qualitative ‘hierarchy of abstraction’

39 of 271

concept analysis method (Sartori, 2009) is justified in spite of its absence from

counselling psychology research, and in light of the unsuitability of proximate qualitative

analytic approaches. I explain the seven stages of concept analysis fashioned here to

manage, organise, analyse and interpret data from interviews with nine people who

recently completed psychotherapy, and data extracted from 68 psychotherapy outcome

studies, to reach a ‘reconstructed’ concept framework. I explain the structure and

colour-coding used to organise this framework over enhanced fishbone diagrams.

In Chapter 3 I also explain steps I took to ensure research rigour and credibility. I then

conclude with consideration of the ethical issues involved in completing this study

including my deliberations on the death of a participant.

Chapter 4: Results

The results chapter examines and explains the contents of the reconstructed concept

hierarchy. It is divided into two sections: helpful and unhelpful client factors. Each

section is subdivided by four temporal stages of change/non-change. In each stage the

component categories and elements are detailed with evidence from participant and

research data. Each section concludes with a summary of key findings.

Chapter 5: Discussion

To commence the discussion I review and answer my four research questions on the

basis of study results. Then, in order to fruitfully challenge the concept reconstruction I

subject it to analytic review using seven criteria for concept adequacy devised by

Gerring (2012). This review comprises the majority of the chapter. Within the review the

most substantive criterion is considered first, i.e. the concept’s utility for counselling

psychology theory. In this part of the discussion I review causal syntax diagrams that

summarise the study’s findings in relation to client contributions to psychotherapeutic

process. I then discuss the remaining six concept adequacy criteria, including domain,

fecundity, differentiation, resonance, consistency and operationalisation. The chapter

then moves on to discuss recommendations. I propose that social learning theory is an

important area of study for counselling psychologists. The discussion concludes with a

non-exhaustive summary of the study’s limitations.

40 of 271

Conclusion

Concluding the study in Chapter 6, I briefly describe three contributions to knowledge: a

provisional client factors process model; a tentative adjustment to change process

theory that requires further testing and validation; and the introduction of concept

analysis as a potentially relevant methodology for counselling psychology theoretical

research. Finally, the results are reviewed in relation to the aims of the study.

41 of 271

CHAPTER 2: LITERATURE REVIEW

2.1 Chapter overview The aim of this literature review is to locate and describe the relevance of client factors

in common factors theory and psychotherapy process theories, and to deepen my

explanation for why I think the client factors concept requires an organising framework.

The chapter unfolds over four sections. The first section provides a brief history of the

development of common factors theory. I contrast Lambert’s (1992) and Wampold’s

(2001; Wampold and Imel, 2015) theories of the relative distribution of various common

factors in psychotherapy outcome variance. I explain why I think the study of client

factors is important to further development of common factors theory. In the second

section I provide a brief history and meanings for the terms ‘client factors’ and

‘extratherapeutic factors,’ and explain why, in this study, the latter is discarded. I

examine some diverse dimensions of the client factors concept and then provide some

examples of under-conceptualisation of client factors in recent studies. In the third

section I review erratic conceptualisations of the related common factor ‘therapeutic

alliance.’ I review some problems with research on the role of therapeutic alliance in

psychotherapy outcomes. I explain why I think therapeutic alliance might be a reflection

of client and therapist relational factors and other client work factors, rather than a

causal factor in its own right. In the fourth and final section I explain the relationship

between client factors and psychotherapeutic change processes. I illustrate further

consequences of under-conceptualisation of client factors via critique of the treatment of

client activity and social factors in Orlinsky’s (1986/2009) transtheoretical process

model. I conclude the chapter with a brief explanation of why Pascual-Leone,

Greenberg and Pascual-Leone’s (2009) ‘task analysis’ method of psychotherapy

process theory development might be better suited to accommodate client factors, and

how this study might form a preliminary stage in such an endeavour.

2.2 Common factors theory Common factors theory proposes that different types of therapy have elements in

common that account for why, according to some research, they produce generally

equivalent results. This phenomenon is coined the ‘dodo bird hypothesis’ based on a

42 of 271

reference to Lewis Carroll’s Alice in Wonderland made by Saul Rosenzweig in a

prescient, and now famous, 1936 paper. In this story, the dodo bird judged that all

competitors in a race had won. Rosenzweig suggested that the equivalent benefits of

different psychotherapies are caused by “unrecognised factors in any therapeutic

situation that may be even more important than those being purposely employed”

(p.412). He posed a question that has inspired many subsequent studies: “what do

therapies have in common that makes them equally successful?” (p.413).

A landmark text in the development of common factors theory is Frank and Frank’s

1993 comparative study of psychotherapy, Persuasion and Healing (following Frank,

1973). Frank and Frank (1993) suggest a wide range of generic patient,

extratherapeutic and therapy factors are responsible for patient healing in various types

of psychotherapy. Patient factors include expecting to improve; remoralisation;

confession; learning, self-reflection; attitude change; and commitment to new values.

Extratherapeutic factors include peer and family influences; and stressful life

circumstances and events. Therapy factors include the patient-doctor relationship;

provision of education and encouragement for patient behaviour modification;

stimulation of patient emotions and self-reflection; and collaborative construction of new

plots for patient life stories that sustain better patient self-image. Frank and Frank

(1993) describe the psychotherapist as not so much an applied behavioural scientist, as

a hermeneutic interpreter of ‘clients as texts’ of their own life stories: a view that

positions the client as the primary author and the therapist as perhaps a temporary

editor. But they also identify psychological healing with the aid of a therapist as just one

of numerous types of ‘ritual’ people might undertake to achieve the same, or similar

aims of healing – activities that express, connect and reinforce one’s belonging in a

wider socio-ideological group. Opportunities for experiential learning that enhance

mastery and self-efficacy are available through such formal and informal social

memberships. Social memberships, it seems to me, in themselves comprise both the

means and the ends of social help-seeking.

43 of 271

Relative contributions of common factors to variability in outcome

Beside Frank and Frank (1993), other researchers have developed conceptual schemes

for understanding the common mechanisms of client change across various schools of

psychotherapy (for example, Goldfried, 1980; Orlinsky and Howard, 1986; Grencavage

and Norcross, 1990; Kolden, 1991; Castonguay, 1993; Garfield, 1995; Asay and

Lambert, 1999; Lambert and Barley, 2002; Castonguay and Beutler, 2006. See

Wampold and Imel, 2015 for a brief history of common factors conceptual model

development). A commonly cited conceptual scheme was devised by Michael Lambert

(1992) (coined ‘Lambert’s pie,’ see for example Cooper, 2008, p.56). Lambert attempted

to quantify the relative contribution of pan-theoretical factors to variability in outcome: in

other words, the determinants of whether therapy works well or not. He proposed that

30% of “improvement in psychotherapy patients as a function of therapeutic factors”

(p.97). These he described as therapist variables including warmth, empathy,

acceptance, which he subsequently relabelled “therapeutic relationship” (Asay and

Lambert, 1999, p.31), although this usage indicates only the therapist’s contribution to

the therapeutic relationship and does not account for the client’s contribution.

Additionally he attributed 15% to techniques of specific therapies; 15% to client

expectancy and 40% to what he described as “extratherapeutic change” (p.97) including

client factors such as ego strength; and contextual factors such as social support.

Lambert’s 1992 ‘percent of variance accounted for’ common factors model is

problematic for a number of reasons. I will touch on four here. First, “no statistical

procedures were used to derive the percentages that appear in [the figure], which

appears somewhat more precise than is perhaps warranted” (Lambert, 1992, p.98).

Compared to subsequent meta-analytic research by Wampold (2001, discussed below)

Lambert’s model considerably over-estimates the variance in outcome attributable to

therapist and techniques. Second, it obscures client agency by inserting client factors

under the label “extratherapeutic change” which includes spontaneous remission and

fortuitous events (Asay and Lambert, 1999). Third, where specific client factors are

mentioned they are only cursorily described, such as capacity for a meaningful

relationship with the therapist, maturity, motivation, quality of social support and

“personality makeup and ego organisation” (Asay and Lambert, 1999, p.32). Finally,

44 of 271

DeRubeis, Gelfand, German, Fournier and Forand (2014) criticise the ‘percent of

variance accounted for’ model because some of the factors, for example, therapist

technical skill, are not necessarily characterised by high variance within and across

studies, although still representing an essential - and potent - ingredient of therapy.

They suggest that this model also does not account for variability amongst clients in the

degree to which the quality of therapy provided to them will effect their outcomes – in

other words, client factors that render them, as DeRubeis et.al. (2014) describe,

‘spontaneous remitter,’ ‘easy,’ ‘pliant,’ ‘challenging’ or ‘intractable’ patients. DeRubeis

et.al. (2014) also note that Lambert’s common factors model does not facilitate

understanding of causal links between therapy factors and outcome, or the potential for

“patients to improve with the passage of time, irrespective of the quality of therapy

provided to them” (p.421). Regardless of these criticisms, Lambert has continued to

describe his 1992 model as offering an “empirical point of view” showing “percent of

improvement in psychotherapy patients as a function of therapeutic factors” (Lambert,

2013, p.200).

In 2001 Wampold re-evaluated common factors outcomes on the basis of rigorous

meta-analysis. Wampold radically re-drew variance attributions, with significantly

diminished variance attributable to therapy factors, which I will now explain in some

detail, as my argument for the potency of client factors in this study rests in part on

Wampold’s influential work.

Wampold’s pie

Wampold (2001, Wampold and Imel, 2015) developed a common factors conceptual

scheme called the ‘contextual model’ which accounts for the benefit of therapy over

three pathways: first, the foundation of a therapeutic relationship; next, through client

expectations of therapy; and finally, through client mobilisation of “treatment actions”

(Wampold and Imel, 2015, p.53).

Wampold’s statistically transparent, stable and robust research on the relative

contribution of therapeutic common factors found, in line with the ‘dodo bird hypothesis’

“small, if not zero” (2001, p.118) differences in outcome between a variety of treatments

in comparison trials. He also proposed that a range of common factors (not just the

45 of 271

therapeutic relationship as in Lambert’s 1992/1999 model) account for a greater

proportion of variance in outcome than specific components of different therapies.

Wampold (2001) says 13% of variability is due to psychotherapy and the remaining 87%

is due to client and extratherapeutic factors. “Variance attributable to clients is great”

(Wampold, 2001, p.204).

Wampold breaks down psychotherapy’s 13% of variance in somewhat complicated

terms. Factors common to most types of therapy account for approximately 70% of the

13% (in other words, approximately 9% of the whole). These include several therapist

factors: competence, congruence, empathy, affirmation (Wampold and Imel, 2015,

p.209, p.258), allegiance/adherence to a “model” (2001, p.205; Wampold and Imel

2015, p.258) and the ability to culturally adapt the model to fit the client (p.209).

Wampold and Imel (2015) also include two client-therapist collaboration factors: goal

consensus and therapeutic alliance; and two client factors: expectations and placebo

effects (p.258).

Factors attributable to specific therapies account for approximately 8% of the 13%

(approximately 1% of the whole) (Wampold, 2001, p.207). The remaining 22% of the

13% (approximately 3% of the whole) Wampold (2001) says is “unexplained variance”

which he attributes to client factors. Client and extratherapeutic factors account for 90%

of variance (87% + 3%).

Figure 8 shows a graphic representation of variance attributions in ‘Wampold’s pie.’

46 of 271

Figure 8: Wampold's pie

While it is useful to see how important client and extratherapeutic factors are to

psychotherapeutic process and outcome, Bandura (1986) commented that “research

aimed at estimating the relative percentage of behavioural variation due to persons or

situations is ill suited for clarifying the transactional nature of human functioning” (p.29).

This suggests that pie charts showing proportional contributions of different variables to

psychotherapy outcome cannot capture the socially interactive determinants of

psychosocial change. Bandura recommends that social research should examine the

interdependency of causal factors; and the conditions in which people both create and

are influenced by their social climates “in an ongoing sequence of events” (p.29). The

client factors reconstruction created in this study achieves this.

Why study client factors

While psychotherapy outcome research continues to focus attention on the relative

effectiveness of competing models of psychotherapy (Wampold and Imel, 2015) “the

increase in knowledge from this vast enterprise has not been commensurate with the

Sources of variability in outcome: Wampold's Pie

Client and extratherapeutic factors 87% Therapeutic common factors 9%

Effects of specific therapies 1% Unexplained variance (client differences) 3%

47 of 271

investment of time, effort and money and the sophistication of the researchers. Different

studies often yield inconsistent findings, and most of the relationships found, even those

reaching the conventional level of statistical significance, have been too weak to

influence practice” (Frank and Frank, 1993, p.18). It follows from this that common

factors research needs a path to influence practice more meaningfully. I have so far

described two important signposts. First, it is clear from Wampold’s research that

variability of outcome due to the ingredients of psychotherapy is relatively slight.

Second, it is clear from Bandura’s criticism of pie charts that, rather than asking which

type of therapy gets better results, it might be more helpful to study key processes in

client extratherapeutic social interaction. I will now add a third signpost: Clarkin and

Levy (2004) assert, in their authoritative scoping review of influential client factors, that

“non-diagnostic client characteristics may be more useful predictors of psychotherapy

outcome than DSM-based diagnoses,” (p.214). They suggest research “focussed on a

constellation of salient [client] variables will be likely to show the greatest impact on

treatment process and outcome” (p.215). They do warn, however, that “one is faced

with an overwhelming number of client variables to consider” (p.214).

So, in calibrating these three signposts to form a direction for this study, it seems

important first to grasp which non-diagnostic client and extratherapeutic factors most

strongly or generally determine psychotherapy outcome. Then it seems important to

represent client contributions as a cohesive framework of social interaction, influence

and transition points. Task complexity and methodological unfamiliarity do not deter my

aim to make progress in this area.

2.3 Client factors In this section I first offer context for the terms ‘client factors’ and ‘extratherapeutic

factors,’ before providing a review of diverse known helpful and unhelpful phenomena

that comprise the term. This section concludes with some observations on the

obscuration of client factors in common factors studies. Further description and

analysis of specific aspects of the client factors concept found in this study comprise

Chapter 4: Results and Chapter 5: Discussion.

48 of 271

Context

The term ‘client factors’ may have entered published psychotherapy research

terminology around the mid-1980s (e.g. Hoffman, 1985), following ground-breaking

meta-analyses of client variables published by Garfield in 1971 and 1978. It contains

and obscures a large constellation of client activities and circumstances that may

influence outcome in psychotherapy. In medicine, psychiatry, occupational and physical

therapies the concept may be termed ‘patient factors,’ or more specifically, ‘patient

health behaviour.’ The concept is similarly used in nursing and midwifery studies (e.g.

Cox, 1982; Forchuk, 1994; Amoakoh-Coleman et.al. 2016); disability studies (e.g.

Scheirs, Blok, Tolhoek, El Aouat, Glimmerveen, 2012) and management consultancy

(e.g. McLachlin, 1999) but does not have much meaning beyond these professional

fields.

The term ‘client factors’ includes, but is not limited to, client demographic background

and socio-cultural identity; quality of social support; expectations and preferences for

therapy; motivation and readiness to change; learning in and outside sessions;

resilience to adverse life events; and severity of impairment (Beutler, 2016; evidence for

each of these areas will be discussed in the review of client factors that follows).

Extratherapeutic factors are more arbitrarily defined. Some authors (e.g. Addis and

Cardemil, 2006) state that client factors and unexplained and error variance are all

‘extratherapeutic.’ Others (e.g. Wampold and Budge, 2012) combine some therapy-

related factors (client motivation and involvement) with some non-therapy related

factors (social support and spontaneous remission) within the term. I argue here and in

more detail later in the study, that the term ‘extratherapeutic factors’ is redundant, as it

forms a part of client-related influence over the process and outcome of therapy.

However, I retain it where referenced authors have used it.

Review of the diversity of client factors

How clients use therapy depends on their beliefs about ‘normal and healthy’

psychosocial functioning. What clients want from therapy depends on how far their

experience deviates from perceived norms, or is causing suffering to themselves or

49 of 271

others (Boorse, 1976). Additionally, how and when clients access therapy is shaped by

availability, culture and context (World Health Organisation, 2013). So, potentially

influential client factors have diverse sources. For clarity I have broadly organised this

section of the literature review into helpful and unhelpful factors. Firstly I review helpful

demographic, social, cognitive and emotional, behavioural, and learning factors,

agency, and resilience with life events. Then I review unhelpful social, cognitive and

emotional and behavioural factors; and severity of impairment.

Demographic factors

In demographic terms, outcome has been found to be influenced in certain

circumstances by client gender and age (Woodhead, Cronkite, Moos and Timko, 2014).

On the other hand, Clarkin and Levy (2004) found no relationship between age and

outcome except for younger clients with substance abuse, who had worse outcomes

than older clients. Other studies have found no difference in outcome for people of

different ethnic groups (Lambert et.al., 2006), marital or employment statuses

(Davidson, et.al. 2004) or socioeconomic backgrounds (Garfield, 1994; Robinson-

Whelen, Hughes, Taylor, Hall, and Rehm, 2007). Garfield (1994) found a small positive

relationship between client educational level and treatment completion but on the other

hand Yu (2011) found a statistically insignificant link between lower client educational

attainment and therapy dropout.

In general, studies of the relationship between a range of demographic factors and

outcome offer inconsistent or contradictory findings (Bohart and Wade, 2013). It might,

then, be extrapolated that demographic factors per se are not necessarily a useful client

factor category.

Social factors

Outcome has been found to be affected by the nature and quality of clients’ social

support. For example, Bankoff’s (1996) quantitative study of engagement with 57

psychodynamic psychotherapy clients found that strong pre-treatment social network

support had a significant positive association with therapeutic bond, client self-

relatedness and therapeutic realisations in the initial phase of therapy.

50 of 271

More recently, Roehrle and Strouse (2008) conducted a meta-analysis of 27 studies on

the effects of social support on psychotherapy outcomes, and found a modest

correlation (r = 0.13) between social support and outcome, which they note is a similar

level of correlation as that found between therapeutic alliance and outcome. They make

a number of assumptions, including that the influences of social support are similar to

therapeutic help in certain aspects such as empathy and attachment, advice, social

coping and social skill development, and encouragement to put change into practice.

They also note negative social etiological factors, where aversive or controlling

interpersonal relationships, or insufficiently helpful social relations may be a reason

people seek help from a therapist. Roehrle and Strouse recommend that further studies

should examine the interaction effect of social support and therapeutic alliance.

Helpful cognitive and emotional factors

In an influential coding review intended to locate therapeutic common factors present in

a sample of 50 research articles, Grencavage and Norcross (1990) identified client

“positive expectancies and hope for improvement” (p.374) as a distinct common factor.

This finding is problematic because it can be argued that ‘hope’ and ‘expectancy’ are

different constructs. The study’s findings are additionally difficult to interpret because

the authors do not include references or specific inclusion/exclusion criteria for the

literature sample they assessed. I therefore focus on expectancy here, as it may have a

wider evidence base than hope (see for example meta-analyses by Kirsch, 1990 and

Constantino, Arnkoff, Glass, Ametrano and Smith, 2011).

The concept of client expectation can be crudely divided in two definitional realms: firstly

the extent to which clients expect to actively work to bring about desired change; and

secondly, how much they expect the therapist’s help will enable change. On the latter

point, Frank and Frank (1993) found that outcome can be influenced by clients’

expectations that they will be helped by the therapist. This finding may be related to how

credible the client found the therapy or therapist as a helping medium (Hardy, et.al.

1990). On the former point, Patterson, Uhlin and Anderson (2008) found that clients

who expected that in their role as client, they would have to work hard and attend

51 of 271

therapy regularly, experienced better alliance. This finding may be related to how well-

socialised the client was to the nature or type of therapy offered, or how much control

they wished to exert on the process (Greenberg, Constantino and Bruce, 2006).

Alternatively it may reflect how well client and therapist mutual role expectations were

matched or suited (Arnkoff, Glass and Shapiro, 2002); or how well client preferences for

type of therapy or therapist behaviour were met (Swift, Callahan and Vollner, 2011).

Other studies, however, have found minimal predictive capacity for expectancy or

treatment credibility (Ladouceur et.al, 2000; Borkovec et.al., 1987, Borkovic et.al. 2002;

referenced in Newman, Crits-Christoph, Gibbons and Erickson, 2006). In summary,

expectation, treatment credibility, or client preference do not consistently relate to

treatment response (Newman, Crits-Christoph, Gibbons and Erickson, 2006).

Looking next at motivation, this too is a multi-faceted concept. Bohart and Wade (2013;

after Donovan and Rosengren, 1999) differentiate motivation to obtain help from

motivation to change. Motivation has elsewhere been described as a set of dimensions

not limited to readiness to change but also including self-efficacy, decision balancing,

goals and values, and reasons not to change (Substance Abuse and Mental Health

Services Administration, 2012). Perhaps not surprisingly, Lombardi, Button and Westra

(2014) assert that motivation to change is difficult to measure. They found that in-

session client language (‘change-talk’) was a better predictor of outcome in cognitive

behavioural therapy for generalised anxiety than formal self-report measures, but that

arguments against change (‘counter-change talk’) more consistently and potently

predicted negative outcome than arguments in favour of change predicted positive

outcome. Furthermore, Vogel, Hansen, Stiles and Götestam (2006) found that high

motivation to change was not significantly related to outcome in cognitive behavioural

treatment for obsessive compulsive disorder. Inconsistent findings may to some extent

be related to wide variability in term definition and operationalisation.

A concept related to readiness to change but more action oriented, is goal-directedness.

People who are more goal-focussed in therapy may experience early change in therapy,

which has in turn been associated with positive outcome (Hansen and Lambert, 2003).

52 of 271

In particular, clients who articulate ‘approach goals’ rather than ‘avoidance goals’ were

found by Elliot and Church (2002) to have better outcomes.

Self-esteem is noted as a client factor by Bohart and Tallman (2010). A study by Rice,

Ashby and Slaney (1998) found that self-esteem mediated the relationship between

maladaptive perfectionism and depression, through a buffering effect. However this was

not a psychotherapy outcome study. Self-esteem is studied more commonly as an

outcome of psychotherapy rather than as a mediator or moderator of outcome (Smith

and Glass, 1977). In short, self-esteem does not have a strong evidence base as a

client factor associated with variability in outcome.

Helpful behavioural factors

The association between quality of client participation in psychotherapy and outcome

has been made by Orlinsky, Grawe and Parks, 1994; Kazdin and Wassel, 1999; and

Littel, Alexander and Reynolds, 2001. Client willingness to talk about and explore

emotions and experiences, along with willingness to try new behaviours are highlighted

by Bohart and Tallman, 2010. A meta-analysis by Mausbach et.al. (2010) found a

significant association between homework completion in CBT and positive outcome,

although their moderator variables were types and sources of homework completion

ratings rather than types of client homework activity. In terms of specific homework

activities, various forms of client self-monitoring have been associated with positive

outcome in studies by Clarke, Rees and Hardy (2004), Stirman et.al. (2018) and Khattra

et.al. (2017). The latter study also found that relaxation skills learned in therapy helped

clients make meaningful changes in coping. Problem-resolution (Paulson, Truscott and

Stuart, 1999) and associated coping strategies (Carver, Scheier and Weintraub, 1989)

may be taught and assisted via therapeutic intervention, but their operation and long-

term adoption relies on client perseverance and resourcefulness.

Learning, agency and resilience to life events

A key element of problem resolution is client learning (Gershefski, Arnkoff, Glass and

Elkin, 1996). Attainment of insight, new realisation or understanding has been found by

numerous researchers to help people adopt a revised view of reality, and of themselves,

53 of 271

which in turn leads to more constructive or congruent lifestyle choices (see for example

Brady, 1967; Elliott, 1985; Frank and Frank, 1993; Garfield, 1994; Hanna and Ritchie,

1995; Timulak, 2007; Bohart, 2007; Higginson and Mansell, 2008). On the other hand, a

study by Llewelyn, Elliott, Shapiro, Hardy and Firth-Cozens (1988; referenced in

Barkham, 2002) compared therapist and client views of helpful events in therapy. The

study found it was therapists who rated the development of insight as the most helpful

element of therapy, whereas clients rated problem evaluation and resolution as most

helpful.

Gibbard (2014) found clients achieved new realisations through writing, reading and

thinking; self-expression, self-reflection and self-acceptance. These findings are echoed

in studies by Elliott (1985), Clarke, Rees and Hardy (2004) and Binder, Holgersen and

Nielsen (2009) whose participants described how reflexively talking about thoughts,

emotions and experiences led to creating new meaning and improved self-

understanding.

A particular form of reflexive insight described by Rennie (1992, 2002, 2010) is client

agency: a sophisticated form of active learning in which clients intentionally participate,

interpret, assess and transform relational interactions in and beyond therapy in order to

achieve their aims and practise new behaviours. An unpublished study by Greaves

[Wade] (2006, noted in Bohart and Wade, 2013) demonstrates how actively clients use

relationship building skills to develop rapport with their therapist, including building a

common language base, honest engagement, expression of vulnerability, hope and

appreciation. Clients also use agency beyond managing their relationship with their

therapist. In a study of client retrospective narratives about therapy experiences, Adler,

Skalina and McAdams (2008) summarise how their participants described “a struggle

with a discrete and personified problem which is …ultimately vanquished by a

reempowered, agentic protagonist….Their therapeutic work continues even though they

have stopped meeting with a therapist” (p.730) [the protagonist being the client

themselves]. It is my interpretation that participants in this study seemed to own their

process of change, and conceptualised it as a larger and longer process in which

therapy had played a smaller and shorter part. However, despite evidence for the

54 of 271

construct of client agency, Coleman and Neimeyer (2015) found, in a review of

psychometric properties of client agency measures, that associated studies showed that

while agency was positively associated with therapeutic alliance, it did not directly affect

therapy outcome. A qualitative study by Mackrill, 2008 examines client reports of

agency and associates these with specific client-reported goal attainment (but not

psychometric outcomes). Mackrill found that therapists sometimes did not notice client-

reported change strategies. It would seem important therefore, where researchers wish

to develop knowledge of client agency and its impact on outcome, that they analyse

direct client reports in relation to specific outcomes, rather than solely depend on

therapist opinion or experience of client agency (for example as described by Williams

and Levitt, 2007).

Resilience is noted as a client factor by Bohart and Tallman (2010) and Beutler (2016)

however, it is not necessarily well-conceptualised in psychotherapy literature (Fougere,

Daffern and Thomas, 2012). It may refer to a personality trait, an outcome, or a process

of adjustment to adversity (Helmreich et.al., 2017). For example, in a study unrelated to

psychotherapy outcome, Ong, Bergeman, Bisconti and Wallace (2006) provide multiple

definitions for resilience including an outcome: achievement of “positive outcomes in the

face of untoward life events” (p.730) and a capability: “to maintain and preserve the

boundaries between positive and negative emotional states” (p.730). Yi, Vitaliano,

Smith, Yi and Weinger (2008) found that resilience capability mediated the relationship

between psychological adjustment and physical health in patients with diabetes through

a buffering effect. However this too was not a psychotherapy outcome study. Resilience

is studied more commonly as an outcome of psychotherapy rather than as a mediator or

moderator of outcome (see for example Helmreich et.al.’s 2017 protocol for a future

Cochrane Review of psychological interventions for resilience enhancement in adults;

and a meta-analysis of resilience training interventions by Joyce, et.al., 2018). As with

self-esteem, resilience does not have a strong evidence base as a client factor that

mediates variability in outcome.

55 of 271

Unhelpful factors

In a literature review seeking to understand client experiences of hindering factors in

therapy, Henkelman and Paulson (2006) note that researchers find it difficult to elicit

information about unhelpful factors from clients. However, there are some hindering

client factors within and beyond psychotherapy that have been acknowledged in

research.

Unhelpful factors are here divided into social, cognitive and emotional, behavioural

factors, and severity of impairment.

Unhelpful social factors

There is possibly clearer evidence of the detrimental impact of unsupportive, neglectful

hostile or stigmatising client social ecology on psychotherapy outcomes than there is for

the beneficial impact of helpful social ecology. In a trial of psychiatric clinic treatment for

generalised anxiety disorders, Yonkers, Dyck, Warshaw, and Keller (2000) found a

diminished likelihood of remission was associated with poor spousal and family

relationships. Probst, Lambert, Loew, Dahlbender and Tritt (2015) found that inpatients

undertaking psychotherapy treatment were more likely to experience extreme negative

deviation from normal outcomes when they experienced poor social support and

adverse life events. Probst et.al. (2015) recommend that, to improve treatment

outcome, therapists should help poorly responding patients to deal with life events and

build social networks.

Social isolation has been identified as a major health risk (Pantell et.al. 2012) that

inhibits regulatory functions necessary for self-reparatory change (Baumeister and

DeWall, 2005). Related to social isolation, Joutsenniemi, Laaksonen, Knekt, Haaramo

and Lindfors (2012) found that lone parents and divorced people were significantly less

likely to benefit from therapy, or else needed long-term therapy. Castonguay and

Beutler (2006) identified associations between financial or occupational difficulties and

worse outcome in therapy. They also identified that clients with dysphoric disorders are

less likely to benefit from therapy when they are from underserved ethnic or racial

groups. Cochran and Cauce (2006) found that drug and alcohol misuse severity may be

exacerbated by client minority sexual or gender identity, and suggest that “coping with

56 of 271

stigmatised sexual identity, dealing with stressors of being a minority group, and

internalising negative feelings towards the self are possible explanations” (p.144).

Unhelpful cognitive and emotional factors

Client hopelessness, passivity, and self-protective resistance to change negatively

impact client efforts to change (Bohart 2010). Carver et.al. (1989) identified emotional or

behavioural disengagement, denial, and substance misuse as unhelpful to outcome

particularly when participants felt they had low control over adverse stressors.

Laaksonen et.al. (2012) found that participant inability to clearly define a problem-focus

contributed to lower levels of symptom change at one-year follow-up. Low goal-

directedness has been linked in older adults with higher levels of stress and lower levels

of engagement with social support (Payne, Robbins and Doughty, 1991). Walker and

Rosen (2004) identified client self-blame or self-criticism as precursors of poor outcome;

and similarly Powers, Milyavskaya and Koestner (2012) found self-criticism to be

negatively associated with goal progress across a variety of domains. Unwanted

thoughts were reported by Llewelyn et.al. (1988) as correlated with negative outcome.

Kim, Zane and Blozis (2012) found that somatic symptoms were negative predictors of

outcome.

Severity of impairment

Outcome may be negatively impacted by client initial symptom severity or functional

impairment. Lorenzo-Luaces, DeRubeis and Webb (2014) found that people with three

or more prior depressive episodes were significantly less likely to benefit from CBT,

regardless of the quality of therapeutic alliance. Castonguay and Beutler (2006) found

evidence that, regardless of diagnosis, clients are less likely to benefit from therapy

when they have higher levels of impairment; have been diagnosed with a personality

disorder; have low internal locus of control; or “negative self-attribution” (p.356). Similar

findings regarding severity of impairment have been described in studies by Keijsers,

Hoogduin and Schaap, 1994; Yonkers, Dyck, Warshaw, and Keller 2000; and in analytic

reviews by Clarkin and Levy, 2004; Keeley, Storch, Merlo and Geffken, 2008; and

Beutler, 2016). However, research findings relating to the impact of severity of

57 of 271

impairment on outcome should be interpreted with caution. Bohart and Wade (2013)

note that “when discussing the relationship between initial levels of disturbance and

outcome it is important to know if outcome means amount of change, or final status

(return to normal levels of functioning)” (p.227). Earliness of change is another

outcome-influencing variable related to severity of impairment. Hansen and Lambert

(2003) found that “a patient’s pretreatment functioning…influenced time to change, with

higher pretreatment distress producing more immediate change than lower pretreatment

distress” (p.7). In summary, severity of impairment is not necessarily a unilaterally

negative client factor. As a client factor associated with outcome, severity can be

mediated by other factors including client motivation to reduce distress; or the curative

effects of client efforts to curb more extreme problem behaviours.

Obscuration of client involvement in therapy

Clients’ views of how and when change occurs, and which changes are important can

differ from psychotherapists’ views (Kazdin, 1999). Similarly, clients’ views of what is

helpful and unhelpful in therapy can differ from therapists’ views (Wark, 1994). When

clients’ views on their role in therapy are considered, their participation can sometimes

be downplayed. For example, a study of helpful therapeutic processes by Watson,

Cooper, McArthur, and McLeod (2012) says the research allows “clients’ contributions

to the change process to come to the fore” (p.77). But in the resulting analytic map,

therapist speaking activity is expanded to 13 discrete categories, while client speaking

activity is reduced to just one, ‘talking’ (p.84).

To conclude this section on known client factors, I suggest that common factors theory

‘short-circuits.’ It shows that the client is the most potent common factor in

psychotherapy outcome, but it does not fully explain how and why that is the case, and

what could be done with this knowledge to improve therapy. Castonguay observed in

1993 that “not all of the common factors identified at a conceptual level have been the

object of thorough empirical investigation….Among the variables that need to be better

understood are the cognitive, emotional and behavioral aspects of client’s [sic]

involvement in therapy” (p.275). Perhaps more research attention has been paid to a

single therapy factor, therapeutic alliance, than to all client factors.

58 of 271

2.4 Therapeutic alliance Common factors theorists have tended to focus on the importance of the therapeutic

relationship, with ensuing debates about how to conceptualise it (Gaston, 1990; Horvath

and Bedi, 2002); measure it (Lambert, 2010; Ardito and Rabellino, 2011) and optimise it

(Wampold, 2008; Rousmaniere, 2016; Castonguay, 2017). This focus may be because

the working relationship is the factor some therapists believe they have the most

responsibility for, or control over.

By way of brief definition, therapeutic relationship is an umbrella term for what some

authors following Greenson (1967) have broken down into three components including

transference, alliance and ‘real relationship.’ Each of these components could be

viewed as inconsistently conceptualised, as I will now explain.

Transference

Transference has been defined as client activation and application of the mental

representation of a significant other to the therapist (Glassman and Andersen, 1999).

However King and O’Brien (2011) suggest that transference is difficult to differentiate

from other aspects of therapeutic relationship and constitutes only “a small part of the

feelings that exist between therapist and client” (p.12).

Alliance

Alliance is contrarily defined across the psychotherapeutic literature. For example Gelso

(2014) says it is the “joining together of the reasonable self or ego of the client and the

therapist’s analysing or therapizing side for the purpose of the work.” Bordin (1979)

suggests it is client-therapist development of a trusting bond, and joint agreement on

goals and tasks of therapy. Some non-directive approaches might not recognise task

assignment as a purpose of therapy. Horvath and Bedi (2002) acknowledge that there

is no universally accepted definition of the alliance concept.

‘Real relationship’

Turning to the third component, the ‘real relationship’ is also nebulously and potentially

controversially defined. Gelso (2014) says it is the combination of client and therapist

59 of 271

‘genuineness’ with their ‘realistic perceptions’ of one another. Genuineness is a

contextually determined construct. On the therapist’s side it is certainly delimited by

judicious self-disclosure required by the professional role; and on the client’s side by

their knowledge that therapy is a temporary working relationship within which they

perform to the therapist’s evaluative gaze (even ‘unconditional positive regard’ has

conditions). ‘Realistic perception’ is impossible to assess or measure as our

intersubjective ideas about others are socially constructed, highly idiosyncratic, and in

constant flux.

Contested research on the role of alliance as a common factor

Major research contributions on alliance (e.g. Horvath, Del Re, Flückiger and Symonds,

2011) are correlation studies, which suggest only an association between alliance and

outcome, not a causal influence. Wampold and Imel (2015) provide a lengthy appraisal

of various other methodological problems with alliance research including halo effects,

rater bias, publication bias, research allegiance, inappropriate timing of measurements,

and differential importance of the alliance within various approaches. While alliance may

be important to most therapists, Bachelor (1995) and Bedi, Davis and Williams (2005)

found that therapists and clients hold divergent understandings of relationship factors,

with some clients paying little regard to collaboration with the therapist.

Studies of therapist and client relative contributions to therapeutic relationship produce

conflicting findings. For example Krupnick et.al. (1996) found that client early-treatment

contributions to the therapeutic alliance were significantly related to therapeutic

outcome, while therapist contributions were not related to outcome. Contradictorily,

Baldwin, Wampold and Imel (2007) found that while therapist variability in therapeutic

alliance affected outcome, client variability in therapeutic alliance made no difference to

outcome, leading Wampold and Imel (2015) to conclude, “there is no relationship

between the patient’s contribution to alliance and outcome….The alliance, with the

possible exception of the bond, is not directly therapeutic. Agreement on the goals and

tasks of therapy are needed to ensure that collaborative work proceeds in therapy”

(pp.190, 194). An inference I make from these findings (which I develop further in

Chapter 5: Discussion), is that the effectiveness of the therapeutic alliance mirrors client

60 of 271

social adjustment, capacity for initiation of relationships, and perseverance in using

relational social learning both within and beyond therapy. Therapeutic alliance is from

this standpoint an indicative client factor rather than a therapy factor. It is my view that

client relational skill generates numerous causal influences on the process and outcome

of therapy, and may go some way to explaining the correlation between alliance and

outcome. This view is supported by discussion of the impact of client ‘social ecology’ on

therapeutic bond by Bankoff (1996), who found that clients who formed a better

therapeutic bond at the beginning of treatment “entered treatment feeling cared for by

each of their significant ties. They are people who discussed entering treatment with

their romantic partner, and who eventually did enter with the support and blessing of

their [parents], their romantic partner and their two best friends. They are people who

also felt pressured by their friends to get help through therapy” (p.58) and ultimately “got

much more out of their therapy” (p.59).

2.5 Change process theory This literature review has so far examined knowledge and conceptualisations of client

factors associated with psychotherapy outcome. The chapter concludes by looking at

how client factors are positioned in change process theory, and in particular, Orlinsky’s

(2009) well-known ‘Generic Model of Psychotherapy.’

Orlinsky’s Generic Model

Orlinsky’s process framework maps the interrelationships of “all the actions, events and

experiences that occur in psychotherapy – and the individual and social contexts in

which psychotherapy occurs” (2009, pp.321-322). Based on meta-analysis of a

substantial number of studies, the first version appeared in Orlinsky and Howard (1986)

and has since been developed in stages over several decades. The purpose of the

review that follows is to explain the model but also to critique its passive conception of

the client and its restricted view of social influences surrounding the therapeutic

process. In this way I further build my argument that client factors require clearer

conceptualisation before they can inform and evolve psychotherapy process theory.

61 of 271

Orlinsky distinguishes between the systems of action that take place in therapy, and

those that take place in the social milieux surrounding therapy, which I explain in turn.

Systems of action within therapy

Orlinsky, Grawe and Parks (1994) describe six aspects of process that appear in all

forms of therapy: contractual implementation (in particular, patient suitability and

therapist skill for therapy); technical operations (in particular, interpretive conversations

about life problems and core personal relationships); interpersonal bond (in particular,

the patient’s perspective on their therapist’s usefulness); intrapersonal self-relatedness

(in particular openness to experience versus defensiveness); clinical in-session impacts

(particularly insight, emotional relief and hope for the future, and negative

‘psychonoxious’ impacts); and sequential flow (the timing of significant events early,

middle or late in therapy).

Systems of action in the social milieux

Orlinsky acknowledges that therapy process is influenced by the patient’s social

conditions and the “adequacy and predictability of support from the economic, political

and social structures that support communal life” (p. 338) prior to entering therapy.

Orlinsky calls these influences “inputs” (p.332). Consequences resulting from the

patient’s engagement in psychotherapy process are “outputs” (p.332). These include

some influence over the quality and character of the patient’s “involvements with family,

friends and intimates and a public life of dealings with teachers, employers, colleagues

and officials in varied organisations” (p.335) but he does not discuss how these

relational involvements might have a reciprocal influence on what happens in therapy.

The powerful therapist and the passive patient

My first critique of Orlinsky’s (2009) process model is that it describes the role of the

patient in explicitly passive and receptive terms. The “compliant and cooperative”

(p.324) patient shows the “ability to perceive and absorb the input stemming from

therapeutic operations and from the quality of the therapeutic bond” (p.327).

Furthermore, Orlinsky’s framework diagram positions the therapist’s contribution to the

therapeutic bond as the most potent ‘active ingredient’ in therapy, while the patient’s

62 of 271

contribution – notwithstanding its essential nature in therapy - is omitted. Nonetheless

Orlinsky obliquely alludes to the power of the client’s contribution, when he proposes

that the ‘dodo bird effect’ in common factors theory is caused by the benefit derived by

“well-disposed, high functioning persons” from “almost any sort of therapeutic

intervention….yet even very skilful and experienced therapists find it difficult to succeed

with patients (e.g. ‘borderlines’ or psychotics) who have difficulty forming a secure

therapeutic bond” (p.330). This interpretation of the ‘dodo bird hypothesis’ illuminates

the flawed logic of privileging the therapist’s contribution to the therapeutic bond if it is

the client’s disposition that more powerfully predicts the outcome of therapy.

Separation of therapy from social context

My second criticism is that Orlinsky distinguishes therapy as a social activity separate

from the rest of the patient’s life, while nonetheless asserting therapy has a primary

impact on the patient’s capacity for engagement with “the ‘never-ending story’ of finding,

fostering or failing in intimate relationships” (p.338). He positions the ‘treatment milieu’

as the therapist’s office rather than the client’s social life. He does not directly consider

the possibility that, for the client, therapy is just another ‘intimate relationship’ which

reflects, more than it changes, their capacity for relational engagement - although he

notes that therapists encounter difficulties forming therapeutic bonds with clients who

have difficulties forming bonds outside therapy.

In summary, Orlinsky’s model in my opinion inadequately accounts for the influence of

client factors on therapy process.

Task analysis: Towards a process model for the active client

Research that seeks to answer ‘how did the client and their social context (in which

therapy plays a part) achieve that outcome?’ might employ task analysis as a method

for mapping and understanding psychotherapy change processes.

Pascual-Leone, Greenberg and Pascual-Leone (2009) describe task analysis as a

three-step research method, involving discovery, validation and dynamic modelling

phases. It starts with a declared set of theoretical assumptions and uses these in the

discovery phase to inform observations from the top down, while using systematically

63 of 271

gathered empirical data to deductively inform the model from the bottom back to theory.

It values the subjective experience of the client and positions client work within internal

(cognitive-affective) and external (social) worlds as the “true causal mechanisms for the

individual expression of internal experience and of outward behaviour” (p.528). Task

analysis places the therapist in the “conceptual background in order to focus on optimal

client processes” (p.532). It aims to illuminate how clients approach specific tasks in

therapy, in order to build more abstract causal models “that hold true for clients in

general.” Empirical concept analysis, as used here, is an essential preliminary step to

provide process context, before task analysis can produce process detail.

I expect that by clarifying the concept of the active client, change processes will also be

observed and clarified. In this case, once the concept analysis is complete, Pascual-

Leone, Greenberg and Pascual-Leone’s (2009) three-phase task analysis approach

informs possible next steps for validation, then dynamic modelling for the development

of measurement tools for further research as potential follow-on projects to this thesis.

64 of 271

CHAPTER 3: METHODOLOGY

3.1 Chapter overview This chapter provides a full explanation of how the concept analysis was carried out,

presented over twelve sections.

The first section explains why a critical realist ontology is useful for exploring the

influence of client factors in research. The second section describes my epistemological

assumption that knowledge creation occurs through social processes. I describe muted

group epistemology and explain why I have chosen it for this study. I counter some

criticism of muted group theory.

In the third section I describe the concept analysis design used here. I first explain

concept analysis as a deductive qualitative method. I then explain the relevance of

concept analysis to counselling psychology and remark on its absence from counselling

psychology literature. I rule out concept mapping, content analysis, framework analysis

and evolutionary concept analysis. I then differentiate replacement from reconstructive

concept analysis and explain why I chose the latter, in particular Sartori’s (1984/2009)

method. The section concludes with an overview of the stages and tasks of concept

reconstruction Sartori recommends.

The fourth, fifth and sixth sections deal with participant data management for creating

the participant-informed concept framework. In particular in the fourth section I explain

participant recruitment, characteristics and data collection. I also cover the semi-

structured interview design, interview style and transcription method. In the fifth section I

explain how I summarised interviews, created a data extraction and analysis

spreadsheet and organised data hierarchically through processes of data cleansing and

data immersion. In the sixth section I explain data interpretation through the creation of

a fishbone framework for organising and representing the data, and the delineation of

four levels of concept abstraction.

The seventh, eighth and ninth sections deal with research data management for

creating the research-informed concept framework. In particular in the seventh section I

explain the study selection process and characteristics of included studies. In the eighth

65 of 271

section I explain how I extracted research data to an extraction spreadsheet. In the

ninth section I explain research data interpretation to its own fishbone framework.

In the tenth section I explain the process of contrasting the participant-informed and

research-informed frameworks, and the removal of elements deemed marginal due to

contradictory evidence. I then explain how the two frameworks were combined into one,

using color-coding to differentiate the elements and categories that were based on (a)

participant-only data; (b) research-only data; and (c) participant and research data.

In the eleventh, penultimate section I explain how I ensured research rigour including

data triangulation, researcher reflexivity, audit trail, participant checks and catalytic

validity.

The chapter concludes in the twelfth section by explaining ethical issues including

approval, consent, confidentiality and risk, as well as a researcher dual role and a

participant death.

3.2 Critical realism and theoretical assumptions The need to balance clarity against flexibility in conceptual research is an ontological

matter: it requires a philosophy of the nature of being and reality. Concepts are abstract,

definitional, theoretical words; but they find their expression and formation in everyday

human activity and conversation which is naturally pragmatic, diverse, local, and

evolving. So, the process of analysing and developing concepts requires the researcher

to take a position on the nature of reality: Do concepts point to fixed, timeless,

decontextualized truths; or are they culturally-rooted linguistic constructions that only

exist insofar as the words and their meanings usefully promote the interests of a

particular power-group?

The former position is compatible with ‘realist’ or ‘essentialist’ philosophies, which hold

that reality exists in a pure or essential form independent of human representations of it.

The latter is more compatible with ‘relativist,’ ‘interpretivist’ or ‘social constructionist’

views which hold that reality is created through shared human activity, values and

languages; or ‘phenomenological’ views, which hold that knowledge is related to human

consciousness and objects of direct experience.

66 of 271

A strength of the ‘realist’ position in concept analysis is that it seeks to precisely define

and locate concepts in order to render them researchable ‘scientific’ entities. A

weakness of the ‘realist’ position in social science more generally is that it can reduce

reality to observable events, and discount theoretical entities with material impact such

as paradigms (Mingers, 2006).

A strength of the ‘relativist’ position is that it respects historically and culturally specific

social processes and language in the creation of knowledge. It allows some scope for

non-dominant representations in the production or sharing of knowledge (Crombie and

Nightingale, 1999). A weakness of the ‘relativist’ position, according to critical realist

philosopher Roy Bhaskar, is that several forms of ‘reality’ cannot be accounted for as

products of discourse. These include materiality (such as atomic structures) and

physical realities (such as gravity), and some moral and political ‘realities’ that ensue

from human activity such as social, financial and military power and oppression; and

more positively, trust and solidarity. A further weakness of the relativist position is that it

cannot fully accommodate a possible intransitive “ontological domain separate from the

activities and cognitions of human beings” (Mingers, 2006, p.20). Consequently relativist

knowledge of reality can be mistaken for reality itself; and thereby ontology confused

with epistemology.

Critical realism

It is of course possible to hold an ontological position in relation to concepts that lies

somewhere between the ‘realist’ and ‘relativist’ realms. Bhaskar’s critical realist

ontological model is useful here. A ‘critical’ ontology is appropriate in this concept

analysis because it attempts to detect and redress beliefs and practices that lead to

psychological therapy research focussing preferentially on therapy factors over client

factors (Usher, 1996). A ‘realist’ ontology is appropriate, because although absolute

knowledge of the real is impossible, it is possible to research social potentialities that

may or may not be galvanised in psychotherapy and the wider material and social

context in which psychotherapy takes place.

Critical realism stands in opposition to empiricism that reduces reality to observable

events, and posits instead that reality exists over three stratified and inter-related levels:

67 of 271

the ‘real,’ the ‘actual’ and the ‘empirical.’ Here I will explain these three realities and how

they apply in critical concept analysis of ‘client factors.’

The Real

Critical realism proposes that reality is the potentiality of causal properties and

mechanisms that shape the world as we experience it. However, this potentiality is

independent of human experience; it is transcendental and extra-linguistic. Its

properties are emergent, and may change their nature. However, we can materially

transact with the mutable nature of reality, and to some extent change its manner of

emergence.

Contextual, relational and individual agency are ‘real’ in that they are intrinsically

implicated in causal processes (Parr, 2015), although these processes may be only

partially – or incorrectly - understood in socially constructed terms. I can assert that

aspects of psychosocial un-wellness, and people’s efforts to address them, point to the

‘real’ in that they make a profound material and social difference in people’s lives,

irrespective of socially constructed debates about diagnostic labels or psycho-medical

mental health theory and treatment.

The Actual

The activation of generative potentialities under certain temporal, local or social

conditions produces changes in states of affairs. It is only in such actual activity of the

world that the unseen mechanisms of the real might be detected. We create and refine

imaginary models (theories) about causal realities on the basis of limited

phenomenological perceptions and language. In the case of client factors research, it is

in the context-bound manifestations of client change-work that we might find clues to

generative properties and mechanisms.

The Empirical

Bhaskar asserts that the move from the ‘actual’ realm of imaginary models to an

adequate account of ‘real’ causality is through empirical testing. “Science is an ongoing

68 of 271

social activity….Its aim is the production of knowledge of the independently existing and

transfactually [structurally generative] active mechanisms of nature” (2008, p.138).

Realism thus becomes critical as it assumes the possibility of an intransitive realm of

reality while also acknowledging the transitive dimension of human knowledge

construction through culture-bound language systems. Concept development and

theory improvement occur synergistically through analytic social research methods. So

in this concept analysis, in order to try to capture the essence of client change

processes and potentialities as they occur in clinical practice and research studies, I am

limited to using the language and existing knowledge of my historical, social and cultural

location, and the obtainable knowledge of others who have either studied or

experienced aspects of the concept. It may also be, as is my intention, that in studying

actual observations of the concept, I change the way the concept is formed in my own

language and perspectives. Additionally, the activity and motivation of my research in

itself may be reflexive evidence of “universals that are themselves implicated in their

own identification” (Scott, 2005, p.637).

Social relationships and first-hand accounts as evidentiary in critical realist

research

Critical realist research is suitable for counselling psychology because it accommodates

social relationships as causal influences. Just as counselling psychology research

prioritises and values individuals’ subjective and intersubjective experiencing (Kasket,

2016), Scott (2005) explains the importance of first-hand accounts in critical realist

social research:

Critical realists prioritise social actors’ descriptions of their experiences,

projects and desires. This does not imply that social actors can always

provide complete and accurate accounts of their activities, plans, projects

and histories. However, critical realists argue that such phenomena are

central to any investigation they may undertake (p.644).

69 of 271

Researcher theoretical assumptions

In terms of research ontology I assume that properties and mechanisms related to client

cognitive, behavioural and relational factors are an actual (in the critical realist sense)

influence in psychotherapy processes. I assume that knowledge of the meaning,

structure and implications of the client factors concept is underdeveloped because (a)

the inherent diversity of its components makes it difficult to draw up what is known into a

single theoretical framework; and (b) there is little professional incentive for

psychotherapists to research clients’ role in making therapy work.

Furthermore, I think it is possible to hold a social constructionist epistemology in my

development of psychological knowledge (elaborated in section 3.3) while also believing

that aspects of the material and mechanical world are not socially constructed. The

material world and its events would exist regardless of whether or not we talked about

them; although language and human relations may be necessary for much of the

material and mechanical world to have been created and activated. So for example, life

and death are real, not socially constructed events - although our knowledge of them is

understood through social means. Importantly, I believe God is real as a force of such

generative magnitude that it (not he or she) remains largely beyond human

comprehension or capability to describe, other than through the term ‘love.’ This

understanding maps onto Bhaskar’s transcendent realm of the real: “this is a beautiful

world. If you go into anything deeply enough, you will find it there. You could describe it

as inner emptiness, but I describe it as bliss…. I call this the ‘cosmic envelope.’ Spirit is

real. At a very deep level we are all spiritual beings” (Bhaskar, 2014, no page number).

Finally, I also believe, like Bhaskar, that “agency is real. We really do make a difference

in the world. Within any social context we as agents must take responsibility for our

action individually and collectively” (Bhaskar, 2014, no page number).

3.3 Muted group epistemology In counselling psychology, a phenomenological perspective leads us to respect human

subjective experience and consciousness as primary sources of knowledge about how

our world works (McLeod, 2003). Bhaskar says, “You start from a human experience.

You start from epistemology. That is where you are” (2014, no page number). While

70 of 271

some qualitative researchers assert that critical realism can function as both ontology

and epistemology (e.g. Fletcher, 2017), in this study I differentiate critical realism as a

relevant ontology of the nature of reality, from my epistemological stance on the nature

of knowledge. I recognise that while the ultimate truth of existence may lie beyond

human comprehension, I also believe human knowledge is constructed from subjective

experience, and in particular, relational activity and language (Gergen, 2015).

Epistemological assumptions

As a practitioner I observe that therapeutic processes and outcomes are shaped by

clients and their social circumstances, far more than by what I do to support their efforts

to make desired psychosocial changes. My thinking about the mechanisms and purpose

of psychotherapy is further influenced by Jordan’s (2010) relational-cultural theory; and

forerunning collaborative theoretical work about women’s development in relationships

from the Wellesley College Stone Center (e.g. Jordan, Kaplan, Miller, Stiver and Surrey,

1991). In particular I rely here on Miller’s idea that “official definitions of reality” (1991,

p.26) do not sufficiently draw upon complex processes of knowledge development that

occur in people’s day-to-day emotional and social connections with each other. While I

acknowledge the practical feminist stance inherent in these epistemological influences,

this study does not explicitly draw on feminist methodology and does not seek to make

a particular contribution to the psychology of women.

While asserting that this study’s epistemology is based on an assumption that

knowledge exists and is created in social interaction, I do not go quite as far as some

postmodern feminists might by saying that “there are no criteria for determining which

version of reality is better” (Fivush, 2000, p.88). If such total relativism were the case

then there would be no point, for example, in qualitative researchers taking steps to

demonstrate the trustworthiness of their work. Advances in knowledge would be

impossible (Parr, 2015).

Lowman (2012) reminds us that the focus of the scientist-practitioner psychologist’s

work should be to ensure that research informs practice in order to improve its

usefulness to clients. I try here to re-present client experience (a) to earn the public trust

placed in psychologists (Donati, 2016); and (b) to respect clients, their activity and

71 of 271

social contexts as integral to the construction of counselling psychology knowledge. I

have tried to create a study that moves beyond therapist experience, and even beyond

established therapy-centric perspectives of how people change in therapy.

This point leads to the relevance in this study of an epistemology derived from muted

group theory (E. Ardener, 1975; Kramarae, 1981). This theory allows client

autobiographic narrative as valid data for knowledge construction (Howard, Maerlender,

Myers and Curtin,1992; Fivush, 2000).

Muted group theory

Muted group theory was developed by anthropologists Edwin Ardener (1975) and

Shirley Ardener (2005) as a theory of communication and language. It highlighted how

socially subordinate groups are not permitted access to participate equally in the

generation of categories of thought, conceptualisation, and ideas about social

structures, and the encoding of these types of knowledge into discourse. The values

and experiences of people in muted groups are thereby kept private, or inadequately

recognised, or mis-represented in dominant communication, including in the design and

presentation of social research. “The astounding deficiency of a method, supposedly

objective, is starkly revealed: the failure to include half the people in the total analysis”

(E. Ardener, 1975, p.4). Another assumption I make, in linking the muted group

epistemology to the critical realist ontology adopted here (after Parr, 2015) is that

despite the socially-constructed nature of knowledge, it is possible for us to distinguish

between more and less successful methods for knowledge development (as Ardener

did in the quote above). Consequently, in seeking to better understand client factors,

this study asks clients directly what they do to make desired changes.

While muted group theory was subsequently adopted and developed as a feminist

theory (Kramerae, 1981), I draw from it in its original form as a theory that befits any

social group that is ‘spoken for’ by those with more social power. Ardener described the

mutually affecting spheres of dominant and muted groups as “simultaneities” (Ardener,

2005, p. 52). Interdependency is acknowledged here because psychotherapists need

clients to lend fundamental credence to their professional identity.

72 of 271

To illustrate how muted group theory epistemologically informs this study, I have

replaced the terms ‘men’ and ‘women’ with ‘psychotherapists’ and ‘clients’ in muted

group’s three main premises as follows (Oregon State University, undated):

1. Therapists cannot validly speak for clients: Psychotherapists and clients

perceive what happens in therapy differently. They have different aims,

perspectives and perception-shaping experiences.

2. Therapists use therapy-centric language: Psychotherapists enact their

language in ways that perpetuate their power, while insufficiently querying, or

being informed by, clients’ experiences about how the work of therapy is

achieved.

3. Client perspectives need translating if the dominant discourse is to adopt

them: Client agency in making desired life-change must be converted into

psychotherapist language and publicised in psychotherapist media in order to be

recognised.

While it may be challenging to practitioners to consider how these premises might apply

in psychotherapy, it been observed that such power imbalances in psychotherapy and

its research are subtle and pervasive (Harrison, 2013; Proctor, 2017).

Criticisms of muted group theory

Here I will address, insofar as they may impact its use in this study, three criticisms

made about muted group theory: it essentialises men and women; it is not empirically

tested; and it is dated (West and Turner, 2003).

Essentialisation

In this research I am not using muted group theory to consider male and female

communication patterns. It is used in a focussed way to address the muting of clients

and the side-lining of other forms of social support in psychotherapy research. I am not

‘essentialising’ that all psychotherapy research is therapy-centric; nor that all clients

behave the same way in therapy. On the contrary, I am seeking to thicken what is

known about the complexities of client influence on psychotherapeutic outcomes.

73 of 271

Empirically untested

An extensive body of literature supports and describes the use of muted group theory

(see Ardener, 2005 for a bibliography).

Dated

If we did not use today ideas that were considered radical in the 1970s we would have

abandoned the work of Carl Rogers and Aaron Beck. The dismissive accusation that

muted group theory is dated possibly betrays a discourse that feminist theories should

not age. “Many in our field have found that muted group theory has helped inform us

about how power functions in our talk and writing, and language….It helps explain

what’s going on in a way that is easily understood, believable and useful (Kramarae,

2005, pp.55, 56).

3.4 Concept analysis design As explained in the introduction on pages 15-19 I have employed a reconstructive

concept analysis method derived from political science (Sartori, 1984; Collier and

Gerring, 2009) to define, map and assess the client factors concept. This work is

intended to represent an initial stage before further research can validly explore or

develop measures for its intrinsic phenomena (Sartori, 1970; Robson, 2002), for

example by using task analysis as described on pages 62-63.

Before explaining concept analysis methodology and design in detail, I first explain the

process by which I found and decided on this method. I initially proposed to create a

client resources survey instrument that therapists might use in routine practice to

understand the role of particular client factors in assisting and hindering client-desired

change. This idea was not supported by my advisors who preferred I attempt a

qualitative thematic analysis of participant interviews around client factors. During this

developmental stage, with further reading into my topic and analytic strategy, it became

clearer to me that the term ‘client factors’ represented an abstract, theoretical shorthand

for a wide range of more specific categories of client experience and activity, and as

such it was a concept (as I have justified on page 14). From here I fixed on the idea that

the vagueness of the concept in itself posed the research problem I should tackle. After

74 of 271

undertaking training in thematic analysis with Virginia Braun, I observed that, despite its

popularity in counselling psychology research, thematic analysis was not the ideal

method for my task. Its principal aim is to identify and analyse patterns in qualitative

data (Braun and Clarke, 2013; 2014), while I needed to create conceptual anatomical

definition. In other words, I sought not simply to put ‘meat on the bones’ of the client

factors concept but to describe its ‘musculoskeletal structure’: a theoretically and

structurally coherent framework linked to existing theory into which to organise my

analysis. I therefore needed an analytic method that respected concepts as multi-

layered abstract linguistic entities, while permitting synthesis of findings from outcome

studies with participant interviews. From here, my ongoing methodological reading

brought me to Gerring’s (2012) appealing and relevant work on social science concept

analysis. Gerring in turn pointed me to Sartori (1984; 2009), whose antipathy for vague

concepts and practical methodology captured my imagination and determined the

direction for the completion and write-up of this study.

Concept analysis as a deductive qualitative method

Hupcey and Penrod (2003) recommend qualitative methods for reconstructive concept

analysis (which I have previously explained, aims to map relevant data units to a

theoretically informed hierarchical definitional framework, then assess the adequacy of

the result). Hupcey and Penrod (2003) combine source material from previous research

with client knowledge. Furthermore, Gilgun (2014) recommends a deductive qualitative

approach when working in theory-guided research with ‘sensitising concepts,’ which, as

with ‘client factors,’ give only a “general sense of reference” and lack “specification of

attributes and consequently [do] not enable the user to move directly to the relevant

content” (Blumer, 1954, p.7). The purpose of deductive analysis is to clarify or revise the

concept’s dimensions, and to show theoretical patterns in a way that “helps the

researcher see what they might not otherwise have noticed” (Gilgun, 2014, p.5). The

process moves from working first with thick descriptions, to interpreting an organising

framework, then assessing the pragmatic utility of the framework against existing

theory.

75 of 271

Concept analysis as a potentially useful method in counselling psychology

Concept analysis straddles academic and clinical knowledge development. Counselling

psychologists could use it to improve the explanatory and predictive nature of theory-

relevant concepts, with the ultimate aim of evolving practice (Forsyth, 1980) as relevant

scientist-practitioner activity (Corrie and Callahan, 2000).

Concept analysis is not used in counselling psychology

A total contents search of The Counseling Psychologist and the Journal of Counseling

Psychology conducted on 11.02.2018 located only one concept analysis study by

Hogan (1975) which describes a Q-sort used to create an empathy scale. A method

more commonly found is concept mapping (e.g. Goodyear, Tracey, Claiborn and

Lichtenberg, 2005) which is used to illustrate knowledge structures underlying concepts

through the use of diagrams of spacial and relational groupings of meanings. In this

regard concept mapping might be useful here as a comparative method. However

concept mapping is ultimately unsuitable for two reasons. First, it is an

expository/descriptive rather than a critical method. Second, it is positioned as a ‘post-

positive’ method that claims to “make use of an objective, impartial researcher and

standardised procedures” (Goodyear et.al., 2005, p. 236). A key premise of this critical

realist study is that no social research can be impartial. “Our theories and methods are

shaped by social forces and informed by interests” (Pilgrim and Bentall, p.262).

Another related method used in counselling psychology research is content analysis.

This may include the identification of conceptual themes relevant to the research topic,

for example in a literature review, but without a specific methodological analysis or

critical appraisal process (e.g. Shin, et.al. 2017; Pilgrim and Bentall, 1999; Olsson,

Bond, Burns, Vella-Brodrick and Sawyer, 2003; Hargons, Mosley and Stevens-Watkins,

2017). Such an approach would be insufficiently rigorous for the purposes of this study

as it is a form of literature review.

Framework analysis (Ritchie and Lewis, 2003) shares a similar method to concept

analysis, but likewise does not appear to be favoured by counselling psychologists. It

76 of 271

starts with a conceptual framework to which data is tagged (an etic process, Morrow

and Smith, 2000), whereas concept analysis (at least as I have used it as a

reconstructive process) starts with organising the data into thematically related groups

which are increasingly differentiated through processes of familiarisation, data

cleansing, and conceptual structuring, before a suitably shaped, theoretically-informed

framework is finalised towards the end (an emic process, Morrow and Smith, 2000).

Framework analysis uses concepts as ‘givens’ against which to locate and organise

appropriate data; while reconstructive concept analysis uses data units as ‘given’

connotations around which to locate and organise appropriate conceptual extensions to

link with existing theory.

Why I did not use a nursing concept analysis method

Nursing concept analysis methods in the tradition of Walker and Avant (2014), Rodgers

(1989) and Chinn and Jacobs (1987) provide clear descriptions of their analytic purpose

and methods. However they were not chosen here for several reasons. Firstly, these

approaches require the creation of ‘model’ or ‘exemplary’ cases. This points to their

realist stance that ‘truth’ is a fixed or universal entity that can be pinned down with

language and specifically located in experience. While this ontology may be suitable for

mathematical and other hard science concepts it is not in my opinion fitting for social

science concepts where meaning is fluidly context-bound (Duncan, Cloutier and Bailey,

2007). A criticism of nursing concept analysis methods, such as Rodgers’ ‘evolutionary’

approach (2000b) that rely solely on literature is that they can replicate professional

bias. They do not actively address language ownership about client experiences

(Wuest, 1994). Beckwith, Dickinson and Kendall (2008) argue that these nursing

concept analysis approaches are uncritical and derivative simplifications of a textbook

(Wilson, 1970) that offers a stepwise strategy for enabling grammar school pupils to

pass Oxbridge university exams. They also propose that these analytic methods lack

processes for evaluation of their findings.

Replacement versus reconstruction

The chapter now moves on to explain in more detail the concept analysis design used

here. In concept analysis an early decision must be made whether the aim is to replace

77 of 271

a defective concept by formulating a new one; or to improve an existing concept through

reconstruction (Sartori, 1984). An example of replacement concept formation can be

found in Pilgrim and Bentall (1999), where concept analysis is described as a “strategy

for replacing biased or misleading concepts with ones that are more useful scientifically

and clinically” (p. 262). However, the purpose here, while still intending to improve the

concept’s scientific utility, is to reconstruct it by clarifying its meaning using multiple

perspectives (Caron and Bowers, 2000). In this way I lend epistemic privilege both to

first-hand participant knowledge and to processed research data on client factors in line

with Hupcey and Penrod (2003).

Sartori’s concept analysis method

Sartori’s concept analysis method (1984; 2009) derives from political science. His

‘ladder of generality’ method used here has been described by Goertz (2006) as a

“classic” (p.69) contribution to the literature on concepts. It fits a critical realist ontology

in that Sartori dismisses any notion that concepts are fixed definitions of essential

phenomena. As objects of language he describes them as “arbitrary conventions”

(2009, p.89). While I argue that language use in an epistemological sense is hardly

arbitrary but results from, and contributes to, complex social structures, the ontology is

appropriate for this study. Sartori devised a process for creating a ‘reconstructed’

information-organising concept hierarchy that values the discriminative use of qualitative

and quantitative data. A hierarchy of abstraction aims to define a concept over several

layers of connotations that are incrementally less ambiguous than their aggregated

‘parent’ terms. Furthermore, Gerring (2012) offers seven criteria with which to evaluate

the adequacy of a reconstructed concept (applied in Chapter 5). The Sartori method is

indicative rather than prescriptive, which is useful here as it offers the researcher some

freedom to adjust the process to the context. I have made alterations and additions in

order to complete this study’s objectives. The following process explains the Sartori

method and how I have modified it:

1. Data collection: To commence, Sartori (1984) calls the researcher to collect a

“representative set of definitions” from “pertinent literature” (p.41). Because

‘representativeness’ has elsewhere been proscribed as too difficult to prove in

78 of 271

qualitative research (Barbour, 2001), I created a purposive literature dataset on

the basis of inclusion/exclusion criteria I devised to locate a diverse range of

client factors in outcome studies. Because ‘client factors’ is a term that

encompasses many other concepts, instead of seeking definitions, I aimed to

locate its component attributes and operations. Additionally Sartori recommends

the researcher strive for “historical depth” (p.41) in the data sample. In this study

this meant reviewing studies back to 1960 although in the included dataset the

oldest article is dated 1983. Study selection is explained in section 3.8 (pp.103-

108). My study is enhanced in line with Hupcey and Penrod (2003)’s advice to

triangulate the concept analysis with participant data, which in this case,

consistent with muted group epistemology, forms the primary structure for the

findings and analysis. The collection of participant data is explained in section 3.5

(pp. 80-87; for an example of a Sartori-inspired concept analysis that follows his

instructions without triangulation, see Knafl and Deatrick, 1990).

2. Data extraction: Sartori then calls the researcher to extract the relevant

characteristics, although he does not explain how this might be done. I achieved

this with the use of spreadsheets which are discussed and illustrated in sections

3.6 (pp. 88-92) and 3.9 (p.109).

3. Data interpretation for concept hierarchy construction: The next step is

creating what Sartori calls a “ladder of abstraction” (p.44) ‘matrix’ diagram that

organises denotative labels for clusters of included phenomena (‘intensions’ or

‘connotations’ of the concept, Sartori, 2009) over levels of abstraction. I replaced

the term ‘matrix’ with ‘framework,’ because ‘matrix’ connotes a total source

context while this study does not aim for comprehensive or conclusive findings.

However I am looking for explanatory mechanisms, associations or tendencies

between client input and therapeutic outcome where possible (Fletcher, 2017;

Sartori, 2009). In terms of framework structure, following the rule that the more

abstract the term, the fewer its connotations, the least abstract terms are

granular context-specific phenomena. These are thematically clustered under

category labels with mid-range abstraction. Further connotative reductions are

made to link mid-range abstractions to the generalised original concept.

79 of 271

Hierarchical conceptualisation strategy yields maximal, mid-range and minimal

definitions of the concept (Gerring, 2012). A further step in the structuring

process is deciding the “conceptual frontiers” (Sartori, 1984, p.43): which

phenomena and denotations are, and are not, part of the concept. In this study,

factors with contradictory or marginal evidence were ruled out (this is explained

further in section 3.11, pp. 115-117). In terms of language use within the concept

framework, the coherence and precision of the “semantic field” (Sartori 1984,

p.51) should be ensured. Jargonistic terms, which Sartori says “add to

obfuscation” (p.54) should be avoided. Neighbouring terms within the hierarchy

should relate well. Synonymies, which Sartori calls “terminological waste” (p.39),

should appear neither laterally nor vertically in the hierarchy. In terms of visually

presenting the findings, in this study I have used an enhanced fishbone graphic

for displaying the interpreted frameworks, and this process is explained in

sections 3.7 (pp. 92-99) and 3.10 (pp.109-115). Encouragingly, Sartori leaves

organising matrices “to the perceptiveness and ingenuity of the analyst” (2009,

p.116).

4. Concept reconstruction: In this study, the meaning of ‘reconstruction’ involves

comparing and combining the participant and research hierarchies into a single

final framework. This process is explained in section 3.11 (pp116-119).

Figure 9 summarises the study design’s parallel process in diagrammatic form over

seven steps. Steps one to three, data collection, extraction and interpretation are

conducted for participant data, then repeated over steps four to six for research data. In

step seven the resulting concept hierarchy frameworks are compared and combined

into one.

80 of 271

Figure 9: Concept analysis method steps flow diagram

Steps one to seven are explained in the following sections 3.5 – 3.11, along with the

labelling of the structural hierarchy at the more abstract levels. The content of the

concept reconstruction is described in detail in Chapter 4.

3.5 Participant recruitment and data collection In explaining how I handled Step 1: Participant data collection (see figure 9, p.80), I

cover the following topics in this section: participant recruitment process; recruitment

objective; participant characteristics; interview design; data collection; power dynamics;

and interview transcription.

Recruitment process

In qualitative research the process of attracting participants is an important contextual

aspect of the study. Participant sites determine to an extent who in included and

therefore the nature of the light shed on the topic (Morrow and Smith, 2000). In this

study I needed to talk to people who had recently finished a course of counselling.

Following Spradley’s (1980) strategies for locating participants, simplicity, permissible

accessibility and unobtrusiveness were key guiding principles. In spring 2016 I obtained

81 of 271

permission to recruit participants from a suburban third sector general counselling

service with a primarily person-centred approach (TS1); and an urban university student

and staff counselling service (US1) with generally short-term CBT and solution-focussed

approaches. An additional suburban site (TS2) is a specialist third sector service I

manage, which predominantly serves clients within an NHS adult community eating

disorder service using integrative, CBT and group-based approaches. This service was

chosen for ease of access and unobtrusiveness. I met with therapy teams from each

service to provide a brief talk about the purpose and method of my research. I asked

them to invite clients who were about to finish counselling to participate. I suggested 4+

sessions of therapy should have been completed, to roughly accommodate advice from

Kadera, Lambert and Andrews (1996) that it takes at least 2 sessions, and for most

clients, 8 or more sessions to achieve significant changes. I also placed recruitment

posters (Appendix 9) in the waiting areas of TS1 and TS2 and the therapy room in US1.

Dataset adequacy and the question of data saturation

I aimed to recruit 12 adults aged 18+ which is the minimum recommended for qualitative

research by Baker and Edwards (2012). However I stopped at nine (a) to manage the

volume of data within project timescale (b) because nine came forward on my first

recruitment exercise; and (c) upon data analysis it was clear that no further interviews

would be needed, because in terms of dataset adequacy I was able to meet the

information requirements of the study with the obtained sample, in that it was large

enough to capture a range of experience but not so large as to be unduly repetitious

(O’Reilly and Parker, 2012).

The question of achieving data saturation with participant interviews as a quality marker

was problematic in this study for two reasons. Firstly, data saturation is described in

grounded theory methodology as a process that assures theoretical categories are fully

accounted for in theory-generating analysis of unstructured participant interviews

(Aldiabat and Le Navenec, 2018). While this may be appropriate for inductive theory-

building research, it is not necessarily a suitable notion in deductive theory-clarifying

research (which is the purpose of this study). The nature and direction of data

collection in deductive research (particularly within a critical realist ontology) means the

82 of 271

researcher cannot capture all possible categories of meaning. This renders the

achievement of saturation “an unrealistic target” (O’Reilly and Parker, 2012, p.194). The

second reason is related to the first. Data saturation is not a recognised aim of

reconstructive concept analysis where instead, data triangulation is recommended (see

p.23 in section 1.4 for Hupcey and Penrod’s 2003 triangulation recommendations).

Fusch and Ness (2015) assert that methodological data triangulation “goes a long way

towards ensuring” data saturation (pp.1411-1412).

In terms of data adequacy, my participant data analysis spreadsheet (figure 11, p.91)

represented what Brod, Tesler and Christiansen (2009) have called a ‘saturation grid,’

albeit modified here for deductive research intended to locate categories rather than

inductively validate categories determined from preceding interviews. The major

categories and elements were listed on the columns and the participants were listed on

the rows. All categories and elements represented in the reported results contained

quotes from at least two participants, in line with Brod, Tesler and Christiansen (2009)

who included data from at least two interviews per category.

Participant characteristics

I hoped for a diverse participant group, but recruitment was non-purposive due to the

exploratory nature of the research design. Because I took no steps to ensure greater

diversity, the actual self-selected group was comprised of white women aged from 18

years into their 30s. With hindsight it was predictable that the services I accessed would

tend to yield participants in this stratum. The composition of this sample was determined

by their having recently accessed counselling and having an interest in the topic of how

clients make therapy work, and/or an interest in participating in research.

A homogenous comparison sample is an advantage in critical realist research where the

investigation probes the mechanism of a generalised concept derived from research to

see if it actually affects clinical outcomes as expected in specific contexts (Miller and

Tsang, 2011). While my research focus is not concerned with specific contexts it could

not evade the contextuality of the settings where I sourced participants.

83 of 271

Participants were predominantly highly educated, childless, working fulltime and in long-

term relationships. Table 1 provides a summary of participant demographics.

Categories were chosen to correspond to those frequently included in the research

literature sample (discussed in section 3.8). I applied culturally appropriate pseudonyms

to avoid the de-humanisation of numerical labels. Although Allen and Wiles (2016)

recommend that participants choose their own pseudonyms at the outset, this issue

only came to my attention at write-up stage.

Table 1: Participant demographic characteristics

Par

tici

pan

t

nu

mb

er

Pse

ud

on

ym

Inte

rvie

w

sch

edu

le

Gen

der

Eth

nic

ity

Age

Edu

cati

on

Emp

loym

ent

Sou

rce

1 Dana A F White

British

31-40 BA Fulltime TS1

2 Ella B F White

British

31-40 BA No TS2

3 Luka B F White non-

European

31-40 Doctorat

e

Fulltime TS2

4 Kyra B F White

British

31-40 Masters Fulltime TS2

5 Lucy A F White

British


6 Zoey A F White

British

31-40 BA Fulltime US1

7 Thea B F Mixed

White

British and

other

White

European

18-25 A-Levels Part

time

TS2

8 Lily B F White

British


9 Esme A F White

British


84 of 271

Interview design

Briefly reviewing the structure of the causal syntax behind the concept analysis, a

change in independent variable X is a mechanism directly or indirectly associated with a

change in outcome Y. In this case X is client factors, which I am seeking to elucidate. Y

is conceptualised here as desired psychosocial life-change, which I am not seeking to

elucidate. A generic outcome is used here for two reasons: (a) participants in this study

had attended psychotherapy for help with a range of different psychosocial issues; (b)

participants could pick the outcome that was meaningful to them, not necessarily the

original reason they sought therapy.

In devising questions for a semi-structured interview to elicit from participants why, from

their point of view, change did and did not happen in therapy, I ruled out some existing

interview schedules. The Client Change Interview (Elliott, 1996) was unsuitable

because it does not explore client perceptions of causality in therapeutic change (e.g.

Klein and Elliott, 2006). I also ruled out the Helpful Aspects of Therapy (HAT, Llewellyn,

1988) and Patients' Perceptions of Corrective Experiences in Individual Therapy

(PPCEIT, Constantino et.al. 2011a) because of their bias in eliciting therapy-validating

information (e.g. Constantino et.al. 2017; Constantino and Angus, 2017). It was

important to the concept reconstruction that participant narratives were not shaped by

therapy-centric questions; and that the starter questions were not leading in terms of

explanatory attributions for change or non-change. After piloting multi-question

interviews with two doctoral student peers, I removed questions relating to whether

participants thought personal demographics had a bearing on the therapy outcome

because (a) they were leading and (b) participant feedback was that they were

irrelevant.

I decided that a semi-structured interview format would provide flexibility for participants

to expand upon topics of relevance to them while being guided by subject-relevance for

the research (Kvale, 2007). This type of interview also permits the researcher freedom

to tailor their responses to the participant (Morrow and Smith, 2000). I chose simply-

framed interview questions that were tightly focussed to the aim of the study (Agee,

2009). These questions assumed that participants and their contexts had exerted

85 of 271

important effects on therapy outcomes, and that participants could identify and describe

what these effects were. The ‘what’ changes and non-changes were not the objects of

enquiry, but needed to be established in order to frame ‘why’ questions aimed to access

participant experience of causal attributions for change and non-change processes

(Swan, 2017).

Schedule A, used with four participants, comprised four open-ended interview

questions:

1. Please describe one positive or desired change that has happened in your life since you

started therapy

2. Why do you think this change happened?

3. Please describe one thing that hasn’t changed in your life as you hoped since you

started therapy

4. Why do you think this change did not happen?

Schedule B, used with five participants, placed the non-change questions (3 and 4) first,

followed by the change questions (1 and 2). This was done to balance the ‘warm-up

effect’ in interviews where participants may provide information more freely in later

stages of the interview (Mellon, 1990).

Data collection

No participant was excluded and none declined to take part once they had read the

information sheet (Appendix 10). Interviews were set up at participant convenience to

occur between one and six months after their last counselling session. This timeframe

was chosen so that there was sufficient interval for participants to reflect on their role in

therapy; while not being so distal that experiential memory might have faded. All

participants were sent the interview questions (Appendix 11) at least a week in advance

so they could think about them before their interview if they wished. All participants gave

consent (Appendix 12) and provided demographic information prior to interviews

commencing (Appendix 22, with results in Table 1, p.83).

86 of 271

Interviews were conducted in summer and autumn of 2016. One interview took place on

Skype and the remainder in-person at TS2 in a confidential consulting room.

Interview lengths were determined by how much the participant had to say on each

question although I more actively wound up two interviews in order to keep to an hour in

line with the participant information sheet which said interviews would be 30-60 minutes.

The shortest interview was 30 minutes (Luka) and the longest was 62 minutes (Esme).

The average interview length was 46.7 minutes.

By way of interview manner my style is warm and relatively informal, although focussed

on the job at hand without engaging in tangential chit-chat or offering information about

myself other than an introduction to my role as student-researcher. Prior to commencing

recording I started interviews with an explanation of the research topic and answered

participant queries in relation to the topic or the information sheet provided. Upon

commencing recording, besides asking the four interview questions my interjections

were restricted to eight types:

Summarising prompts intended to invite participant to discuss a relevant topic

further (for example: So all of these things you are attributing to helping you get

to the point where things are really good now.)

Summarising interpretations seeking correction (Kvale, 2007) (for example, So

there was a sense of escape there…? Dana: Yeah! I think so. Well, not escape,

but, just, new beginnings. Kind of, ‘yes I want to go away from here but I do

actively want to go where I’m going.’)

Questions inviting further discussion of a specific topic (for example: How did you

decide upon each one of those helpful activities?)

Questions seeking deeper meaning or relevance of the topic in the participant’s

life (for example: What does that mean for you making that really big decision,

something that you’d been thinking about for a long time, that you were actually

able to do it?)

Questions seeking clarification of a process (for example, Do you think that’s as

a result of sleeping better, feeling better, having taken the steps that you’ve

taken? Or did that actually contribute to your being able to sleep better?

87 of 271

Open questions testing for further information (for example: Is there anything else

about this process of change that you feel is important to tell me about?)

Directing the conversation to areas of relevance (for example: Well, moving

house was another decision there.)

Brief joining affirmations, emotionally mirroring something the participant said (for

example: Definitely! [Laughing with participant])

Power dynamics

I was aware of mild but not detrimental role power imbalances fluctuating between me

and the participants (Vähäsantanen and Saarinen, 2013). Participants understood from

the information sheet that it was my role to ask questions and theirs to share relevant

personal experience (Kvale, 2007). Implicit in this arrangement was my role to listen,

understand, interpret and learn (although this was not, and probably should have been

spelled out in the information sheet). In terms of researcher power, I devised the

questions, managed the commencement and conclusion of the interviews, the timing

and introduction of the four questions, and I sometimes guided conversations to optimal

relevance for the research topic (Kvale, 2007). In terms of participant power, I was

dependent on their willingness to donate time and to provide analysable information. It

is my observation that the interviews unfolded in a spirit of mutual cooperation, in which

power balanced itself by homeostasis-seeking social grace. I felt it would have been

inappropriate, in terms of my ethical obligation not to burden or exploit participants, to

de-stabilise this power balance, for example by introducing my personal experiences

into the conversations (Karnieli-Miller, Strier and Pessach, 2009).

Interview transcription

I transcribed all interviews verbatim within four weeks of their recording date while they

were still fresh in my memory (Sreenan, Smith and Frost, 2015). Visible body language,

facial expressions and hand gestures were not noted. Audible sighs, laughter and crying

were noted, as were pauses lasting five seconds or longer. Participant anonymity was

protected by the use of codes for diagnoses, services attended and named staff,

geographical places. The names of significant others and types and places of work

were removed. A partial transcript is available in Appendix 13. My decision-making

88 of 271

regarding removing diagnostic terms is presented in Appendix 14, along with how I

handled transcribing local pronunciations, punctuation and speech fillers (Oliver,

Serovich and Mason, 2015).

3.6 Participant data extraction Step 2 (see figure 11, p.91) was carried out over the following stages: Interview

summaries; data extraction spreadsheet; data cleansing; and data immersion. I will now

explain each of these stages.

Interview summaries

While transcribing, I simultaneously kept a notebook of hand-written observations.

These briefly summarised each participant’s answers to the four interview questions. I

also noted some key themes along with initial interpretations which helped me monitor

and limit my “intrusion into the text” (Borland, 1991, p.70). Figure 10 shows my notes on

points Ella makes in her interview, plus observations, for example, that she did not

seem to give herself credit for the changes she’d achieved; and that she said she

sought security. I also jotted down ideas for working up the analysis, for example, that

‘personal factors’ and ‘relational factors’ are interdependent.

89 of 271

Once all interviews were transcribed I typed up summaries which are presented in

Appendix 15. Six participants chose as their desired change the primary reason they

attended therapy, while the remaining three chose other issues. In terms of changes not

achieved, four participants said they had not satisfactorily reached the goal for which

they attended therapy; and four identified other issues that remained unresolved. One

participant said there were no outstanding issues after therapy.

Data extraction spreadsheet

While transcribing I simultaneously built a spreadsheet in which I organised and

categorised all conceptual data related in any way to helpful and unhelpful factors

participants described. Over a lengthy data extraction and analysis process these data

were organised and re-organised into increasing levels of abstraction:

Figure 10: Example notes of preliminary interview summary

90 of 271

(a) Elements are mildly abstract, person-level, labelled groupings of direct activities

and experiences associated with processes of change and non-change.

Elements are built up from, and illustrated by non-abstract participant quotes.

Quotes are the ‘thick descriptions’ with which Gilgun (2014) encourages the

deductive researcher to start the analysis. I used at least one and in many cases,

several quotes from one or more participants to illustrate each element. The

labelling of elements represented the first stage of interpretation. Labels were

revised several times over the course of data extraction to best summarise the

meaning of the element.

(b) Categories are groups of elements that belong best together in classes of action

or inaction, although it was only after some time working with the data that I

discerned that action and inaction were defining pattern characteristics at the

categorical level. Categories are labelled with terms that indicate the particular

‘state of doing’ indicated by their component elements on the helpful factors side;

and the ‘state of being’ indicated by component elements on the unhelpful factors

side. Abstracted away from actual experience, categories form chains of inter-

related factors comprising processes of change on the helpful side, and non-

change on the unhelpful side. The creation of categories represented a further

stage of interpretation.

(c) Super-factors are highly abstract features of the most abstract term ‘client

factors.’ Without a specific definition, super-factors depend on reader epistemic

experience to extrapolate their meaning. These factors were derived from my

interpretation and management of the data, informed non-specifically by

counselling theory, and were imposed on the data at the outset to provide a

preliminary deductive basis for concept categorisation.

Data was used at ‘face value’ as honest explanations of experience. I did not, as other

qualitative researchers might, listen for ‘distortions,’ ‘self-deception,’ or ‘groundless

assertions’ (Elliott, 2010). While my previous knowledge from personal and

professional observations and reading about how people make desired life-changes

informed this interpretive activity, I tried to minimize seeking, interpreting, interposing or

inventing meaning beyond what participants said. In this way I assumed there was no

91 of 271

failure (or veiling) in participant communication of meaning nor in my listening and

understanding. This was done to reduce risk of ‘therapy-centrism’ in my use of

participant data. So my guiding questions as I familiarised myself with the interviews

and developed the data extraction spreadsheet was, ‘What factors enabled this

participant to make desired changes? What factors posed obstacles to change?’

Figure 11 shows how the hierarchy of super-factors, categories and elements were

constructed within descending rows of the spreadsheet, with illustrative participant

quotes within each element’s column.

Figure 11: Participant data extraction spreadsheet sample

Data cleansing

Once notations from all transcripts were added into element columns, I printed out the

labelling rows of the entire spreadsheet (Appendix 16). This was then cut into

manageable sections for close review and data cleansing, which involved collapsing

Super-factors

Categories Elements

Participant quotes

92 of 271

duplicates; re-positioning those that seemed to be under the wrong super-factor;

renaming those that required more accurate labels; and deleting entries which, upon

consideration, were not elements because, for example, they were not precisely

evidenced by participant data.

Figure 12 shows my editorial notations on a section of the spreadsheet print-out.

Figure 12: Hand notations for spreadsheet data cleansing

Data immersion

To further immerse myself in the data I carefully re-read each transcript several times to

locate exact quotes for the conceptual themes and added these to the spreadsheet

(figure 11 on p.91 shows location of quotes). Over 270 quotes were added, although

only 68 were chosen to present in this write-up. All participants gave clear and

descriptive answers to each of the four interview questions. Material from all participants

was well-dispersed across the spreadsheet. There were no participants whose data was

sparsely represented.

93 of 271

3.7 Participant data interpretation Step 3 (see figure 9, p.80) interpretation was carried out over two interrelated tasks:

development of a fishbone framework diagram to illustrate the interrelationship of

helpful and unhelpful client factors over time; and the delineation and expression of four

levels of conceptual abstraction within the framework. While more detailed discussion of

the contents of the framework is presented in Chapter 4, I explain here how I used the

interview data as the basis for the fishbone structure. First I explain the central spine of

the framework, then follow this with explanation of how the ribs were built up.

Fishbone spine: Process of change over time

All participants, in providing detailed accounts of how they made desired life change,

naturally embedded significant events in unfolding time. The ‘time-stamp’ on key

activities and experiences in client narratives permitted me to design the concept

framework using ‘process of change over time’ as the spine. I could then position

change and non-change elements in a general chronological order based on where they

were positioned in participant life-narratives.

Temporal descriptions had various purposes. Most participants provided a timeline to

orient me to the history of their problems as well as their decisions and efforts to

change. For example, Kyra explained that she had recently left a job she had held for

seven years, in which colleagues had accommodated her mental and physical health

problems. She did not want these to carry on in her new job, So I decided at the start of

[the current year] that I probably needed some additional support to try and break the cycle for

good….There’s been probably a lot of things over the past six to twelve months, that have

accumulated in me thinking that I need to, I need to make some real life changes. Zoey

explained how several geographical moves over time were part of her increasing

recognition of the need for help: And then shortly after moving to [English city 2], about six

months, I went to the doctor and asked for some more counselling.

Change processes were described as requiring active work over extended periods of

time during and after psychotherapy. For example, Luka said, I’ve done probably quite a

lot of work since it was a really, really bad relapse in Christmas, so last six months I’ve done

94 of 271

quite a lot. And it just gradually happened. Kyra said, [my therapist] sent me an app to

download, that I’ve used. I’ve been using that for a good few months now [since therapy ended],

to look at, and then you can pull off your sheets of progress to see patterns of how you’ve been

feeling at the same time, and when you’ve had blips. Thea explained, Each time [I’ve had

psychotherapy] I’ve been a bit more able to try and get on board with [therapist suggestions].

Although it did take five-ish years.

Contrasting how slowly positive changes came about for some participants, Lily

described how quickly problems emerged: I just didn’t get…how things could change so

quickly from one month, in January, feeling completely fine and [asymptomatic] on Christmas

Eve, then five months later, [symptomatic of D3] and terrified of anything. Others reported a

chronic pattern of difficulties going back to childhood. For example, Lucy said, I want to

break away from all the rules and restrictions that I’ve imposed on myself for so many

years….I’m confronting years and years and years of behaviours that have become so

engrained. Zoey explained, I went to see…my mum, and I said, ‘I’m not coming round here

anymore.’ There’s been over the years, so many conversations.

Some participants recounted time lost within developmental trajectories due to the

impact of recurrent or debilitating problems. For example, Dana explained how she had

to take time out from her training course and work: I was missing a lot of time off work and I

had major stomach upsets and, that I’ve now recognised was [D2] and [D1]. Ella said, I’d gone

from [less improved to more improved]…. That was last Christmas, not the Christmas just gone,

the Christmas before. Or was it this Christmas? No it was last Christmas. So then I [relapsed]

beginning of the year. Thea explained how she had had to defer starting university a few

years previously, due to mental health problems: I couldn’t do my personal statement so

that’s why I couldn’t apply to Uni the first year.

Another form of reflective time description related to how long participants had recently

enjoyed the fruits of their change efforts. For example Dana said because I’ve been

sleeping well for a good couple of months I can now watch the news and then go to bed. Luka

said, I very rarely, actually, for the last few weeks I think, I’ve not thought that I hate myself.

Some participants remarked how long it had been since they had last functioned as

95 of 271

well. For example, Ella said, I saw people I’ve not seen for twelve months on Saturday night.

We had a big reunion. Thea noted, A few months into the relationship with [girlfriend] I

managed to go for a month without [specific difficulty occurring]. Which was the longest it had

been in six years.

Some participants described reflecting back to the past to see how far their life had

since changed for the better, to motivate themselves to keep going. For example, Ella

said, All I have to do is look back at those times and think, ‘do you really wanna be there? Do

you really want to be the ones that you felt sorry for in [Service A programme]? Do you really

want to feel as low as them, as isolated as them?’ On the other hand, Esme projected into

the future to motivate herself to make changes: So in ten years, do I want to be childless,

friendless, still [unwell], miserable? No.

In summary, participants described taking a series of large and small, complementary

steps over quite lengthy periods of time before they achieved their desired aim; and in

aggregate these steps brought about wider changes in their quality of life and

relationships.

Fishbone ribs

Once the data extraction spreadsheet was complete, using A3 paper I sketched out and

worked up a provisional framework in order to clarify and organise all aspects of the

hierarchy and its component categories and elements. Meaning was created from

participant material at the elemental level. Quotes from at least two different participants

were required to qualify each element. Meaning elements were grouped via a mildly

interpretive process into categories of activity and inactivity. Categories and elements

were reviewed and adjusted in terms of contents and labels until best fit was achieved.

Categories were then further condensed and abstracted to stages of change, which

were then abstracted to super-factors, in a process similar to content analysis (see for

example Erlingsson and Brysiewicz, 2017). Figures 13 and 14 show some of this work

in progress.

96 of 271

Figure 13: Initial sketching for participant-informed concept hierarchy

97 of 271

Figure 14: Intermediate sketch of participant informed hierarchy

Through data extraction a fundamental distinction was made between helpful and

unhelpful factors. Because these largely mirrored each other, a fishbone structure fit for

interpreting and presenting client factors as components of a process that unfolds over

time. The finalised participant-informed client factors hierarchy is shown in figure 15. It

should be noted that this diagram is intended to be viewed in A3/200% size. It is

reproduced here at A4 size to give the reader an indication of the fishbone but it is

acknowledged that detail at the elemental level is lost in this format. Final elements are

explained in detail in Chapter 4.

98 of 271

Figure 15: Participant-informed concept framework

99 of 271

Hierarchy of abstraction

The following description explains how the concept framework was built and labelled.

This is a structural, methodological explanation which is distinguished from the results

presented in Chapter 4, although for illustrative purposes I explain here the labels I used

and why.

The blue-spined participant-informed concept framework shows participant social,

behavioural, cognitive and identity factors organised into holistic chains of change and

non-change occurring over time. The area above the spine of the fishbone shows

helpful factors and the corresponding area below the spine shows unhelpful factors. The

helpful and unhelpful sides are each divided vertically into four levels of conceptual

abstraction from highly abstract super-factors, to moderately abstract stages of

change/non-change, to less abstract categories of activity/inactivity and finally specific

elements which are non-exhaustive connotations of each category. Elements include

illustrative participant quotes and research findings which are the “real world

counterparts” (Sartori 1984, p.22) of conceptual terms.

I will now explain the four levels of abstraction in turn. Each of the concept labels is

shown in bold throughout the remainder of the write-up for highlighting purposes.

Level Four: Super-factors

Super-factors are the most abstract level of intension -or connotation- of the concept, of

which there are two. Super-factor 1, social support is subdivided as (A) a moderator

(inspiration), and (B) a mediator (facilitation and reinforcement) for super-factor 2, which

is reflective learning. These helpful super-factors form a concatenated trajectory that

leads to ongoing inter- and intrapersonal growth cycles across the lifespan.

On the unhelpful side of the framework, super-factor 1: inadequate social support

refers to various types of non-generative relational impairment or injustice in the familial,

community or cultural context. An impoverished social environment contributes to, and

is reinforced by super-factor 2: psychosocial stagnation. Here this term refers to living

in a state of cognitive and social inflexibility that hinders reflective learning and adaptive

growth.

100 of 271

Level Three: Stages of change/non-change

Connotations of the super-factors are visible at the mid-range level of abstraction, which

I have conceptualised as stages of change/non-change. There are four stages A-D on

both the helpful and unhelpful sides of the fishbone. An emphasis on a dynamic ‘state of

doing’ is apparent in stages on the helpful side, as well as gradual social resource

accumulation over the lifespan.

Super-factor 1A (social support moderators) is connoted by Stages A and B. Stage A is

problem recognition prompted and assisted by social contacts. Stage B is a pivotal

process of committing to change initially through recognising the importance of one’s

relationships with others, then deciding to make changes to improve one’s situation for

oneself.

Super-factor 1B (social support mediators) is connoted by Stage C, which is a

particularly active process of engaging with sources of targeted help and social

activity which confer interrelated benefits in reinforcing movement into and through

Stage D. Stage D is the connotation for super-factor 2 (reflective learning) and is

characterised by exploratory psychosocial change and growth.

Figure 16 shows the relationship across helpful super-factors and between super-

factors and stages of change.

Figure 16: Helpful super-factors and stages of change

Facilitative trajectory

Stage A and B as connotations of Super-factor 1A

101 of 271

On the unhelpful side of the framework, a less mobile ‘state of being’ is apparent in

stages of non-change. Unhelpful super-factor 1 (Inadequate social support) is connoted

by Stages A and B. Stage A Non-change is a problem-maintaining social

environment. From this ensues Stage B Non-change which is characterised by no

commitment to change.

Unhelpful super-factor 2 (psychosocial stagnation) is connoted by Stages C and D. In

stage C Non-change, no new generative social connections compounds the effects

of the previous two stages to propel the individual into stage D Non-change, a cycle of

resource depletion, which re-connects with stage A Non-change. Here I define

resource depletion as adverse socio-cultural or emotional circumstances that overtax

one’s health, social support, coping skills, self-efficacy or sense of identity, to the point

that one can no longer reflect, learn, explore or develop one’s potential.

Figure 17 shows the self-maintaining relationship across unhelpful super-factors and

between super-factors and stages of non-change.

Figure 17: Unhelpful super-factors and stages of non-change

Level Two: Categories of activity/inactivity

The connotations of the stages of change are visible in the next less abstract level

which comprises nine categories of activity on the helpful side, and nine obverse

categories of inactivity on the unhelpful side. On the positive side, most categories

are labelled with participles to capture the ‘state of doing’ that characterises participant

work at this level.

Self-maintaining relationship

Stages A and B non-change as connotations of unhelpful super-factor 1

102 of 271

Figure 18 shows categories of activity as connotations of respective stages of change.

Figure 18: Categories of activity

Figure 19 shows categories of inactivity as connotations of respective stages of non-

change.

Figure 19: Categories of inactivity

Level One: Elements

Elements are the densest and least abstract level of the concept hierarchy because

they are built from non-conceptual raw data. However, as interpretations from the data,

they are mild abstractions. They are illustrated by constituent participant quotes and

research references to assist the reader understand their formation. There are 38

helpful and 22 unhelpful elements, 60 in total. Element contents are described in

Chapter 4.

Categories 7 to 9 as connotations of change Stage D

Category 5 as a connotation of change Stage B and C

Categories of inactivity 1 and 2 as connotations of Stage A non-change

Inactivity category 5 as a connotation of stages B and C non-change

103 of 271

Figure 20 shows a narrated screenshot of helpful elements that connote helpful

categories 7, 8 and 9 (which in turn are connotations of change stage D).

Figure 20: Elements of helpful activity categories 7 to 9

Figure 21 shows a narrated screenshot of unhelpful elements that connote unhelpful

categories 7, 8 and 9 (which in turn are connotations of non-change stage D).

Elements connoting their respective helpful categories

104 of 271

Figure 21: Unhelpful elements connoting non-change categories 7 to 9

3.8 Research data collection Step 4 (see figure 9, p.80) research data collection is now presented over two parts:

study selection; and a summary of characteristics of the 68 included studies.

Study selection

Study selection took place over four phases which are shown in the flow chart in figure

22 (adapted from Moher, Liberati, Tetzlaff and Altman, 2009). This chart is included to

assist the reader in following the stages of the literature selection process. Although this

‘PRISMA’ style chart is commonly seen in systematic reviews it is stressed that the

review conducted here was selective in line with concept analysis methodology (Sartori,

1984; Rodgers, 2000b) not systematic. Phases included identification; screening;

application of eligibility criteria; and inclusion. Each phase is next described in further

detail.

Elements connoting their respective non-change categories

105 of 271

Figure 22: Study selection flow chart

Articles located in search of

PsycINFO (n-677), ScienceDirect

(n=410), Mendeley Open Access

(n=2056) & Sage Journals (n=197)

(Total n=3340)

2. S

cree

nin

g 4

. In

clu

ded

3

. Elig

ibili

ty

1.

Iden

tifi

cati

on

Articles located in existing files and

textbook reference lists

(n=299)

Articles screened for relevance

by title (n=3639) Records excluded (n=3564)

Full-text articles assessed for

eligibility against quality control

criteria (n=124)

Full-text articles excluded (n=56)

Measures/ timepoints not specified (n= 10)

Client factor not found, not described or not

a significant predictor (n= 9)

Predictor or moderator not clearly a client

factor (n= 8)

Outcome or predictor judged by therapist

(n=7)

Client factor is an aspect of presenting

problem (n= 6)

Design or analysis does not explain change

mechanisms (n= 5)

Client factor not operationally defined (n= 3)

Involuntary treatment participants (n=2)

Analytic focus on therapist action (n= 2)

Minimum 3 sessions not specified (n= 2)

Therapy not described (n= 1)

Therapy outcome not specified (n= 1)

Studies included in concept

analysis (n=68)

Abstracts screened for relevance

(n= 697)

Articles located in systematic

reviews and meta-analyses

(n=49)

106 of 271

Identification

The search for relevant research about client factors associated with process and

outcome was conducted in January 2018. Three sources were used: databases, select

textbook bibliographies, and my existing collection of articles.

In considering which databases to search, although Penrod and Hupcey (2004) advise

that in concept analysis, multidisciplinary sources should be sampled in order to capture

the concept’s use in different contexts, in this study, due to the discipline-specific nature

of the client factors concept, databases were restricted to those likely to yield articles

about client activity in psychotherapy. These included Sage Journals, Science Direct,

PsycINFO and the Mendeley open access catalogue. The Science Direct search was

limited to 2017-18 for manageability and to yield up-to-date indications of the concept

while providing some balance with the text book searches which yielded older studies.

Search terms were searched for within ‘all fields’ rather than restricted to ‘keywords’

and/or ‘title’ to increase the scope and credibility of the search (Rodgers, 2000b). Terms

included ‘client factors,’ client characteristics,’ ‘extratherapeutic factors,’ ‘client agency’

(altered in PsycINFO to ‘empowerment’), ‘client variables,’ ‘moderators’ ‘mediators,’ and

‘predictors of therapeutic change.’ In PsycINFO these terms were exploded to additional

relevant terms and then associated with exploded terms for ‘counselling.’ Appendix 1

shows screen shots of the searches conducted.

Textbook reference lists were searched including Garfield 1978, 1994; Clarkin and

Levy, 2004; and Bohart and Wade 2013. These were chosen to ensure the inclusion of

studies that preeminent researchers relied on for the concept’s development over time

(a precedent for this type of hand search can be found in Bernd and Strouse, 2008).

Relevant articles from my existing library of research articles (collected in preparatory

work from 2015-2018) were included.

Screening

3340 study titles were reviewed and 2643 were excluded as irrelevant. 697 were

retained for abstract review against primary inclusion criteria (Appendix 2), and a further

608 studies were excluded. 89 studies remained.

107 of 271

Among the 89 remaining studies were 14 meta-analyses/systematic reviews. Because

the final sample could include only primary research, 49 relevant primary studies were

retrieved from the bibliographies of these studies, which were themselves removed

(detail of studies included from removed meta-analyses is shown in Appendix 3). 124

studies remained for full text review.

Eligibility

Under full-text review the 124 remaining studies were subjected to quality control. The

baseline quality threshold was the study’s publication in a peer-reviewed journal. Then,

because I required data relating as exclusively as possible to client or extratherapeutic

factors (rather than therapist, technique or diagnostic factors); and because I was

seeking either statistical inference of mediation, moderation or prediction (rather than

simple correlation); or client asserted association between client factor and outcome,

further exclusion criteria were applied (detailed in Appendix 5. Results of exclusion are

in Figure 22 ‘full text articles excluded’ text box).

Inclusion

After the application of these exclusion criteria, 68 studies remained for full data

extraction and concept analysis. A distinction was made between those where outcome

was analysed from follow-up data gathered at least 1 month after therapy finished

(41.2%, n=28/68), and those where outcome was measured at the end of therapy

(58.8%, n=40/68). Client factors reported from follow-up data might be less likely to be

temporary artefacts of the therapeutic process. However due to the relatively small

number of studies in each subgroup, I decided not to create concept analyses

comparing any differences, but treated the 68 studies as a complete dataset.

Dataset adequacy

Study sample dataset adequacy in concept analysis is indicated by its capacity to

support the study. The appropriateness of the sample is indicated by its capacity to offer

a sketch of how the concept is operationalised (Penrod and Hupcey, 2005). Both

requirements were attained with this sample. Penrod and Hupcey (2005) note ‘modest’

single-researcher concept analysis studies have included 83 and 107 studies, whereas

108 of 271

Rodgers (2000b) notes that an effective concept analysis may include anywhere from

30 to over 4,000 articles in studies that do not triangulate with qualitative interviews.

The data from 68 included studies was triangulated in this study with nine interviews as

discussed in sections 3.5–3.7 above.

Characteristics of 68 included studies

The following characteristics of included studies are summarised here: nationality,

journals, participant demographics, presenting issues and type of therapy, study design

and type of analysis.

Nationality and journals

15 countries were represented in the complete set of 68 studies although more than half

(55.9%, n=38) were from USA. They appeared in a total of 35 journals with the most

appearing in the Journal of Consulting and Clinical Psychology (42.9%, n=15).

Publication dates spanned 1983 – 2018.

Participant demographics

The number of participants in each study ranged from 2 to 1383. Gender was specified

in 86.8% of studies (n=59) from 0% - 100% female, with 92.6% of studies (n=63)

reporting cohorts > 50% female. Participant ethnic background was reported by 42.6%

of studies (n=29), ranging from 19.5% to 100% white/causasian. Participant mean age

was reported in 91% of studies (n=62) and ranged from 20.5 years to 69.2 years.

Participant cohabiting/marital status was specified in 54.4% (n=37) studies, ranging

from 0.8% to 100% cohabiting/married, with 37.8% (n=14) studies reporting more than

half the participants were cohabiting/married. Education levels were specified in 42.6%

(n=29) of studies either as mean years (ranging from 11.3 to 16.8 years) or percentage

who had some university education or more (ranging from 14.8% to 85.3%).

Employment status was specified in 23.2% (n=16) studies, ranging from 33.3% to 100%

of participants in part-time or full-time employment.

109 of 271

Presenting issues and types of therapy

14 specific problem areas were covered as well as cohorts with mixed diagnoses. The

largest problem segment was depression (30.9%, n=21). In terms of types of therapy,

22 studies compared therapies using multiple orientations and the remaining 44 studies

examined single therapies, with variants of cognitive and behavioural therapy

predominating (77.2%, n=32). Therapy modality was stated in 53 (77.9%) of studies,

including individual (58.5%, n=31), multimodal (22.6%, n=12), group (n=9) and couples

(n=1). Treatment setting was mentioned in 48 (70.6%) of studies, the largest segment

being university hospitals and research training clinics (56.3%, n=27). Participant

cohorts in more than half the studies (n=38) were treated on an outpatient basis. 60%

(n=41) of studies mentioned therapists’ professional disciplines including

multidisciplinary teams (26.8%, n=11); qualified and trainee psychologists (24.4%,

n=10), qualified and trainee psychological therapists (19.5%, n-8), and one study each

for CBT therapist, psychiatrist and generic mental health staff.

Study design and analysis

The majority of studies (85.3%, n=58) were quantitative designs, with the remainder

mixed methods (7.35%, n=5) and qualitative designs (7.35%, n=5). The qualitative

studies used grounded theory (n=3), consensual (n=1) and thematic phenomenological

(n=1) analyses. The majority of the remaining studies (76.2%, n=48) used variants of

regression analysis including hierarchical linear, logistic, univariate, multiple, stepwise

and Cox proportional hazards. “Hierarchical linear modelling allows coefficients to vary

contingent on moderating and mediating effects at other levels of analysis and as such

is [a] tool suitable for critical realist interest in testing the effects of intervening

mechanisms on theoretical relations” (Miller and Tsang, 2011, p.151). In terms of types

of causal relationships between client factor and outcomes, the majority (72%, n=49)

were predictive. Associations, including those relying on participant reports, were

identified in 14 (20.6%). The remaining five studies analysed mediators (n=4) and

moderators (n=1) of outcome.

110 of 271

3.9 Research data extraction For step 5 (see figure 9, p.80) I created a colour-coded data extraction spreadsheet in

which I entered and organised relevant data from the 124 studies identified at eligibility

stage. Appendix 4 shows a screenshot of a portion of this spreadsheet and lists all

column headings. Identified client factors were listed according to whether the research

found they were predictors, mediators, moderators or in other ways associated with

outcome, although this distinction ultimately proved to be of limited utility because the

majority (72%, n=49) were prediction studies.

3.10 Research data interpretation Research data interpretation for step 6 (see figure 11) took place over the following

phases: manual interpretation of findings tags; labelling the tags; entering the tags into

the analysis spreadsheet; creating the research-informed concept hierarchy framework;

and comparison of this with the participant-informed framework. I now explain these five

phases.

Manual interpretation of tags

In order to work with the extracted client factor findings from the 68 studies included in

the research sample, I printed these out from the data extraction spreadsheet, cut each

into a tag, and arranged these over the large sketch which had formed the basis for the

participant-informed concept framework (see figures 15, 16 and 17). In this way I

privileged the participant framework as the organising principle for interpretation. Figure

23 shows this work in progress.

111 of 271

Figure 23: Manual interpretation of study data tags

112 of 271

Labelling the tags

Once all 68 research tags were organised into elements, I labelled them with sticky

notes. Elements contained from one to seven studies. Figure 24 shows this job when it

was finished.

Entering the tags onto data analysis spreadsheet

I created a research data analysis spreadsheet labelled identically to the participant

data extraction sheet, and entered the tags and associated references. A screenshot of

a portion of the data analysis spreadsheet is shown in figure 25.

Figure 24: Labelled research data tags

113 of 271

Figure 25: Portion of data analysis spreadsheet

114 of 271

Creating the research-informed concept hierarchy framework

The red-spined research-informed framework, shown in figure 26, is structured over the

same hierarchy of abstraction as the participant-informed framework. It should be noted

that this diagram is intended to be viewed in A3/200% size. It is reproduced here at A4

size to give the reader an indication of the fishbone but it is acknowledged that detail at

the elemental level is lost in this format. Final elements are explained in detail in

Chapter 4.

115 of 271

Figure 26: Research-informed concept framework

116 of 271

Comparison of participant and research frameworks

Here I revisit the advice given by Hupcey and Penrod (2003; noted in Chapter 1), that

the process of reconstruction involves comparing the participant- and research-informed

frameworks to observe differences. The reader can do this by comparing figure 15

(p.98) to figure 26 (p.115).

In the research-informed framework, the super-factors remain the same as the

participant-informed framework on both the helpful and unhelpful sides of the fishbone.

There is generally good concordance between the participant and research-based

frameworks on the other three levels of the hierarchy as well. However, the following

differences appear at abstraction Level Three (stages of change).

Stage A (problem recognition) lacks connotative Category 2Positive: Hitting rock

bottom. Stage B (commitment to change) is less dynamic, as it lacks connotative

Category 3Positive: Precluding relational loss. Impoverishment of connotation in the

early stages of change is because most included studies investigated what happened

in, and in some cases, after therapy, not what prompted clients to access therapy.

The connotations for Stage C (new social connections) and Stage D (exploratory

learning) replicate their counterparts in the participant-informed framework.

On the unhelpful side of the fishbone, Category 4Negative: Unready to change is

missing from stage B Non-change (no commitment to change), because people

uncommitted to change generally do not start or complete therapy, and my research

sample excluded drop-out studies.

3.11 Combining participant and research frameworks The seventh and final stage of the method (see figure 9, p.80) involves removing

marginal denotations that do not fit an operational definition of the concept (Sartori

1984), then combining the frameworks into one.

117 of 271

Removed denotations

I removed four client factors comprised of contradictory evidence, all of which appeared

in the research-informed framework (figure 26). These include education and age;

therapeutic alliance and experiencing emotions.

Education and age

Two demographic factors, education and age appeared in the research-informed

framework (categories 10 and 11 in figure 26 on p.115) but not in the participant-

informed framework. However, the findings in both these categories were contradictory.

For example, more education was associated with positive outcome in one study

(Kyrios, Hordern and Fassnacht, 2015) and negative outcome in another (McCrady,

Hayaki, Epstein and Hirsch, 2002). Likewise younger age was found to predict better

outcome in three studies (McCrady, Hayaki, Epstein and Hirsch, 2002; Fournier, et.al.

2009; and Reddy and Jagannathan, 2017) potentially because of the advantages of

early intervention. On the other hand Hendriks et.al. (2012) and Chambless et.al.

(2017) found that later age of illness onset also predicted better outcomes, potentially

because earlier-onset illness is more likely to become part of a person’s identity. Given

the contradictory and treatment-dependent findings in the education and age

categories, they were cancelled out of the framework.

Therapeutic alliance

In category 5, counselling appears at element 5F in the participant-informed framework

(figure 15 on p.98) and 5B in the research-informed framework (figure 26 on p.115). As

an aspect of counselling, therapeutic alliance was not included in either framework due

to contradictory evidence in the research data and insufficient evidence in the

participant data. Evidence in the research data for the predictive role of client

contributions include studies by Hartmann, et.al. (2010) and Goldman, Greenberg and

Pos (2007) who found therapeutic alliance was of little significance to outcome.

Mallinckrodt (1996) found it was associated with outcome in so far as it promoted

change in client’s social support outside therapy. Findings by Shahar, et.al. (2004),

Hawley et.al., (2006), Liebert, Smith and Agaskar (2011) and Hartzler et.al. (2011)

118 of 271

suggested that participant contribution to the therapeutic alliance replicated participant

relational engagement or difficulties in other areas of life and that these social

influences in turn had a mediating impact on outcome. Castonguay et.al. (1996) and

Zuroff et.al. (2007) reported a simple positive association between working alliance and

outcome.

In terms of participant data on therapeutic alliance, although all participants mentioned

the helpful and unhelpful influence of things therapists said, did and did not do, as well

as their learning from regular counselling contact, only one mentioned their relationship

with their therapist having an impact on the changes they made, which did not meet the

threshold of quotes from two participants for element inclusion.

Element 8PositiveG: Experiencing emotions in the research-informed framework

(figure 26, p.115) was cancelled out due to inconsistent evidence. Goldman, Greenberg

and Pos (2007), and Castonguay, Goldfried, Wiser, Raue, and Hayes (1996) found that

experiencing emotions in therapy predicted positive outcome. Somewhat contradictory

findings were presented by Dingemans, Spinhoven, and van Furth (2007), who found

lower expression of emotion after treatment predicted better outcome at 1-year follow-

up. Participant data lent further support to my decision to remove this category. No

participants credited being emotional in therapy for helping them achieve change. On

the contrary, one said she preferred online therapy because it helped contain

emotionality.

Combining participant-informed and research-informed frameworks into one

The reconstructed framework (green spine) appears in figure 27. It should be noted that

this diagram is intended to be viewed in A3/200% size. It is reproduced here at A4 size

to give the reader an indication of the fishbone but it is acknowledged that detail at the

elemental level is lost in this format. Final elements are explained in detail in Chapter 4.

119 of 271

Figure 27: Reconstructed concept framework

120 of 271

The reconstructed concept framework repeats the structure presented in Section 3.7

(see in particular pp. 99-104) for the participant-informed concept hierarchy. It relies on

colour-coding at the elemental and categorical levels of the concept hierarchy to

differentiate their evidence bases. Green categories and elements are those that

contain both participant and research data. Blue categories and elements are those that

contain only participant data. Pink elements are those that contain only research data.

Figure 28 provides a screenshot from a portion of the (non-finalised) unhelpful side of

the framework to show examples of colour-coding with explanations.

Figure 28: Example of color-coding on reconstructed concept framework

Blue categories and

elements =

participant-only data

Pink elements = research-only data

Green categories and

elements = participant

and research data

121 of 271

3.12 Rigour Rigour in this study is defined as the quality control measures I took to ensure credibility

and trustworthiness. These include providing sufficient raw data to evidence my

interpretations; data triangulation; researcher reflexivity; an audit trail; and participant

checks.

Raw data to back interpretations

In Chapter 4, I present the findings that comprise the categorical and elemental levels of

the helpful and unhelpful factors sides of the reconstructed concept framework. All

elements are explained with data examples: either participant quotes or key research

findings and references, or both in many cases. Faithful inclusion of original material

provides an indication of the interpretive process they have been subjected to (White,

Woodfield and Ritchie, 2003).

Data triangulation

While triangulation in qualitative research has various definitions, in this study it means

the methodological choice to use two data sources for contrast and complementariness

(Jensen, 2008; see also section 1.4 p.78 above).

Researcher reflexivity

In this study I am careful to differentiate researcher reflectivity from reflexivity.

Reflectivity is here defined as the introspective activity of reviewing important episodes

of personal or professional activity in order to make sense of how things happened and

why certain choices were made (Dallos and Stedman, 2009). While my mind has

certainly been preoccupied with this project for much of the past two years, I did not, as

some counselling psychology researchers do, keep a contemporaneous reflective

journal as a qualitative research tool (e.g. Demarco and Willig, 2011) due to uncertainty

of its worth. I did however create hundreds of pages of ephemeral reflective notes in the

writing-up process (example in Appendix 17). Overall, for me as I assume for others,

academic writing involves innumerable reflective decisions about discursive expansion

and editorial contraction.

122 of 271

Reflectivity is a thinking activity that both informs and critically reviews the action of

reflexivity, which I discuss next.

Reflexivity in a critical realist paradigm involves critically attending to what our data,

theories and concepts are claiming about the world, focusing particularly on explanatory

mechanisms in social processes (Archer, et.al. 2016), as I am in this study. It also

involves making conscious choices to either follow existing cultural practices, or else

make efforts to change them (Archer, 2007). In this regard reflexivity is demonstrated in

this study where, for example, I have transparently justified methodological choices to

either align with, or divert from, the work of previous researchers, particularly other

concept analysts. I have also assessed and explained the reflexive influence of my

assumptions and experiences on this study. In its totality this study is a large reflexive

act built on thousands of undocumented reflective actions. Reflexive practice demands

honesty, courage and wisdom in order to make “changes, alterations and adjustments”

(Pope, 1999, p.180) and to divert from taken-for-granted practices, which I have done

by pursuing concept analysis in the first place.

Audit trail

Confirmability, as an aspect of credibility in qualitative research is demonstrated by an

audit trail (Anney, 2014). This type of evidence makes the “building blocks used by

researchers in arriving at their interpretations clearly visible to the reader” (Snape and

Spencer, 2003, p.21). I have taken this instruction literally by providing in the body and

appendices of this study, a number of screenshots and photos of work in progress,

including the literature selection process (Appendices 1-5, 24). Additionally, regular,

formalized debriefing sessions with fellow doctoral students helped me to describe and

justify the developmental stages in completing this study, and deepen my reflexive

analysis of the work (Anney, 2014; See Appendix 18 for anonymised minutes of a peer

meeting).

123 of 271

Participant checks

To promote data dependability as an aspect of study credibility (Anney, 2014), I asked

each participant to review their interview transcript for accuracy. This process is

elsewhere referred to as a “member check” (e.g. Guba and Lincoln, 1982, p.247; Elliott,

Rennie and Fischer, p.218). However, the term ‘member’ did not fit the context of this

study, as participants did not identify as belonging to a common group (Billig, 1999).

I attached their transcript in an email to the participant with thanks for their participation

and a request for transcript review and correction. All participants replied. Three added

comments for clarity which were adopted into the transcripts.

In line with the muted group epistemology I requested participants do a follow-up check

of my use of their quoted material in the results chapter. Five replied, two were happy

with the text and three made minor clarifications that were incorporated.

Catalytic validity

Stiles (1993) suggests that research with catalytic validity will re-orient, focus or

energise the participants. Regarding transcript checks, Kyra said, “It has been really

interesting to reflect back on the interview with you.” Lily said, “That was such an

interesting [read] and to see how things have changed (positively!) over the past few

months.”

Regarding the interpretation checks, Zoey said, “It’s been quite a revelation reading this

to confirm how much progress I have made since I did this interview with you.” Thea

said, “I think reading this was very good for me, I will probably read it again.”

3.13 Ethics This section explains how I complied with relevant British Psychological Society (2014)

and University of Manchester (2014) codes of ethics for the conduct of research with

human subjects. It includes my consideration of ethical issues related to the death of a

participant during the completion of this study.

124 of 271

University approval

This study was approved as medium risk by the University Research Ethics Committee

(UREC) in November 2015 with minor amendments approved in March 2016 (Appendix

19). No adults classified as vulnerable by UREC were recruited.

Informed consent

Prior to interviews, all participants were given an information sheet (Appendix 10) which

presented information about the topic and purpose of the research, what they were

being asked to do, method of data collection, risks of participating, confidentiality and

right to withdraw. Consent was obtained from all participants (Appendix 12).

Confidentiality

Participant information and consent sheets were stored in a secure filing cabinet in my

home. Interviews were audio-recorded to MP3 and stored in my secure dropbox.

Transcription adjustments were made to protect participant anonymity. No person-

identifiable information was used in the write-up of this study.

Risk to participants

Participants were informed there was a small chance of feeling upset when discussing

the interview questions. Most interviews unfolded relatively unemotionally but Kyra cried

when thinking about the potential loss of relationships she faced if she did not address

her problems. Similarly Esme cried as she expressed regret about the impact of her

illness on her life and relationships. Zoey cried when recounting family indifference to

her poor mental state, and also when describing the impact of an adverse event on her

life. I provided empathic responses. These participants self-stabilised and continued

with the interview without further distress.

Risk to participants of becoming upset when reading the transcript came to light after I

sent the first transcript to Dana for review. She was bothered by the number of speech

fillers such as ‘um’ and ‘you know’ appearing in the verbatim script. I explained I would

remove these. With all subsequent transcripts I removed these before sending to

participants in line with the editing notes provided in Appendix 14.

125 of 271

Mitigating researcher dual role

In consideration of my dual role as researcher and manager at TS2 I prepared an action

plan should a TS2 participant express a complaint about their therapy (Appendix 20; in

the event no complaints arose). Although the focus of the research was not clients’

therapy experiences, I also considered the risk with TS2 participants knowing I was the

service manager, that they might avoid saying if they found TS2 therapy or staff

unhelpful in any regard which might skew the results. However I deemed this risk low

because I was not asking about therapy experience. In the event 5/7 TS2 participants

mentioned helpful therapy experiences and 1/7 mentioned a non-adverse experience of

under-engagement with therapy.

Participant death

Thirteen months after her interview, while I was still working on study write-up, I was

informed by a family member that a TS2 participant had suddenly and unexpectedly

died. I had two ethical concerns related to this tragic event. The first was whether I

ought to mention her death in the study. The second related to appropriate use of her

data, because at that time participants had not yet been asked to review how I used

their quotes in the analysis.

I discussed the matter with my research supervisor who did not feel university

involvement was required in this instance. Neither the British Psychological Society

Code of Human Research Ethics (2014) or the Health and Care Professions Council

Standards of Conduct, Performance and Ethics (2016) mention participant death.

Qualitative research around participant death is sparse, and what I found did not speak

to my concerns. For example Burles (2017) discusses the impact of predictable

participant death on the welfare of the psycho-oncology researcher, but she does not

deal with ethical use of data from participants who die before the completion of a study.

I considered the following matters in favour of not mentioning her death in the study:

News that she died did not come to me in my role as researcher, but in my role

as director of the service where she had received care in previous years.

126 of 271

Although I did not know the cause of death it clearly was in no way related to

her participation in this study.

Her death neither invalidated her data nor impacted the findings and analysis.

On the other hand, these facts were over-ridden by my belief that it would be

disrespectful to not mention her death in the study. In critical realist terms, death is real.

In personal and professional terms, it was terribly sad, and especially so for TS2

colleagues and clients who knew her well. Furthermore it was necessary to address in

some way the difficulty that I could not ask her to review the final draft. For this reason I

considered removing her data from the study, but as the analysis was at that point

almost finished, it would have been difficult to un-weave her valuable contributions. A

family member told me in a conversation following her death that it would have been

very important to her that her contribution be included.

On balance, her original consent to participate in my mind over-rode her not being able

to review the write-up. To further protect her anonymity I have deliberately not disclosed

which participant she was.

127 of 271

CHAPTER 4: RESULTS

4.1 Chapter overview The aim of this chapter is to explain in detail the content of the categories and elements

in reconstructed concept framework. The chapter is organised over two sections: helpful

client factors and unhelpful client factors. Each section is subdivided at the hierarchy

level ‘stage of change’ of which there are four in each section. Within each stage I

explain the meaning of the categories that comprise each stage. I then explain the

elements that comprise each category, including whether they are evidenced by

participant-only, research-only or participant and research data. I provide illustrative

examples from participant quotes and research references for the 38 elements that

comprise the helpful factors, and 22 elements that comprise the unhelpful factors. Each

section concludes with a summary of key findings.

The results are presented with coded references to diagnoses to protect participant

confidentiality. In line with the muted group epistemology, the results are presented with

close reference to participant experience across all relevant elements, which makes

their exposition lengthy.

4.2 Helpful client factors In this section I provide detailed descriptions the elements for each category of action

along the helpful factors side of the reconstructed framework, which for reader

convenience is shown in figure 29, with the caveat previously noted that this should be

viewed at A3/200% for clarity).

128 of 271

Figure 29: Helpful factors in the reconstructed concept framework

129 of 271

The following section is broken down by stages of change, of which there are four:

Stage A: Problem recognition; Stage B: Self-commitment; Stage C: New social

connections; and Stage D: Exploratory learning.

Stage A: Problem recognition Stage A: Problem recognition is connoted by two categories: Trusting family and

friends; and hitting rock bottom. I explain the meaning of each along with its connotative

elements.

Category 1Positive: Trusting family and friends which means allowing and engaging

in conversations with supportive family and friends around problem recognition. It is

connoted by four elements: partnership quality, speaking to others about the problem;

others speak to me about the problem; and spirituality.

1PositiveA: Partnership quality (participant and research data)

Five participants spoke about the importance of a loving long-term partner or friend in

helping them overcome their problem. Dana said, my partner massively supported me,

forever. Similarly, Zoey credited my best friend who has been an absolute rock to me

throughout everything. The idea of her partner being a ‘rock’ was echoed by Ella: I’m

lucky, my fiancé is an absolute rock. He’s an absolute star.

The association of clients’ marital quality with positive outcomes was supported by client

reports in five studies, including Fournier et.al. (2009); Jarrett, Eaves, Grannemann, and

Rush, A.J. (1991); and Hooley and Teasdale (1989). Studies by Renneberg,

Chambless, Fydich and Goldstein, (2002) and McCrady, et.al. (2002) also made this

finding, the latter at 6 months follow-up.

1PositiveB: Speaking to family/friends (participant-only data)

Opening up to trusted people was identified as an early step in problem recognition by

all nine participants. Luka said, My sister was over and she noticed that I was really ill. And I

did ask her, ‘do you think I’m ill?’ because I don’t think [I had] the realisation that I was actually

really ill. And she said, ‘yeah, I think you are.’ And I did ask her, ‘Do you think I need to do

130 of 271

something about it?’ And when I decided…I really wanted to do something, then I think the

changes slowly started to happen.

1PositiveC: Family/friends raised the problem (participant-only data)

Six participants spoke about the helpful impact of trusted people being prepared to raise

and discuss the problem before they had fully acknowledged it to themselves. Lily said,

The first time [boyfriend] ever met me, he said, ‘that’s not [healthy behaviour].’ And then from

then I felt like, he gets it. He encouraged me then to go to the doctor. Esme said, [My friends]

learned they had to go, ‘You know what? You’re being ridiculous!’ And actually confront me

with it. A lot of the time I just needed a good shake.

Category 2Positive: Hitting rock bottom means experiencing a crisis as a result of

the problem where ensuing contacts with healthcare professionals spur resolve to

address the problem. It is connoted by two elements: accessing crisis services; and GP

services.

2PositiveA: Crisis services (participant-only data)

Four participants identified health or psychiatric crises that propelled them to attend

emergency services for help. These experiences were identified as turning points in

problem recognition. Lucy said, I was actually taken from work into hospital. And I think that

was the wake-up call.

2PositiveB: GP services (participant-only data)

Seven participants explained how conversations with GPs represented turning points

either in problem recognition or in accessing help. Kyra said, So I was referred through to

[service A] by my GP because for quite a long time I’d had probably underlying [D3] and [D2]

issues. Zoey recalled, [My GP] said, ‘just try [counselling] once more and see how you get on.

And if it doesn’t work again, I’ll get you to somebody else.’ And she, the doctor, was amazing.

Stage B: From other-focussed to self-focussed commitment to change The active Stage B: From other-focussed to self-focussed commitment to change

is connoted by three ‘state of doing’ categories of action: Precluding relational loss; self-

131 of 271

reclamation; and persistently accessing help. Each is now explained along with its

connotative elements.

Category 3Positive: Precluding relational loss means wishing to preserve and

protect relationships from damage the problem has caused (or could potentially cause).

It is connoted by two elements: current relationships and future relationships.

3PositiveA: Preserving current relationships (participant-only data)

The risk of one’s unaddressed problems damaging important relationships was

highlighted as a motivating factor by six participants. Dana explained how her partner

warned her: She was like, ‘just stop it. It’s really, making me want to leave.’ I thought, ‘Ohh hoo

– oh okay!’ Ella described her partner and father’s reactions when they visited her in

hospital during a crisis: And that gave me another kick just to say, ‘Look, you’ve got to get

right.’ I can’t see him broken again. Really wrecked me, that. And my dad, his face just coming

in.

3PositiveB: Protecting future relationship opportunity (participant-only data)

In this participant sample, the wish to preclude loss also applied to future relationships.

The importance of not losing the chance of parenthood was mentioned by four

participants as motivating change. Kyra said, I’m at an age where we might think about a

family. And I worry about the impact of all the years of self-abuse on my chances of being able

to have children. And I thought that if we even want to consider that angle, then I need to do

something to change.

Category 4Positive: Self-reclamation is a pivotal transition point of committing

responsibility to salvage oneself from the problem. It is connoted by three elements:

autonomous responsibility; intention; and maturity.

4PositiveA: Autonomous responsibility (participant and research data)

Stage B pivots from making changes for others to making changes for oneself.

Autonomous responsibility requires a degree of self-worth that translates the value of

preserving important relationships into self-mobilisation. Three clients expressed

reaching a point of readiness to change for one’s own sake. Ella, who was initially

132 of 271

motived by a desire not to hurt loved ones, said, I wouldn’t have come out of it until I

decided for myself.

Two studies were represented in this element. In a qualitative study based on client

report at end of treatment and 12-month follow-up, Khattra et.al. (2017) found that a

cognitive shift from other-focussed to self-focussed motivation was associated by

participants with leading to meaningful adjustments in coping. A similar finding was

made by Zuroff et.al. (2007) based on end-of-treatment data.

4PositiveB: Intention (participant and research data)

This element encompasses self-determining motivation and commitment to change,

expectancy that help will be effective, and willingness to do what is necessary to create

a better life. Six participants described aspects of this experience. Luka said, What made

it possible? I think I was just really committed to get better, because I was really in a very bad

place. Lily said, I was desperate, willing to do anything.

Six studies are represented in this element. Hartmann, Orlinsky, Weber, Sandholz and

Zeeck (2010) found that client positive expectancy early in therapy predicted symptom

remission at 3-month follow-up. Similar expectancy findings based on end of therapy

data are presented by Chambless, et.al. (2017) and Safren, Heimberg, and Juster

(1997). Meyer at.al. (2002) found that the benefits of the therapeutic alliance on

outcome were mediated by client expectancy.

Motivation was conceptualised by Keijsers, Hoogduin and Schaap (1994) as willingness

to participate in therapy, and was associated with symptom reduction in therapy. Burns

and Nolen-Hoeksema (1991) found that client willingness to learn new coping strategies

was correlated with being less unwell at the start of therapy and with being more

improved at 12 weeks. This suggests that people who are more prepared to change

may seek help sooner.

4PositiveC: Maturity (participant-only data)

Seven participants asserted that an increasing sense of maturity in the form of their

advancing age assisted their motivation to change. Kyra described it like this: I’ve

133 of 271

decided that I can’t go on in that cycle, I’m in my mid-thirties now, can’t continue to keep going

round the same old vicious circle. And maybe (laughing), with age comes wisdom! Thea

described how coping with mental illness also promoted maturity: I think the [D1] itself has

made me wiser. So the [D1] in a weird way helped me get over it because I would not have got

as mature without it.

Category 5Positive: Persistently accessing help is characterised by resourcefully

tapping different types of social help, over time, with repeated attempts where

approaches may initially fail or fall short. It is connoted by six elements, two of which

contain participant and research data and four, participant-only data.

5PositiveA: Help from family/friends (participant and research data)

Four participants described receiving valuable emotional and practical help from family

and friends after they had commenced a period of active change. For example, Ella

said, My dad’s an angel. He worries as much as I do….He will say, ‘You alright, love? You sound

a bit down.’ I’m like, ‘Dad, you alright?’ ‘Oh,’ he says, ‘I’m having a shit day!’ (laughing). Dana’s

mother lent money to help her move, an important step in improving her situation.

Six studies support the element of help given by family and friends. Steinmetz,

Lewinsohn and Antonuccio (1983) and Toth, et.al. (2013) found that client experience of

family social support was predictive of better outcomes at 6-8 months post-therapy.

Mallinckrodt (1996) made a similar finding without follow-up data. Drilling into specific

types of helpful support, Davidson, Borg, Marin, Topor and Sells (2005) found

participants’ significant others’ focus on their capabilities, interests and successes was

associated with positive outcome, as was their help in making important decisions.

Leibert, Smith, and Agaskar (2011) found that participants with higher subjective social

support had less severe symptoms when starting therapy. The authors suggested this

means that socially supported clients may seek help earlier. Mallinckrodt (1996) found

that the therapeutic alliance mediated participant social support by promoting it, but that

the association between change in alliance and symptoms was not significant. This

suggests that therapy outcome may be related to how much it helps people improve

their social support, rather than directly linked to therapeutic alliance.

134 of 271

5PositiveB: Help from staff at school/work

5PositiveC: Self-help and social media including apps

5PositiveD: Support groups

5PositiveE: Medication

Use of each of these elements was endorsed by several participants, and all

participants used at least one of them. Three participants received supportive help from

staff at school or work. For example, Thea said: The school nurse supported me for the five

years. I could always go to her for a hot chocolate or just sit in her room. Seven participants

used a range of self-help media, for example Zoey wrote blogs and Dana read blogs for

advice; Luka and Kyra used apps; Lucy and Thea used social media; and Zoey, Esme

and Thea used books or other reading material. Five participants used support groups.

Dana, Luka and Thea mentioned psychotropic medication as useful.

No studies are included in these helpful elements, although this does not indicate a lack

of evidence for their role in supporting psychotherapy outcomes. The absence of

studies in these areas is related to my study selection process. Self-help and support

groups were excluded from the research sample, which only included professionally-

facilitated treatments (further study in this area should include self-help activity).

Medication outcomes were not included in data extraction, because in those studies

where it was used (Hendriks, et.al. 2012; Fournier, et.al. 2009) medication was

positioned as a comparison treatment factor rather than a client factor.

5PositiveF: Persistence with counselling (participant-only data)

This element refers to participant persistence with accessing counselling and also

tenacity with forming and appreciating therapeutic relationships, sometimes in spite of

previous experiences of unsatisfactory therapy encounters. All participants discussed

persistence, which I have here divided into four areas to elucidate its dimensions.

135 of 271

Seeking and attending counselling

Participants demonstrated considerable perseverance in seeking and attending

counselling. Ella, Kyra and Thea recounted having attended multiple services over time.

Thea said: Starting with [service D] then going to [service E then service A]. And the same

things have come up. And each time I’ve been a bit more able to try and get on board with

them. Although it did take five-ish years.

Giving counselling another try

Kyra, Lucy and Zoey sought different types of therapy after previous sub-optimal

experiences. Zoey recalled, I said [to my GP] I needed more than a counsellor. Because I

wasn’t coping very well. The counsellors were nice, and they listened, and they tried to show

support, but I needed someone…to give me direction and sort myself out. Then I started seeing a

CBT therapist….I started to make big steps forward. She was really good.

Trying counselling alongside other types of help

Dana credited her mental health improvement to using a range of different types of help

simultaneously: As well as going to see the counsellor, I saw a homeopath, I’ve

changed my diet, started doing yoga every day and I really did make changes. And I

was on the medication as well.

On the other hand using several types of therapy simultaneously left Luka unsure which

had been helpful. When not wholly satisfied with online therapy, she simultaneously

sought and attended face-to-face CBT, and a support group. However, she said, Really

hard for me to tell…what actually worked. I don’t know what specific, why, and what time.

Steinmetz, Lewinsohn and Antonuccio (1983) found participant reports of being in

concurrent therapy predicted negative variance in outcome at 6-month follow-up.

Appreciating therapist contributions

Although the focus of this research is on client contributions to outcome it should be

noted that all participants mentioned helpful or unhelpful contributions of therapy and

therapists either in the therapy just concluded, or in previous therapy. Successful

136 of 271

therapist relational engagement was highlighted particularly where the therapist

matched their approach to the participant’s preferred way of working. For example, Kyra

said, The thing with [therapist], she sussed me out quite quickly. So when we talked about

methodology at the start, of how I can take some practical steps, she delved quite deep about

my professional role and about how I work and how I think. And then approached the next few

sessions in that way.

Stage C: New social connections fuel change Stage C: New social connections fuel change is connoted by two categories, first,

accessing forms of help (category 5Positive above). Category 6: improving social

activity also occurs at this stage and is explained here along with its connotative

elements.

Category 6Positive: Improving social activity has the double meaning of actively

extending, enriching or re-engaging with social life; and self-improving through

broadened social contact. Social reinforcement and social self-efficacy energises

motivation to make further changes. This category synergistically links participant efforts

to access help with specific further self-improvement pursued in the categories that

connote stage D: Exploratory learning. Category 6Positive is connoted by six elements:

expanded sociality; engagement; secure home; new job; proximalising; and role

models.

6PositiveA: Expanded sociality (research and participant data)

This element means the commencement of new (or resumption of old) social activity,

including pro-socially helping others. Six participants described benefits of expanded

sociality, Ella particularly persuasively: The amount of sense of security I’ve had around me

over these last two years [from family, friends, therapy and social life]. Without that whole

structure I don’t think I’d have come this far. I’ve been going out loads at the moment….It’s

probably helping as well. I saw people I’ve not seen for twelve months on Saturday night….So I

felt more of the group.

Three included studies supported this element. Binford et.al. (2005) and Zywiak,

Longabaugh and Wirtz (2002) confirmed that participants who sought more social

137 of 271

contact directly after therapy had better outcomes at 6-month follow-up. Davidson et.al.

(2005) found that clients associated prosocial behaviour with positive outcome.

6PositiveB: Re-engagement (participant and research data)

This element refers to client efforts to improve existing relationships with better

communication skills, along with positive reception of these efforts by others. Four

participants reported this element. For example, Dana said she had become more

patient and respectful with her partner. Similarly Esme reported controlling her partner

less. Both participants said these changes were appreciated by their partners and

contributed to desired change.

Two studies supported this element. Constantino et.al. (2017) found that participants

credited ‘corrective experiences’ for leading to positive outcome. These included

reduced need for control, and improved communications such as disclosure,

assertiveness and accommodation of others’ views. Khattra et.al. (2017) found that

participants who reported becoming more adaptive in interpersonal relationships also

reported better therapy outcomes; but the authors solely credit therapy for this change

without acknowledging that it could only have happened in the context of welcoming

social milieux.

6PositiveC: Secure private home (research and participant data)

Dana and Zoey reported that improving their housing was important to the change

process. This is supported by a study by Davidson et.al. (2005) who found that

participants associated better outcome with having safe, private accommodation.

6PositiveD: New job (or leaving bad job) (participant-data only)

Six participants described how either getting a new job or leaving a bad job was a

helpful step that indicated improvement in their situation and also propelled self-concept

development (which is an element of Category 9Positive: Reintegrating). Dana said,

I’ve found a job that is perfect for me. I’m good at it….I can see very clear steps up the ladder.

Kyra explained how her new job helped her adopt a healthier identity: It’s a positive thing

that I’ve moved jobs. I leave quite a lot of the baggage…I’m trying to draw a line in the sand

138 of 271

with the [D2] that I’ve suffered and how it’s impacted on my health and wellbeing….There are

too many ghosts in the closet from where I’ve been for the last 7-8 years.

Ella explained how leaving a bad job opened space for therapeutic work: I didn’t realise

how exhausting it can be, getting better (laughing). But it is. And I’m so glad I’ve not been at

work through it. She was an absolute cow I was working for. So, what did I do? She offered me

voluntary redundancy because she knew damn well I wasn’t happy. So I took that and I thought

‘I’ll just have some months, me.’

6PositiveE: Proximalising and distalising (participant-data only)

Three participants were attached to this element.

Proximalisation is a vascular surgery term for an arterial attachment and grafting

procedure. Here it is useful for describing how participants moved geographically closer

to well-loved family (‘my blood’) or to places of personal significance (‘my roots’) to

improve or repair their sense of belonging. For example Lily said about moving near her

family, I’m moving down to [English city 1] in a few weeks and…I just feel so much more

positive. And my sister said, ‘it’s so nice to have my sister back because you were just a different

person a few years ago.’ Dana observed, My roots are up here. We were so miserable in

[English city 1]. And we have family up here. So we decided to move. I love it here.

Distalising is the converse action of taking important steps to move away from unhelpful

family or places associated with bad memories. Zoey said, I needed to get out of [English

city 6] because of just too many negative things and I didn’t know how to sort myself out living

in it all. When I moved away…I came into my own.

6PositiveF: Role models (participant-data only)

Four clients described using either positive or negative role models in a helpful way.

Lucy described the importance of having recovered people in her social network who

she could emulate: People who are living…who are where your goal is. Who are living life, and taking

more of an advocate role. Three participants described consciously endeavouring to not be

like people they pitied or disliked. For example, Dana said, I used my step-mum as a kind

139 of 271

of image of everything that I don’t want to be. I’m really horrified (laughing) to think anyone

would think that I was like her.

Stage D: Reflective and exploratory learning Stage D: Reflective and exploratory learning is connoted by three categories that

represent inter-related types of work on oneself: Self-helping behaviours; helpful

thinking; and re-integrating. Each category is now explained along with its connotative

elements.

Category 7Positive: Self-helping behaviours are actions identified by participants as

being directly helpful in obtaining desired change, primarily through experiential

learning. Some, but by no means all can occur in the context of therapy. The

occurrence of these actions depends on preceding conditions of social support in order

to understand what might be helpful and to promote willingness to try them. There is an

inherent interplay between self-helping behaviours and the subsequent category,

8Positive: Helpful thinking.

Category 7Positive is connoted by five elements: Journaling; insight-led change;

exposure; exercise; and pacing.

7PositiveA: Journaling (participant and research data)

Five participants endorsed the value of reflective writing, including self-monitoring,

journaling and blogging, to improving insight. Lily described how self-monitoring helped

her externalise and de-fuse worry. It also helped her respond pro-actively to stress: On

Monday when I got really lonely, I knew the triggers, where they were going to be, which I didn’t

know before. And it meant I felt in control of what I was doing. It was so eye-opening.

Four studies endorsed this element. Participant reports in studies by Clarke, Rees and

Hardy (2004), Constantino, et.al. (2017), and Stirman et.al. (2018) verifed the

therapeutic value of reflective writing. These results are further supported by Khattra

et.al. (2017) in a study using 12-month follow-up data, who found that keeping thought

records was associated by participants with achieving meaningful differences in coping.

7PositiveB: Insight-led behavioural change (participant and research data)

140 of 271

This element comprises a synergistic interplay of self-awareness, new behaviour and

improved self-guidance and was described by eight participants. Thea described it like

this: I sometimes watch candles to self-soothe or listen to music. I was taking care of myself, it

felt sort of like a mum taking care of you. But you’re doing it for yourself, making sure you get to

bed nice and properly. Lily observed, I recognise where things might go wrong during the day,

and I can just think, ‘right, you need to change this.’

Four studies are included in this element. Simons, Lustman, Wetzel and Murphy (1985)

describe ‘learned resourcefulness’ as an insight-led ability to minimise undesirable

behaviours. Their study found that participant pre-therapy learned resourcefulness

accounted for 57% variance in therapy outcome.

Matos et.al. (2009) describe ‘re-conceptualisation’ as a metacognitive ability to observe

and describe personal change processes over time, which was associated in their study

with better outcomes.

‘Mentalization’ is a label for the ability to understand how one’s own and others’ mental

states and behaviour effect each other. It was operationalised as ‘reflective functioning’

in a study by Ekeblad, Falkenstrom and Holmqvist (2016) who found that participant

higher baseline and mid-therapy reflective functioning predicted better outcome.

Hanna and Ritchie (1995) found that insight-led change was the most potent factor

reported by participants in recounting helpful therapeutic experiences.

7PositiveC: Exposure and flexibility (participant and research data)

Exposure means pushing oneself out of ‘comfort zones’ to face and manage fears.

‘Psychological flexibility’ is defined by Gloster at.al. (2014) as the ability to mindfully

accept cognitions and emotions to promote a meaningful life. Flexibility must be

activated to prevent exposure strategies from backfiring.

In this participant group, exposure was demonstrated by five participants who described

both honest self-examination, and courageous behavioural experimentation. Lucy said

about the self-reflecting aspect of therapy, It’s been a very difficult journey (laughing) and

it’s probably been one of the hardest things I’ve had to do in my life. Because I’m confronting

141 of 271

years…of behaviours that have become so engrained. Esme said that through planned

behavioural experimentation she pushed herself to a position where I was uncomfortable

and had to let go of control. Each time I did, it became easier.

Ella described psychological flexibility in this way: I think it’s pushing myself to go [to social

activity] for the first time. Seeing it was fun, it was great. The world didn’t collapse because Ella

left the house (laughing).

Exposure and behavioural activation were well-evidenced for predicting positive

outcome by four studies in the sample, including Mörtberg, Hoffart, Boecking and Clark

(2015); Ryba, Lejuez and Hopko (2014); Mörtl and Wietersheim (2008); and Leung and

Heimberg (1996).

7PositiveD: Exercise (participant-only data)

Three clients mentioned how an adjusted relationship with exercise promoted desired

change. For Thea starting a new sport was not simply a means to fitness but an

accomplishment: It was really good to be able to practise [new sport]. The achievement was

quite, well, instantaneous. I was strong and getting stronger. For Lily, reducing her athletics

was a precursor to improved self-concept: I like [sport E] being a smaller part of my life. I

can go to work and…forget about [sport E] and not over-analyse it.

Category 8Positive: Helpful thinking includes imagining how and what to do

differently; reflecting on the results of therapeutic and non-therapeutic behaviours; and

changing thought climates to realise potential for emotional and relational growth. This

category is connoted by five elements: helpful self-talk; thought stopping and

restructuring; recognising accomplishments and self-efficacy; hopes; and gratitude.

8PositiveA: Helping self-talk (participant and research data)

All nine participants gave detailed descriptions of helpful inner dialogue, such as

debates that spurred their ‘better self’ to make a decision. Some included using an

‘internal therapist,’ as Lucy described: There’s two sides to my thought process. There is the

[D4] voice…and there is me, who actually is very logical….When those [D4] voices start to take

over, it’s having that conversation…you’d have with the therapist, but having it with yourself! In

142 of 271

the sense of, ‘right, now this is what you need to do, actually this is the right decision and this is

the reasons why, regardless of how you may feel.’

Two studies including Hartmann et.al. (2010) and Zeeck and Hartmann (2005) found

that participant recreation of therapeutic dialogue predicted better outcomes.

8PositiveB: Thought-stopping and restructuring (participant and research data) Seven participants described ways they actively averted unhelpful thinking. Ella said,

I’ve been doing diversion. Trying to get my mind off. I’ll take the dog out around the block.

You’ve got to find a way to ride over it. Esme said, I read about trying to recognise when your

thoughts are [D3] thoughts, and then go, ‘well, you can ignore that then.’

Thea said she used cognitive restructuring. Neimeyer and Feixas (1990) define this as

identifying and scaling intensity of upset, then identifying associated negative automatic

thoughts, then developing adaptive alternatives. They found that participant use of this

skill at the end of therapy predicted maintenance of treatment gains six months later.

8PositiveC: Recognising accomplishments and self-efficacy (participant and

research data)

Eight participants expressed admiration for themselves, particularly in relation to

achieving desired change. Ella said: When I noticed…that I was actually being logical, and I

wasn’t feeling as down, I was like, ‘sommat’s good!’ I make steps myself to enjoy life, believe in

myself more.

Self-efficacy, rather than being the retrospective confidence participants described, is a

prospective expectation of being able to achieve specific goals in future. Interestingly no

participants described this expectation. However, four studies noted its predictive power

on outcome including Gomes and Pascual-Leone (2015); Kavanaugh and Wilson,

(1985); Hartzler, Witkiewitz, Villarroel, and Donovan (2011) and Khattra et.al. (2017)

(the latter three based on follow-up data).

143 of 271

8PositiveD: Hope for the future (participant and research data)

Hopes were voiced by seven participants. They are similar to self-efficacy in intention,

but more broadly aspirational. Dana intended to get married and have children. Lucy

intended to return to work. Luka hoped she would fully recover. Esme used a negative

life-forecast to prompt changes: So in ten years, do I want to be childless, friendless, still

[unwell], miserable? No.

Davidson et.al. (2005) in their qualitative study of recovery processes found hope

offered participants the possibility of an improved life.

8PositiveE: Spirituality (research-only data)

No participants mentioned that faith or spirituality played a part in their change process.

However, three studies were included here. Rosmarin et.al. (2013a) found that belief in

God improved client expectancy in therapy. Rosmarin et.al. (2013b) found religious

coping predicted a positive therapy outcome. Similarly a qualitative study by Mohr et.al.

(2011) found that at 3-year post-therapy follow-up, the subjective importance of religion

in coping was a predictive factor for participant global wellbeing.

8PositiveF: Gratitude (research-only data)

Krentzman (2017) found baseline trait gratitude (a life-orientation towards noticing and

appreciating the positive) was a predictor of positive outcome at six months after

treatment but not 12 months. No participants expressed quotable gratitude although

several had appreciative attitudes towards aspects of their life.

9Positive: Reintegrating has a dual meaning. The first is ‘coming into oneself’ in an

evolved form; and the second, negotiating new relationships with one’s social contexts,

including accepting the inevitability of discordance in life. While this category could be

viewed as a plateau of attainment resulting from preceding participant efforts to change,

the findings here suggest that it is actually a dynamic springboard for participants’

anticipated future personal and relational growth. Only one participant did not have data

in this category but perhaps not coincidentally she decided to return to therapy around

144 of 271

the time she was interviewed. This category is connoted by five elements: owning

evolution; self-acceptance; life on life’s terms; minding regret; and rejecting stereotypes.

9PositiveA: Owning evolution (research and participant data)

This element means a sense of ownership of a lifespan evolutionary process of ‘coming

into oneself’ through new or reframed relationships. Seven participants discussed this

element. Kyra was redefining herself using a preferred healthy identity in her new job:

And it’s leaving that [troubled identity] behind, because in my new role I don’t want people to

know that about me. In Lily’s case this process meant negotiating family resistance to

her evolving identity: I can be anyone I want. I said to my mum ‘I can call myself Lily rather

than [nickname]’ and she said, ‘no you’re [nickname], don’t do that!’ (laughing). But I thought,

it’s nice, it’s so refreshing to think I can.

In Constantino et.al. (2017)’s qualitative study of participant ‘corrective experiences,’ the

findings are organised into two parts: (a) participant views of what was corrected; and

(b) of how correction was caused. In the former section under the heading ‘A balanced

outlook’ a participant is quoted, “I finally found that balance where I can be me and I can be

part of society and finding this mindset has made things a lot better” (p.143). I interpret this

quote as more than merely evidence of personal evolution, but a client factor belonging

on the causal side of the analysis, where wider integration has the potential to promote

further growth.

9PositiveB: Self-acceptance (research and participant data)

Four participants provided evidence of the role of improved self-acceptance in coming

to terms with the flawed nature of life experiences. Luka said, I am happy with what I am

now, more or less….Just having my internal self-confidence constant and not being affected

by…something else. Ella invoked the ‘serenity prayer’ when she said, Everything that I am

able to change, I have changed. Things that I am unable to change I have to live with or find a

coping mechanism.

Goldin et.al. (2013) found that participants’ increased self-endorsement mediated

therapy’s effect on outcome. It was also correlated with sustained improvement in

145 of 271

mental health at 1-year follow-up. However the authors did not examine the mechanism

of action of self-endorsement on outcome, which I propose is, in part, its role in facing

life on life’s terms, which forms the next element.

9PositiveC: Life on life’s terms (participant-only data)

This expression from the 12-Step tradition means acceptance that life is never problem-

free and that one will continually encounter challenges that require personal adaptation.

Three clients noted this element, with Lucy expressing it this way: Ownership…is the

foundation of continuing [recovery], outside of the support network, when you…get back into

society and you get back onto your life. Because I don’t know whether all these thoughts or

voices will completely disappear….But it’s about coping with it.

Esme echoed this: You go to counselling, so you’ve obviously got a problem. But [after

therapy] you’ve just got normal-people problems, that I’m going to have to deal with…like I did

before.

9PositiveD: Minding regret (participant-only data)

This element represents an advanced accommodative step in personal evolution that

involves coming to terms with, and seeking restoration from mental illness’ devastating

toll on one’s life. Two participants expressed this point. Ella said: I feel…like I’ve wasted a

big chunk of my life. And I don’t think I’ll ever get rid of that feeling, but I’ve got to just look

forward. I can’t keep saying, ‘Right, you’ve wasted thirteen years.’ Now…is a hell of a better place

than where I was.

Esme said, It’s the idea that you wasted your time and other people’s time, and hurt people

and yourself, and put yourself through stuff that just didn’t need to happen. And then you come

out of it the other end in the same place that you would have been if you’d just been like you

were. Ultimately achieved nothing. I mean I probably have achieved something in a

way….You’ve got to take the good things from it.

146 of 271

9PositiveE: Rejecting unhelpful stereotypes (participant-only data)

This final helpful element, endorsed by two participants, represents being aware of and

consciously disputing taken-for-granted social norms that might contribute to mental un-

wellness. Esme expressed this, with the following observation about body image

consciousness: But I think I’m hyper-sensitive and opinionated on things about women

and…well, not just women and body image - people and body image. And media and shit. And I

think it’s almost sad that you’re in a culture where even once you’re healthy, you’re not happy.

And I do think that’s the culture, it’s not people. Kyra said in rejection of a social expectation

she felt that she should be preoccupied with dieting, I’m hoping I’ve thrown away all the

whole problems I had with disciplined eating etcetera.

I will briefly summarise the key findings from the helpful elements, including a density

table which shows the number of participants and studies that yielded findings for each

helpful element.

I will then move on to describe in detail the unhelpful categories and elements.

Summary of key findings: Helpful client factors 1. Helpful client factors are conceptualised as a chain of causal influences over

time. The chain is characterised by discernible transition points in client

awareness, commitment to self and others, learning activity and relational

engagement.

2. At the most abstract level of the hierarchy, various forms of engagement with

social resources play a predominant role in promoting and supporting participant

reflective and exploratory learning, which is both the product of, and the catalyst

for, psychosocial growth across the lifespan.

3. I applied active, doing-as-being labels at the low-abstraction categorical level

which captured the personal and relational evolutionary nature of participant

change process narratives.

4. Even though the participant group was homogenous (therapy-seeking white

women under the age of 40) and the research sample was heterogeneous, there

was considerable overlap between the two data pools in the combined

147 of 271

(reconstructed) conceptual framework. To briefly summarise the non-matching

areas, there was no participant evidence to support research evidence of

demographic client factors, or the elements of spirituality and gratitude. There

was no research evidence to support participant evidence of speaking with

trusted family/friends to promote problem recognition; experiencing crisis as a

turning point; sensing maturity; using self-help media; making occupational

adjustments; relocating; using role models; addressing regrets; or rejecting social

stereotypes.

5. Contradictory findings were rare, both within and across the participant and

research pools. However, studies of the role of education, age, therapeutic

alliance, and experiencing emotions on outcome yielded contradictory findings.

These factors were therefore cancelled out of the reconstructed framework.

6. Participants were resourceful and persistent in making desired mental health

changes, which for all, took several years, different forms of help, and/or multiple

attempts. In narrating how they achieved their desired changes, all participants

described using wide-ranging social gambits in which attending therapy took a

partial, intermittent, minor and sometimes fallible role.

7. For participants, psychological therapy represented an occasional, partial or

limited form of assistance in these long change trajectories. All participants

described therapeutic work linked with their social contexts. In contrast to this,

only 18/68 (26%) of the included studies found participant social factors predicted

or were otherwise associated with outcome (see Appendix 24 for the list of these

studies).

The helpful elements density table in Appendix 25 shows the number of participants and

studies that yielded findings for each element.

I now explain the elements on the unhelpful side of the reconstructed framework.

4.3 Unhelpful client factors In this section I provide detailed descriptions of the elements for each category of action

along the unhelpful factors side of the framework, which for reader convenience is

shown in its entirely in the screenshot in figure 30 (best viewed at A3/200%) ::

148 of 271

Figure 30: Unhelpful factors in the concept reconstruction framework

149 of 271

The section is broken down by stages of non-change, of which there are four.

Stage A Non-change: Problem-maintaining social environment Stage A Non-change: Problem-maintaining environment is connoted by two

categories: Compromised support and no problem attribution. Each is now explained

along with its connotative elements.

Category 1Negative: Compromised support means the presence of family care and

communication characteristics in which the participant’s mental health problem is

unrecognised or efforts to address it are unfulfilled. It is connoted by two elements:

Difficult family relationships, and family mental health issues.

1NegativeA: Difficult family/friend relationships (research and participant data)

Four participants gave examples of family or friends not being helpful and/or not

understanding their difficulties, which compounded their problems or posed obstacles to

working on them. Zoey explained: Every time I go speak to mum and try to get to the bottom

of why I feel the way I feel, I’m constantly going round in circles, I’m crying all the time, I’m

confused, I’m distressed. And every time I went to see her, I’d come out…more confused…more

distracted and upset.

Thea said, We are not as close as most families….We had two family therapy sessions. They

were very eye-opening but extremely uncomfortable….I didn’t enjoy those and I’d prefer my

parents weren’t there or that I wasn’t there.

The four studies offering research findings in this category fell into two distinct types of

family difficulty that predicted less, or slower, symptom change. The first was hostile or

dysfunctional family climates (Beevers, Wells and Miller, 2007; and Hooley and

Teasdale, 1999). The other was fear of abandonment (Yen, Johnson, Costello and

Simpson, 2009) or separation (Aaronson et.al. 2008) associated with fear of danger to

significant others.

1NegativeB: Family member with mental health issues (participant-only data)

Three participants described how dealing with a family member with mental illness

made it considerably more difficult for them to work on their own problems.

150 of 271

Category 2Negative: Ignored problem means that in spite of experiencing crises as a

result of the problem, the problem is not attributed by the participant as contributory; or

no decision to address the problem is made in spite of crises. It is connoted by only one

element.

2NegativeA: Unrecognised crisis (Research-only data)

A study by Barnett et.al. (2010) found that patients who did not attribute problem

behaviour to a problem-related event for which they were treated in accident and

emergency did not reduce problem behaviour at 12-month follow-up.

Only one participant endorsed this element, which fell below the minimum to be counted

as a participant-endorsed element. However it seems worth mentioning that Kyra

described how her own and others’ habituation to crises precluded change: I’d had a few

being (laughing) carted away in ambulance and things when I’ve lost consciousness. It almost,

to some extent, to some people, defined me, …a group of people that knew the background to

the illness that I had…for quite a long time.

Stage B Non-change: No commitment to change The non-dynamic Stage B: No other- or self-directed commitment to change is

connoted by three corresponding ‘state-of-being’ categories: socially disconnected;

unready to change; and not minded to use help. Each is now explained along with its

connotative elements.

Category 3Negative: Socially disconnected means the experience of social isolation

but may also involve participants taking more or less active steps to cut themselves off

from others. It is connoted by three elements: alienation; guardedness; and social

difficulties.

3NegativeA: Alienation (participant and research data)

This element, mentioned four participants, means feeling one does not belong culturally,

relationally or geographically which compounded or led to problems. Lily described it

like this: Dana recalled, I wasn’t going anywhere in [English city 1]. A lot of my friends had

moved away. My parents had moved away. So expensive. We were absolutely stifled. This type

151 of 271

of alienation can for some people provide an impetus for element 6Positive E:

Proximalising/Distalising. None of the participants spoke about experiencing stigma

as a form of alienation, Ritsher and Phelan (2004) who found that participants who felt

out of place or like “less than a full member of society” (p.260) particularly in mental

illness stigma predicted greater subjective demoralisation and dysphoria at 4-month

follow-up.

3NegativeB: Guardedness (research and participant data)

All participants, when well, experienced no difficulty talking with hindsight about their

problems and feelings. But this category means being reticent about problems when

unwell. It is the counterpoint of 1PositiveB: Speaking with others about the problem.

Six participants described experiencing it in the past, including Lucy who said, I found it

very secretive, the [D4] side of me wanted to protect what was happening, and I think that’s

where you shut everybody out and not realise you’re doing it. And stop your life, basically.

Leweke, Bausch, Leichsenring, Walter, and Stingl (2009) found that alexithymic

difficulty in verbalising feelings mildly predicted poorer outcome.

3NegativeC: Social difficulty (research and participant data)

This category encompasses social withdrawal and social maladjustment. Six

participants described this as an aspect of being unwell. For example, Zoey said, I cut

myself off. I’d go aggressive, didn’t want anyone near me.

Four studies, all conducted with follow-up data, found that social difficulty predicted

worse outcome or treatment failure (Ritsher and Phelan, 2004; Dow et.al. 2007; Nakano

et.al. 2008 and Hartmann et.al. 2010).

Category 4Negative: Unready to change is the sequel to deficits in social connection.

It is likely for some that mental health problems are artefact of these deficits, but

problems remain unperceived, unacknowledged, or are deemed too formidable to

change. It is connoted by two elements: fear and denial.

152 of 271

4NegativeA: Fear (participant-only data)

Three participants explained how fear held them back from making desired changes.

Zoey explained why she did not want to mention a serious problem: Because I thought if I

say it then it becomes real and it needs to be addressed, and it’s out there, then I’m scared of

doing that. Lucy explained fear of change this way: I’ve not come much further [in

wellness]. So I’m preventing myself from doing that, in some ways. So why can’t I do the next

one? It’s breaking through the fear barriers…because it’s the unknown.

4NegativeB: Denial (participant-only data)

Six participants described being in degrees of denial about their problems along the

way. Lily explained how fear and denial can be connected: I don’t think I acknowledged I

had an issue until I actually said it out loud. I think I was almost scared to admit something was

wrong.

Category 5Negative: Unable to use help means not wholeheartedly pursuing social

help; or, where help is accessed, limited commitment or capacity to engage. It is

connoted by just one element.

5NegativeA: Find counselling fruitless (participant and research data)

This element encompasses experiences of therapy that doesn’t ‘work.’ It includes (a)

participant lack of motivation and (b) therapist inability to strike a useful working

alliance. It was mentioned by four participants. It also includes how people may be less

inclined or capable of engaging in therapy when demoralised or immobilised by mental

illness.

Kyra described the effects of being insufficiently motivated in previous therapy: I half-

heartedly went to counselling and things like that but never really felt like I got…what I needed

to out of it….Sometimes because the fit wasn’t right between me and the counsellor. Or possibly

because I wasn’t in the right frame of mind to make positive changes in my life.

Therapist inability to strike a useful working relationship tended to be framed in terms of

relational clumsiness. Zoey recalled, I saw this guy. The first session didn’t go very well. I

153 of 271

was ready to sit down and tell somebody. And he went all around the houses, talking about all

the admin and I was starting to get panicky. And it went really badly, because I…when I was

telling him things, the expression on his face, I think he was in a lot of shock about what was

coming out of my mouth. Similarly Ella recalled, The first counsellor I had said, ‘I don’t know

what your problem is. Just pull yourself together. I had [D3] as a child.’ And I’m like, ‘Are you a

counsellor?’ It wasn’t really the right thing to say. Thea explained about an earlier course of

therapy, I didn’t like my therapist, which was the first bad thing. And so, I didn’t like the work

we did together and I viewed it as something stupid that didn’t make sense, and I was very

cynical of it.

Therapy may also be fruitless if not sensitively tailored to the client capabilities or level

of functioning. A study by Scheel, Seaman, Roach, Mullin and Mahoney (1999) found

that client perception of the difficulty of the therapist’s recommendation predicted non-

implementation.

Stage C Non-change: No new connections Stage C Non-change: No new connections is connoted by two categories, first,

unable to use help (category 5Negative above). Category 6: Oppressed also occurs at

this stage and is explained here along with its connotative elements.

Category 6Negative: Feeling oppressed refers to participant experience of a socially

or culturally oppressive environment that impedes psychosocial growth. It is connoted

by two elements: adverse work, and adverse partnership.

6NegativeA: Adverse work (participant-only data)

Seven participants spoke about being in situations where they felt strained by too much,

or unpleasant occupational or academic work, which caused or worsened their

problems. Dana said about her circumstances at university, leading up to becoming

unwell: Basically, the course was really stressful, and everything around it. It just wasn’t

plausible for a human being to get that amount of work done.

154 of 271

6NegativeB: Adverse partnership (research-only data)

A study of narrative therapy for women in abusive partnerships by Gonçalves, et.al.

(2011) found that failed therapy (indicated by participant ongoing ‘problem narratives’)

was associated, unsurprisingly, with ongoing abuse in their relationships.

Stage D Non-change: Resource-depletion cycle Stage D non-change: resource-depletion cycle is connoted by three categories that are

counterpoints to their positive counterparts: retained behaviours, retained thinking and

unwell identity. These are explained here along with their connotative elements.

Category 7Negative: Retained unhelpful behaviours are identified by participants as

things they do that tend to keep themselves stuck; or that represent areas where further

work is needed. Research refers to hard-to-change unhelpful behaviours or coping

styles. Retained behaviours reinforce retained thinking (the next category). This

category is connoted by two elements: unhelpful reactions, and avoidance.

7NegativeA: Unhelpful reactions (research and participant data)

Participants described different types of unhelpful reactivity when stressed. For example

Zoey described regressive anxiety: I tend to react in a childlike way when something bad

happens. Or I panic. I’m not calm like a grown woman…I’m quite immature. Both Thea and

Lucy experienced complacency, which Thea described like this: What felt challenging in

the…early stages of recovery is now very easy…I’ve got to a point where I’ve become very, very

comfortable. So it’s about challenging, re-challenging yourself. Lily described indecisiveness:

I don’t trust myself with decisions sometimes, because…I over-analyse everything so I don’t

know why I’m making a particular decision. Esme found her tendency for organising and

leading could turn to unhelpful control of herself and others.

Dingemans, Spinhoven and van Furth (2007) noted another kind of unhelpful stress

response which they call ‘palliative reacting:’ turning to food, alcohol, drugs, smoking

(and, oddly, relaxation) for distraction and comfort. Their study found that palliative

reacting was a predictor for treatment failure at 12-month follow-up.

155 of 271

7NegativeB: Avoidance (research and participant data)

Six participants described how being unwell shielded them from facing potentially

growth-inducing social experiences. Kyra put it like this: In a weird sort of way, it’s another

crutch in my life, that I can let myself get absorbed by the feelings that you have with [D2]. And

sometimes that means you don’t have to make the really tough decisions. If I think about my

relationships with good friends...I can use it as a cop-out.

Three studies supported this element. An experientially avoidant coping style was found

to predict worse outcome in studies by Kim, Zane, and Blozis, (2012) and Berking,

Neacsiu, Comtois, and Linehan (2009). Also, in counterpoint to positive expectancy,

Hartmann et.al. (2010) found that where participants expected painful challenge in

therapy, this predicted treatment failure at 3-month follow-up.

Category 8Negative: Retained unhelpful thinking means unchanged and sometimes

engrained ways of seeing oneself, and of interpreting internal and external stimuli, that

perpetuate psychosocial stagnation. It is a later consequence of social disconnection

and can be experienced as ‘living in my own world with my own rules.’ This category

also refers to hard-to-change, unhelpful thinking styles and cognitive events. It is

connoted by five elements: perfectionism; negative self-talk; obsessions; low

metacognition; and dissociation.

8NegativeA: Perfectionism (research and participant data)

Five participants described how variants of perfectionism were implicated in their

problems. For example, Lily said, Before I even had any [D3] issues…I’d obsessively plan all

my clothes I was going to wear. And…I’m so competitive. [But] I told my mum, ‘I feel like a failed

perfectionist’ because I think I am, but I’m not, because sometimes I have a vision and I don’t do

it.

Three studies supported this element, Hawley, Ho, Zuroff and Blatt (2006) found that

perfectionism slowed the rate of participant symptom reduction, while Kyrios, Hordern

and Fassnacht (2015) found that it predicted treatment failure at 3-month follow-up.

Shahar, Blatt, Zuroff, Krupnick and Sotsky (2004) provide one explanation for

156 of 271

perfectionism’s hindering properties. They found that pre-treatment perfectionism

impeded social relations, which in turn predicted persistence of mental ill-health

symptoms after therapy.

8NegativeB: Unhelpful self-talk (research and participant data)

Six participants gave detailed descriptions of different types of low self-worth, self-put-

downs and inner criticism. Thea interpreted symptoms of mental illness like this: At the

time I just thought ‘I’m lazy, and a bad person which is why I don’t do these things.’ …I thought I

was a terrible person. Generalised ‘negative cognition’ was found by Beevers, Wells and

Miller (2007) to predict slower change in symptoms.

Ella recreated therapeutic dialogue in an unhelpful way: I won’t take compliments.

[Therapist] says, ‘You’re doing so well!’ And I’m like, ‘No I’m not, it’s you! Will you stop!’…In my

head I took, ‘That’s because I’m not good enough. I’m not a good person.’ Altimir et.al. (2010)

found that participant recreation of therapeutic dialogue with negative emotions

predicted non-remission at 3-month follow-up. Although shame is a different construct

than self-criticism, it too can induce a sense of worthlessness. Brown, Linehan,

Comtois, Murray and Chapman (2009) found shame was found to predict ongoing

symptom severity during therapy.

8NegativeC: Obsessions (research-only data)

Alonso, Menchon, Pifarre, Mataix-Cols and Torres (2001) found that greater sexual or

religious obsessions predicted poorer outcome at long term (1-5 years) follow-up.

8NegativeD: Low metacognition (research-only data)

Metacognition refers to awareness and ability to distinguish thoughts and other mental

events from the self. A study by Teasdale at.al. (2002) found that lower metacognitive

processing of memories of depressing events occurring in the 5 months preceding

therapy assessment significantly predicted earlier relapse by 11-month follow-up.

157 of 271

8NegativeE: Dissociation (research-only data)

Kleindienst et.al. (2011) found that dissociation predicted poor improvement in general

psychological functioning at one month follow-up. They proposed that dissociative

disturbances in psychotherapeutic learning may account for this.

Only one participant endorsed this element, which fell below the minimum to be counted

as a participant-endorsed element. However it seems worth mentioning that Zoey

described why dissociative experiences delayed her accessing help: I’d been suffering

from flashbacks and I’d not told anybody. I kept it to myself for six-seven years, because I was

terrified as to what it was, what it meant.

Category 9Negative: Unwell identity means, at varying levels of awareness, living

with psychosocial debility and decreasing hope of recovery. It is connoted by three

elements: endorsed stereotypes; problem-enmeshment and long illness. These are now

explained along with their connotative elements.

9NegativeA: Endorsed unhelpful stereotypes (research and participant data)

This element refers to the extent to which participants aligned themselves with unhelpful

cultural mores or beliefs. Five endorsed the social expectation that women should be

dissatisfied with their bodies, although only Esme noted its cultural origins and

pervasive impact.

Ritsher and Phelan (2004) found that when participants endorsed negative 'mental

health patient' self-stigmatising stereotypes for themselves, this predicted greater

subjective demoralisation and dysphoria at 4-month follow-up.

9NegativeB: Problem-enmeshment (participant-only data)

This element refers to the merging of personal and problem identity. Subtle evidence

was present in the use of possessive pronouns by eight participants to describe aspects

of the problem, such as ‘my episodes,’ ‘my relapse,’ ‘my tablets’ ‘my [diagnostic label]’

and so on. Problem identity enmeshment was clearly voiced by Lily, who said, I didn’t

realise how much my [D3]…had become a part of me.

158 of 271

9NegativeC: Long illness (research and participant data)

This element encompasses hopelessness of overcoming mental health problems when

people have suffered repeated, severe or unremitting problems for years. It is a key

factor in the cycle of resource depletion, and six participants described it, including Ella

who said, I don’t think I’ll ever get rid of it basically. My [D2], I’ve had that since I was a kid.

Luka said, I’m quite realistic about the fact that I can go back into relapse at some point. So I

don’t think it’s been cured. I think I just happened to go into remission. Whether because of

counselling or not. Lucy poignantly described the impact of having had problems since

childhood: It’s always been there, from a very young age and I didn’t recognise it, never got any

help or support. And I think that makes it…tricky going into adulthood, and getting treatment

when you’re in adult services rather than as a child. Because actually you’ve founded your whole

life on something you thought was normal. When now, I can see, it’s not.

Studies by Ramnerö and Ost, (2004) and Nakano et.al. (2008) both found that

participants with longer duration of illness had worse outcomes at 12-month follow-up.

Nakano et.al. (2008) suggest that people with long-term mental health problems may

have more impaired social role functioning which in turn reduces the benefit of therapy.

Summary of key findings: Unhelpful client factors

1. Unhelpful client factors are conceptualised as a cycle of gradual deterioration in

psychosocial functioning over time, if unchecked.

2. At the most abstract ‘super-factor’ level of the hierarchy, inadequate social

support leads to psychosocial stagnation, where difficulties on an individual and

social level obstruct reflective and exploratory learning (Hendry and Kloep,

2002).

3. I applied passive adjectival labels at the categorical level to represent participant

narratives about experiencing states of non-change (or insufficient positive

change). The personally overwhelming nature of these circumstances leads, for

some, to the adoption of an ‘unwell identity’ which is both a signal and promoter

of further personal and social resource depletion.

159 of 271

4. Categories of unhelpful client factors form thematic counterpoints to identified

helpful factors.

5. To briefly summarise the non-matching areas on the unhelpful side of the

reconstructed framework, there was no participant evidence to support research

evidence for difficult counselling recommendations not being followed by clients;

or for the unhelpful impact of partner abuse, religious or sexual obsessions or low

metacognition. There was no research evidence to support participant evidence

of fear or denial preventing change or the unhelpful impact of adverse work

conditions.

An unhelpful elements density table is available in Appendix 26, which shows the

number of participants and studies that yielded findings for each element.

160 of 271

CHAPTER 5: DISCUSSION

5.1 Chapter overview The purpose of this discussion is to answer the four research questions and to review

the adequacy of the concept reconstruction. The chapter is organised over twelve

sections.

In the first four sections I answer each of the research questions set out in the

introduction (p.24), recapped here:

1. What do participants describe as salient factors in their efforts to make desired

psychosocial life-changes, and how are these best conceptualised? This data

informed the primary concept framework.

2. How are client factors represented in psychotherapy research? This data

informed the comparator framework.

3. What are the key helpful and unhelpful components of a reconstructed client

factors framework?

4. What is the utility of the reconstructed concept to counselling psychology theory?

The first two questions are answered relatively briefly as a lead-in to lengthier

discussion of the third question which represents the first aim of the study, to

reconstruct the client factors concept.

The answer to the fourth question regarding the utility of the concept to counselling

psychology theory addresses the first of seven criteria for concept adequacy suggested

by Gerring (2012). Each of Gerring’s remaining six criteria is discussed over a further

six sections. These comprise domain (scope); fecundity (richness and depth);

differentiation (sufficiently contrasted from neighbouring concepts); resonance

(familiarity with existing terminology and meanings; avoidance of neologism);

consistency (same meaning in different contexts) and operationalisation (ease of

observation and measurement)

The chapter rounds out with two further sections: recommendations for further study;

and study limitations.

161 of 271

5.2 Study question 1: Factors salient to participants For the participants in this study, client factors that directly facilitated therapeutic change

included capacity for, and persistence with, reflective and exploratory learning; insight-

led behavioural and cognitive change; and personal and social reintegration and

adaptation. These occur within wider social contexts, assisted by attuned professionals

when participants are in the right frame of mind to engage and benefit from problem-

specific therapeutic psychoeducation. It could be argued that these client factors might

be exclusive to the young, educated, female participant sample that volunteered for this

study. But there were clear similarities and considerable overlap between these

participant results and those yielded by 68 international studies with broad participant

demographic strata (discussed on pp.108-109 in section 3.8 above and also in section

5.3 below).

In terms of factors that posed barriers to desired change reported by participants in this

study, psychosocial stagnation and inadequate social support are linked in cycles of

mutual influence that can for some lead to increasing social isolation or withdrawal

which in turn compounds emotional and behavioural suffering, and can for some have

profoundly negative consequences on self-identity.

5.3 Study question 2: Client factors in psychotherapy research The research sample in this study offered a picture of the client factors concept which

mapped well onto the participant concept framework but with smaller subject-territory

boundaries (with exceptions as explained further below). My study exclusion criteria

account for the absence of research on self-help resources, which feature prominently

in the participant framework (as noted in study limitations in section 5.13 below, future

work on conceptualisation of client factors ought to include research on self-help

resources and strategies). A larger research sample might have extended or deepened

the conceptual framework. However a possible explanation for the narrower footprint in

the research-informed framework is that psychotherapy research tends to focus on how

therapy works, and client factors that account for it not working, rather than on how

clients work to make desired changes within broader social systems, over periods of

time that for some, extend before and after they are ‘in therapy.’

162 of 271

Where the research contributed information that was not present in the participant

framework, it was more on the side of unhelpful client factors, particularly in the area of

hard-to-change cognitive features associated with psychological distress, including

obsessionality, perfectionism and low metacognition. Demographic features proved to

be a contested area with contradictory findings within the research sample and no

findings in the participant data sample, and were thus removed. It can be argued

however that people are unlikely to know, or to assert, how demographic characteristics

such as their gender, age, intelligence, education, social class and so forth have a direct

bearing on how they achieve desired psychosocial changes.

5.4 Study question 3: Key components of the reconstructed concept I have organised the discussion of question 3 to first consider what the reconstructed

concept framework tells us about the resources, processes and activities that comprise

the client factors concept. I then propose a revised title for client factors: social learning

factors. I conclude this section with discussion of conceptual boundaries and scope with

reference to Bronfenbrenner’s (2005) bioecological model.

Resources

Resources for therapeutic change with this participant and research sample fall into two

overarching factors which are represented as super-factors on the concept hierarchy.

The primary resource is reliable, long-term, generative social support that inspires and

reinforces adaptive change over the lifespan, and across close interpersonal and wider

cultural fields of social influence. Caring family members, enduring friendships, loving

life-partnerships, meaningful work and leisure activities, and a sense of belonging are

examples here. Educational, recreational, health care and counselling services also

play an important role, as do a range of self-help media. The availability and

accessibility of these resources is a wider social issue. While the determination to

access and make the most of these resources is a personal issue, it depends in part,

according to the information shared by the participants in this study, on the

encouragement of caring others.

163 of 271

The other resource found to be relevant in this study is the capacity for persistent and

resilient reflective learning. This is facilitated by a range of naturally occurring, and self-

and other-fostered activities that stretch self-knowledge and one’s ability to deal with

life’s ‘slings and arrows.’

Insufficiencies in either of these two factors were implicated in this model in the

psychosocial stagnation that characterises the inability to make desired psychosocial

life-change.

Processes

Processes of therapeutic change for all in the participant sample took place over years,

and for most involved passage through overlapping stages and turning points that are

represented at Level 3 of the concept hierarchy. Starting with problem recognition, this

process evolves through commitment, to tapping social resources for assistance and

reinforcement (of which therapy is only one of a range of elements), to reflective and

exploratory learning, which in turn promotes personal and social integration and realised

potential. The helpful process can be summarised in lifespan terms, as ‘coming into

oneself as a social being.’ Processes of non-change can be described, on the basis of

information provided by participants in this study, as a cycle of personal and social

resource depletion. Compromised social or familial support can for some, lead to social

disenfranchisement and mental ill-health, with concomitant difficulties recognising,

obtaining help for, or finding energy to address problems. Ensuing psychosocial

stagnation is a result of unchanged or unhelpful thinking and behaviour that can

become entrenched over time as ‘coming into an unwell identity.’ Mental health stigma

and detrimental cultural expectations can propel personal deterioration within this cycle.

Activities

Activities that participants identified as most helpful in making desired change are

represented at the categorical and elemental levels of the concept hierarchy. Helpful

activities included trusting and using social resources; persistently seeking and trying

different forms of help; engaging in a range of insight-led and insight-promoting

behaviours; consciously being open to changing thought-climates; and consolidating

164 of 271

personal gains and losses into a new outlook on self and life. Conversely, participants

described difficulties engaging in these activities when they were in a resource-depleted

state, particularly if they had been there for years or since childhood. They also

described how this state deteriorates, for example through mistrust of self and others,

fear, and avoidance of challenge.

Retitling ‘client factors’

In the reconstructed concept framework I found social learning processes were centrally

associated with psychosocial change. Fields of learning included family of origin,

partners, friends; self-help materials; and educational, recreational, health and

counselling service resources. This leads me to wonder if the concept title might be

fruitfully changed so that its definition is no longer predicated on the semantic

component ‘client,’ which indicates dependence on the ‘practitioner.’ ‘Social learning

factors’ might more accurately point to the potentially major constellation of wider client

and extratherapeutic social contributions to psychological therapy outcome. Concept

reconstruction has thus brought me to the brink of concept replacement (see pp. 76-77

above for related discussion).

Conceptual boundaries

I discarded the term ‘extratherapeutic factors’ in this concept reconstruction because

client work and outcomes within therapy are ultimately inseparable from engagement

with wider social resources. Bronfenbrenner’s bioecological systems theory (2005) is

useful here in considering conceptual boundaries. He describes human biopsychosocial

development as occurring via ‘proximal processes’ of regular interaction between

individuals and their immediate social environments over extended periods of time. This

is illustrated in the concept reconstruction diagram (figure 27, p.119) particularly in

Category 1Positive: Trusting family and friends; 5Positive: Persisting with help;

6Positive: Improving social activity and 9Positive:Reintegrating. These socially-

oriented categories are tessellated with self-oriented categories 2Positive: Hitting rock

bottom; 3Positive: Precluding relational loss; 7Positive: Self-helping behaviours

and 8Positive: Helpful thinking. I would go further by suggesting that the definition of

‘client factors’ includes interactions taking place across Bronfenbrenner’s (1993) entire

165 of 271

‘bioecological’ system of person, process, context and time. This system is laid out over

four levels. I will illustrate how each fit with the concept reconstruction using just one

spine, 2Positive: Hitting rock bottom as an example, although any spine might

similarly fit the ‘bioecological’ system:

The ‘microsystem’ context involves direct activities and interactions, which can be seen

for example in element 2PosB: GP services which involves conversations between

doctors and patients.

The ‘mesosystem’ involves networked interactions between parts of the microsystem

including direct effects of indirect events such as accidents and health incidents and

consequential interactions with care services that sparked participant awareness that

one had ‘hit rock bottom.’

The ‘exosystem’ involves strengths and weaknesses in social policies and services

surrounding the microsystem, for example the relative availability, quality, relevance and

accessibility of care services.

The ‘macrosystem’ involves enveloping social structures such as shared values, beliefs,

hazards, lifestyles and opportunities within temporal cultural-political shifts. These can

be seen in the life courses available to persons repeatedly experiencing helpful versus

unhelpful factors. The macrosystem is also responsible for creating and expanding the

diagnostic labelling system applied to people experiencing disturbance, loss,

disconnection or inequality (Bronfenbrenner, 1967; Walker and Rosen 2004; Wilkinson

and Pickett, 2010).

5.5 Study question 4: Concept utility for counselling psychology theory As one of Gerring’s (2012) seven criteria for concept adequacy, causal utility refers to

how well the concept serves as a component of a larger causal argument regarding the

relationship between independent and dependent variables (for want of better terms to

describe ‘the changing’ and ‘the changed’). It is important to note that the research

studies upon which the concept reconstruction is based were predominantly predictive

designs. This means that (within the considerable limitations in scope and size of this

study as discussed in section 5.13 below) any causality implied here is not materially

166 of 271

but socially mechanistic in that there may be a potential or emergent causal

interdependence between the variables in the arguments I now discuss.

The causal utility discussion evolves over two parts in order to establish a learning

theory into which the concept reconstruction fits. In the first part I contrast evidence for

the contributions of therapeutic alliance to therapy outcomes, against the tentative

causal argument made in this study’s concept reconstruction, for reflective and

experiential learning mediated by broader sources of social support. In the second part

I examine aspects of two social learning theories that the concept reconstruction maps

onto well: Margaret Archer’s (2007) ‘internal conversations’ model; and Albert Bandura’s

(1986) ‘social cognitive’ theory. My fourth and final research question, about the causal

utility of the concept to counselling psychology is answered here as I explain how they

speak differentially to the causal influences I have interpreted from the findings in this

study.

Causal influence A: Therapeutic alliance

Some psychotherapy research positions the therapeutic relationship, alliance or bond

as a primary healing factor in therapeutic outcomes (e.g. Orlinsky, Graves and Parks,

1994). Further to this positioning, the therapist is credited by some authors for the

relational skill required to maintain a therapeutic relationship, while the client’s role is

side-lined or rendered void. An example is found in this statement by Norcross (2002),

“The [therapeutic] relationship does not exist apart from what the therapist does in terms

of technique, and we cannot imagine any techniques that would not have a relational

impact” (p.8). Recalling from Chapter 2 that in causal syntax, X is the independent

variable input that facilitates change in dependent variable outcome Y, the causal

influence diagram in the case of Norcross’ statement is set out in figure 31:

167 of 271

Figure 31: Therapist relational skill causal influence

However, the concept reconstruction suggests a differently structured causal syntax.

Therapeutic alliance is not present in the concept framework as its role is disputed by

contradictory and null research findings (for further discussion of the contested

relationship between therapist interpersonal skills and therapy outcome see Truax,

1963; Lambert, DeJulio and Stein, 1978; and Wampold and Imel, 2015). Conducting

this study has led me to propose that the potency of the therapeutic alliance is a

reflection of client capacity for forming and using relationships more generally to assist

in ‘learning by doing.’ To illustrate this reasoning with an example, Lorenzo-Luaces, De

Rubeis and Webb (2014) found the predictive relationship between alliance and

outcome was related to a critical client factor, namely, the number of depressive

episodes the client had previously had. For those with more depressive episodes, the

alliance had less predictive value. Perhaps this finding was due to the likelihood that

recurrent depressive episodes might cause a person to be more socially resource-

depleted and therefore over time less able to form and use relationships, including

therapeutic relationships.

168 of 271

This argument does not say that the collaborative bond between client and therapist

(Krupnick et.al. 2014) is unimportant. On the contrary: without it, clients might

understandably quit therapy, or not engage well in joint efforts of goal formulation,

action planning and strategy exploration that forms some therapy models (see for

example Cooper and McLeod’s 2011 pluralistic approach, which, on a side note, does

not use the term ‘alliance’). But the amount of research, practise and theoretical interest

paid to therapist contributions to alliance may be disproportionate to their causal

association with outcome.

Causal influence B: Client relational and learning resources

The causal influence of helpful client factors arising from the reconstructed concept is

set out in figure 32:

Figure 32: Reconstructed helpful client factors causal influence

As discussed on p.117 in section 3.11 above), in this framework background

demographic factors B (e.g. age, gender, level of education, socioeconomic heritage)

appear to have an indirect, contradictory, contested or hard-to-measure effect on causal

169 of 271

factors involved in the achievement of desired psychosocial change (dependent

variable) Y. Causal super-factor M (mediating/moderating variable) is social support

from proximal sources. Adequate social support facilitates and mediates second super-

factor X (independent variable) which is the persistent and committed efforts

participants made, using a range of relational, social and self-help resources, to learn

via exploration and reflection how to make generative relational, behavioural, cognitive

changes, and to self-redefine; which in turn led to outcome Y. This model therefore

suggests why therapy works well for people who are socially well-prepared to help

themselves. Murphy, Cramer and Lillie (1984) found that, in the context of an

understanding therapeutic relationship, helpful advice given by the therapist was clients’

most valued curative factor derived from attending therapy. Centralising client learning

in this way, according to cognitive psychologist Albert Bandura, can “place the therapist

in a less glamorous role, and this may create some reluctance on the part of therapists

to part with the procedures currently in use” (1961, p.156).

Turning now to review the results for unhelpful client factors, the causal influence of

client factors in the resource-depletion cycle is set out in figure 33:

Figure 33: Reconstructed unhelpful client factors causal influence

170 of 271

Persistently inadequate social support, particularly from proximal sources over time

(independent variable X) leads to alienation, psychosocial stagnation, non-change, and

social resource depletion (dependent variable Y) which can for some, if not ameliorated,

turn to a chronic cycle of un-wellness. The model suggests why therapy tends not to

work as well for people who are socially distressed, disadvantaged or resource-depleted

(Brunner, 2017), because they cannot as effectively ‘convert’ social support to create

the critical ‘independent variable’ of experiential and reflective learning. Counselling is

chiefly concerned with supporting client strategies for psychosocial change. However,

altering or improving clients’ wider social learning environment, would, in this model, be

a helpful prerequisite to priming the capacity for change.

Learning theory and counselling psychology

To recap, in theoretical terms, the concept reconstruction positions client reflective and

exploratory learning enabled by, and interdependent with, social support, as the area

where psychological therapy makes its contribution, via active therapist behaviours such

as providing appropriately tailored psychoeducation and encouraging client growth-

inducing relational connections outside therapy. I now turn therefore to discuss the

relevance of learning theory in counselling psychology.

In addressing the position of learning theory in psychotherapy, cognitive psychologist

Edward Shoben said in 1948, “Psychotherapy is essentially a learning process and

should be subject to study as such” (p.112). He decried the lack of attempts at that time

to formulate therapy in terms of learning theory. Subsequently, Albert Bandura said in

1961, “While it is customary to conceptualise psychotherapy as a learning process, few

therapists accept the full implications of this position.”

To check if, half a century later, this might still be the case in contemporary British

counselling psychology, I searched articles published in Counselling Psychology

Review over the past twelve years (2006 – 2018) for clues to social learning theories

that counselling psychologists might use to inform their work. I found articles in the

areas of cognitive behaviour therapy (Boucher, 2006; Davis, McCabe and Winthrop,

2010; Beaumont, Galpin and Jenkins, 2012); pluralistic therapy (Shorrock, 2012);

solution-focussed group work (Proudlock and Wellman, 2011); neuroscience (Rizq,

171 of 271

2007; Goss, 2015); working with older adults (McIntosh and Sykes, 2016), and people

with HIV diagnosis (Harris and Larson, 2008) that mentioned the transformative role of

client psychosocial or relational learning in therapy outcomes. However, none

referenced a learning theory that might underpin or support their observations. One

article by Gianakis and Carey (2008) reviewed four theories of psychotherapeutic

change including Stiles’ (2002) ‘assimilation of problematic experiences’ model;

Prochaska and Norcross’ (2001) ‘stages of change’ model; Powers’ (2005) ‘perceptual

control theory’; and Bohart and Tallman’s (1999) ‘inherent self-healing’ theory. Key

aspects of all these theories echo those that appear in the reconstructed concept,

including problem awareness/clarification, experiential ‘learning by doing,’ and personal

reorganisation through metacognitive awareness. However, Gianakis and Carey point

out that none of these models reveal what enables these processes to occur. In the

reconstructed concept, the enabling factors are interpreted as the interdependencies of

reflective and experiential learning and social support, as previously discussed.

I also conducted an index search of all four Handbooks of Counselling Psychology

(Woolfe and Dryden 1996; Woolfe, Dryden and Strawbridge, 2003; Woolfe,

Strawbridge, Douglas and Dryden (2010); Douglas, Woolfe, Strawbridge, Kasket and

Galbraith, 2016) and found learning theory referenced as follows: John McLeod

describes client experiential learning in therapy as a relational process including

behavioural change, insight and reframing, and emotional catharsis. The concept

reconstruction resonates with these ideas. He and Donati (2016) reference David Kolb’s

(1984/2014) experiential learning cycle. However, Jarvis (2018) has criticised Kolb’s

(2014) model for not emphasising enough the interactive and social aspects of

experiential and reflective learning, which were fundamental aspects of the concept

reconstruction. It will therefore not be further explored here.

Ontologies of the ‘learning self’

Before unpacking relevant learning theories to compare fit with the concept

reconstruction, it is helpful to first set out three ontological bases for the notion of the

‘learning self,’ to understand how learning theory might describe learning process.

172 of 271

Perspectives on the nature of ‘the self’ can be described as falling along a spectrum

from individualism to social diffusion. On the individualistic side is the humanistic idea of

the ‘separate self.’ This philosophy considers the self to be a functional reality. It values

freedom, self-reliance and independence from the influence of culture and relationships.

Aspects of Rogers’ ideas about learning can be seen to fit on this end of the self-

ontological spectrum. Rogers (1983) describes how good teachers create encouraging

conditions in which students become responsible for self-directed learning via personal

creativity, curiosity and independence. On the opposite side is the social constructionist

notion of the ‘dependent self’ who is entirely a product of relationships, society, culture

and contingency. For example Gergen (2009) dispatches notions of the individual

‘coherent self.’ Instead he describes ‘multi-selves:’ human constellations of potentials

for action existing within the relational dimension.

Lying on the middle ground between these two philosophies is the notion of a

‘transactional self’ who has personal concerns, choice and agency that impact the

material and social world in a real way, and whose sense of identity and self-knowledge

are mediated through interactions with the material and social world. Here the ‘self’ and

the material world are mutually formative.

This ontological middle ground is consistent with a critical realist ontology of the self that

acknowledges the neurological basis of the mind and consciousness from which arises

epistemologically created selfhood and personal identity, from which in turn emerges

social agency (O’Mahoney, 2011; although I would say that personal identity does not

precede social agency in order of emergence, but they are interdependent processes).

Next I examine two learning theories that lie in this self-ontology middle ground:

Archer’s ‘reflexive social action’ theory, and Bandura’s ‘social cognitive’ theory.

Archer: reflexive social action

Relational sociologist Margaret Archer (2007) defines reflexivity as internal

conversations in which one is both subject and object (speaker and listener). She

positions inner dialogue as a vital (and empirically overlooked) means by which people

achieve social and occupational mobility. She describes three types of reflexive self-

talk: communicative, autonomous and meta-reflexivity.

173 of 271

Communicative reflexivity is intersubjective in that it involves sharing one’s inner

dialogue with others for affirmation, information, deliberation, or ascertaining others’

needs. Autonomous reflexivity is task-oriented and kept to oneself, because input from

others is not required or desired for self-injunctions or self-directed action. Meta-

reflexivity is value-driven and involves reflecting on one’s reflections, for example

through self-monitoring, self-analysis, daydreaming, or staging imaginary conversations

with others. Meta-reflexivity involves engaging with the possibility that there are at least

“two sides to the story” (2007, p.131).

Participants in my study voiced all three types of self-talk and all gave examples of

helpful internal dialogue that described and assisted their changing outlook, behaviours

and self-understanding (captured in element 8PositiveA: Helpful self-talk). Some also

gave examples of unhelpful self-talk that tended to confirm or reinforce a stagnant or

resource-depleted state (captured in element 8NegativeB: Negative self-talk).

Archer’s theory fits well with the reconstructed client factors concept, in that she

ascribes causal power to self-talk as a method people use for responding to

circumstances and enacting desired life-changes:

[Reflexivity enables] subjects to design and determine their responses to

the structured circumstances in which they find themselves, in the light of

what they personally care about most….Outcomes vary enormously with

agents’ creativity in dreaming up brand new responses, even to situations

that may have occurred many times before. Ultimately the precise

outcome varies with subjects’ personal concerns, degrees of commitment,

and with the costs different agents will pay to see their projects through in

the face of structural hindrances. Equally they vary with subjects’

readiness to avail themselves of enablements (pp. 11-12).

Archer offers a three-stage model of reflexive learning that sits within a larger

ontological philosophy of ‘relational realism’ (2014; Tilly, 2002) which proposes that

social interactions constitute society:

174 of 271

1. Advantageous or disadvantageous social features exist across four orders of reality

(Archer, 2014). These include (a) the natural environment including human

embodiment; (b) evolving technological resources; (c) the social realm including

education, politics, economics, culture and personal relationships; and (d)

transcendental reality, which Archer describes as belief in God and leads to

detachment from over-riding concern with the previous three realms of reality. This

bears similarities both with Rogers’ ego-less universe and Bhaskar’s transcendent-

emergent realm of the real – although Archer (2014) assigns individual formative

moral drive, free will and determinism to this order of reality.

One or more of these orders of reality becomes a generative interactive power in

relation to people’s projects when…

2. …they ask themselves, ‘what do I want?’ and ‘how do I go about getting it?’ Archer

proposes that, following these questions comes three phases of self-determining life-

long internal conversations: (a) discernment of predominant satisfactions and

dissatisfactions with one’s current way of life, and which of the realms of reality it

falls into (which maps onto Stage A: Problem recognition in the concept

reconstruction); (b) deliberation over which concerns, in which realm of reality one

should prioritise and commit to pursuing (which maps onto Stage B: Other-

focussed and self-focussed commitment to change); and (c) dedication to

persevere with a particular area of concern even with the possibility that one’s

actions might fail; while giving up concerns in other areas (which maps onto

Category 5Positive: Persisting with help).

3. Then the chosen course of action is reviewed, adapted, adjusted, abandoned or

enlarged with the aim of becoming “what we care about most in society” (p.21)

(which maps onto Super-factor 2: Reflective learning with the aim of achieving

Category 9Positive: Reintegrating). Where people are unable to conceive or

execute their concerns in any realm, they tend to experience shame, withdrawal and

“loss of relational goods” (2014, no page number) with which to engage in social

interaction. This maps onto Stage D Non-change: resource depletion cycle.

Archer’s theory and my concept reconstruction also overlap where the catalyst for

action is desire for a particular way of being that is different than one’s current state.

175 of 271

The major mapping gap is where the reconstructed client factors concept ascribes some

causal influence to social relations in inspiring and reinforcing psychosocial change,

whereas Archer solely ascribes this agency to individual inner dialogue. Archer’s

theoretical framework positions constructs similar to psychosocial change (social

actualisation and occupational mobility) as the outcome of reflexive learning, but she

describes reflexive learning as if it originates and develops in a mental vacuum,

“relatively autonomous from…structural or cultural properties” (p.15) and originating

from “first-personhood…as distinct from any social ‘concept of the self’” (p.27).

Reflexivity is examined in Archer’s model insofar as it mediates the intra-personal

relationship between deliberation and action in people’s social lives. But in my opinion

(which is heavily influenced by relational-cultural theory e.g. Jordan, 2010), the contents

of reflexive inner dialogue (the what, why and how one talks to oneself) originate in, and

are structured and limited by the socio-cultural foundations of language, relationships,

resources and potentialities in which one’s life is embedded. Furthermore, the power of

relationality can sometimes override autonomous reflexive agency as a primary

determinant of behaviour. These two points are voiced by Archer’s research

participants. Some described self-talk that addressed inherently relational, interactive

problems in their work (and working identities). Others described occupational decision

making based on the influence of trustworthy ‘similar and familiar’ friends; or based on

the emotional pull to maintain contextual and relational continuity, over and above

following one’s own reflexive inclinations. In this way, ‘learning by doing’ was shown to

be a relationally-informed rather than solely reflexivity-informed activity. Nonetheless,

Archer does not view self-talk as dependent on social moderators and mediators.

Bandura: Social cognitive theory

The ‘self’ in Albert Bandura’s social cognitive theory is a stable single entity that can

perform multiple cognitive and behavioural functions, although not simultaneously. He

ascribes five reflective capabilities to human nature, which he sees as being formed

neither solely by inner mental forces nor by external stimuli, but by a mutually

determinative reciprocity between personal behavioural and cognitive factors, and

environmental factors. Interestingly, although Bandura does not describe it this way,

176 of 271

four of the five capabilities depend on forms of inner dialogue (as Archer argues):

symbolizing (translating experience into internal models that serve as guides for future

action); forethought (symbolic representation of future possibilities exerting causal

impact on present action); self-regulation (attempting to bring oneself in line with internal

standards as well as the expectations of others); and self-reflection (knowledge

development and self-alteration through self-analysis). The fifth capability is vicarious

learning through modelling and imitative observation.

All five of these capabilities are clearly found in elements and categories of the concept

reconstruction. For example, symbolising was found in element 7PositiveB: Insight-led

change. Reflexive forethought was seen in 8PositiveD: Hopes for the future as well

as Category 3: Precluding relational loss and Category 4: Reclaiming self. Self-

regulation was seen in Category 8: Self-helping behaviours and Category 9: Self-

helping cognitions. Self-reflection was evident throughout all categories and stages of

the helpful factors side of the concept analysis, particularly in Stage D: Exploratory

learning. Vicarious learning was present particularly in element 6PositiveF: Role

models and more implicitly in some forms of help participants accessed such as

5PositiveD: Support groups and 5PositiveF: Counselling.

Bandura also describes how the individual and their social environment reciprocally

influence each other. Idiosyncratic individual behaviour brings about certain interactive

responses in others. People select and create environments in a way that can

reproduce generative change or stagnation. This is evident on the concept

reconstruction in Unhelpful superfactors 1 and 2: Inadequate social support and

psychosocial stagnation. Group behaviour can empower individuals. But in different

circumstances groups can form detrimental (e.g. coercive) reciprocal systems, or

impacts that are felt to be coercive by member participants.

Bandura also describes the determining influence of “chance encounters” in people’s

life paths, well as the influence of wider historical and political events. On an individual

level he describes how self-evaluative standards (acquired from proximal social

influences) give direction to life by guiding choices. This is seen on the reconstruction in

categories 9Positive: Reintegrating; and 9Negative: Unwell identity.

177 of 271

Bandura also discusses incentive to action as motivated by two-way interpersonal

interaction, and the accommodation of social influence. This is captured in the concept

reconstruction in category 1Positive: Trusting family and friends particularly in the

elements of communication to and from supportive people as an aspect of problem

recognition.

Two areas included in the concept reconstruction that are not explicitly covered by

Bandura’s theory are (a) the recognition of a personal crisis by oneself and/or trusted

others as a catalyst for decisive action in seeking life-change; and (b) the requirement of

trust for generative social relations.

Utility of the concept reconstruction to counselling psychology

The concept reconstruction framework suggests that counselling psychology might

beneficially develop learning theories to help clients make therapy work through

engagement with social resources beyond the consulting room. Bandura’s social

learning theory is explicitly devised to hypothesise mechanisms at work in psychological

therapy. It provides a clear, detailed and deep account of how dependent psychosocial

learning is on pre-existing and concurrent social resources, the importance of enacted

and reflective learning, and, consequently the role of insufficient social support in

obstructing the natural human drive to learn and adapt over time.

A potential lack of utility in the concept reconstruction is that it does not explicitly label

empathy anywhere within the hierarchy. Empathy has been positioned as a change-

producing relational element by a number of researchers (Davis, 1996; Bohart and

Greenberg, 1997; Bohart, 2004; Jordan, 2010; Martinez, 2017). However Bandura

proposes that empathy is not a unitary phenomenon, but comprises three constructs

which are not particularly closely related to one another, as I discuss further here.

Three constructs of empathy

The first construct, social perspective taking, can be seen across category 9Positive:

Reintegrating. The second, imaginative self-involvement, can be seen in reflective

activities including elements 5PositiveF: Counselling and 7PositiveA: Journaling.

The third, emotional responsiveness, is present across a number of categories,

178 of 271

particularly 1Positive: Trusting family and friends; 3Positive: Precluding relational

loss; 6Positive: Improving social activity and 9Positive: Re-integrating. In the

concept reconstruction, empathy is subsumed (and therefore assumed to exist) within

Super-factor 1: Social support.

I now move on to discuss Gerring’s (2012) remaining six criteria for concept adequacy:

domain, fecundity, differentiation, resonance, consistency and operationalisation.

5.6 Domain Gerring recommends that the researcher stipulate the conceptual domain, or ‘language

communities’ the concept relates to, for example, lay or academic audiences; or

‘schools’ and ‘cultures’ within a discipline. Domain also refers to the distance from lay

usage the author has travelled to create scientifically precise terminology.

In this concept reconstruction I have attempted to rely on everyday language rather than

psychotherapeutic terminology in line with the ‘muted group’ epistemology that suggests

the concept should be meaningful to those who experience it (Wuest, 1994). Tight

space on the framework diagrams leaves some labels requiring explanation for example

4PositiveC: Maturity means participant sense of ‘with age comes responsibility and

wisdom’ rather than their numerical age. No participants offered suggestions to change

my conceptual labels after second participant review, but this does not necessarily

indicate that the linguistic domain was entirely relevant for them.

I have also tried to retain non-pathologising and context-informed terminology in line

with the feminist paradigm in counselling psychology (Taylor, 1996). The reconstructed

concept also reflects relational-cultural theory (Jordan 2010) that positions amelioration

of social disconnection as the purpose of therapy rather than the removal of ‘symptoms.’

5.7 Fecundity Descriptive concepts help explain causal influences in human science by interpreting

and illustrating intersections of circumstances, events, intentions and reactions over a

set of cases. Good concepts “reveal the structure within the realities they attempt to

describe” (Gerring, 2012, p. 125). The large size of the conceptual fishbone frameworks

permits the fecund inclusion of detail to the least abstract level. This in turn helps to

179 of 271

clarify the types of elements and categories that belong to the concept; to show

coherent relationships between levels of abstraction and over temporal stages; and to

delimit conceptual borders. Such delimitation help with the next criterion, differentiation.

5.8 Differentiation This criterion refers to how well the concept is differentiated from contiguous or similar

concepts. The concept reconstruction collapsed extratherapeutic factors into client

factors because they were insufficiently distinguished. Then, because the concept

includes a wide range of activities and causal influences beyond therapy, I have

proposed that the title is inaccurate and could be replaced with ‘social learning factors’

(see discussion on p.164 in section 5.4 above). In this way the concept may be clearly

differentiated from contiguous concepts such as therapist or therapy factors.

5.9 Resonance In assessing the linguistic resonance and conceptual familiarity of the reconstructed

client factors concept against existing usage, it is useful to compare and contrast how

client factors are presented in this study against how they are represented in well-

researched ‘treatment matching’ theory.

Treatment matching approaches shape therapy around client factors known to mediate

or moderate therapy processes (which I here differentiate from ‘outcome informed’

approaches that match therapeutic approaches to elicited client preferences for working

style or approach e.g. Duncan, Miller and Sparks, 2004). A number of such approaches

have evolved over the past quarter century, for example culturally adapted treatments

(Benish, Quintana and Wampold, 2011; Smith, Rodríguez and Bernal, 2011); phase-

based therapies based on clients’ ‘stage of change’ (Kanfer and Schefft, 1988;

Prochaska, DiClemente and Norcross, 1992; Prochaska, Norcross and DiClemente,

1994) or stage of ‘assimilation of problematic experiences’ (Stiles, et.al., 1990; Stiles,

2002, 2005). Norcross and colleagues have developed a treatment matching approach

called ‘customized/tailored therapy relationship’ (TTR) (Norcross, 2002; Norcross and

Lambert, 2010, 2011; Norcross and Wampold, 2010; Norcross and Wampold, 2011;

also called ‘treatment adaptation’ and ‘prescriptive matching’ in Norcross and Karpiak,

2012).

180 of 271

Systematic treatment selection (STS)

The treatment matching approach I focus on here is Larry Beutler and colleagues’

‘systematic treatment selection’ (STS) (Beutler and Consoli, 1992; Castonguay and

Beutler, 2006; Beutler et.al., 2016; also called ‘prescriptive therapy’ in Beutler 2016).

STS has been chosen because it conceptualises client factors as impacting outcome in

their own right, where other approaches (e.g. TTR) conceptualise client dimensions only

in so far as they inform therapist contributions which are in turn viewed as the primary

outcome mediators.

Based on decades of direct and meta-analytic research work, STS is a set of technically

eclectic treatment principles (Beutler, 1983, 2016) that are tailored to clients’ non-

diagnostic ‘predisposing trait-like tendencies’ and problem dimensions. The range of

salient client dimensions identified by Beutler and colleagues has been adjusted over

time (for development of STS see Beutler 1983; Beutler, Patterson, Jacob, Shoham,

Yost and Rohrbaugh, 1994; Beutler and Clarkin, 1999; Beutler, Clarkin and Bongar,

2000; Beutler and Harwood, 2000; Groth-Marnat, Roberts and Beutler, 2001;

Castonguay and Beutler, 2006; Beutler, Harwood, Kimpara, Verdirame and Blau, 2011;

Beutler et.al. 2011b; Beutler and Forrester, 2014).

Client factors in STS

Beutler et.al. (2016) mention eight client factors that comprise STS (the first four from

Beutler, Clarkin and Bongar 2000 and the second four from Beutler at.al. 2006). They

explain that these dimensions are not equally well-established in the research literature,

and that they are descriptive and atheoretical. At the end of the list of eight I have added

a further two relevant client factors that appear in Beutler and Clarkin (1999) but for

unknown reasons were subsequently dropped from the model. I describe only the client

factors and not the associated treatment recommendations. I concurrently map these

factors against the labels and the conceptual footprint of the reconstructed framework to

check its resonance with established use in STS.

1.Functional impairment: This dimension includes symptom intensity,

chronicity and comorbidity as well as the presence of family problems and/or

181 of 271

isolation from social, school or occupational activity and/or lack of friendships. In

previous iterations (e.g. Beutler 1983) this dimension was called ‘symptom

complexity’ and highlighted negative social reinforcement for the continuation of

client symptom patterns. In the reconstructed concept this relatively vague

dimension is redrawn as unhelpful super-factors 1 and 2: Inadequate social

support and psychosocial stagnation. It is also associated with non-change

stages A and D: Problem-maintaining environment and resource-depletion.

The main difference is that Beutler et.al. use terminology that locates the problem

within the individual whereas the framework here tends to identify problem

sources as transactions between the individual and their social environment.

2. Coping Style is characterised by internalising or externalising traits (people can

display aspects of both). Internalising traits include self-responsibility; self-blame

and constricted affect. Externalising traits include self-dramatisation, and

experiencing anxiety “only when deprived of a direct, active method of coping

with stress or when blame cannot be transferred” (Beutler and Clarkin, 1999.

p.78). Neither internalising nor externalising traits are located on the

reconstructed concept because they did not appear in this study as contributing

to outcome.

3. A. Resistance is defined in STS as a psychopathological trait based in

unconscious avoidance of threatening personal material. It may involve passive

obstructionism or outright refusal to engage in aspects of therapy (Beutler,

Harwood, Michelson et.al. 2011). It can be seen in the reconstructed framework

in Stage B Non-change: No commitment to change. Beutler et.al. use

terminology that ascribes to clients more detrimental causal force than the

passive description I used.

B. Reactance within normal personality expression is the forcefulness of one’s

efforts to resist threats to one’s freedom from the surrounding environment.

Reactance is not found on the reconstructed concept.

182 of 271

4. Degree of subjective distress may be a feature of reactance or functional

impairment. More distress was experienced by participants when they were

anywhere within the cycle of non-change featured on the unhelpful factors side

of the concept reconstruction.

5. Stage of readiness: Prochaska and Norcross’ (1999) transdiagnostic theory of

change proposes that people use different attitudes, intentions and behaviours to

effect change over time in five stages: pre-contemplation, contemplation,

preparation, action and maintenance. STS refers to stage of readiness theory to

inform therapy (Beutler et.al., 2016). Contemplation, preparation and action

stages are found in the reconstructed concept in Stages A-D: problem

recognition, commitment to change, new social connections and

exploratory learning.

6. Preference for type of therapist: Although this dimension is included in Beutler

et.al’s (2016) list of salient client factors, I have not been able to find any

research by Beutler that discusses the role of client preference for type of

therapist. It appears to have been adopted into STS from Norcross’ (2002;

Norcross and Wampold, 2011) TTR. The suitability of the match between type of

therapist/therapy and client is found on the reconstructed concept framework in

the element 5PositiveF: Counselling and 5NegativeA: Find counselling

fruitless.

7 & 8. The interactive effect of demographic and symptom variables:

Castonguay and Beutler (2006) examine the impact of age, gender, ethnicity,

socioeconomic status and religion in studies on diagnostic categories including

dysphoria, anxiety, substance misuse and personality disorders. The chart for

interactive effects of demographic client factors on symptom variables that I

extracted from Castonguay and Beutler (2006) is in Appendix 21. Socioeconomic

status had the strongest evidence-base as a cross-diagnostic, outcome-

mediating client factor. It is not found on the reconstructed concept, because

183 of 271

neither the participant or research data pools highlighted social class,

employment status or income as having a bearing on outcome.

9. Client expectation as Beutler uses the term refers to considerably more than

expectations about the type of therapy and the therapist (e.g. Arnkoff, Glass and

Shapiro, 2002), the outcome (e.g. Constantino, Arnkoff, Glass, Ametrano and

Smith, 2011) or even of themselves as clients (e.g. Beitel et.al., 2009). In

summary, they are the “accumulation of wisdom gained through observed or

experienced cause and effect relationships. [They] may be situation specific

(states) or general (traits) that bear a relationship to one’s age, gender, social

position and personal background” (p.58). In the concept reconstruction this

dimension is Super-factor 2: Reflective learning. The multiple meanings of the

term ‘expectations’ in psychotherapy lexicon reduces the resonance of Beutler’s

use of the term here.

10. Environmental resources refers to social support systems including work,

school, friends, partners and family. People with well-established social and

familial support systems were found by Beutler and Clarkin (1999) to respond

less pathologically to stressful events, and to have higher self-concepts than their

counterparts who are more socially isolated. In the reconstructed framework this

dimension is Super-Factor 1 A and B: Social support moderators and

mediators.

Summary of resonance

In terms of conceptual resonance, abstraction levels four (super-factors) and three

(stages of change) on the helpful and unhelpful sides of the reconstructed framework

were well-covered by STS dimensions 1 (functional impairment), 3B (resistance), 5

(stage of readiness), 9 (expectations) and 10 (environmental resources). Dimension 6

(preference for type of therapist) matched helpful and unhelpful counselling elements.

Four STS dimensions were not present in the concept reconstruction (2:

internalising/externalising coping style; 3B: reactance; 4: subjective distress; and 7/8

184 of 271

socioeconomic status). The main feature of the concept reconstruction that is missing in

STS is Stage C non-change: No new social connections as a crucial maintaining

aspect of the psychosocial stagnation cycle.

There was little terminological resonance between STS and the reconstructed concept.

Some terms in STS hold specialist definitions whereas the reconstructed framework

generally uses lay terms (with the exception of the term ‘proximalising’), in line with

Gerring’s advice to avoid technical jargon and neologisms. STS terminology tends to

adopt a ‘deficit perspective’ in which client factors are framed as problems upon which

the therapist operates; whereas the reconstructed concept takes a much broader and

more strengths-based view, in line with counselling psychology philosophy (Gelso and

Fretz, 1992). However the primary terminological dissonance is minimal mention of

social support in STS, in comparison to the central stage it takes across the

reconstructed concept. This may be because STS focusses on the therapist’s role in

adapting therapy to best help remediate client dysfunctional stress responses, rather

than helping clients harness healing resources beyond the consulting room, which in

turn may be viewed as falling beyond the therapist’s preferred or perceived domain of

influence. Consequently, the client factors concepts upon which treatment matching

approaches are based are skewed by a focus on aspects of client functioning that are

within direct awareness and influence of the psychotherapist. The mediating effects of

extratherapeutic client factors such as social support and culture tend not to be

represented as directly influencing outcome.

Gerring (2012) provides reassurance to the researcher of specialised concepts that they

can break with established usage, and thus, as shown above, incur a cost to resonance.

Reckoning with this cost requires embedding the concept in concrete, representational

data, in other words, at a low level of abstraction. The hierarchy of abstraction method

is therefore particularly useful for working with a concept that is relevant within a limited

context. I will discuss context further with the following criterion, consistency.

5.10 Consistency “A concept ought to carry the same meaning…in each empirical context in which it is

applied” (Gerring, 2012, p.121). The consistency of the client factors concept is easy to

185 of 271

assure because it only appears in psychotherapeutic literature that relies on common

factors theory.

5.11 Operationalisation “A concept that cannot be measured cannot be tested – at least not very precisely”

(Gerring, 2012, p. 205). The process of linking the more- and less-abstract connotations

in the reconstructed concept to empirical indicants poses numerous challenges. I will

touch on just two here. Firstly, causal events and processes in psychosocial change,

and their manifestations in subjective experience, can be difficult to observe, let alone to

capture, because they are best described in “actor-defined meanings and motivations”

(Gerring, 2012, p.157) via personal, subjective, narrative material that requires careful

interpretation to convert for use in research or practice. Qualitative and mixed methods

research do make this obstacle surmountable. However, a second challenge is that

ideally, the causal factor should be amenable to manipulation by the researcher to

understand its impact on outcome. Social support from family, friends, colleagues and

so on, is not particularly amenable to manipulation. This may go some way to explaining

why social support is rarely considered in psychotherapy outcome research. Likewise,

reflective and exploratory learning as the other major client factor relies on mental

states and conditions that “stem partly from…free will” (Gerring, 2012, p.208). Such

volitional states are not particularly amenable to direct or ethical manipulation.

5.12 Recommendation for further study On the basis of the results of the client factors concept reconstruction I here make a

tentative recommendation for theory development in counselling psychology, in the area

of relational learning networks.

It has been proposed by Goertzen (2010, after Royce, 1987) that psychology produces

too much empirical data and not enough integrative theoretical research. Evolving

theory is a difficult, time-consuming, trial-and-error process that inevitably courts

controversy. The idea that therapy works best for people who are already socially

positioned to help themselves is not new (Orlinsky, 2009). Likewise, the related idea

that clients make therapy work is not new (Bohart and Tallman, 1996), but it has not had

the impact it could have had because, I propose here, descriptions of client

186 of 271

presentations have tended to be limited to features that lend purpose to 20th century

therapy models that position the ‘problem’ within the individual and the ‘cure’ in the

techniques or personal attributes of the psychotherapist. Furthermore, the theory that

therapy is a social learning process is not new (e.g. Bandura, 1961) but counselling

psychology does not seem to actively embrace – or develop – theories that explicitly

embrace client social learning. I further contend that academic circles tend to produce

theory in esoteric, self-referential language that is incomprehensible and irrelevant to

the ‘labouring class’ of practitioners who have a vital need for theory to inform and

develop their work (Beck, 2013).

I therefore suggest that an interesting route for theory development in counselling

psychology is in the area of learning exchange via relational networks. From a

sociological perspective Crossley (2015) considers the focal points of various networks

in which social ties are made. These might include shared values, preferences and

interests; ethnic or cultural backgrounds; or professional, educational and income

brackets and so forth. Furthermore, I propose that simple proximity determines relatively

superficial yet nonetheless influential social exchanges, although deeper affinities may

endure tests of time, social mobility and geographic distance. Social learning results

from active and interdependent participation in, and identification with, networks and

power-alliances within networks. Trust and empathy are essential inputs and outputs in

successful relational interaction histories. Crossley (2011) suggests that personal

reflexive dialogue tends to be characterised by greater or lesser degrees of interactional

trust and empathy, just as social networks are.

Even more theoretically relevant might be the integration of social network analysis with

relational-cultural theory (RCT: Jordan, 2000, 2010). RCT proposes that people grow

through and towards trusting, empathic, ‘authentic’ relationships over the lifespan. A

goal of relational-cultural therapy is the development of relational competence and

capability by promoting flexible relational expectations and reducing the isolation,

disempowerment, shame and marginalising impact of relational disconnections and

relational abuse. Therapy formed from these theories would of necessity be rooted in a

formulation process that, insofar as this is acceptable and makes sense to clients,

187 of 271

attends closely to relationship stories; accounts for actual and potential relational-

cultural resources within social networks; identifies areas of intra-psychic and relational

resource depletion or power inequalities; and sets out a strategy for exploratory and

reflective learning through social resource enrichment. This process would de-

centralise the therapist and even the therapeutic relationship in favour of focus on the

client’s social world. This recommendation that mental health providers develop and

embrace methods for increasing extratherapeutic social support for clients identified as

socially resource-depleted is in line with studies by Turner, 1981; Kelly, 2006;

Radomski, 2014; Budge and Wampold, 2015; Probst, Lambert, Loew, Dahlbender and

Tritt, 2015; and Amati, Banks, Greenfield and Green, 2017.

This chapter concludes with discussion of some of the study’s limitations.

5.13 Study limitations In this section I provide a non-exhaustive overview of the study’s limitations in six areas.

Ontological limitation

Critical realist research in some cases, as here, does not seek to make final judgements

or generalisable truth claims on an issue but simply to make revisions to previous

understandings (Pawson, Greenhalgh, Harvey and Walshe, 2005). Additionally, Scott

(2005) brings our attention to the way all social research inevitably lags behind the

evolving and emergent nature of reality. However, when we create new knowledge, we

can in turn change some small aspect of reality through adjusted human action, but in

such evolution, reality once again makes our revised actions obsolete. Over the two

years it has taken me in this study to make the obvious-sounding recommendation that

therapists can enhance their work by focussing on helping clients enrich experiential

learning through active use of social resources - further work has been done in the area

of relational sociology, for example into art space and social spaces (Crossley, 2015).

Epistemological limitation

Given the assumptions set out in the introduction to this study, it could be argued that

my results confirm what I already believed: that emotional wellness, and recovery from

episodes of mental un-wellness depend on social resources and generative relational

188 of 271

events; and that psychological therapy is only as useful as the client’s social resources

(or social potentialities) outside the consulting room. Gerring (2012) reminds us that in

social science, “descriptive inferences draw from a standard itinerary of tropes” (p.141),

and in the case with this study the trope is the recent ‘relational turn’ in social science

(see for example, Eacott, 2018). But to balance my partiality I have been clear about

my assumptions and views; and I have provided a methodological audit trail so the

reader can see how the contents and boundaries of the concepts have been drawn up.

And although the result does perhaps reflect the ‘relational turn’ in the social sciences,

this does not necessarily reduce the value of its message, that counselling psychology

perhaps needs to engage with learning theory for direction.

Procedural limitation

Concept analysis has been described as “best done as a collaborative endeavour”

(Knafl and Deatrick, 2000, p. 43). This recommendation resonates with my belief that

knowledge construction proceeds through social interaction. While my mental

interaction with the literature and participant data surrounding my topic could, at a

stretch, be considered a social process, this study’s potential scope, richness and reach

is restricted by its being of necessity a single-author project. The generalisability of the

findings are limited by the small scale of the study, its qualitative design, and its

epistemological assumption that knowledge is “situated and perspectival” (Hekman,

1997). The timing of my decision to replace thematic analysis with concept analysis in

the design of this study does not have a material bearing on the generalisability of the

findings.

Data limitations: Literature sample

The research literature sample excluded comparison studies that found different client

factors to be salient predictors of outcome dependent on the type of therapy received.

This was because the analytic strategies in those studies were designed to demonstrate

differential treatment effects on clients rather than client effects on treatment and

outcome. It could therefore be argued that it is impossible to completely separate, or

distinguish, client factors from treatment and therapist factors, especially when outcome

data is collected in or at the end of therapy. I had considered conducting the concept

189 of 271

analysis only with studies based on long-term follow-up data, where time elapsed after

therapy provides some assurance that the client factor measured is not an immediate

‘halo effect’ of the therapeutic encounter. However, after an extensive and time-

consuming literature search, only 28 relevant studies were found with follow-up data,

which was insufficient for a concept analysis. This led to my decision to include the

remaining 40 studies that were based on end-of-therapy data. Future research in this

area might see if any difference is found if solely studies using follow-up data are used.

Studies included in the literature concept analysis were chosen on the basis of the

presence of client factors associated with outcome. Other studies may find that the

same client factors are not associated with outcome, but such studies were not included

or analysed due to their null findings relative to this study.

In this study I did not explore client predictors of premature termination, non-completion

or drop-out from therapy; nor whether there is a difference in salient client factors

between people who do and do not access psychological therapy for help. While drop-

out is certainly an ‘outcome’ of therapy that is mediated by client factors, I made this

decision for two reasons. First, the process of drop-out did not directly address the

research questions (although it might have contributed valuable data to the unhelpful

factors side of the research framework). Second, the participant sample had all

completed the course of therapy about which they spoke in their interview, so it made

sense to match this with the published material.

My literature search excluded self-help resources, which turned out to feature

prominently in the participant framework. Future work on conceptualisation of client

factors ought to include research on self-help resources and strategies.

Data limitations: Participant sample

The white, young, mostly European, female participant sample was relatively socially

homogenous. This raises questions about whether people in other demographic groups

might approach psychosocial change using different activity and social factors; or may

experience different obstacles to change. Further studies might use the same general

client factors concept derived from international research and reconstruct it with data

190 of 271

from groups with other socio-cultural identities (insofar as this is possible given the

middle-class western phenomenon of individual talking therapy as socially approved

help). While the heterogeneity of the research literature sample led to some research-

informed elements (client age, education, therapeutic alliance and experiencing

emotions) being cancelled out of the framework due to contradictory findings, no

contradictions arose between participants in participant-led elements, so none were

cancelled out. It might be possible that a more heterogeneous participant sample could

produce contradictory findings that cancel out some participant-informed elements.

It is possible that the homogeneity of the participant group led to a certain loss of

individuality in their voices in the results chapter. There was considerable harmony

across their narratives in terms of the elements and categories their interviews

endorsed. None were found to report factors that were particularly outstanding from the

rest of the cohort.

A further limit in the participant sample is that they all accessed counselling for help with

debilitating and in some cases potentially life-threatening mental health problems. It

could be argued that the ‘recovery trajectory’ these participants followed required a

different kind of work – and thus results in a differently shaped client factors concept,

than had they, for example, sought counselling for personal development.

Limitation in scope: There are doubtlessly other salient client factors that might belong

on the detailed elemental level of the concept hierarchy that this study, given its limited

data pools, did not pick up. Also, at the most abstract level, the key mediating super-

factor ‘reflective and experiential learning’ may have been differently constructed had

the participant sample been comprised of people with learning or communication

difficulties. This study also does not consider macro-system political, economic, legal or

cultural conditions involved in super-factor social resource enrichment or depletion.

These are however extremely important considerations, particularly for counselling

psychology theory, training and practice development, for example in the area of

culturally informed work (Chung and Bemak, 2012). Lastly, this study does not

specifically consider differential client factors that might be at work in group or couple

therapy (see Perkins, 2010 for consideration of client factors in couple therapy).

191 of 271

As all that lives keep movin' on When all is said and done

The breeze come down and touch our domes The raindrops fall on every home

Cause there ain't nothin' set in stone.

Stereo MCs, Breeze, Deep Down & Dirty, 2001

CHAPTER 6: CONCLUSION The conclusion is set out over two sections. In the first, I propose three contributions to

knowledge offered by the concept reconstruction. In the second review the study’s

results in relation to its aims.

6.1 Contributions to knowledge This study contributes a provisional process model of client factors; a tentative

adjustment to psychotherapeutic change process causal theory; and it introduces

concept analysis as a potentially relevant methodology to advance theoretical research

in counselling psychology. Each of these contributions is framed under a rubric of ‘risky

territory’ as I will now explain further.

A provisional process model of client factors

This study’s main argument is that ‘client factors’ is an under-conceptualised term in

psychotherapeutic literature. It requires a coherent theoretical structure to bring forth its

meaning and ensure its grounding in lived experience. Based on my analysis and

interpretation of participant reports and carefully selected research articles, this study

presents a graphic concept framework and description that depicts a system client

factors associated with process and outcome in psychotherapy. The model enhances

the contextual clarity and definitional specificity of the client factors concept. A practical

risk with this model, however, is its size. Its A3 format is not reproducible in any

publication format (as the reader will observe in figures 15, 26 and 27 above). On the

other hand, while the model is provisional due to its numerous limitations, it is unique in

several ways. I touch on four here:

1. It is derived from open participant interviews, rather than questionnaires that

force participants to confirm existing client factor theories (as in Thomas, 2006);

or that ask about corrective experiences of therapy (as in Constantino and

192 of 271

Angus, 2017). The interview questions used here permitted participants to

introduce and discuss factors that were evident and important to them (rather

than factors that are important to therapists). This widened the client factors

scope to include a diverse range of potent social resources that exert influence

beyond therapy, as well as obstacles to achieving desired change.

2. Participant information is complemented with information from published

research to enhance the pragmatic utility of the concept analysis (Penrod, 2007).

Both types of data are considered valuable but in this study the participant data,

as ‘muted group’ evidence (in line with study epistemology), formed the primary

framework template against which the research evidence was compared.

3. Arranged over a conceptual hierarchy, the model offers ‘close up’ ‘midrange’ and

‘overview’ perspectives that could assist with further theorising about key

mechanisms and stages in client change and non-change processes over time.

In particular, it highlights the vital influence of social support before and during

participants’ active change processes. It also highlights the role of social

resource depletion that makes psychosocial change difficult.

4. The model positions therapy and the therapeutic relationship and learning

resources, while client agency within therapy and wider social contexts as the

primary driver of change.

A tentative adjustment to change process causal theory

A second contribution to knowledge is this model’s spotlight on the causal potentialities

of client reflective and exploratory learning within social contexts. This in turn highlights

that the experience of being supported with trust, solidarity and empathy may help

people to make desired psychosocial changes, particularly where these relational

resources are available outside therapy. Without the presence of these extratherapeutic

resources there is a risk that for some clients, the effects of therapy may last only as

long as therapy does.

Causal theory is risky territory for counselling psychology research (whether

quantitative, qualitative, or mixed methods) because from a critical realist ontological

standpoint, human intelligence may be unable to fully grasp causality, and can at best

193 of 271

only interpret its signs through faulty epistemologies and contestable concepts. So,

through doing this study I have come to understand better how, in the words of

philosopher Alan Watts, “your concepts will be attempts to catch water in a sieve”

(1960, no page number). I may be no closer to knowing at the end of this work the

reality of what client factors are, than I was at the beginning. I have however learned a

tremendous amount about the representation of client factors. My learning is illustrated

by the Stereo MCs lyrics that open this chapter: Nothing is set in stone, including the

little flower with its roots deep in the soil in the rock crack, which this study commenced.

We struggle to explain with words the rain and wind of life and relationships.

Introducing concept analysis as a relevant method for counselling psychology

research

Abstract, complex or contested concepts represent epistemological quagmires: poor

places to build a research project. Concepts that are vague and important require

critical examination and meaningful reconstruction as a foundational step for any

research. I have found concept analysis to be a helpful methodological tool for this job.

I am puzzled why, when it enjoys a reputable position in nursing scholarship and social

science (both disciplinarily adjacent to counselling psychology) concept analysis is not

featured on counselling psychology’s standard menu of favoured qualitative research

methodologies (for example, ethnography; empirical phenomenology; grounded theory;

narrative, discourse and conversation analysis; protocol analysis; and thematic analysis,

Creswell, 1998; Elliott and Timulak, 2005; Braun and Clarke, 2013; Finlay, 2015).

Sartori (2009) asserts that because “thinking is language-wrapped” (p.101) concept

analysis “plays a non-replaceable role in the process of thinking.…At no stage of the

methodological argument does…taxonomical unpacking lose weight and importance”

(p.19). I worry that lacking a repertoire of critical concept analysis approaches on the

‘acceptable’ list of counselling psychology research methods we struggle to evolve

concept-based theories in psychological therapy. This in turn risks counselling

psychologists relying on 20th century individualistic theories and therapies that neither

acknowledge nor tackle unequal access to social resources in the aetiology of

psychosocial distress (Vermes, 2017).

194 of 271

Learning surprises

While the results are provisional and limited, the concept analysis process has proven

to be gratifyingly effective for organising and representing the interrelationship of

potentially powerful helpful and unhelpful client factors across planes of abstraction. I

could never have imagined creating the resulting graphics when I started the study.

However, my most important learning is the eye-opening relevance of Bandura’s social

learning theory to my findings. Shockingly, given how long I have been a therapist and

how many advanced trainings I have undertaken, I do not recall reading or being taught

much, if anything, about Bandura’s theories prior to doing this study. Despite my

previous ignorance of his work, the extensive mapping of Bandura’s theories to my

concept framework may point to the enormous influence and absorption of Bandura’s

theories into the zeitgeist of much counselling theory I have read over the past 25 years.

The comprehensiveness and accuracy of the mapping also, for me, indicate that my

framework is reassuringly ‘on the right track’ and makes me determined to read and rely

on Bandura’s work more extensively in future.

6.2 Closing remarks: Results in relation to aims In closing, I return to the concerns with which I opened the study. Bohart and Tallman

(1996, 1999) asserted that the common factor that makes therapies equally effective, is

the client working in a therapeutic relationship. With this study, however, I conclude that

client work outwith the therapeutic relationship largely accounts for therapeutic

outcome. People need generative, long-term and evolving learning relationships with

people, places and cultures, not therapy per se, to be as well as possible. In closing

their chapter Tallman and Bohart (1999) argue that “truly taking seriously the active,

generative nature of how clients change in therapy would fundamentally re-structure

how we view the process of therapy…Finally, researchers would focus their study on

clients rather than on therapists” (p.119). Here, I have proposed that if we are to

account for the active role of the client, we require theories about the mechanisms and

purpose of therapy that centre on client factors. And if this theory is ultimately to

improve therapy, we must not only conceptualise how the active client achieves change,

but also why clients don’t manage to achieve change. This study has produced one

195 of 271

such conceptualisation of the ‘grand theatre’ of how people do, and don’t make desired

psychosocial life-changes, in which psychotherapy and its practitioners are bit players

amongst a large cast of more important characters. Wampold and Imel (2015) describe

“the therapist as an agent of change and patients as active participants” in therapy

(p.34). On the basis of participant narratives in this study I have played with reversing

this proposition: Clients and their social resources are agents of change, while

therapists may be active participants in limited ways for partial aspects of the process.

196 of 271

REFERENCES References with an asterisk indicate those included in the concept analysis. Those with a double asterisk included client follow-up data.

*Aaronson, C. J., Shear, M. K., Goetz, R.R., Allen, L.B., Barlow, D. H. and White, K.S. (2008). Predictors and time course of response among panic disorder patients treated with cognitive-behavioral therapy. The Journal of Clinical Psychiatry, 69(3), 418-424. DOI: 10.4088/JCP.v69n0312.

Addis, M.E. and Cardemil, E.V. (2006). Does manualization improve therapy outcomes? In J.C. Norcross, L.E. Beutler and R.F. Levant (Eds), Evidence-Based Practices in Mental Health (131-160). Washington DC: American Psychological Association. ISBN: 1-59147-290-3.

Adler, J.A., Skalina, L.M. and McAdams, D.P. (2008). The narrative reconstruction of psychotherapy and psychological health. Psychotherapy Research, 18(6), 719-734. DOI: 10.1080/10503300802326020.

Agee, J. (2009). Developing qualitative research questions: A reflective process. International Journal of Qualitative Studies in Education, 22(4), 431-447. DOI: 10.1080/09518390902736512.

Aldiabat, K.M. and LeNavenec, C-L. (2018). Data saturation: The mysterious step in grounded theory method. The Qualitative Report, 23(1), 245-261. Downloaded from https://nsuworks.nova.edu/cgi/viewcontent.cgi?article=2994&context=tqr on 05.01.2019.

Allen, R.E.S. and Wiles, J.L. (2016). A rose by any other name: Participants choosing research pseudonyms. Qualitiative Research in Psychology, 13(2), 149-165. DOI: 10.1080/14780887.2015.1133746.

**Alonso, P., Menchon, J.M., Pifarre, J., Mataix-Cols, D., Torres, L., Salgado, P. and Vallejo, J. (2001). Long-term follow-up and predictors of clinical outcome in obsessive-compulsive patients treated with serotonin reuptake inhibitors and behavioral therapy. Journal of Clinical Psychiatry, 62(7). DOI: 10.4088/JCP.v62n07a06.

Altimir, C., Capella, C., Nunez, L., Abarzua, M. and Krause, M. (2017). Meeting in difference: Revisiting the therapeutic relationship based on patients’ and therapists’ experiences in several clinical contexts. Journal of Clinical Psychology, 73(11), 1510-1522. DOI: 10.1002/jclp.22525.

**Altimir, C., Krause, M., Parra, G. De, Dagnino, P., Tomicic, A., Valdés, N., Perez, J.C., Echavarri, O. and Vilches, O. (2010). Clients’, therapists’, and observers' agreement on the amount, temporal location, and content of psychotherapeutic change and its relation to outcome. Psychotherapy Research, 20(4), 472-487. DOI: 10.1080/10503301003705871.

Amati, F., Banks, C., Greenfield, G. and Green, J. (2017). Predictors of outcome for patients with common mental health disorders receiving psychological therapies in community settings: A systematic review. Journal of Public Health, 39(4), 1-13.

American Psychological Association Task Force on Psychological Intervention Guidelines (1995). Template for developing guidelines: interventions for mental disorders and psychosocial aspects of physical disorders. Washington, DC: American Psychological Association.

Amoakoh-Coleman, M., Agyepong, I.A., Zuithoff, N.P.A., Kayode, G.A., Grobbee, D.E., Klipstein-Grobusch, K., and Ansah, E.K. (2016). Client factors affect provider adherence to clinical guidelines during first antenatal care. PLoS ONE, 11(6) e0157542. DOI: 10.1371/journal.pone.0157542.

Anney, V.N. (2014). Ensuring the quality of the findings of qualitative research: Looking at trustworthiness criteria. Journal of Emerging Trends in Educational Research and Policy Studies, 5(2), 272-281. ISSN: 2141-6990.

Archer, M. (2007). Making Our Way Through The World: Human Reflexivity and Social Mobility. Cambridge: Cambridge University Press. ISBN: 978-0-521-69693-7.

https://nsuworks.nova.edu/cgi/viewcontent.cgi?article=2994&context=tqr

197 of 271

Archer, M. (2014). We are what we love: Love and identity. Accessed on www.socialontology.org on 11.06.2018.

Archer, M., Decoteau, C., Gorski, P., Little, D., Porpora, D., Rutzou, T., Smith, C., Steinmetz, G. and Vendenberghe, F. (2016). What is Critical Realism? Downloaded from http://www.asatheory.org/current-newsletter-online/what-is-critical-realism on 30.05.2018.

Ardener, E. (1975). Belief and the problem of women. In S. Ardener, S. (Ed.) Perceiving Women. London: Malaby Press, 1-17. ISBN: 0-460-14006-x.

Ardener, S. (1975). Introduction. In S. Ardener, S. (Ed.) Perceiving Women. London: Malaby Press, vii-xxiii. ISBN: 0-460-14006-x.

Ardener, S. (2005). Ardener’s ‘Muted groups’: The genesis of an idea and its praxis. Women and Language, 28(2), 50-54.

Ardito, R.B. and Rabellino, D. (2011). Therapeutic alliance and outcome of treatment: Historical excursus, measurements and prospects for research. Frontiers in Psychotherapy, 2:270, 1-11. DOI: 10.3389/fpsyg.2011.00270.

Arnkoff, D.B., Glass, C.R. and Shapiro, D.A. (2002). Expectations and preferences. In In J.C. Norcross (Ed.), Psychotherapy Relationships That Work: Therapist Contributions and Responsiveness to Patients (335-356). Oxford: Oxford University Press. ISBN: 978-0-19-514346-1.

Asay, T.P. and Lambert, M.J. (1999). The empirical case for the common factors in therapy: Quantitative findings. In M.A. Hubble, B.L. Duncan and S.D. Miller (Eds), The Heart and Soul of Change: What Works in Therapy (23-55). Washington, D.C: American Psychological Association. ISBN: 1-55798-557-X.

Aveyard, H., Payne, S. and Preston, N. (2016). A Post-Graduate’s Guide to Doing a Literature Review in Health and Social Care. Maidenhead: Open University Press. ISBN: 978-0-33-526368-4.

Bachelor, A. (1995). Clients’ perception of the therapeutic alliance: A qualitative analysis. Journal of Counseling Psychology, 42(3), 323-337. DOI: 10.1037/0022-0167.42.3.323.

Baker, S.E. and Edwards, R. (2012). How many qualitative interviews is enough? National Centre for Research Methods. DOI: 10.1177/1525822X05279903.

Baldwin, S.A., Wampold, B.E. and Imel, Z.E. (2007). Untangling the alliance-outcome correlation: Explaining the relative importance of therapist and patient variability in the alliance. Journal of Consulting and Clinical Psychology, 75(6), 842-852. DOI: 10.1037/0022-006X.75.6.842.

Bandura, A. (1961). Psychotherapy as a learning process. Psychological Bulletin, 58(2). 143-159.

Bandura, A. (1978). The self system in reciprocal determinism. American Psychologist, 33(4), 344-358. Downloaded from https://www.uky.edu/~eushe2/Bandura/Bandura1978AP.pdf on 02.08.2018.

Bandura, A. (1986). Social Foundations of Thought and Action. Englewood Cliffs, NJ: Prentice-Hall. ISBN: 0-13-815614-X.

Bankoff, E.A. (1996). Pre-treatment social support and effective psychotherapy process: A panel study. Psychotherapy: Theory, Research, Practice, Training, 33(1), 51-60. ISSN: 00333204.

Barbour, R.S. (2001). Checklists for improving rigour in qualitative research: A case of the tail wagging

the dog? British Medical Journal, 322. DOI: 10.1136/bmj.322.7294.1115.

Barkham, M. (2002). Methods, outcomes and processes in the psychological therapies across four successive research generations. In W. Dryden (Ed.), Dryden’s Handbook of Individual Therapy (4th Ed.). London: Sage, 373-433. ISBN: 978-0761969433.

http://www.socialontology.org/

http://www.asatheory.org/current-newsletter-online/what-is-critical-realism

http://www.asatheory.org/current-newsletter-online/what-is-critical-realism

https://dx.doi.org/10.3389%2Ffpsyg.2011.00270

https://www.uky.edu/~eushe2/Bandura/Bandura1978AP.pdf

https://doi.org/10.1136/bmj.322.7294.1115

198 of 271

**Barnett, N.P., Apodaca, T.R., Magill, M., Colby, S.M., Gwaltney, C., Rohsenow, D. J. and Monti, P.M. (2010). Moderators and mediators of two brief interventions for alcohol in the emergency department. Addiction, 105(3), 452-465. DOI: 10.1111/j.1360-0443.2009.02814.x.

Baron, R.M. and Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. DOI: 10.1037/0022-3514.51.6.1173.

Beaumont, E., Galpin, A. and Jenkins, P. (2012). Being kinder to myself. Counselling Psychology Review, 27(1), 31-43. ISSN: 0269-6975.

Beck, J. (2013). Powerful knowledge, esoteric knowledge, curriculum knowledge. Cambridge Journal of Education, 43(2), 177-193. DOI: 10.1080/0305764X.2013.767880.

Beckwith, S., Dickinson, A. and Kendall, S. (2008). The “con” of concept analysis: A discussion paper which explores and critiques the ontological focus, reliability and antecedents of concept analysis frameworks. International Journal of Nursing Studies, 45(12), 1831-1841. DOI: 10.1016/j.ijnurstu.2008.06.011.

Bedi, R., Davis, M. and Williams, M. (2005). Critical incidents in the formation of the therapeutic alliance from the client’s perspective. Psychotherapy: Theory, Research, Practice, Training, 42(3), 311-323. DOI: 10.1037/0033-3204.42.3.311.

*Beevers, C.G., Wells, A. T. and Miller, I.W. (2007). Predicting response to depression treatment: The role of negative cognition. Journal of Consulting and Clinical Psychology, 75(3), 422-431. DOI: 10.1037/0022-006X.75.3.422.

Beitel, M., Hutz, A., Sheffield, K., Gunn, C., Cecero, J., and Barry,D. (2009). Do psychologically minded clients expect more from counselling? Counselling and Psychotherapy: Theory, Research and Practice, 82, 369-383. DOI: 10.1348/147608309X436711.

Benish, S.G., Quintana, S. and Wampold, B.E. (2011). Culturally adapted psychotherapy and the legitimacy of myth: A direct-comparison meta-analysis. Journal of Counseling Psychology, 58(3), 279-289. DOI: 10.1037/a0223626.

*Berking, M., Neacsiu, A., Comtois, K.A. and Linehan, M.M. (2009). The impact of experiential avoidance on the reduction of depression in treatment for borderline personality disorder. Behavior Research and Therapy, 47(8), 663-670. DOI: 10.1016/j.brat.2009.04.011.

Bernd, R. and Strouse, J. (2008). Influence of social support on success of therapeutic interventions: A meta-analytic review. Psychotherapy: Theory, Research, Practice and Training, 45(4), 464-476. DOI: 10.1037/a0014333.

Beutler, L.E. (1983). Eclectic Psychotherapy: A Systematic Approach. New York: Pergamon. ISBN: 978-0-08028-842-0.

Beutler, L.E. (1991). Have all won and must all have prizes? Revisiting Luborsky et. al.’s verdict. Journal of Consulting and Clinical Psychology, 59(2), 226-232. ISSN: 0022-006X.

Beutler, L.E. (2016). Selecting the most appropriate treatment for each patient. Accessed at https://www.youtube.com/watch?v=BpmKDqpbLzo on 15.09.2017.

Beutler, L.E., & Clarkin, J.F. (1999). Systematic treatment selection: Toward targeted therapeutic interventions. New York, NY: Routledge. ISBN: 978-0-87630-576-8.

Beutler, L.E., Clarkin, J.F. and Bongar, B. (2000). Guidelines for the systematic treatment of the depressed patient. New York: Oxford University Press. ISBN: 978-0756763800.

Beutler, L.E. and Consoli, A. (1992). Systematic Eclectic Psychotherapy. In J.C. Norcross and M.R. Goldfried (Eds). Handbook of Psychotherapy Integration (264-299). New York: Basic Books

https://www.youtube.com/watch?v=BpmKDqpbLzo

199 of 271

Beutler, L.E. and Forrester, B. (2014). What needs to change: Moving from “research informed” to “empirically effective” practice. Journal of Psychotherapy Integration, 24, 168-177. DOI:10.1037/a0037587.

Beutler, L.E. and Harwood (2000). Prescriptive psychotherapy: A Practical Guide to Systematic Treatment Selection. New York: Oxford University Press. ISBN: 978-0-19513-669-2.

Beutler, L.E., Harwood, M., Kimpara, S., Verdirame, D. and Blau, K. (2011a). Coping Style. Journal of Clinical Psychology, 67(2), 176-183. DOI: 10.1002/jclp.20752.

Beutler, L.E., Harwood, M., Michelson, A., Song, X. and Holman, J. (2011b). Resistance/reactance level. Journal of Clinical Psychology, 67(2), 135-142. DOI: 10.1002/jclp.20753.

Beutler, L. E., Patterson, K. M., Jacob, T., Shoham, V., Yost, E., & Rohrbaugh, M. (1994). Matching treatment to alcoholism subtypes. Psychotherapy: Theory, Research, Practice, Training, 30, 463– 472. DOI:10.1037/0033-3204.30.3.463

Beutler, L., Someah, K., Kimpara, S. and Miller, K. (2016). Selecting the most appropriate treatment for each patient. International Journal of Clinical and Health Psychology, 16, 99-108. DOI: 10.1016/j.ijchp.2015.08.001.

Bhaskar, R. (2008). A Realist Theory of Science. London: Routledge. ISBN: 978-0-203-89263-3.

Bhaskar, R. (2014). Critical realism. Accessed on https://www.youtube.com/watch?v=TO4FaaVy0Is on 24.08.2017.

Bickman, L. and Salzer, M. S. (1996). Dose-response, disciplines, and self-help: Policy implications of Consumer Reports findings. Paper presented as part of a symposium on ‘Consumer Reports Mental Health Survey Results: Practice and Policy Implications’, 104th Annual Convention of the American Psychological Association, Toronto, Canada, August 1996.

Billig, M. (1999). Whose terms? Whose ordinariness? Rhetoric and ideology in conversation analysis. Discourse and Society, 10(4), 543-558. Downloaded from http://www.jstor.org/stable/42888872 on 03.06.2017.

Binder, P.E., Holgersen, H. and Nielsen, G.H. (2009). Why did I change when I went to therapy? A qualitative analysis of former patients’ conceptions of successful psychotherapy. Counselling and Psychotherapy Research, 9(4), 250-256. DOI: 10.1080/14733140902898088.

Binder, P.E., Holgersen, H. and Nielsen, G.H. (2010). What is a ‘good outcome’ in psychotherapy? A qualitative exploration of the former patient’s point of view. Psychotherapy Research, 20(3), 285-294. DOI: 10.1080/10503300903376338.

**Binford, R.B., Mussell, M.P., Crosby, R.D., Peterson, C.B., Crow, S.J, and Mitchell, J.E. (2005). Coping strategies in bulimia nervosa treatment : Impact on outcome in group cognitive – behavioral therapy. Journal of Consulting and Clinical Psychology, 73(6), 1089-1096. DOI: 10.1037/0022-006X.73.6.1089.

Blumer, H. (1954). What is wrong with social theory? American Sociology Review, 18, 3-10.

Bohart, A.C. (2004). How do clients make empathy work? Person-Centred and Experiential Psychotherapies, 3(2), 102-116. DOI: 10.1080/14779757.2004.9688336.

Bohart, A.C. (2007). Insight and the active client. In L.G. Castonguay and C.E. Hill (Eds), Insight in Psychotherapy (257-277). Washington, DC: American Psychological Association. ISBN: 978-1-59147-477-7.

Bohart, A.C. and Greenberg, L.S. (1997). Empathy Reconsidered: New Directions in Psychotherapy. Washington DC: American Psychological Association. ISBN: 978-1557984104.

https://doi.org/10.1016/j.ijchp.2015.08.001

https://doi.org/10.1016/j.ijchp.2015.08.001

https://www.youtube.com/watch?v=TO4FaaVy0Is

http://www.jstor.org/stable/42888872

http://psycnet.apa.org/doi/10.1080/14779757.2004.9688336

200 of 271

Bohart, A.C. and Tallman, K.L. (2010). Clients: The neglected common factor. In B. Duncan, S.D. Miller, B.E. Wampold and M.A. Hubble (Eds). The Heart and Soul of Change: Delivering What Works in Therapy (2nd Ed; 83-111). Washington, DC: American Psychological Association. ISBN: 978-1433807091.

Bohart, A.C. and Tallman, K.L. (1996). The active client: Therapy as self-help. Journal of Humanistic Psychology, 36, 7-30. DOI: 10.1177/00221678960363002.

Bohart, A.C. and Tallman, K.L. (1999). How Clients Make Therapy Work: The Process of Active Self-Healing. Washington, DC: American Psychological Association. ISBN: 978-1557985712.

Bohart, A.C., Tallman, K.L., Byock, G. and Mackrill, T. (2011). The “Research Jury” Method: The application of the jury trial model to evaluating the validity of descriptive and causal statements about psychotherapy process and outcome. Pragmatic Case Studies in Psychotherapy, 7(1), 101-144. Accessed from https://ejbe.libraries.rutgers.edu/index.php/pcsp/article/viewFile/1075/2524 on 29.08.2018.

Bohart, A. and Wade, A.G. (2013). The client in psychotherapy. In M.J. Lambert (Ed.) Bergin and Garfield’s Handbook of Psychotherapy and Behaviour Change (6th Edition, 219-257). Hoboken, NJ: Wiley. ISBN 978-1-118-03820-8.

Boorse, C. (1976). What a theory of mental health should be. Journal for the Theory of Social Behavior, 6(1), 61-84. DOI: 10.1111/j.1468-5914.1976.tb00359.x.

Bordin, E.S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research and Practice, 16, 252-260. DOI: 10.1037/h0085885.

Borkovec, T.D., Mathews, A.M., Chambers, A, Ebrahimi, S, Lytle, R and Nelson, R. (1987). The effects of relaxation training with cognitive or nondirective therapy and the role of relaxation-induced anxiety in the treatment of generalized anxiety. Journal of Consulting and Clinical Psychology, 55, 883-888.

Borkovec, T.D., Newman, M.G., Pincus, A.L. and Lytle, R. (2002). A component analysis of cognitive-behavioral therapy for generalized anxiety disorder and the role of interpersonal problems. Journal of Consulting and Clinical Psychology, 70, 288-298. DOI: 10.1037//0022-006X.70.2.288.

Borland, K. (1991). That’s not what I said: Interpretive conflict in oral narrative research. In S.B. Gluck and D. Patai (Eds), Womens’ Words: The Feminist Practice of Oral History. New York: Routledge, 63-75. ISBN: 978-0415903721.

Boucher, T. (2006). In cognitive therapy, how would the therapist understand and work with difficulties that arise in the therapeutic relationship. Counselling Psychology Review, 21(3), 12-18. ISSN: 0269-6975.

Brady, J.P. (1967). Psychotherapy, learning theory and insight. Archives of General Psychiatry, 16, 304-311.

Braun, V. and Clarke, V. (2013). Successful Qualitative Research. London: Sage. ISBN: 978-1-84787-582-2.

Braun, V. and Clarke, V. (2014). What can “thematic analysis” offer health and wellbeing researchers? International Journal of Qualitative Studies on Health and Wellbeing, 9. DOI: 10.3402/ghw.v9.26152.

British Psychological Society (2014). Code of Human Research Ethics. Downloaded from https://www.bps.org.uk/sites/bps.org.uk/files/Policy%20-%20Files/BPS%20Code%20of%20Human%20Research%20Ethics.pdf on 29.05.2018.

Brod, M., Tesler, L. E., & Christiansen, T. L. (2009). Qualitative research and content validity: Developing best practices based on science and experience. Quality of Life Research, 18(9), 1263-1278. doi:10.1007/s11136-009-9540-9.

Bronfenbrenner, U. (1967). The psychological costs of quality and inequality in education. Child Development, 38, 909-925.

https://ejbe.libraries.rutgers.edu/index.php/pcsp/article/viewFile/1075/2524

https://www.bps.org.uk/sites/bps.org.uk/files/Policy%20-%20Files/BPS%20Code%20of%20Human%20Research%20Ethics.pdf

https://www.bps.org.uk/sites/bps.org.uk/files/Policy%20-%20Files/BPS%20Code%20of%20Human%20Research%20Ethics.pdf

201 of 271

Bronfenbrenner, U. (1993). The ecology of cognitive development: Research findings and fugitive findings. In R. Wonziak and K. Fisher (Eds). Development in Context: Acting and Thinking in Specific Environments (3-44). Hillsdale, NJ: Erlbaum.

Bronfenbrenner, U. (2005). The ecobiological theory on human development. In U. Bronfenbrenner (Ed). Making Human Beings Human: Bioecological Perspectives on Human Development (1-15). London: Sage.

Brown, L.S. (1994). Subversive Dialogues: Theory in Feminist Therapy. New York: Basic Books. ISBN: 978-0-46508-321-3.

*Brown, M.Z., Linehan, M.M., Comtois, K.A., Murray, A. and Chapman, A.L. (2009). Shame as a prospective predictor of self-inflicted injury in borderline personality disorder: A multi-modal analysis. Behaviour Research and Therapy, 47(10), 815-822. DOI: 10.1016/j.brat.2009.06.008.

Brunner, R. (2017). Why do people with mental distress have poor social outcomes? Four lessons from the capabilities approach. Social Science and Medicine, 191, 160-167. DOI: 10.1016/j.socscimed.2017.09.016 0277.

Budge, S.L. and Wampold, B.E. (2015). The relationship: How it works. In O.C.G. Gelo, O. Pritz and B. Rieken (Eds). Psychotherapy Research (213-228). DOI: 10.1007/978-3-7091-1382-1_11.

Burles, M.C. (2017). Negotiating post-research encounters: Reflections on learning from participant deaths following a qualitative study. Mortality, 22(2), 170-180. DOI: 10.1080/13576275.2017.1291605.

*Burns, D.D. and Nolen-Hoeksema, S. (1991). Coping styles, homework compliance, and the effectiveness of cognitive-behavioral therapy. Journal of Consulting and Clinical Psychology, 59(20), 305-311. DOI: 10.1037/0022-006X.59.2.305.

Brush, B.L., Kirk, K., Gultekin, L. and Baiardi, J.M. (2011). Overcoming: A concept analysis. Nursing Forum, 46(3), 160-168. DOI: 10.1111/j.1744-6198.2011.00227.x

Carey, T.A., Carey, M., Stalker, K., Mullan, R.J., Murray, L.K. and Spratt, M.B. (2007). Psychological change from the inside looking out: A qualitative investigation. Counselling and Psychotherapy Research, 7(3), 178-187. DOI: 10.1080/14733140701514613.

Caron, C.D. and Bowers, B.J. (2000). Methods and application of dimensional analysis: A contribution to concept and knowledge development in nursing. In B.L. Rodgers and K.A. Knafl (Eds). Concept Development in Nursing: Foundations, Techniques and Applications (2nd Edition, 285-319). Philadelphia, PA: Saunders/Elsevier. ISBN: 978-0-7216-8243-3.

Cartwright, D.A. and Roth, I. (1957). Success and satisfaction in psychotherapy. Journal of Clinical Psychology, 13(1), 20-26. ISSN: 0021-9762.

Carver, C.S., Scheier, M.F. and Weintraub, J.K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56 (2), 267-283. DOI: 10.1037/0022-3514.56.2.267. Castonguay, L.G. (1993). ‘Common factors’ and ‘non-specific variables’: Clarification of the two concepts and recommendations for research. Journal of Psychotherapy Integration, 3, 267-286. DOI: 10.1037/h0101171. Castonguay, L.G. (2017). How and Why are Some Therapists Better Than Others? Understanding Therapist Effects. Washington, DC: American Psychological Association. ISBN: 978-1-433-82771-6.

Castonguay, L.G. and Beutler, L.E. (2006). Common and unique principles of therapeutic change: what do we know and what do we need to know? In L.G. Castonguay and L.E. Beutler (Eds.), Principles of Therapeutic Change that Work. New York: Oxford University Press, 353-370. ISBN: 978-0-19-515684-3.

http://psycnet.apa.org/doi/10.1037/0022-3514.56.2.267


http://psycnet.apa.org/doi/10.1037/h0101171

202 of 271

Castonguay, L.G., Boswell, J.F., Zack, S.E., Baker, S., Boutselis, M.A., Chiswick, N.R., Damer, D.D., Hemmelstein, N.A., Jackson, J. S., Morford, M., Ragusea, S.A., Roper, J.G., Spayd, C, Weiszer, T., Borkovec, T.D., Grosse Holtforth, M. (2010). Helpful and hindering events in psychotherapy: a practice research network study. Psychotherapy: Theory, Research, Practice and Training, 47, 327-344. DOI: 10.1037/a0021164.

*Castonguay, L.G., Goldfried, M.R., Wiser, S., Raue, P.J., and Hayes, A.M. (1996). Predicting the effect of cognitive therapy for depression: A study of unique and common factors. Journal of Consulting and Clinical Psychology, 64(3), 497-504. DOI: 10.1037/0022-006X.64.3.497.

*Chambless, D. L., Milrod, B., Porter, E., Gallop, R., McCarthy, K. S., Graf, E., Rudden, M., Sharpless, B.A. and Barber, J.P. (2017). Prediction and moderation of improvement in cognitive-behavioral and psychodynamic psychotherapy for panic disorder. Journal of Consulting and Clinical Psychology, 85(8), 803-813, DOI: 10.1037/ccp0000224.

Chinn, P. and Jacobs, M. (1987). Theory and Nursing: A Systematic Approach (2nd Ed). St Louis: C.V. Mosby. ISBN: 978-0-80-160988-6.

Chinn, P.L. and Kramer, M.K. (1994). Theory and Nursing: A Systematic Approach (3rd Ed.). St Louis MO: Mosby Year Book. ASIN: B00IF1YR96

Chung, R.C. and Bemak, F.P. (2012). Social Justice Counseling. London: Sage. ISBN: 978-1-4129-9952-6.

Claes, C., Van Hove, G., Vandevelde, S., van Loon, J., Schalock, R. (2012). The influence of supports strategies, environmental factors and client characteristics on quality of life related personal outcomes. Research in Developmental Disabilities, 33, 96-103. DOI: 10.1016/j.ridd.2011.08.024.

*Clarke, H., Rees, A. and Hardy, G.E. (2004). The big idea: clients' perspectives of change processes in cognitive therapy. Psychology and Psychotherapy: Theory, Research and Practice, 77, 67-89. DOI: 10.1348/147608304322874263.

Clarkin, J.E. and Levy, K.N. (2004). The influence of client variables on Psychotherapy. In M.J. Lambert (Ed.), Bergin and Garfield’s Handbook of Psychotherapy and Behaviour Change (5th Edition, 194-226). New York: Wiley. ISBN: 0-471-37755-4.

Cochran, B.N. and Cauce, A.M. (2006). Characteristics of lesbian, gay, bisexual and transgender individuals entering substance abuse treatment. Journal of Substance Abuse Treatment, 30(2), 135-146. DOI: 10.1016/j.jsat.2005.11.009.

Coleman, R.A. and Neimeyer, R.A. (2015). Assessment of subjective client agency in psychotherapy: A review. Journal of Constructivist Psychology, 28(1), 1-23. DOI> 10.1080/10720537.2014.939791.

Constantino, M.J. and Angus, L. (2017). Clients’ retrospective accounts of corrective experiences in psychotherapy: An international multisite collaboration. Journal of Clinical Psychology, 73(2), 153-167. DOI: 10.1002/jclp.22427.

Constantino, M.J., Angus, L., Friedlander, M.L., Messer, S.B. and Moertl, K. (2011a). Patients’ perceptions of corrective experiences in individual therapy. Unpublished interview protocol.

Constantino, M.J., Arnkoff, D.B., Glass, C.R., Ametrano, R.M. and Smith, J.Z. (2011). Expectations. Journal of Clinical Psychology, 67(2), 184-192. DOI: 10.1002/jclp.20754.

*Constantino, M.J., Morrison, N.R., Coyne, A.E., Goodwin, B.J., Santorelli, G.D. and Angus, L. (2017). Patients' perceptions of corrective experiences in naturalistically delivered psychotherapy. Journal of Clinical Psychology, 73(2), 139-152. DOI: 10.1002/jclp.22428.

Collier, D. and Gerring, J. (2009). Introduction. In D. Collier and J. Gerring (Eds), Concepts and Method in Social Science. London: Routledge (1-10). ISBN: 978-0-415-77578-6.

https://doi.org/10.1002/jclp.22427

203 of 271

Cooper, M. (2008). Essential Research Findings in Counselling and Psychotherapy: The Facts are Friendly. London: Sage. ISBN: 978-1847870438.

Cooper, M. and McLeod, J. (2007). A pluralistic framework for counselling and psychotherapy: Implications for research. Counselling and Psychotherapy Research, 7 (3). pp. 135-143. DOI: 10.1080/14733140701566282.

Corrie, S. and Callahan, M.M. (2000). A review of the scientist-practitioner model: Reflections on its potential contributions to counselling psychology within the context of current health care needs. British Journal of Medical Psychology, 73, 413-427. DOI: 10.1348/000711200160507.

Cox, C.L. (1982). An interaction model of client health behaviour: Theoretical prescription for nursing. Advances in Nursing Science. 5(1), 41-56.

Creswell, J.W. (1998). Qualitative Inquiry and Research Design: Choosing Among Five Traditions. London: Sage. ISBN: 0-7619-0144-2.

Crits-Christoph, P. (1997). Limitations of the dodo bird verdict and the role of clinical trials in psychological research: Comment on Wampold et.al. (1997). Psychological Bulletin, 122(3), 216-220. DOI: 10.1037/2F0033-2909.122.3.216.

Crossley, N. (2011). Towards Relational Sociology. Abingdon, Oxford: Routledge. ISBN: 978-0-415-48014-7.

Crossley, N. (2015). Relational sociology and culture: A preliminary framework. International Review of Sociology, 25(1), 65-85. DOI: 10.1080/03906701.2014.997965.

Dallos, R. and Stedmon, J. (2009). Reflective Practice in Psychotherapy and Counselling. Open University Press. ISBN: 978-0335233618.

*Davidson, L., Borg, M., Marin, I., Topor, A. and Sells, D. (2005). Processes of recovery in serious mental Illness: Findings from a multinational study. American Journal of Psychiatric Rehabilitation, 8(3), 177-201. DOI: 10.1080/15487760500339360.

Davidson, J. R. T., Edna, B., Hupperty, J. D., Keefe, F. J., Franklin, M. E., Compton, J. S., Zhao, N., Connor, K.M., Lynch, T.R. and Gadde, K. S. (2004). Fluoxetine, comprehensive cognitive behavioral therapy, and placebo in generalized social phobia. Archives of General Psychiatry, 61, 1005–1013. doi:10.1001/archpsyc.61.10.1005

Davis, M.H. (1996). Empathy: A Social Psychological Approach. New York: Routledge. ISBN: 978-0-8133-3001-3.

Davis-Mccabe, C. and Winthrop, A. (2010). Computerised CBT: University students’ experiences of using an online self-help programme. Counselling Psychology Review, 25(4), 46-55. ISSN: 0269-6975.

De Hoog, N., Bolman, C., Bernt, N., Kers, E., Mudde, A., De Vries, H. and Lechner, L. (2016). Smoking cessation in cardiac patients: the influence of action plans, coping plans and self-efficacy on quitting smoking. Health Education Research, 31(3), 350-362. DOI:10.1093/her/cyv100.

Department of Health (2018). The Common Law Duty of Confidence. Accessed at https://www.health-ni.gov.uk/articles/common-law-duty-confidentiality on 29.05.2018.

DeRubeis, R., Gelfand, L.A., German, R.E., Fournier, J.C. and Forand, N.R. (2014). Understanding processes of change: How some patients reveal more than others – and some groups of therapists less – about what matters in psychotherapy. Psychotherapy Research, 24(3), 419-428. DOI: 10.1080/10503307.2013.838654.

**Dingemans, A.E., Spinhoven, P. and van Furth, E.F. (2007). Predictors and mediators of treatment outcome in patients with binge eating disorder. Behaviour Research and Therapy, 45(11), 2551-2562. DOI: 10.1016/j.brat.2007.06.003.

https://www.health-ni.gov.uk/articles/common-law-duty-confidentiality%20on%2029.05.2018

https://www.health-ni.gov.uk/articles/common-law-duty-confidentiality%20on%2029.05.2018

204 of 271

Donald, I. and Carey, T. (2017). Improving knowledge about the effectiveness of psychotherapy. Psychotherapy and Politics International, e.1424. DOI: 10.1002/ppi1424.

Donati, P. (2016). The ‘relational subject’ according to a critical realist relational sociology. Journal of Critical Realism, 15(4), 352-375. DOI: 10.1080/14767430.2016.1166728.

Donovan, D.B. and Rosengren, D.B. (1999). Motivation for behavior change and treatment among substance abusers. In J.A. Tucker, D.M. Donovan and A.G. Marlett (Eds), Changing Addictive Behavior: Bridging Clinical and Public Service Strategies (127-159). New York: Guilford. ISBN: 978-1572306776.

Douglas, B., Woolfe,R., Strawbridge, S., Kasket E. and Galbraith, V. (Eds.), The Handbook of Counselling Psychology (4th Ed, 169-184). London: Sage. ISBN: 978-1-4462-7632-7.

**Dow, M.G.T., Kenardy, J.A., Johnston, D.W., Newman, M.G., Taylor, C.B. and Thompson, A. (2007). Prognostic indices with brief and standard CBT for panic disorder: II. Moderators of outcome. Psychological Medicine, 37(10), 1503-1509. DOI: 10.1017/S0033291707000682.

Drapela, V.J. (1990). The value of theories for counselling practitioners. International Journal for the Advancement of Counselling, 13, 19-26. DOI: 10.1007%2FBF00154639.

Duncan, B., Miller, S.D., Wampold, B.E. and Hubble, M. (2010). The Heart and Soul of Change: Delivering What Works in Therapy (2nd Ed). Washington, DC: American Psychological Association. ISBN: 978-1433807091.

Duncan, B., Miller, S.D. and Sparks, J.A. (2004). The Heroic Client. San Francisc, CA: Jossey Bass. ISBN: 0-7879-7240-1.

Duncan, C., Cloutier, J.D. and Bailey, P.H. (2007). Concept analysis: The importance of differentiating the ontological focus. Journal of Advanced Nursing, 58(3), 293-300. DOI: 10.1111/j.1365-2648.2007.04277.x.

Eacott, S. (2018). Beyond Leadership: A Relational Approach to Organizational Theory in Education. London: Springer. ISBN: 978-981-10-6568-2.

*Ekeblad, A, Falkenström F. and Holmqvist R. (2016). Reflective functioning as predictor of working alliance and outcome in the treatment of depression. Journal of Consulting and Clinical Psychology, 84(1), 67-78. DOI: 10.1037/ccp0000055.

Elkins, A. (2009). The medical model in psychotherapy: Its limitations and failures. Journal of Humanistic Psychology, 49(1), 66-84. DOI: 10.1177/0022167807307901.

Elliot, A.J. and Church, M.A. (2002). Client-articulated avoidance goals in the therapy context. Journal of Counseling Psychology, 49, 243-254. DOI: 10.1037//0022-0167.49.2.243.

Elliott, R. (1985). Helpful and non-helpful events in brief counselling interviews: An empirical taxonomy. Journal of Counseling Psychology, 32, 307-322. Elliott, R. (1996). Client change interview schedule. Unpublished research instrument. Department of Psychology, Toledo University.

Elliott, R. (2010). Psychotherapy change process research: Realizing the promise. Psychotherapy Research, 20(2), 123-135. DOI: 10.1080/10503300903470743.

Elliott, R., Fischer, C.T. and Rennie, D.L. (1999). Evolving guidelines for publication of qualitative research studies in psychology and related fields. British Journal of Clinical Psychology, 38(3), 215-229. DOI: 10.1348/014466599162782.

Elliott, R. and Timulak, L. (2005). Descriptive and interpretive approaches to qualitative research. In J. Miles and P. Gilbert (Eds.), A Handbook of Research Methods for Clinical and Health Psychology. Cary, N.C: Oxford University Press USA, 147-159. ISBN: 978-0198527565.

https://www.ncbi.nlm.nih.gov/pubmed/?term=Ekeblad%20A%5BAuthor%5D&cauthor=true&cauthor_uid=26594944

https://www.ncbi.nlm.nih.gov/pubmed/?term=Falkenstr%C3%B6m%20F%5BAuthor%5D&cauthor=true&cauthor_uid=26594944

205 of 271

Erlingsson, C. and Brysiewicz, P. (2017). A hands-on guide to doing content analysis. African Journal of Emergency Medicine, 7, 93-99. DOI: 10.1016/j.afjem.2017.08.001 2211-419X.

Farley, L. (2017). The replacement of child figure: An allegory for scholars of education, psychology and childhood. Critical Discourses in the Academy Seminar Series, University of Manchester, 08.06.2017.

Fauconnier, G. and Turner, M. (1998). Conceptual integration networks. Downloaded from http://www.cogsci.ucsd.edu/~faucon/BEIJING/CIN.pdf on 29.08.2018.

Feltham, C. (2013). Counselling and Counselling Psychology: A Critical Examination. Ross-on-Wye: PCCS Books. ISBN: 978-1-906254-58-2.

Finlay, L. (2015). Qualitative methods. In A. Vossler and N. Moller (Eds). The Counselling and Psychotherapy Research Handbook. London: Sage. ISBN: 978-1-4462-5527-8.

Fivush, R. (2000). Accuracy, authority and voice: Feminist perspectives on autobiographical memory. In P.H. Miller and E.K. Scholnick (Eds), Towards a Feminist Developmental Psychology (85-105). New York: Routledge. ISBN: 0-415-92177-5.

Fleetwood, S. (2004). The ontology of organisation and management studies. Downloaded from http://eprints.uwe.ac.uk/16278/2/Download.pdf on 15.09.2017.

Fletcher, A.J. (2017). Applying critical realism in qualitative research: Methodology meets method. International Journal of Social Research Methodology, 20(2), 181-194. DOI: 10.1080/13645579.2016.1144401.

Forchuk, C. (1994). The orientation phase of the nurse-client relationship: testing Peplau’s theory. Journal of Advanced Nursing, 20(3), 532-537.

Forsyth, G.L. (1980). Analysis of the concept of empathy: Illustration of one approach. Advances in Nursing Science, 2(2), 33-42. ISSN: 0161-9268.

Fougere, A., Daffern, M. and Thomas, S. (2012). Towards an empirical conceptualisation of resilience in young adult offenders. Journal of Forensic Psychiatry and Psychology, 23(5-6), 706-721. DOI: 10.1080/14789949.2012.719536.

*Fournier, J.C., De Rubeis, R.J., Shelton, R.C., Hollon, S.D., Amsterdam, J.D. and Gallop, R. (2009). Prediction of response to medication and cognitive therapy in the treatment of moderate to severe depression. Journal of Consulting and Clinical Psychology, 77(4), 775-787. DOI: 10.1037/a0015401.

Frank, J.D. (1973). Persuasion and Healing: A Comparative Study (2nd Ed.). Baltimore, MD: Johns Hopkins University Press. ISBN: 978-0801814433.

Frank, J.D. and Frank, J.B. (1993). Persuasion and Healing: A Comparative Study (3rd Ed.). Baltimore, MD: Johns Hopkins University Press, ISBN: 978-0-801-84636-6.

Frazier, P.A., Tix, A.P. and Barron, K.E. (2004). Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology, 51(1), 115-134. DOI: 10.1037/0022-0167.51.1.115.

Fredericks, M. and Miller, S.I. (1987). The use of conceptual analysis for teaching sociology courses. Teaching Sociology,15(4), 392-398. DOI: 10.2307/1317995.

Fusch, P.I. and Ness, L.R. (2015). The Qualitative Report, 20(9), 1408-1416. Downloaded from https://cpb-us-e1.wpmucdn.com/sites.nova.edu/dist/a/4/files/2015/09/fusch1.pdf on 05.09.2019.

Garfield, S.L. (1971). Research on client variables in psychotherapy. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of Psychotherapy and Behaviour Change: An Empirical Analysis. New York: Wiley.

http://www.cogsci.ucsd.edu/~faucon/BEIJING/CIN.pdf

http://eprints.uwe.ac.uk/16278/2/Download.pdf%20on%2015.09.2017

https://cpb-us-e1.wpmucdn.com/sites.nova.edu/dist/a/4/files/2015/09/fusch1.pdf

206 of 271

Garfield, S.L. (1978). Research on client variables in psychotherapy. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of Psychotherapy and Behaviour Change (2nd Edition, 191-232). New York: Wiley. ISBN: 978-0-471-06968-3.

Garfield, S.L. (1994). Research on client variables in psychotherapy. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of Psychotherapy and Behaviour Change (3rd Edition, 190-228). New York: Wiley. ISBN: 0-471-54513-9.

Garfield, S.L. (1995). Psychotherapy: An Eclectic-Integrative Approach (2nd Ed.). Hoboken, NJ: Wiley. ISBN: 978-0471595564.

Gaston, L. (1990). The concept of the alliance and its role in psychotherapy: Theoretical and empirical considerations. Psychotherapy: Theory, Research, Practice, Training, 27(2), 143-153. DOI: 10.1037/2F0033-3204.27.2.143.

Gelso, C.J. (2014). A tripartite model of the therapeutic relationship: Theory, research, practice. Psychotherapy Research, 24(2), 117-131. DOI: 10.1080/10503307.2013.845920.

Gelso, C.J. and Fretz, B.R. (1992). Counseling Psychology. Fort Worth, TX: Harcourt-Brace. ISBN: 0-03-027858-9.

Gergen, K.J. (2009). Relational Being: Beyond Self and Community. New York: Oxford Universtiy Press. ISBN: 978-0-19-984626-9.

Gergen, K.J. (2015). Towards a relational humanism. Journal of Humanistic Counselling, 54(2), 149-165. DOI: 10.1002/johc.12009.

Gerring, J. (1997). Ideology: A definitional analysis. Political Research Quarterly, 50(4), 957-994. DOI: 10.2307/448995.

Gerring, J. (1999). What makes a concept good? A criterial framework for understanding concept formation in the social sciences. Polity, 31(3), 357-393. Downloaded from http://blogs.bu.edu/jgerring/files/2013/06/Conceptformation.pdf on 29.09.2017.

Gerring, J. (2012). Social Science Methodology: A Unified Framework (2nd Ed.). Cambridge University Press. ISBN: 978-0-521-13277-0.

Gershefski, J.J., Arnkoff, D.B. Glass, C.R. and Elkin, I. (1996). Clients’ perceptions of treatment for depression: I Helpful aspects. Psychotherapy Research, 6, 233-247. DOI: 10.1080/10503309612331331768. Gianakis, M. and Carey, T.A. (2008). A review of the experience and explanation of psychological change. Counselling Psychology Review, 23(3), 27-38. ISSN: 0269-6975.

Gibbard, I. (2014). Clients’ experiences of change in cognitive behavioural therapy and person-centred therapy. Unpublished thesis. Available from: https://ethos.bl.uk/OrderDetails.do;jsessionid=31AD135B7A46811D92C206685830938C?uin=uk.bl.ethos.632280.

Gilgun, F. (2014). An introduction to deductive qualitative analysis. International Congress on Qualitative Inquiry. Downloaded from https://www.slideshare.net/JaneGilgun/an-introduction-to-deductive-qualitative-analysis on 06.09.2018.

Glaser, B.G. (1998). Doing Grounded Theory: Issues and Discussions. Mill Valley, CA: Sociology Press. ISBN: 978-1884156113.

Glaser, B.G. and Strauss, A.L. (2000). The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago, IL: Aldine. ISBN: 978-0202302607.

http://blogs.bu.edu/jgerring/files/2013/06/Conceptformation.pdf%20on%2029.09.2017

https://www.slideshare.net/JaneGilgun/an-introduction-to-deductive-qualitative-analysis

https://www.slideshare.net/JaneGilgun/an-introduction-to-deductive-qualitative-analysis

207 of 271

Glassman, N.S. and Andersen, S.M. (1999). Transference in social cognition: Persistence and exacerbation of significant-other-based inferences over time. Cognitive Therapy and Research, 23(1), 75-91. DOI: 10.1023%2FA%3A1018762724798.

**Gloster, A.T., Klotsche, J., Gerlach, A.L., Hamm, A., Ströhle, A., Gauggel, S., Kircher, T., Alpers, G.W., Deckert, J. and Wittchen, H. (2014). Timing matters: change depends on the stage of treatment in cognitive behavioral therapy for panic disorder with agoraphobia. Journal of Consulting and Clinical Psychology, 82(1), 141-153. DOI: 10.1037/a0034555.

Goertz, G. (2006). Social Science Concepts: A User’s Guide. Princeton University Press. ISBN: 978-0-691-12411-7.

Goertz, G. and Mazur, A.G. (Eds), (2007). Politics, Gender and Concepts: Theory and Methodology. Cambridge University Press. ISBN: 978-0-521-72342-8.

Goertzen, J.R. (2010). Dialectical pluralism: A theoretical conceptualisation of pluralism in psychology. New Ideas in Psychology, 28, 201-209. DOI: 10.1016/j.newideapsych.2009.09.013.

Goldfried, M.R. (1980). Toward the delineation of therapeutic change principles. American Psychologist, 35, 991-999.

**Goldin, P.R., Jazaieri, H., Ziv, M., Kraemer, H. Heimberg, R.G. and Gross, J.J. (2013). Changes in beliefs about the social competence of self and others following group cognitive-behavioral treatment. Clinical Psychological Science, 1(3), 301-310. DOI: 10.1177/2167702613476867.

*Goldman, R.N., Greenberg, L.S. and Pos, A.E. (2007). Depth of emotional experience and outcome. Psychotherapy Research 15(3), 248-260. DOI: 10.1080/10503300512331385188.

*Gomes, K. and Pascual-Leone, A. (2015). A resource model of change: Client factors that influence problem gambling treatment outcomes. Journal of Gambling Studies, 31(4), 1651-1669. DOI: 10.1007/s10899-014-9493-y.

*Gonçalves, M.M., Ribeiro, A. P., Stiles, W. B., Conde, T., Matos, M., Martins, C., and Santos, A. (2011). The role of mutual in-feeding in maintaining problematic self-narratives: Exploring one path to therapeutic failure. Psychotherapy Research, 21(1), 27–40. DOI: 10.1080/10503307.2010.507789.

Goodyear, R., Tracey, T.J.G., Claiborn, C.D. and Lichtenberg, J. (2005). Ideographic concept mapping in counselling psychology research. Journal of Counseling Psychology, 52(2), 236-242. DOI: 10.1037/0022-0167.52.2.236.

Gopen, G.D. and Swan, J.A. (1990). The science of scientific writing. American Scientist, 78(6), 550-558.

Gordon, R. (2012). Where oh where are the clients? The use of common factors in counselling psychology. Counselling Psychology Review, 27 (4), 8-17.

Goss, D. (2015). Affective neuroscience. Counselling Psychology Review, 30(1), 52-63.

Goss, D. and Parnell, T. (2017). Integrating neuroscience into counselling psychology. Counselling Psychology Review, 32(4), 4-17.

Greaves, A.L. (2006). The active client: A qualitative analysis of thirteen clients’ contributions to the psychotherapy process. Unpublished doctoral dissertation, University of Southern California.

Greenberg, R.P., Constantino, M.J. and Bruce, N. (2006). Are patient expectations still relevant for psychotherapy process and outcome? Clinical Psychology Review, 26, 657-678. DOI: 10.1016/j.cpr.2005.03.002.

Greenson, R.R. (1967). The Technique and Practice of Psychoanalysis (vol. 1). New York: International Universities Press. ASIN: B00M0QWR0I.

208 of 271

Grencavage, L. and Norcross, J. (1990). Where are the commonalities among the therapeutic common factors? Professional Psychology: Research and Practice, (21)5, 372-378. DOI: 10.1037/0735-7028.21.5.372.

Groth-Marnat, G., Roberts, R. and Beutler, L.E. (2001). Client characteristics and psychotherapy: Perspectives, support, interactions and implications for training. Australian Psychologist, 36(2), 115-121. DOI: 10.1080/00050060108259643.

Guba, E.G. and Lincoln, Y.S. (1982). Epistemological and methodological bases of naturalistic enquiry. Educational Communication and Technology Journal, 30(4), 233-252. DOI: 10.1126/science.146.3642.347.

**Hanna, F.J. and Ritchie, M.H. (1995). Seeking the active ingredients of psychotherapeutic change: Within and outside the context of therapy. Professional Psychology: Research and Practice, 26(2), 176-183. DOI: 10.1037/0735-7028.26.2.176.

Hansen, N.B. and Lambert, M. (2003). An evaluation of the dose-response effect in naturalistic treatment settings using survival analysis. Mental Health Services Research, 5. 1-11. DOI: 10.1023/A:102175130.

Harding, S. (2004). A socially relevant philosophy of science? Resources from standpoint theory’s controversiality. Hypatia, 19(1), 25-47. Downloaded from https://muse.jhu.edu/article/53922/pdf on 10.09.2017.

Hardy, G.E., Barkham, M., Shapiro, D.A., Reynolds, S., Rees, A. and Stiles, W.B. (1995). Credibility and outcome of cognitive-behavioral and psychodynamic-interpersonal therapy. British Journal of Clinical Psychology, 34, 555-569. DOI: 10.1111/j.2044-8260.1995.tb01489.x

Hargons, C., Mosley, D.V. and Stevens-Watkins, D. (2017). Studying sex: A content analysis of sex research in counselling psychology. The Counseling Psychologist, 45(4), 528-546. DOI: 10.1177/0011000017713756.

Harris, G.E. and Larson, D. (2008). High-risk behaviours following an HIV diagnosis, Counselling Psychology Review, 23(3), 48-61. ISSN: 0269-6975.

Harrison, K. (2013). Counselling psychology and power: Considering therapy and beyond. Counselling Psychology Review, 28(2), 107-117. ISSN: 0269-6975.

**Hartmann, A., Orlinsky, D., Weber, S., Sandholz, A., Zeeck, A. (2010). Session and intersession experience related to treatment outcome in bulimia nervosa. Psychotherapy Theory, Research, Practice and Training, 47(30), 355-370. DOI: 10.1037/a0021166.

**Hartzler, B., Witkiewitz, K., Villarroel, N. and Donovan, D. (2011). Self-efficacy change as a mediator of associations between therapeutic bond and one-year outcomes in treatments for alcohol dependence. Psychology of Addictive Behavior, 25(2), 269-278. DOI: 10.1037/a0022869.

*Hawley, L.L., Ho, M.R., Zuroff, D.C and Blatt, S.J. (2006). The relationship of perfectionism, depression, and therapeutic alliance during treatment for depression: Latent difference score analysis. Journal of Consulting and Clinical Psychology, 74(5), 930-942. DOI: 10.1037/0022-006X.74.5.930.

Hayes, A.M., Laurenceau, J-P., Feldman, G., Strauss, J.L. and Cardaciotto, L. (2007). Change is not always linear: The study of non-linear and discontinuous patterns of change in psychotherapy. Clinical Psychology Review, 27, 715-723. DOI: 10.1016/j.cpr.2007.01.008.

Health and Care Professions Council (2016). Standards of Conduct, Performance and Ethics. Downloaded from https://www.hcpc-uk.org/assets/documents/10004EDFStandardsofconduct,performanceandethics.pdf on 09.09.2018.

Hekman, S. (1997). Truth and method: Feminist standpoint revisited. Signs, 22(2), 341-365. Stable URL: http://www.jstor.org/stable/3175275.


https://muse.jhu.edu/article/53922/pdf%20on%2010.09.2017

https://muse.jhu.edu/article/53922/pdf%20on%2010.09.2017

https://doi.org/10.1111/j.2044-8260.1995.tb01489.x

https://doi.org/10.1177%2F0011000017713756

https://www.hcpc-uk.org/assets/documents/10004EDFStandardsofconduct,performanceandethics.pdf

https://www.hcpc-uk.org/assets/documents/10004EDFStandardsofconduct,performanceandethics.pdf

209 of 271

Helmreich, I., Kunzler, A., Chmitorz, A., König, J., Binder, H., Wessa, M., Lieb, K. (2017). Psychological interventions for resilience enhancement in adults (Protocol). Cochrane Database of Systematic Reviews Issue 2. Art. No.: CD012527. DOI: 10.1002/14651858.CD012527.

*Hendriks, G.J., Keijsers, G. P.J., Kampman, M., Hoogduin, C. A.L. and Voshaar, R. C.O. (2012). Predictors of outcome of pharmacological and psychological treatment of late-life panic disorder with agoraphobia. International Journal of Geriatric Psychiatry, 27(2), 141-150. DOI: 10.1002/gps.2700.

Hendry, L. B. and Kloep, M. (2002). Lifespan Development: Resources, Challenges and Risks. Cengage Learning. ISBN: 978-1861527547.

Henkelman, J. and Paulson, B. (2006). The client as expert: researching hindering experiences in counselling. Counselling Psychology Quarterly, 19(2), 139-150. DOI: 10.1080/09515070600788303. Higginson, S. and Mansell, W. (2008). What is the mechanism of psychological change? A qualitative analysis of six individuals who experienced personal change and recovery. Psychology and Psychotherapy: Theory, Research and Practice, 81, 309-328. DOI: 10.1348/147608308X30125. Hill, C. and Lambert, M. (2004). Methodological issues in studying psychotherapy processes and outcomes. In M.J. Lambert (Ed), Bergin and Garfield's Handbook of Psychotherapy and Behavior Change (5th ed., pp. 84-135). 0-471-37755-4.

Hoddy, E. (2018). Critical realism in empirical research employing techniques from grounded theory methodology. International Journal of Social Research Methodology. DOI: 10.1080/13645579.2018.1503400.

Hoffman, J.J. (1985). Client factors related to premature termination of psychotherapy. Psychotherapy: Theory, Research, Practice, Training 22(1), 83-85. DOI: 10.1037/2Fh0088530.

Hogan, R. (1975). Empathy: A conceptual and psychometric analysis. The Counseling Psychologist. 5(2), 14-18. DOI: 10.1177/001100007500500204.

*Hooley, J.M. and Teasdale, J.D. (1999). Expressed emotion and behavior therapy outcome: A prospective study with OCD and agoraphobic outpatients. Journal of Consulting and Clinical Psychology, 67(5), 658-665.

*Hooley, J.M. and Teasdale, J.D. (1989). Predictors of relapse in unipolar depressives: Expressed emotion, marital distress, and perceived criticism. Journal of Abnormal Psychology, 98(3), 229-235. DOI: 10.1037/0021-843X.98.3.229.

Horvath, A.O. and Bedi, R.P. (2002). The alliance. In J.C. Norcross (Ed.), Psychotherapy Relationships That Work: Therapist Contributions and Responsiveness to Patients (37-70). Oxford: Oxford University Press. ISBN: 978-0-19-514346-1.

Horvath, A.O., Del Re, A.C., Flückiger, C. and Symonds, D. (2011) Alliance in individual psychotherapy. In J.C. Norcross (Ed.), Psychotherapy Relationships That Work: Therapist Contributions and Responsiveness to Patients (2nd Ed., 25-69). Oxford: Oxford University Press. ISBN: 978-0199737208.

Hougaard, E. (1994). The therapeutic alliance – a conceptual analysis. Scandinavian Journal of Psychology, 35(1), 67-85. DOI: 10.1111/j.1467-9450.1994.tb00934.x

Howard, G.S., Maerlender, A.C., Myers, P.R. and Curtin, T.D. (1992). In stories we trust: Studies of the validity of autobiographies. Journal of Counseling Psychology, 39(3), 398-405. DOI: 10.1037/0022-0167.39.3.398.

Hoyle, R.H. and Robinson, J.C. (2004). Mediated and moderated effects in social psychological research: Measurement, design and analysis issues. In C. Sansone, C. Morf and A.T Panter (Eds). The Sage Handbook of Methods in Social Psychology (213-234). Thousand Oaks, CA: Sage. ISBN: 978-0-761925354.

https://doi.org/10.1080/13645579.2018.1503400

https://doi.org/10.1177/001100007500500204


210 of 271

Hubble, M.A., Duncan, B.L. and Miller, S.D. (1999a). The Heart and Soul of Change: What Works in Therapy. Washington, DC: American Psychological Association. ISBN: 1-55798-557-X.

Hubble, M.A., Duncan, B.L. and Miller, S.D. (1999b). Directing attention to what works. In M.A. Hubble, B.L. Duncan and S.D. Miller (Eds), The Heart and Soul of Change: What Works in Therapy (407-448). Washington, DC: American Psychological Association. ISBN: 1-55798-557-X.

Hupcey, J.E. and Penrod, J. (2003). Concept advancement: Enhancing inductive validity. Research and Theory for Nursing Practice, 17, 19-30.

Iacoboni, M. (2008). Mirroring People: The New Science of How We Connect with Others. New York: Farrar, Straus and Giroux. ISBN: 978-0-374-21017-5.

Ivey, A.E. and Zalaquett, C.P. (2011). Neuroscience and counselling: Central issue for social justice leaders. Journal for Social Action in Counselling and Psychology, 3(1), 103-116. Downloaded from http://psysr.org/jsacp/ivey-v3n1-11_103-116.pdf on 02.12.17.

Jackson, R.H. (1984). Ethnicity. In G. Sartori (Ed). Social Science Concepts (206-238). Beverley Hills, CA: Sage. ISBN: 0-8039-2177-2.

*Jarrett, R.B., Eaves, G.G., Grannemann, B.D. and Rush, A.J. (1991). Clinical, cognitive, and demographic predictors of response to cognitive therapy for depression: A preliminary report. Psychiatry Research, 37, 245-260. DOI: 10.1016/0165-1781(91)90061-S.

Jarvis, P. (2018). Learning to be a person in society. In K. Illeris (Ed). Contemporary Theories of Learning (2nd Ed., 15-28). London: Routledge. ISBN: 978-1-138-55049-0.

Jensen, D. (2008). Credibility. In L.M. Gi ven (Ed.) The Sage Encyclopedia of Qualitative Research Methods (138-139). London: Sage. ISBN: 978-1412941631.

Johansson, P. and Høglend, P. (2007). Identifying mechanisms of change in psychotherapy: Mediators of treatment outcome. Clinical Psychology and Psychotherapy, 14(1), 1-9. DOI: 10.1002/cpp.514.

Jordan, J.V. (2000). The role of mutual empathy in relational/cultural therapy. Journal of Clinical Psychology. In Session 56(80), 1005-1016. DOI: 10.1002/1097-4679(200008)56:8<1005::AID-JCLP2>3.0.CO;2-L

Jordan, J.V. (2010). Relational-Cultural Therapy. Washington, DC: American Psychological Association. ISBN: 978-1-4338-0463-2.

Jordan, J.V., Kaplan, A.G., Miller, J.B., Stiver, I.P. and Surrey, J.L. (Eds). Women’s Growth in Connection. New York: Guilford. ISBN: 0-89862-465-7.

Joutsenniemi, K., Laaksonen, M.A., Knekt, P., Haaramo, P. and Lindfors, O. (2012). Prediction of the outcome of short- and long-term psychotherapy based on social-demographic factors. Journal of Affective Disorders, 141, 331-342. DOI: 10.1016/j.jad.2012.03.027. Joyce, S., Shand, F., Tighe, J., Laurent, S.J., Bryant, R.A. and Harvey, S.B. (2018). Road to resilience: A systematic review and meta-analysis of resilience training programmes and interventions. BMJ Open,

8(6),1-9. DOI: 10.1136/bmjopen-2017-017858. Kadera, S.W., Lambert, M.J. and Andrews, A.A. (1996). How much therapy is really enough? A session-by-session analysis of the psychotherapy dose-effect relationship. Journal of Psychotherapy Practice and Research 5(2), 132-151. Kaldo, V., Ramnerö, J. and Jernelöv, S. (2015). Involving clients in treatment methods: A neglected interaction in the therapeutic relationship. Journal of Consulting and Clinical Psychology, 83(6), 1136-1141. DOI: 10.1037/ccp0000039.

http://psysr.org/jsacp/ivey-v3n1-11_103-116.pdf%20on%2002.12.17

https://doi.org/10.1002/1097-4679(200008)56:8%3C1005::AID-JCLP2%3E3.0.CO;2-L

https://doi.org/10.1002/1097-4679(200008)56:8%3C1005::AID-JCLP2%3E3.0.CO;2-L

https://doi.org/10.1016/j.jad.2012.03.027

http://dx.doi.org/10.1136/bmjopen-2017-017858

211 of 271

Karnieli-Miller, O., Strier, R. and Pessach, L. (2009). Power relations in qualitative research. Qualitative Health Research, 19(2), 279-289. DOI: 10.1177/1049732308329306.

Kanfer, F.H. and Schefft, B.K. (1998). Guiding the Therapeutic Process of Change. Champaign, IL: Research Press. ISBN: 978-0-87822-285-8.

Kasket, E. (2016). Carrying out research. In B. Douglas, R. Woolfe, S. Strawbridge, E. Kasket and V. Galbraith (Eds.), The Handbook of Counselling Psychology (4th Ed, 228-243). London: Sage. ISBN: 978-1-4462-7632-7.

**Kavanagh, D.J. and Wilson, P.H. (1985). Prediction of outcome for group cognitive therapy for depression. Behavior Research and Therapy. 27(4), 333-343.

Kazdin, A.E. (1999). The meanings and measurement of clinical significance. Journal of Clinical and Consulting Psychology, 67, 322-339. DOI: 10.1037/0022-006x.67.3.332.

Kazdin, A. E. (2009). Understanding how and why psychotherapy leads to change. Psychotherapy Research, 19(4-5), 418-428. DOI: 10.1080/10503300802448899.

Kazdin, A.E. and Nock, M.K. (2003). Delineating mechanisms of change in child and adolescent therapy: Methodological issues and research recommendations. Journal of Child Psychotherapy and Psychiatry, 44(8), 1116-1129. DOI: 10.1111/1469-7610.00195.

Kazdin, A. E. and Wassel, G. (1999). Barriers to treatment participation and psychological change among children referred for conduct disorder. Journal of Clinical Child Psychology, 28, 160-172.

Keeley, M.L. , Storch, E.A., Merlo, L.J. and Geffken, G.R. (2008). Clinical predictors of response to cognitive-behavioral therapy for obsessive-compulsive disorder. Clinical Psychology Review, 28, 118-130. DOI: 10.1016/j.cpr.2007.04.003.

*Keijsers, G.P.J, Hoogduin, C.A.L. and Schaap, C.P .D .R. (1994). Predictors of Treatment Outcome in the Behavioural Treatment of Obsessive-Compulsive Disorder. British Journal of Psychiatry, 165, 781-786. DOI: 10.1192/bjp.165.6.781.

Kelly, J. (2006). Being ecological. An Expedition into Community Psychology. New York, NY: Oxford University Press. ISBN: 978-0-19517-379-6.

Kenniston, K, Boltax, S. and Almond, R (1971). Multiple criteria of treatment outcome. Journal of Psychiatry Research, 8, 107-118. DOI: 10.1016/0022-3956(71)90012-4.

**Khattra, J., Angus, L., Westra, H., Macaulay, C., Moertl, K, and Constantino, M. (2017). Client perceptions of corrective experiences in cognitive behavioral therapy and motivational interviewing for generalized anxiety disorder: An exploratory pilot study. Journal of Psychotherapy Integration, 27(1), 23-34. DOI: 10.1037/int0000053.

*Kim, J.E., Zane, N.W. and Blozis, S.A. (2012). Client predictors of short-term psychotherapy outcomes among Asian and White American outpatients. Journal of Clinical Psychology, 68(12), 1287-1302. DOI: 10.1002/jclp.21905.

Kindt, M., Buck, N., Arntz, A. and Soeter, M. (2007). Perceptual and conceptual processing as predictors of treatment outcome in PTSD. Journal of Behavior Therapy and Experimental Psychiatry, 38(4), 491-506. DOI: 10.1016/j.jbtep.2007.10.002.

King, R. and O’Brien, T. (2011). Transference and countertransference: Opportunities and risks as two technical constructs that migrate beyond their psychoanalytic homeland. Psychotherapy in Australia, 17(4). Downloaded from http://www.psychotherapy.com.au/fileadmin/site_files/pdfs/Transference_and_Countertransference.pdf on 28.07.2018.

Kirsch, I. (1990). Changing Expectations: A Key to Effective Psychotherapy. Belmont, CA: Brroks/Cole. ISBN: 978-0534126483.

https://doi.org/10.1192/bjp.165.6.781

http://www.psychotherapy.com.au/fileadmin/site_files/pdfs/Transference_and_Countertransference.pdf

212 of 271

Klein, M.J. and Elliott, R. (2006). Client accounts of personal change in process-experiential psychotherapy: A methodologically pluralistic approach. Psychotherapy Research, 16(1), 91-105. DOI: 10.1080/10503300500090993

**Kleindienst, N., Limberger, M.F., Ebner-Priemer, U.W., Keibel-Mauchnik, J., Dyer, A., Berger, M., Schmahl, C. and Bohus, M. (2011). Dissociation predicts poor response to dialectial behavioral therapy in female patients with borderline personality disorder. Journal of Personality Disorders, 25(4), 432-447. DOI: 10.1521/pedi.2011.25.4.432.

Knafl, K.A. and Deatrick, J.A. (1990). Family management style: Concept analysis and development. Journal of Pediatric Nursing, 5(1), 4-14.

Knafl, K.A. and Deatrick, J.A. (2000). Knowledge synthesis and concept development in nursing. In B.L. Rodgers and K.A. Knafl (Eds). Concept Development in Nursing: Foundations, Techniques and Applications (2nd Edition, 39-54). Philadelphia, PA: Saunders/Elsevier. ISBN: 978-0-7216-8243-3.

Kolb, D.A. (1984). Experiential Learning. Englewood Cliffs, NJ: Prentice-Hall.

Kolb, D.A. (2014). Experiential Learning: Experience as the Source of Learning and Development. Pearson Education. ISBN: 978-0-13-389240-6.

Kolden, G.G. (1991). The generic model of psychotherapy: An empirical investigation of patterns of process and outcome relationships. Psychotherapy Research, 1(1), 62-73. DOI: 10.1080/10503309112331334071.

Kraemer, H.C., Wilson, G.T., Fairburn, C.G. and Agras, W.S. (2002). Mediators and moderators of treatment effects in randomised clinical trials. Archives of General Psychiatry, 59(10), 877-883. DOI: 10.1001/archpsyc.59.10.877.

Kramarae, C. (1981). Women and Men Speaking: Frameworks for Analysis. Rowley, MA: Newbury House. ISBN: 978-0-883-77179-2.

Kramarae, C. (2005). Muted group theory and communication: Asking dangerous questions. Women and Language, 28(2), 55-61.

**Krentzman, A.R. (2017). Gratitude, abstinence, and alcohol use disorders: Report of a preliminary finding. Journal of Substance Abuse Treatment, 78, 30-36. DOI: 10.1016/j.jsat.2017.04.013.

Krupnick, J.L., Sofsky, S.M., Simmens, S., Moyer, J., Elkin, I., Watkins, J. and Pilkonis, P.A. (1996). The role of the therapeutic alliance in psychotherapy and pharmacotherapy outcome: Findings in the NIMH Treatment of Depression Collaborative Treatment Programme. Journal of Consulting and Clinical Psychology, 64, 532-539.

Kuyken, W, Watkins, E., Holden, E., White, K., Taylor, R.S., Byford, S., Evans, A., Radford, S., Teasdale, J.D. and Dalgleish, T. (2010). How does mindfulness based cognitive therapy work? Behavior Research and Therapy, 48, 1105-1112. DOI: 10.1016/j.brat.2010.08.003.

Kvale, S. (2007). Doing Interviews (2-10). London: Sage. ISBN: 978-0-76194-977-0.

**Kyrios, M., Hordern, C. and Fassnacht, D.B. (2015). Predictors of response to cognitive behaviour therapy for obsessive-compulsive disorder. International Journal of Clinical and Health Psychology, 15(3), 181-190. DOI: 10.1016/j.ijchp.2015.07.003.

Laaksonen, M.A., Lindfors, O., Knekt, P. and Aalberg, V. (2012). Suitability for Psychotherapy Scale and its reliability, validity and prediction. British Journal of Clinical Psychology, 51, 351-375. DOI: 10.1111/j.2044-8260.2012.02033.x. Ladouceur, R., Dugas, M.J., Freeston, M.H., Leger, E., Gagnon, F. and Thibodeau, N. (2000). Efficacy of a cognitive-behavioral treatment for generalised anxiety disorder: Evaluation in a controlled clinical trial. Journal of Consulting and Clinical Psychology, 68, 957-964. DOI: 10.1037/0022-006X.68.6.957.

https://doi.org/10.1111/j.2044-8260.2012.02033.x

http://dx.doi.org/10.1037/0022-006X.68.6.957

213 of 271

Lambert, M.J. (1992). Implications of outcome research for psychotherapy integration. In J.C. Norcross and M.R. Goldfried (Eds.), Handbook of Psychotherapy Integration (94-129). New York: Basic Books. ASIN: B01A64LX0A.

Lambert, M.J. (2010). Preventing Treatment Failure: The Use of Measuring, Monitoring and Feedback in Clinical Practice. Washington, DC: American Psychological Association. ISBN: 978-1433807824.

Lambert, M.J. (2013). The efficacy and effectiveness of psychotherapy. In M.J. Lambert (Ed), Bergin and Garfield’s Handbook of Psychotherapy and Behaviour Change (6th Edition, 169-218). Hoboken, N.J: Wiley. ISBN 978-1-118-03820-8.

Lambert, M.J. and Barley, D.E. (2002). Research summary on the therapeutic relationship and psychotherapy outcome. In J.C. Norcross (Ed.), Psychotherapy Relationships that Work: Therapist Contributions and Responsiveness to Patients (17-32). New York: Oxford University Press. ISBN: 973-0-19-514346-1.

Lambert, M.J., DeJulio, S.S. and Stein, D. (1978). Therapist interpersonal skills: Process, outcome, methodological considerations and recommendations for future research. Psychological Bulletin, 85, 467-489. AN: 00006823-197805000-00002.

Lambert, M.J., Smart, D.W., Campbell, M.P., Hawkins, E.J., Harmon, C. and Slade, K.L. (2006). Psychotherapy outcome as measured by the OQ-45 in African American, Asian/Pacific Islander, Latino/a and Native American clients compared to matched Caucasian clients. Journal of College Student Psychotherapy, 20, 17-29. DOI: 10.1300/J035v20n04_03.

Larkin, M., Watts, S. and Clifton, E. (2006). Giving voice and making sense in interpretive phenomenological analysis. Qualitative Research in Psychology, 3, 102-120. DOI: 10.1191/1478088706qp062oa.

Laurence, S. and Margolis, E. (1999). Concepts and cognitive science. In E. Margolis and S. Laurence (Eds), Concepts: Core Readings. MIT Press, 3-81. ISBN: 978-0262631938.

*Leibert, T. W., Smith, J. B. and Agaskar, V. R. (2011). Relationship between the working alliance and social support on counseling outcome. Journal of Clinical Psychology, 67(7), 709-719. DOI: 10.1002/jclp.20800.

Lemmens, L.H.J.M., Müller, V.N.L.S., Arntz, A. and Huibers, M.J.H. (2016). Mechanisms of change in psychotherapy for depression: An empirical update and evaluation of research aimed at identifying psyshcological mediators. Clinical Psychology Review, 50, 95-107. DOI: 10.1016/j.cpr.2016.09.004.

*Leung, A.W. and Heimberg, R.G.(1996). Homework compliance, perceptions of control and outcome of cognitive-behavioral treatment of social phobia. Behavior Research and Therapy, 34(5), 423-432. DOI: 10.1016/0005-7967(96)00014-9.

*Leweke, F., Bausch, S., Leichsenring, F., Walter, B. and Stingl, M.(2009). Alexithymia as a predictor of outcome of psychodynamically oriented inpatient treatment inpatient treatment, Psychotherapy Research, 19(3), 323-331. DOI: 10.1080/10503300902870554.

Lilienfeld, S., Ritschel, L.A, Lynn, S.J., Cautin, R.L. and Latzman, R.D. (2014). Why ineffective psychotherapies appear to work: A taxonomy of causes of spurious therapeutic effectiveness. Perspectives on Psychological Science, 9(4), pp.355–387. 10.1177/1745691614535216.

Littel, J., Alexander, L. and Reynolds, W. (2001). Client participation: Central and underinvestigated elements of intervention. Social Service Review, 75(1), 1-28. DOI: 10.1086/591880.

Llewelyn, S. (1988). Psychological therapy as viewed by clients and therapists. British Journal of Clinical Psychology, 27(3), 223-238. DOI: 10.1111/j.2044-8260.1988.tb00779.x.

214 of 271

Llewelyn, S.P., Elliott, R., Shapiro, D.A., Hardy, G.E. and Firth-Cozens, J. (1988). Client perceptions of significant events in prescriptive and exploratory periods of individual therapy. British Journal of Clinical Psychology. 27, 105-114. DOI: 10.1111/j.2044-8260.1988.tb00758.x. Lombardi, D.R., Button, M. and Westra, H.A. (2014). Measuring motivation: Change talk and counter-change talk in cognitive behavioral therapy for generalised anxiety. Cognitive Behavioral Therapy, 43(1) [no page numbers]. DOI:10.1080/16506073.2013.846400. Lorenzo-Luaces, L., De Rubeis, R.J. and Webb, C.A. (2014). Client characteristics as moderators of the relationship between the therapeutic alliance and outcome in cognitive therapy for depression. Journal of Consulting and Clinical Psychology, 82(2), 368-373. DOI: 10.1037/a0035994.

Lowman, R.L. (2012). The scientist-practitioner consulting psychologist. Consulting Psychology Journal: Practice and Research. 64(3), 151-156. DOI: 10.1037/a0030365.

Luborsky, L., Crits-Christoph, Mintz, J. and Auerbach, A. (1988). Who Will Benefit from Psychotherapy?

New York: Basic Books. ISBN: 0-465-09189-X.

Luborsky, L., Singer, B. and Luborsky, L. (1975). Comparative studies of psychotherapies: Is it true that ‘Everyone has won and all must have prizes’? Archives of General Psychiatry, 32(8), 995-1008. DOI: 10.1001/archpsyc.1975.01760260059004.

Luborsky, L., Rosenthal, R., Diguer, L., Andrusyna, T.P., Berman, J.S., Levitt, J.T., Seligman, D.A. and Krause, E.D. (2002). The dodo bird verdict is alive and well – mostly. Clinical Psychology: Science and Practice, 9(1), 2-12. DOI: 10.1093/clipsy.9.1.2/pdf.

Lunnen, K. M., & Ogles, B. M. (1998). A multiperspective, multivariable evaluation of reliable change. Journal of Consulting and Clinical Psychology, 66, 400–410. DOI: 10.1037/0022-006X.66.2.400.

Lynch, J. (2017). Reframing inequality: the health inequalities turn as a dangerous frame shift. Journal of

Public Health, 4(1), 653-660. DOI: 10.1093/pubmed/fdw140.

Mackrill, T. (2008). Exploring psychotherapy clients’ independent strategies for change while in therapy. British Journal of Guidance and Counselling, 36(4), 441-453. DOI: 10.1080/03069880802343837.

*Mallinckrodt, B. (1996). Change in working alliance, social support, and psychological symptoms in brief therapy. Journal of Counseling Psychology, 43(4), 448-455. DOI: 10.1037/0022-0167.43.4.448.

Mansell, W. (2011). Core processes of psychopathology and recovery: ‘Does the dodo bird have wings’? Clinical Psychology Review, 31(2), 189-192. DOI: 10.1016/j.cpr.2010.06.009.

Martin, P. (2009). Book review of ‘Reflections on Human Potential: Bridging the Person-centered Approach and Positive Psychology’ by B.E. Levitt. British Journal of Guidance and Counselling, 37(3), 387-402. DOI: 10.1080/03069880903007844.

Martinez, R. (2017). Empathy. London: Cannongate. ISBN: 9781786892379

**Matos, M., Santos, A., Gonçalves, M., Martins, C. (2009). Innovative moments and change in narrative therapy. Psychotherapy Research, 19(1), 68-80. DOI: 10.1080/10503300802430657.

Mausbach, B.T., Moore, R., Roesch, S., Cardenas, V. and Patterson, T.L. (2010). The relationship between homework compliance and therapy outcomes: An updated meta-analysis. Cognitive Therapy

and Research, 34(5), 429-438. DOI: 10.1007/s10608-010-9297-z.

**McCrady, B.S., Hayaki, J., Epstein, E.E. and Hirsch, L.S. (2002). Testing hypothesized predictors of change in conjoint behavioral alcoholism treatment for men. Alcoholism: Clinical and Experimental Research, 26(4), 463-470. DOI: 10.1111/j.1530-0277.2002.tb02562.x.

McIntosh, M. and Sykes, C. (2016). Older adults’ experience of psychological therapy, Counselling Psychology Review, 31(1), 20-30.


https://doi.org/10.1093/pubmed/fdw140

https://dx.doi.org/10.1007%2Fs10608-010-9297-z

215 of 271

McLachlin, R. (1999). Factors for consulting engagement success. Management Decision, 37(5), 354-404.

McLaughlin, H. (2009). What's in a name: ‘Client’, ‘patient’, ‘customer’, ‘consumer’, ‘expert by experience’, ‘service user’—What's next? British Journal of Social Work, 39(6), 1101-1117. DOI: 10.1093/bjsw/bcm155.

McLeod, J. (2003). Qualitative research methods in counselling psychology. In R. Woolfe, W. Dryden and S. Strawbridge (Eds). Handbook of Counselling Psychology (2nd Ed., 74-92). London: Sage.

McLeod, J. and Balamoutsou, S. (2006). A method for qualitative narrative analysis of psychotherapy transcripts. In J. Frommer and D.L. Rennie (Eds), Qualitative Psychotherapy Research: Methods and Methodology (2nd Ed, 128-152). Lengerich, Germany: Pabst Science.

Mellon, C.A. (1990), Naturalistic Inquiry for Library Science: Methods and Applications for Research, Evaluation, and Teaching. New York: Greenwood Press. ISBN: 978-0313256530.

*Meyer, B., Pilkonis, P.A., Krupnick, J.L., Egan, M.K., Simmens, S.J. and Sotsky, S.M. (2002). Treatment expectancies, patient alliance and outcome: Further analyses from the National Institute of Mental Health Treatment of Depression Collaborative Research Program. Journal of Consulting and Clinical Psychology, 70(4), 1051-1055. DOI: 10.1037/0022-006X.70.4.1051.

Miller, J.B. (1991). The development of women’s sense of self. In J.V. Jordan, A.G. Kaplan, J.B. Miller, I.P Stiver and J.L. Surrey (Eds). Women’s Growth in Connection (11-26). New York: Guilford.

Miller, K.D. and Tsang, E.W.K. (2011). Testing management theories: Critical realist philosophy and research methods. Strategic Management Journal, 32(2), 139-158. Downloaded from http://www.jstor.org/stable/41000104 on 27.05.2018.

Mingers, (2006). Realising Systems Thinking: Knowledge and Action in Management Science. Springer. ISBN: 978-0387281889.

Mintz, J. (1972). What is “success” in psychotherapy. Journal of Abnormal Psychology, 81(1), 11-19. DOI: 10.1037/h0033225.

Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA Statement. PLoS Med 6(6): e10000097. DOI:10.1371/journal.pmed10000097.

**Mohr, S., Perroud, N., Gillieron, C., Brandt, P.Y., Rieben, I., Borras, L. and Huguelet, P. (2011). Spirituality and religiousness as predictive factors of outcome in schizophrenia and schizo-affective disorders. Psychiatry Research, 186(2-3), 177-182. DOI: 10.1016/j.psychres.2010.08.012.

Mones, A.G. and Schwartz, R.C. (2007). The functional hypothesis: A family systems contribution towards an understanding of the healing process of the common factors. Journal of Psychotherapy Integration, 17(4), 314-329. DOI: 10.1037/1053-0479.17.4.314.

Morrow, S.L. and Smith, M.L. (1995). Constructions of survival and coping by women who have survived childhood sexual abuse. Journal of Counseling Psychology, 42(1), 24-33. DOI: 10.1037/0022-0167.42.1.24.

Morrow, S.L. and Smith, M.L. (2000). Qualitative research for counseling psychology. In S.D. Brown and R.W. Lent (Eds), Handbook of Counseling Psychology (3rd Ed., 199-230). New York: Wiley and Sons. ISBN: 0-471-25458-4.

Morse, J.M. and Penrod, J. (1999). Linking concepts of enduring, uncertainty, suffering and hope. Journal of Nursing Scholarship, 31, 145-150. DOI: 10.1111/j.1547-5069.1999.tb00455.x.

*Mörtberg, E., Hoffart, A., Boecking, B. and Clark, D.M. (2015). Shifting the focus of one's attention mediates improvement in cognitive therapy for social anxiety disorder. Behavioural and Cognitive Psychotherapy, 43(1), 63-73. DOI: 10.1017/S1352465813000738.

http://www.jstor.org/stable/41000104%20on%2027.05.2018




216 of 271

**Mörtl, K. and Wietersheim, J.V. (2008). Client experiences of helpful factors in a day treatment program: A qualitative approach. Psychotherapy Research, 18(3) 281-293. DOI: 10.1080/10503300701797016.

Murphy, P.M., Cramer, D. and Lillie, F.J. (1984). The relationship between curative factors perceived by patients in their psychotherapy and treatment outcome: An exploratory study. British Journal of Medical Psychology, 57, 187-192. 10.1111/j.2044-8341.1984.tb01599.x.

Najavits, L.M. and Strupp, H. (1994). Differences in the effectiveness of psychodynamic therapists: A process-outcome study. Psychotherapy, 31, 114-123. DOI: 10.1037/0033-3204.31.1.1114.

**Nakano, Y., Lee, K., Noda, Y., Ogawa, S., Kinoshita, Y., Funayama, T., Watanabe, N., Chen, J., Noguchi, Y. and Furukawa, T.A. (2008). Cognitive-behavior therapy for Japanese patients with panic disorder: Acute phase and one-year follow-up results. Psychiatry and Clinical Neurosciences, 62(3), 313-321. DOI: 10.1111/j.1440-1819.2008.01799.x.

**Neimeyer, R.A. and Feixas, G. (1990). The Role of Homework and Skill Acquisition in the Outcome of Group Cognitive Therapy for Depression. Behavior Therapy, 21, 281-292. DOI: 10.1016/S0005-7894(05)80331-4.

Newman, M.G., Crits-Christoph, P., Gibbons, M.B.C. and Erickson, T.M. (2006). Participant factors in treating anxiety disorders. In L.G. Castonguay and L.E. Beutler (Eds), Principles of Therapeutic Change that Work. New York: Oxford University Press, 121-153. ISBN: 978-0-19-515684-3.

Nilsson, T., Svenson, M., Sandell, R. and Clinton, D. (2007). Patients’ experiences of change in cognitive-behavioral therapy and psychodynamic therapy: A qualitative comparative study. Psychotherapy Research, 17(5), 553-566. DOI: 10.1080/10503300601139988.

Norcross, M.J. (2002). Empirically supported therapy relationships. In J.C. Norcross (Ed.), Psychotherapy Relationships that Work: Therapist Contributions and Responsiveness to Patients (3-16). New York: Oxford University Press. ISBN: 973-0-19-514346-1.

Norcross, J.C. and Karpiak, C.P. (2012). Teaching clinical psychology: four seminal lessons that all can master. Teaching of Psychology, 39(4), 301-307. DOI: 10.1177/0098628312461486.

Norcross, J.C. and Lambert, M.J. (2010). Evidence-based therapy relationships. In J.C. Norcross (Ed.), Evidence-based Therapy Relationships (1-4). Downloaded from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.463.7645&rep=rep1&type=pdf on 26.11.2017.

Norcross, J.C. and Lambert, M. (2011). Psychotherapy relationships that work II. Psychotherapy, 48, 4-8. DOI: 10.1037/a0022180.

Norcross, J.C. and Wampold, B.E. (2010). Adapting the relationship to the individual patient. In J.C. Norcross (Ed.), Evidence-based Therapy Relationships (27-31). Downloaded from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.463.7645&rep=rep1&type=pdf on 26.11.2017.

Oliver, D.G., Serovich, J.M. and Mason, T.L. (2005). Constraints and opportunities with interview transcription: Towards reflection in qualitative research. Social Forces, 84 (2), 1273-1289. DOI: 10.1353/sof.2006.0023.

Olsson, C.A., Bond, L., Burns, J.M., Vella-Brodrick, D.A., and Sawyer, S.M. (2003). Adolescent resilience: A concept analysis. Journal of Adolescence, 26, 1-11. DOI: 10.1016/S0140-1971(02)00118-5.

O’Mahoney, J. (2011). Critical realism and the self. Journal of Critical Realism, 10(1), 122-129. DOI: 10.1558/jcr.v10i1.122.

O’Reilly, M., & Parker, N. (2012). Unsatisfactory saturation: A critical exploration of the notion of saturated sample sizes in qualitative research. Qualitative Research Journal, 1-8. DOI:10.1177/1468794112446106.


http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.463.7645&rep=rep1&type=pdf

https://doi.org/10.1037%2Fa0022180

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.463.7645&rep=rep1&type=pdf

217 of 271

Ong, A.D., Bergeman, C.S., Bisconti, T.L. and Wallace, K.A. (2006). Psychological resilience, positive emotions and successful adaptation to stress in later life. Journal of Personality and Social Psychology, 91(4), 730-749. DOI: 10.1037/0022-3514.91.4.730.

Oregon State University (undated). Muted group theory. Downloaded from http://oregonstate.edu/instruct/theory/mutedgrp.htmlon 04.11.2017.

Orlinsky, D.E. (2009). The ‘Generic Model of Psychotherapy’ after 25 years: Evolution of a research-based metatheory. Journal of Psychotherapy Integration, 19(4), 319-339. DOI: 10.1037/a0017973.

Orlinsky, D.E. and Howard, K.I. (1986). Process and outcome in psychotherapy. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of Psychotherapy and Behaviour Change (3rd Edition, 331-381). New York: Wiley. ISBN: 0-471-54513-9.

Orlinsky, D.E., Grawe, K. and Parks, B.K. (1994). Process and outcome in psychotherapy: Noch einmal [once again]. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of Psychotherapy and Behavior Change (4th Ed, 270-376). New York: Wiley. ISBN: 978-0471545132.

Oxford English Dictionary (2017). Accessed at https://en.oxforddictionaries.com/definition/client on 10.09.2017.

Paley, J. (1996). How not to clarify concepts in nursing. Journal of Advanced Nursing, 24, 572-578. DOI: 10.1046/j.1365-2648.1996.22618.x.

Pantell, M., Rehkopf, O., Jutte, D., Syme, S.L., Balmes, J. and Adler, N. (2012). Social isolation: A predictor of mortality comparable to traditional clinical risk factors. American Journal of Public Health, 103 (11), 2056–2062. DOI: 10.2105/AJPH.2013.301261. Parker, I. (Ed.) (1998). Social Constructionism, Discourse and Realism. London: Sage. ASIN: B01JXPARY6

Parker, I. (2007). Critical psychology: What it is and what it is not. Social and Personal Psychology Compass, 1(1), 1-15. DOI: 10.1111/j.1751-9004.2007.00008.x.

Parr, S. (2015). Integrating critical realist and feminist methodologies: Ethical and analytical dilemmas. International Journal of Social Research Methodology, 18(2), 193-207. DOI: 10.1080/13645579.2013.868572.

Patterson, C.L., Uhlin, B. and Anderson, T. (2008). Clients’ pretreatment counselling expectations as predictors of the working alliance. Journal of Counseling Psychology, 55, 528-534. DOI: 10.1037/a0013289.

Paulson, B.L., Truscott, D. and Stuart J. (1999). Clients’ perceptions of helpful experiences in counseling. Journal of Counseling Psychology, 46, 317-324. ISSN: 0022-0167. Pawson, R., Greenhalgh, T., Harvey, G. and Walshe, K. (2005). Realist review: A new method of systematic review designed for complex policy interventions. Journal of Health Services Research and Policy, 10(1), 21-34. DOI: 10.1258/1355819054308530.

Payne, E.C., Robbins, S.B. and Doughty, L. (1991). Goal-directedness and older adult adjustment. Journal of Counseling Psychology, 38 (3), 302-308. DOI: 10.1037/0022-0167.38.3.302. Pekarik, G. and Guidry, L. (1999). Relationship of satisfaction to symptom change, follow-up adjustment and clinical significance in private practice. Professional Psychology: Research and Practice, 27(2), 202-208. DOI: 10.1037//0735-7028.27.2.202.

Penrod, J. (2007). Living with uncertainty: Concept advancement. Journal of Advanced Nursing, 57(6), 658-667. DOI: 10.1111/j.1365-2648.2006.04008.x.

http://oregonstate.edu/instruct/theory/mutedgrp.htmlon%2004.11.2017

https://en.oxforddictionaries.com/definition/client


218 of 271

Penrod, J. and Hupcey, J. (2005). Enhancing methodological clarity: Principle-based concept analysis. Journal of Advanced Nursing, 50(4), 403-409, DOI: 10.1111/j.1365-2648.2005.03405.x.

Perkins, S.N. (2010). Influential client factors: Understanding and organising therapists’ perceptions of client factors that influence reported outcome of therapy. Unpublished PhD thesis. Virginia Polytechnic Institute and State University. Downloaded from https://vtechworks.lib.vt.edu/bitstream/handle/10919/27631/sperkins_dissertation_2010.pdf?sequence=1&isAllowed=y on 23.09.2017.

Pilgrim, D. (2017). Key Concepts in Mental Health (4th Ed). London: Sage. ISBN: 978-1-4739-7301-5.

Pilgrim, D. (2016). Diagnosis and formulation in medical contexts. In B. Douglas, R. Woolfe, S. Strawbridge, E. Kasket and V. Galbraith (Eds.), The Handbook of Counselling Psychology (4th Ed, 169-184). London: Sage. ISBN: 978-1-4462-7632-7.

Pilgrim, D. and Bentall, R. (2009). The medicalisation of misery: A critical realist analysis of the concept of depression. Journal of Mental Health, 8(3), 261-274. DOI: 10.1080/09638239917427.

Pintrich, P.R., Marx, R.W., Boyle, R.A. (1993). Beyond cold conceptual change: the role of motivational beliefs and classroom contextual factors in the process of conceptual change. Review of Educational Research, 63(2), 167-199. Downloaded from http://links.jstor.org/sici?sici=0034-6543%28199322%2963%3A2%3C167%3ABCCCTR%3E2.0.CO%3B2-1 on 16.11.2017.

Pope, C.A. (1999). Reflection and refraction: A reflexive look at an evolving model for methods instruction. English Education, 31(3), 177-200.

Porter, N. (2002). Contextual and developmental frameworks in diagnosing children and adolescents. In M. Ballou and L.S. Brown (Eds.), Rethinking Mental Health and Disorder, New York: Guilford, 262-278. ISBN: 978-1-57230-799-5.

Powers, W.T. (2005). Behavior: The Control of Perception. San Diego, CA: Benchmark Publications. ISBN: 987-0-96971-217-1.

Powers, T.A., Milyavskaya, M. and Koestner, R. (2012). Mediating the effects of self-criticism and self-directed perfectionism on goal pursuit. Personality and Individual Differences, 52, 765-770. DOI:

10.1016/j.paid.2011.12.031. Probst, T., Lambert, M.J., Loew, T.H., Dahlbender, R.W. and tritt, K. (2015). Extreme deviations from expected recovery curves and their associations with therapeutic alliance, social support, motivation and life events in psychosomatic in-patient therapy. Psychotherapy Research, 25(6), 714-723. DOI: 10.1080/10503307.2014.981682.

Prochaska, J.O., DiClemente, C.C. and Norcross, J.C. (1992). In search of how people change: Applications to addictive behaviors. American Psychologist, 47, 1102-1114. DOI: 10.1037/0003-066X.47.9.1102.

Prochaska, J.O. and Norcross, J.C. (1999). Systems of Psychotherapy: A Transtheoretical Analysis (4th Ed). Pacific Grove, CA: Brooks/Cole. ISBN: 0-534-35704-0.

Prochaska, J.O., Norcross, J.C. and DiClemente, C.C. (1994). Changing for Good: A Revolutionary Six-Stage Program for Overcoming Bad Habits and Moving Your Life Positively Forward. New York: Harper Collins. ISBN: 0-380-72572-X.

Proctor, G. (2017). The Dynamics of Power in Counselling and Psychotherapy: Ethics, Politics and

Practice (2nd Ed.). Ross-on-Wye: PCCS Books. ISBN: 978-1910919187.

Proudlock, S. and Wellman, N. (2011). Solution-focussed groups - the results look promising. Counselling Psychology Review 26(3), 45-54. ISSN: 0269-6975.

https://vtechworks.lib.vt.edu/bitstream/handle/10919/27631/sperkins_dissertation_2010.pdf?sequence=1&isAllowed=y

https://vtechworks.lib.vt.edu/bitstream/handle/10919/27631/sperkins_dissertation_2010.pdf?sequence=1&isAllowed=y

http://links.jstor.org/sici?sici=0034-6543%28199322%2963%3A2%3C167%3ABCCCTR%3E2.0.CO%3B2-1

http://links.jstor.org/sici?sici=0034-6543%28199322%2963%3A2%3C167%3ABCCCTR%3E2.0.CO%3B2-1

https://doi.org/10.1016/j.paid.2011.12.031

http://psycnet.apa.org/doi/10.1037/0003-066X.47.9.1102

http://psycnet.apa.org/doi/10.1037/0003-066X.47.9.1102

219 of 271

Radomski, M.V. (2014). Assessing context: Persona, social and cultural. In M.V. Radomski and C.A. Trombly Latham (Eds). Occupational Therapy for Physical Dysfunction. Lippincott, Williams and Wilkins. ISBN: 9781451189216.

**Ramnerö, J. and Ost, L. (2004). Prediction of outcome in the behavioural treatment of panic disorder with agoraphobia. Cognitive Behaviour Therapy, 33(4), 176-180. DOI: 10.1080/16506070410031691.

*Reddy, A.S. and Jagannathan, A. (2017). Predictors of coping and perceived expressed emotions in persons with alcohol dependence in India: a pilot study. Asian Journal of Psychiatry, 28, 38-40. DOI: 10.1016/j.ajp.2017.03.011.

*Renneberg, B., Chambless, D.L., Fydich, T. and Goldstein, A.J. (2002). The relationship of affect balance in family interactions to behaviour therapy outcome: A study with agoraphobic outpatients and their relatives. Clinical Psychology and Psychotherapy, 9(2), DOI: 10.1002/cpp.306.

Rennie, D.L. (1992). Qualitative analysis of the client’s experience of psychotherapy: The unfolding of reflexivity. In S.G. Toukmanian and D.L. Rennie (Eds), Psychotherapy Process Research: Paradigmatic and Narrative Approaches (211-233). London: Sage. ISBN: 978-0803943551.

Rennie, D.L. (2002). Experiencing psychotherapy: Grounded theory studies. In D.J. Cain and J. Seeman (Eds), Humanistic Psychotherapies: Handbook of Research and Practice (117-144). Washington, DC: American Psychological Association. ISBN: 978-1557987877.

Rennie, D.L. (2010). Humanistic psychology at York University: Retrospective: Focus on clients’ experiencing in psychotherapy: Emphasis of radical reflexivity. Humanistic Psychologist, 38, 40-56. DOI: 10.1080/08873261003635856.

Rennie, D.L. (2012). Qualitative Research as Methodological Hermeneutics. Psychological Methods, 17(3), 385-398. DOI: 10.1037/a0029250.

Rice, K.G., Ashby, J.S. and Slaney, R.B. (1998). Self-esteem as a mediator between perfectionism and depression: A structural equations analysis. Journal of Counseling Psychology, 45(3), 304-314.

Ridner, S. (2004). Psychological distress: Concept analysis. Journal of Advanced Nursing, 45(5), 536-545. DOI: 10.1046/j.1365-2648.2003.02938.x.

Ritchie, J. and Lewis, J. (2003). Qualitative Research Practice. London: Sage. ISBN: 978-0-7619-7110-8.

**Ritsher, J.B. and Phelan, J.C. (2004). Internalized stigma predicts erosion of morale among psychiatric outpatients. Psychiatry Research, 129(3), 257-265. DOI: 10.1016/j.psychres.2004.08.003.

Rizq, R. (2007). Tread carefully: Counselling psychology and neuroscience. Counselling Psychology Review, 22(4), 5-18. ISSN: 0269-6975.

Robinson-Whelen, S., Hughes, R.B., Taylor, H.B., Hall, J.W. and Rehm, L.P. (2007). Depression self-management programme for rural women with physical disabilities. Rehabilitation Psychology, 52, 254–262. doi:10.1037/0090-5550.52.3.254.

Robson, C. (2002). Real World Research. Malden, MA: Blackwell. ISBN: 978-0-631-21304-8.

Rodgers, B.L. (1991). Using concept analysis to enhance clinical practice and research. Dimensions of Critical Care Nursing (10)1, 28-34. ISSN: 0730-4625.

Rodgers, B.L. and Knafl, K.A. (Eds). Concept Development in Nursing: Foundations, Techniques and Applications (2nd Edition, 77-102). Philadelphia, PA: Saunders/Elsevier. ISBN: 978-0-7216-8243-3.

Rodgers, B.L. (2000a). Philosophical foundations of concept development. In B.L. Rodgers and K.A. Knafl (Eds). Concept Development in Nursing: Foundations, Techniques and Applications (2nd Edition, 7-37). Philadelphia, PA: Saunders/Elsevier. ISBN: 978-0-7216-8243-3.

http://dx.doi.org/10.1080/08873261003635856

220 of 271

Rodgers, B.L. (2000b). Concept analysis: An evolutionary view. In B.L. Rodgers and K.A. Knafl (Eds). Concept Development in Nursing: Foundations, Techniques and Applications (2nd Edition, 77-102). Philadelphia, PA: Saunders/Elsevier. ISBN: 978-0-7216-8243-3.

Roehrle, B. and Strouse, J. (2008). Influence of social support on success of therapeutic interventions: A meta-analytic review. Psychotherapy: Theory, Research, Practice, Training, 45(4), 464-476. DOI: 10.1037/a0014333.

Rogers, C.R. (1983). Freedom to Learn for the 80’s. Columbus, OH: Charles E. Merrill. ISBN: 978-0-67-509519-8.

Rosenzweig, S. (1936). Some implicit common factors in diverse methods of psychotherapy. American Journal of Orthopsychiatry, 6(3), 412-415. DOI: 10.1111/j.1939-0025.1936.tb05248.x.

*Rosmarin, D.H., Bigda-Peyton, J.S., Kertz, S.J., Smith, N., Rauch, S.L. and Björgvinsson, T. (2013a). A test of faith in God and treatment: The relationship of belief in God to psychiatric treatment outcomes. Journal of Affective Disorders, 146(3), 441-446. DOI: 10.1016/j.jad.2012.08.030.

*Rosmarin, D.H., Bigda-Peyton, J.S., Öngur, D., Pargament, K.I. and Björgvinsson, T. (2013b). Religious coping among psychotic patients: Relevance to suicidality and treatment outcomes. Psychiatry Research, 210(1), 182-187. DOI: 10.1016/j.psychres.2013.03.023.

Rounsaville, B.J. and Carroll, K. (2002). Commentary on ‘Dodo bird revisited’: Why aren’t we dodos yet? Clinical Psychology: Science and Practice, 9(1), 17-20. DOI:10.1093/clipsy.9.1.17.

Rounsaville, B.J., Weissman, M.M. and Prusoff, B.A. (1981). Psychotherapy with depressed outpatients: Patient and process variables as predictors of outcome. British Journal of Psychiatry, 138, 67-74. PMID: 7023593.

Rousmaniere, T. (2016). Deliberate Practice for Psychotherapists: A Guide to Improving Clinical Practice. New York: Routledge. ISBN: 978-1138203204.

Royce, J.R. (1987). A strategy for developing unifying theory in psychology. In A.W. Staats and L.P. Mos (Eds). Annals of Theoretical Psychology, Vol 5, 275-286.

Rundle, B. (1995). Concept. In T. Honerich (Ed.). The Oxford Companion to Philosophy (146). Oxford: Oxford University Press.

*Ryba, M. M., Lejuez, C. W. and Hopko, D. R. (2014). Behavioral activation for depressed breast cancer patients: The impact of therapeutic compliance and quantity of activities completed on symptom reduction. Journal of Consulting and Clinical Psychology, 82(2), 325-333. DOI: 10.1037/a0035363.

Ryle, G. (1971). Collected Papers (Volume 2). London: Hutchinson.

Sacks, H. (1972). An initial investigation of the usability of conversational data for sociology. In D.N. Sudnow (Ed.), Studies in Social Interaction (31-74). New York: Free Press. ISBN: 978-0029323601.

*Safren, S.A., Heimberg, R.G. and Juster, H.R. (1997). Clients' expectancies and their relationship to pretreatment symptomatology and outcome of cognitive-behavioral group treatment for social phobia. Journal of Consulting and Clinical Psychology, 65(4), 694-698.

Sartori, G. (1970). Concept misinformation in comparative politics. American Political Science Review. 64, 1033-1046. DOI: 10.2307/1958356. Sartori, G. (1984). Guidelines for concept analysis. In G. Sartori (Ed). Social Science Concepts: A Systematic Analysis (15-88). London: Sage. ISBN: 0-8039-2177-2.

Sartori, G. (2009). Part 1: Sartori on concepts and method. In D. Collier and J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni Sartori. London: Routledge (11-178). ISBN: 0-415-77578-7.

http://psycnet.apa.org/doi/10.1037/a0014333

221 of 271

*Scheel, M.J., Seaman, S., Roach, K., Mullin, T. and Mahoney, K.B. (1999). Client implementation of therapist recommendations predicted by client perception of fit, difficulty of implementation and therapist influence. Journal of Counseling Psychology, 46(3) 308-316. DOI: 10.1037/0022-0167.46.3.308.

Scheirs, J.G.M., Blok, J.B., Tolhoek, M.E.; El Aouat, F., Glimmerveen, J.C. (2012). Client factors as predictors of restraint and seclusion in people with intellectual disabilities. Journal of Intellectual and Developmental Disability, 37(2), 112-120. DOI: 10.3109/13668250.2012.682357.

Shoben, E.J. (1949). Psychotherapy as a problem in learning theory. Psychological Bulletin, 46, 366-392. DOI: 10.1037/h0061680.

Scott, D. (2005). Critical realism and empirical research methods in education. Journal of Philosophy of Education, 39(4), 633-646. DOI: 10.1111/j.1467-9752.2005.00460.x.

*Shahar, G., Blatt, S.J., Zuroff, D.C., Krupnick, J.L. and Sotsky, S.M. (2004). Perfectionism impedes social relations and response to brief treatment for depression. Journal of Social and Clinical Psychology, 23(2), 140-154. DOI: 10.1521/jscp.23.2.140.31017.

Shin, R.Q., Welch, J.C., Kaya, A.E., Yeung, J.G., Obana, C., Sharma, R., Vernay, C.N. and Yee, S. (2017). The intersectionality framework and identity intersections in the Journal of Counseling Psychology and The Counseling Psychologist: A content analysis. Journal of Counseling Psychology, 64(5), 458-474. DOI: 10.1037/cou0000204.

Shorrock, M. (2012). The pragmatic case of Ed. Counselling Psychology Review, 27(2), 23-35. ISSN: 0269-6975.

Siegel, D.J. (2012). Pocket Guide to Interpersonal Neurobiology. New York: Norton. ISBN: 978-0-393-70713-7.

*Simons, A. D., Lustman, P.J., Wetzel, R.D., Murphy, G.E. (1985). Predicting response to cognitive therapy of depression: The role of learned resourcefulness. Cognitive Therapy and Research, 9(1), 79-89. DOI: 10.1007/BF01178752.

Smith, J.A., Flowers, P. and Larkin, M. (2009). Interpretive Phenomenological Analysis: Theory, Method and Research. London: Sage. ISBN: 978-1412908344.

Smith, J.A. and Osborn, M. (2007). Interpretive Phenomenological Analysis. In J.A. Smith (Ed), Qualitative Psychology: A Practical Guide to Research Methods, 53-80. ISBN: 978-141-293084-0.

Smith, M.L., Glass, G.V. and Miller, T.I. (1980). The Benefits of Psychotherapy. Baltimore: Johns Hopkins University Press. ISBN: 0-8018—382-8.

Smith, T. B., Rodríguez, M. D., & Bernal, G. (2011). Culture. In J. C. Norcross (Ed.), Psychotherapy Relationships that Work (2nd ed., 316-335). New York: Oxford University Press. ISBN: 978-0-19973-720-8.

Snape, D. and Spencer, L. (2003). The foundations of qualitative research. In J. Lewis and J. Ritchie (Eds). Qualitative Research Practice: A Guide for Social Science Students and Researchers (1-23). Thousand Oaks, CA: Sage. ISBN: 0761971092.

Spradley, J.P. (1980). Participant Observation. Cengage Learning. ISBN: 978-0030445019.

Sreenan, B., Smith, H. and Frost, C. (2015). Student top tips. In A. Vossler and N. Moller (Eds). The Counselling and Psychotherapy Research Handbook, 245-256. London: Sage. ISBN: 978-1-4462-5527-8.

Stavros, C., & Westberg, K. (2009). Using triangulation and multiple case studies to advance relationship marketing theory. Qualitative Market Research, 12(3), 307-320. doi:10.1108/13522750910963827.

**Steinmetz, J.L., Lewinsohn, P.M. and Antonuccio, D.O. (1983). Prediction of individual outcome in a group intervention for depression. Journal of Consulting and Clinical Psychology, 51(3), 331-337. DOI: 10.1037/0022-006X.51.3.331.

http://psycnet.apa.org/doi/10.1037/h0061680

222 of 271

Stiles, W.B. (1993). Quality control in qualitative research. Clinical Psychology Review, 13, 593-618. DOI: 10.1016/0272-7358(93)90048-Q.

Stiles, W.B. (2002). Assimilation of problematic experiences. In J.C. Norcross (Ed.), Psychotherapy Relationships That Work: Therapist Contributions and Responsiveness to Patients (357-365). Oxford: Oxford University Press. ISBN: 978-0-19-514346-1.

Stiles, W.B. (2003). When is a case study scientific research? Psychotherapy Bulletin, 38, 6-11. Downloaded from file:///C:/Users/Admin/Downloads/Stiles2003PsychotherapyBulletin.pdf on 01.01.2019.

Stiles, W.B. (2005). Extending the Assimilation of Problematic Experiences Scale: Commentary on the Special Issue. Counselling Psychology Quarterly, 18, 85-94. DOI: 10.1080/09515070500136868.

Stiles, W.B., Elliott, R., Llewelyn, S., Firth-Cozens, J, Margison, F.R., Shapiro, D.A. and Hardy, G. (1990). Assimilation of problematic experience by clients in psychotherapy. Psychotherapy, 27, 411-420. Downloaded from https://www.researchgate.net/profile/William_Stiles/publication/232578783_Assimilation_of_problematic_experiences_by_clients_in_psychotherapy/links/09e415089403714e77000000/Assimilation-of-problematic-experiences-by-clients-in-psychotherapy.pdf on 02.12.2017.

Stiles, W.B., Shapiro, D.A. and Elliott, R. (1986). Are all psychotherapies equivalent? The American Psychologist, 41(2), 165-180. DOI: 10.1037/0003-066X.41.2.165.

Stiles, W.B., Shapiro, D., Firth-Cozens, J. (1990). Correlations of session evaluations with treatment outcome. British Journal of Clinical Psychology, 29(1), 13-21. DOI: 10.1111/j.2044-8260.1990.tb00844.x.

*Stirman, S.W., Gutner, C.A., Suvak, M. K., Adler, A., Calloway, A. and Resick, P. (2018). Homework Completion, Patient Characteristics, and Symptom Change in Cognitive Processing Therapy for PTSD. Behavior Therapy, In press as of 06/2018. DOI: 10.1016/j.beth.2017.12.001.

Strauss, A. and Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. London: Sage. ISBN: 978-0803932517.

Strauss, A. and Corbin, J. (1998). Basics of Qualitative Research: Grounded Theory Procedures and

Techniques (2nd Ed). London: Sage. ISBN: 978-0803959408.

Street, E. (2003). Counselling psychology and naturally occurring systems (Families and couples). In R. Woolfe, W. Dryden and S. Strawbridge (Eds). Handbook of Counselling Psychology (2nd Ed., 419-441). London: Sage.

Substance Abuse and Mental Health Services Administration (2012). Enhancing Motivation for Change in Substance Abuse Treatment. Rockville, MD: United States department of health and Human Servoces. Downloaded from https://www.ncbi.nlm.nih.gov/books/NBK64967/pdf/Bookshelf_NBK64967.pdf on 05.01.2019.

Swan, E. (2017). How to use a fishbone diagram. GoLeanSixSigma. Accessed on https://www.youtube.com/watch?v=OPxSbRuv3Cw on 03.06.2017.

Swift, J. K., Callahan, J. L., & Vollmer, B. M. (2011). Preferences. In J. C. Norcross (Ed.), Psychotherapy Relationships that Work (2nd ed., pp. 301---315). New York: Oxford University. ISBN: 978-0-19973-720-8.

Tallman, K. and Bohart, A.C. (1999). The client as a common factor: Clients as self-healers. In M.A. Hubble, B.L. Duncan and S.D. Miller (Eds), The Heart and Soul of Change: What Works in Therapy (91-132). Washington, DC: American Psychological Association. ISBN: 1-55798-557-X.

Taylor, M. (1996). The feminist paradigm. In In R. Woolfe and W. Dryden (Eds). Handbook of Counselling Psychology (201-239). London: Sage.

**Teasdale, J. D., Moore, R.G., Hayhurst, H., Pope, M., Williams, S. and Segal, Z.V. (2002). Metacognitive awareness and prevention of relapse in depression: Empirical evidence. Journal of Consulting and Clinical Psychology, 70(2), 275-287. DOI: 10.1037//0022-006X.70.2.275.

https://doi-org.manchester.idm.oclc.org/10.1016/0272-7358(93)90048-Q

file:///C:/Users/Admin/Downloads/Stiles2003PsychotherapyBulletin.pdf

https://www.researchgate.net/profile/William_Stiles/publication/232578783_Assimilation_of_problematic_experiences_by_clients_in_psychotherapy/links/09e415089403714e77000000/Assimilation-of-problematic-experiences-by-clients-in-psychotherapy.pdf



https://www.ncbi.nlm.nih.gov/books/NBK64967/pdf/Bookshelf_NBK64967.pdf%20on%2005.01.2019


https://www.youtube.com/watch?v=OPxSbRuv3Cw

223 of 271

Thelen, E. and Smith, L.B. (1996). A Dynamic Systems Approach to the Development of Cognition and Action. Cambridge, MA: MIT Press. ISBN: 978-0-26-270059-7.

Thomas, M.L. (2006). The contributing factors of change in a therapeutic process. Contemporary Family Therapy, 28, 201-210. DOI 10.1007/s10591-006-9000-4.

Tilly, C. (2002). Stories, Identity and Political Change. Lanham, MD: Rowman and Littlefield. ISBN: 978-1-461-642-602.

Timulak, L. (2007). Identifying core categories of client-identified impact of helpful events in psychotherapy: A qualitative meta-analysis. Psychotherapy Research, 17(3), 305-314. DOI: 1080/10503300600608116.

**Toth, S. L., Rogosch, F. A., Oshri, A., Gravener-Davis, J., Sturm, R. and Morgan-Lopez, A.A. (2013). The efficacy of interpersonal psychotherapy for depression among economically disadvantaged mothers. Development and Psychopathology, 25, 1065-1078. DOI: 10.1017/S0954579413000370.

Truax, C.B. (1963). Effective ingredients in psychotherapy: An approach to unravelling the patient-therapist interaction. Journal of Counseling Psychology, 10, 256-263.

Tufford, L. and Newman, P. (2010). Bracketing in qualitative research. Qualitative Social Work, 11(1), 80-96. DOI: 10.1177/1473325010368316.

Turner, J. (1981). Social support as a contingency in psychological wellbeing. Journal of Health and Social Behaviour, 22, 357-367. Downloaded from http://www.jstor.org/stable/pdf/2136677.pdf on 11.12.2017.

University of Auckland (no date). Questions about thematic analysis. Accessed on https://www.psych.auckland.ac.nz/en/about/our-research/research-groups/thematic-analysis/frequently-asked-questions-8.html on 26.08.2018.

University of Manchester (2014). Manchester Institute of Education Ethical Practice Policy Guidance. Downloaded from http://documents.manchester.ac.uk/display.aspx?DocID=22801 on 29.05.2018.

Usher, R. (1996). A critique of the neglected epistemological assumptions of educational research. In D. Scott and R. Usher (Eds), Understanding Educational Research. London: Routledge. ISBN: 0-415-13131-6.

Vähäsantanen, K., & Saarinen, J. (2013). The power dance in research interview: Manifesting power and powerless. Qualitative Research, 13 (5), 493-510. DOI:10.1177/1468794112451036

Vayrynen, P. (2015). The Lewd, The Rude and The Nasty: A Study of Thick Concepts in Ethics. Oxford: Oxford University Press. ISBN: 978-0-1902-6217-4.

Vermes, C. (2017). The individualism impasse in counselling psychology. Counselling Psychology Review, 32(1), 44-53.

Vogel, P.A., Hansen, B., Stiles, T.C. and Götestam, K.G. (2006). Treatment motivation, treatment expectancy and helping alliance as predictors of outcome in cognitive behavioral treatment of OCD. Journal of Behaviour Therapy, 37, 247-255. DOI: 10.1016/j.jbtep.2005.12.001.

Walker, L.O. and Avant, K.C. (2014). Strategies for Theory Construction in Nursing (5th Ed.). Harlow, Essex: Pearson Education Limited. ISBN: 978-1-2920-2776-0.

Walker, M. and Rosen, W.B. (2004). How Connections Heal. New York: Guilford Press. ISBN: 1-59385-032-8. Wampold, B.E. (2001). The Great Psychotherapy Debate. Mahwah NJ: Earlbaum. ISBN: 0-8058-3202-5.

Wampold, B.E. (2008). Qualities and actions of effective therapists. American Psychological Association. Accessed from http://www.apa.org/education/ce/effective-therapists.pdf on 18.09.2017.

http://www.jstor.org/stable/pdf/2136677.pdf

https://www.psych.auckland.ac.nz/en/about/our-research/research-groups/thematic-analysis/frequently-asked-questions-8.html

https://www.psych.auckland.ac.nz/en/about/our-research/research-groups/thematic-analysis/frequently-asked-questions-8.html

http://documents.manchester.ac.uk/display.aspx?DocID=22801

https://doi.org/10.1016/j.jbtep.2005.12.001

http://www.apa.org/education/ce/effective-therapists.pdf

224 of 271

Wampold, B.E. and Budge, S.L. (2012). The 2011 Leona Tyler Award Address: The relationship – and its relationship to the common and specific factors of psychotherapy. The Counseling Psychologist, 40(4), 601-623. DOI: 10.1177/0011000011432709.

Wampold, B.E. and Imel, Z. (2015). The Great Psychotherapy Debate: The Evidence for What Makes Psychotherapy Work (2nd Ed). Hove, East Sussex: Routledge. ISBN: 978-0-8058-5709.

Wark, L. (1994). Therapeutic change in couples’ therapy: Critical change incidents perceived by therapists and clients. Contemporary Family Therapy, 16(1), 39-52. DOI 10.1007/BF02197601.

Watson, V.C., Cooper, M., McArthur, K. and McLeod, J. (2012). Helpful therapeutic processes: Client activities, therapist activities and helpful effects. European Journal of Psychotherapy and Counselling, 14(1), 77-89. DOI: 10.1080/13642537.2012.652395.

West, R. and Turner, L.H. (2003). Muted group theory. Introducing Communication Theory: Analysis and Application. ISBN: 0767430344. Downloaded from http://highered.mheducation.com/sites/0767430344/student_view0/chapter28/index.html on 18.11.2017.

Westra, H.A., DoZois, D.J. and Marcus, M. (2007). Expectancy, homework compliance and initial change in cognitive-behavioural therapy for anxiety. Journal of Consulting and Clinical Psychology, 75(3), 363-373. DOI: 10.1037/0022-006X.75.3.363.

Watts, A. (1960). The World as Emptiness. Accessed from http://project.unicorn.holtof.com/watts/the_world_as_emptiness.htm on 05.09.2018.

White, C., Woodfield, K. and Ritchie, J. (2003). Reporting and presenting qualitative data. In J. Lewis and J. Ritchie (Eds). Qualitative Research Practice: A Guide for Social Science Students and Researchers (287-300). Thousand Oaks, CA: Sage. ISBN: 0761971092.

Williams, D.C. and Levitt, H.M. (2007). Principles for facilitating agency in psychotherapy. Psychotherapy Research, 17(1), 66-82. DOI: 10.1080/10503300500469098.

Wilson, G.T., Fairburn, C.G., Agras, W.S., Walsh, B.T. and Kraemer, H. (2002). Cognitive-behavioral therapy for bulimia nervosa: Time course and mechanisms of change. Journal of Consulting and Clinical Psychology, 70(2), 267-274. DOI:10.1037//0022-006X.70.2.267.

Wilson, J. (1970). Thinking with Concepts. Cambridge: Cambridge University Press. ISBN: 978-0-52109-6010.

Woodhead, E.L., Cronkite, R.C., Moos, R.H. and Timko, C. (2014). Coping strategies predictive of adverse outcomes among community adults. Journal of Clinical Psychology, 70(12), 1183-1195. DOI: 10.1002/jclp.21924.

Woolfe, R. (2016). Mapping the world of helping: The place of counselling psychology. In B. Douglas, R. Woolfe, S. Strawbridge, E. Kasket and V. Galbraith (Eds.), The Handbook of Counselling Psychology (4th Ed, 5-19).

Woolfe R. and Dryden, W. (Eds). Handbook of Counselling Psychology. London: Sage. ISBN: 0-8039-8992-X.

Woolfe, R., Dryden, W. and S. Strawbridge (Eds). (2003) Handbook of Counselling Psychology (2nd Ed). London: Sage. ISBN: 0-7619-7207-2.

Woolfe, R., Strawbridge, S., Douglas, B. and Dryden, W. (Eds), Handbook of Counselling Psychology (3rd Ed), London: Sage. ISBN: 978-1-84787-079-7.

World Health Organisation (2013). Counselling for Maternal-Newborn Health: A Handbook for Building Skills. Geneva: World Health Organisation. Downloaded from https://www.ncbi.nlm.nih.gov/books/NBK304190/pdf/Bookshelf_NBK304190.pdf on 28.10.2017.

http://highered.mheducation.com/sites/0767430344/student_view0/chapter28/index.html%20on%2018.11.2017

http://project.unicorn.holtof.com/watts/the_world_as_emptiness.htm


225 of 271

Wuest, J. (1994). A feminist approach to concept analysis. Western Journal of Nursing Research, 16(5), 577-586. DOI: 10.1177/019394599401600509.

*Yen, S., Johnson, J., Costello, E. and Simpson, E.B. (2009). A 5-day dialectical behavior therapy partial hospital program for women with borderline personality disorder: Predictors of outcome from a 3-month follow-up study. Journal of Psychiatric Practice, 15(3), 173-182. DOI: 10.1097/01.pra.0000351877.45260.70.

Yi, J.P., Vitaliano, P.P., Smith, R.E., Yi, J.C. and Weinger, K. (2008). The role of resilience on psychological adjustment and physical health in patients with diabetes. British Journal of Health Psychology, 13, 311-325. DOI: 10.1348/135910707X186994.

Yonkers, K.A., Dyck, I.R., Warshaw, M. and Keller, M.B. (2000). Factors predicting the clinical course of generalised anxiety disorder. British Journal of Psychiatry, 176, 544-549. Downloaded from http://bjp.rcpsych.org/content/bjprcpsych/176/6/544.full.pdf on 28.10.2017.

Yu, J.J. (2011). Predicting psychotherapy client dropout from in-treatment client-reported outcome. Texas A&M University Unpublished doctoral thesis. Downloaded from http://oaktrust.library.tamu.edu/bitstream/handle/1969.1/ETD-TAMU-2011-12-10356/YU-DISSERTATION.pdf?sequence=2 on 30.12.2018.

*Zeeck, A. and Hartmann, A. (2005). Relating therapeutic process to outcome: Are there predictors for the short-term course in anorexic patients? European Eating Disorders Review, 13, 245-254. DOI: 10.1002/erv.646.

*Zuroff, D.C., Koestner, R., Moskowitz, D.S., Mcbride, C., Marshall, M. and Bagby, M.R. (2007). Autonomous motivation for therapy: A new common factor in brief treatments for depression. Psychotherapy Research, 17(2), DOI: 10.1080/10503300600919380.

**Zywiak, W.H., Longabaugh, R. and Wirtz, P. (2002). Decomposing the relationships between pretreatment social network characteristics and alcohol treatment outcome. Journal of Studies on Alcohol, 63(1), 114-121. DOI: 10.15288/jsa.2002.63.114.

http://bjp.rcpsych.org/content/bjprcpsych/176/6/544.full.pdf%20on%2028.10.2017

http://oaktrust.library.tamu.edu/bitstream/handle/1969.1/ETD-TAMU-2011-12-10356/YU-DISSERTATION.pdf?sequence=2

http://oaktrust.library.tamu.edu/bitstream/handle/1969.1/ETD-TAMU-2011-12-10356/YU-DISSERTATION.pdf?sequence=2

226 of 271

APPENDICES

Appendix 1: Database search screenshots

Science Direct

PsycInfo

Step one

Step two

227 of 271

Step Three

228 of 271

Sage Journals

Appendix 2: Primary study inclusion criteria (a) quantitative, qualitative, mixed method outcome studies, meta-analyses or

systematic reviews of outcome studies that identify client factors that mediate,

moderate, predict or are otherwise associated with psychotherapeutic outcomes

(no theory papers)

(b) dated 1960 or later

(c) published in English in a peer-reviewed journal, and freely obtainable through the

University of Manchester library or Google Scholar

(d) participants aged 18+

(e) studies of treatment participants or completers (not predictors/mediators of early

termination or drop-out)

(f) face-to-face therapy modalities including group, family, couples and individual

psychological therapy (excluding on-line modalities)

(g) bona fide therapy and psychoeducation (excluding pure self-help)

(h) delivered by trainee or qualified practitioners

(i) inpatient, day patient, outpatient, community, home, educational and

organisational settings

(j) clients in voluntary therapy (excluding people receiving mandatory or

probationary treatment or those legally detained for mental health treatment)

229 of 271

(k) psychological or behavioural presenting problems including addictions

(excluding health promotion therapy such as HIV programmes or pain

management)

(l) client factor is demographic, behavioural, cognitive, emotional or experiential

(excluding neurological, learning, communication or health conditions)

(m) client factor is a clearly defined personal variable (e.g. demographic

characteristics), process variable (i.e. activity or learning initiated within therapy)

or extra-therapeutic variable (e.g. events and circumstances) whose positive or

negative effects (i.e. mediation, moderation, prediction, or other association) on

outcome of the intervention were demonstrated through the use of recognised

statistical methods in quantitative studies; or client quotes in qualitative studies

(n) the interaction between outcome and the associated client variable in

quantitative studies was statistically significant to p < 0.05

(o) the predictor/mediator is a client-originating circumstantial, demographic,

psychological, cognitive, or behavioural variable, not a variable that is dependent

on therapy configuration, for example, client satisfaction with therapy. To further

illustrate this distinction, for example where client marital happiness is a positive

predictor for reduced risk of depression relapse, this is a client-originating factor;

but where client minority status is a positive predictor of outcome only in therapy

groups comprised of more than half minorities, this is a treatment process factor

230 of 271

Appendix 3: 49 studies extracted from 14 reviews/analyses

231 of 271

Appendix 4: Research data extraction spreadsheet All column headings in data extraction spreadsheet

Date of extraction No. of P’s Setting Significance of

interaction b/w

OM and

predictor

variables

Search source % white/cauc Inpt/day pt/

outpt

Negative

predictors

Included/excluded % female Staff profile Positive

predictors

Reason for

exclusion

Mean age Time points for

OMs

Negative

mediators

Authors %

Married/cohab

Type of design Positive

mediators

Year of pub. Education Analysis Negative

moderators

Title % in

employment

Primary OM Positive

moderators

Journal Presenting

issues

Outcome

variable

Associated

w/neg outcome

Issue, pages, DOI Therapy model Threshold of

improvement

Associated

w/pos outcome

Country Therapy

modality

Predictor OM TA not assoc

w/OM

Quant/qual/mixed # sessions/tx

length

Predictor

variable

TA assoc w/OM

Screenshot of a portion of the research data extraction spreadsheet

Green = excluded after full-text review

White = included after full-text review – studies with no follow-up data

Yellow = included after full-text review – studies including follow-up data

Pink = column titles

232 of 271

233 of 271

Appendix 5: Quality control exclusion criteria (1) The predictive client factor was an aspect of the presenting problem, for example

specific symptoms (e.g. avolition as a predictor of outcome in treatment for

schizophrenia), because these conflate the independent and dependent variables in the

causal argument

(2) No therapy was mentioned; less than 3 sessions of therapy were provided; or length of

therapy is not mentioned

(3) The outcome and/or predictor variable or measures were not specified

(4) The outcome or client variable was partially or wholly judged by the treating therapist,

introducing unacceptably high risk of bias

(5) the relationship between client factors and treatment, or client factors and outcome, was

not discussed

(6) where studies by author(s) with same findings were already included

Additionally in quantitative studies further exclusions were made where:

(7) outcome or predictor measures were not named

(8) outcomes were measured using tests with disputed validity (i.e. Rorschach)

(9) treatment comparison studies that found the predictor variable had differential effects

depending on type therapy (indicating the client factor was confounded with a co-

occurring process mediator)

(10) the predictor variable could not be sufficiently distinguished from a process

variable, for example where a predictor was not a baseline measurement; or where a

mediator represented a therapeutic technique (for example, mindfulness skills as a

predictor of outcome in mindfulness-based cognitive therapy)

(11) the design (e.g. cross-sectional) or statistical analysis (e.g. simple correlation)

inadequately explained the relationship between client factor and outcome

(12) time points for outcome measures were not provided

Additionally, in qualitative studies further exclusions were made where:

(13) the analytic focus was on therapist not client action

(14) the analytic focus was on the ‘meaning of change’ rather than participants’

understanding of how they made change happen

234 of 271

Appendix 6: Tests, variants and causal syntax chart for moderators Moderators are qualitative or quantitative variables that influence the direction or

strength of relationship between the independent and dependent variables in

experimental psychological therapy studies (Baron and Kenny, 1986). Conceptually,

moderators help identify for whom and under what circumstances therapies have

different effects (Kraemer, Wilson, Fairburn and Agras, 2002). Moderators tend to be

contextual factors such as client or therapist demographics, state/trait characteristics,

competencies, enduring attitudes, or extratherapeutic events and circumstances.

In causal syntax, moderators exist prior to the commencement of the independent

variables they interactively moderate, in the following order:

[Moderator A: Extratherapeutic, client or therapist state/trait characteristic] pre-exists the

application of [Independent variable X: client and/or therapist therapeutic activity] and

exerts [specific impedance or facilitation quality C in relation to a control or comparison

group] on B’s potential to achieve [Dependent variable Y: outcome of interest]

Using variables from an actual study (Beutler et.al. 2011b) moderation causal syntax is

illustrated here:

235 of 271

Tests for moderation

Baron and Kenny (1986) describe moderator data types as either continuous, for

example age; or categorical, for example gender. It can however be argued that a

variety of psychologically relevant observable human traits are categorised according to

potentially oppressive, oversimplified, biased, irrelevant or value-laden social

constructs. But for statistical purposes, tests for moderation where the independent and

moderator variables are categorical or continuous are conducted using hierarchical

multiple regression. Where both variables are categorical, ANOVA can be used. The

significance of continuous data moderators can be calculated with single degree of

freedom F-tests; and of categorical moderators with multiple degrees of freedom

omnibus F-tests (Frazier, Tix and Barron, 2004). More detailed statistical procedures for

moderator-outcome research can be found in Baron and Kenny (1986); Hoyle and

Robinson (2004); and Frazier, Tix and Barron (2004).

Variants

236 of 271

Frazier, Tix and Barron (2004) describe three moderator interaction patterns: (a)

‘enhancing,’ where treatment (or other independent variable) and moderator effect the

outcome in the same direction, and together lend a stronger-than-additive effect; (b)

‘buffering,’ where the moderator weakens the effect of the independent variable; and (c)

‘antagonistic,’ where treatment (or other independent variable) has the same effect on

outcome but in the opposite direction. Hoyle and Robinson (2004) add a further

interaction, (d) ‘crossover,’ where the independent variable has opposite effects on the

dependent variable at the two extreme ends of the moderator’s expression.

Categorical moderators are variables that have certain effects in certain therapies and

not others. For example a client who likes observing and documenting their daily

activities might gain more insight into how to change certain behaviours by doing self-

monitoring in CBT, than might someone who finds self-monitoring difficult or boring.

Continuous moderators are variables that have the same effects regardless of the type

of therapy. For example a client who has a firm belief that God loves him may be

emotionally protected by this in difficult times regardless of secular theories of change

his therapist might offer him.

The trajectory of moderator effects might be described as ‘linear’ (i.e. increase in effect

is proportional to increase in application of the moderator); ‘quadratic’ (i.e. moderator

effects are compounded either by time or strength of application); or ‘stepped’ (i.e. time

or strength of application yields a smaller then larger effect, or vice versa) (Baron and

Kenny 1986).

Causal syntax charts

The following chart shows illustrative examples of moderator relationships from four

actual studies.

Study

A. Moderator

(pre-existing

circumstance

that impacts

relationship

between B and

D)

B.

Independent

variable (input

element)

C.

Direction or

strength of

moderator

impact

D.

Dependent

variable

(measured

outcome of

interest)

237 of 271

Example 1:

Beutler et.al.

2011b

Therapist

directiveness

[Therapist factor]

Client reactivity

[Client factor]

Facilitates

Client tardiness

for appointments

Example 2:

Lorenzo-Luaces,

DeRubeis and

Webb, 2014

Number of

previous

depressive

episodes

[Client factor]

Therapeutic

alliance

[Process factor]

Reduces

Depressive

symptom change

Example 3:

Burns and

Nolen-

Hoeksema, 1991

Client

willingness to

learn new coping

strategies

[Client factor]

CBT for affective

disorders

[Therapy factor]

Enhances

Degree of

improvement in

mood

Example 4:

Leibert, Smith

and Agaskar

(2011)

Level of social

support

[Extratherapeutic

factor]

Lower initial

symptom level

[Client factor]

Slows

Change in

symptom level

Appendix 7: Tests, variants and causal syntax chart for mediators Mediators are transformational mechanisms which are activated or changed by events

in the therapeutic process in a way that at least partially accounts for the outcome of the

therapy. Some researchers define mediating processes as correlated with treatment

type, and that mediators are of necessity treatment effects (e.g. Wilson, Fairburn,

Agras, Walsh and Kraemer, 2002). However other researchers allow for mediators to

represent client response or activity in relation to treatment effects (e.g. Berking,

Neacsiu, Comtois, and Linehan 2009), such as cognitive (e.g. insight), emotional (e.g.

dissonance) or behavioural factors (e.g. task diligence) (Baron and Kenny, 1986) that

offer explanatory value about why therapy works (Johansson and Høglend, 2007).

Descriptions of mediation should explain the relational role of the mediator between

relevant variables, for example, “early homework compliance mediates the relationship

between expectancy for anxiety change and initial change in CBT” (Westra, Dozois and

Marcus, 2007, p. 363). Descriptions should also indicate that mediation occurred prior to

outcome, thereby ensuring that the mediator was a result of therapeutic activity.

238 of 271

Building on the A-X-C-Y causal sentence developed in Appendix 6 above, mediators

are introduced as follows:

[Moderator A: Extratherapeutic, client or therapist state/trait characteristic] pre-exists the

application of [Independent variable X: client and/or therapist therapeutic activity].

Moderator exerts [specific impedance or facilitation quality C in relation to a control or

comparison group] on X’s potential to promote [Mediator D: client transformational

thoughts, feelings and behaviours] which partially or completely cause [Dependent

variable Y: outcome of interest]

It is important to note that the syntax above is idealised. The impact of moderators and

mediators are not commonly examined in the same study, although they undoubtedly

simultaneously exert influence on each other and on outcome. The syntax diagram

shown here is not based on a specific study:

Lemmens, Muller, Arntz and Huibers (2016) explain various difficulties in demonstrating

causal relationships between change in the mediator and change in outcome. Where

researchers describe a mediator only in relation to its outcome but not the antecedent

239 of 271

variable it mediates, the relationship is predictive rather than mediating. The majority of

quantitative studies included in this concept analysis describe a predictive relationship

between independent and dependent variables. Only 9.5% (n=6) identified mediators of

change and, interestingly, none identified moderators.

Tests for mediation

Tests of mediation are generally carried out via multiple linear (for quantitative

measures) or logistic (for binary measures) regression analyses; multiple regression

over four steps, or structural equation modelling. These methods are explained in detail

in Frazier, Tix and Baron (2004). In order to establish that change in the mediator

precedes change in the outcome it is important that mediator and outcome are

assessed at multiple points in therapy, not just pre-and post (Johansson and Høglend,

2007).

Mediation is suggested if (a) outcome is significantly related to treatment; (b) treatment

is significantly related to changes in the mediating variable; and (c) the relationship

between treatment and outcome decreases or goes to zero when change in the

mediating variable is entered into the equation (Baron and Kenny, 1986; Dingemans,

Spinhoven and van Furth, 2007).

Variants

While moderators and mediators have distinct and separate roles in influencing

outcomes, Frazier, Tix and Barron (2004) describe two ways mediators and moderators

can interact in certain circumstances: moderated mediation, and mediated moderation.

In moderated mediation, a mediated relation between input and outcome varies across

levels of a moderator. Using smoking cessation in this fictional example, the moderator

is client quit-self-efficacy based on the number of previous failed attempts to quit. The

client undertakes a group-based smoking cessation programme (the input). One of the

treatment mediators is social encouragement to quit provided by group members. The

higher the number of previous failed quit attempts, the less beneficial impact social

encouragement has on the client, and the lower the likelihood of permanent abstinence.

240 of 271

In mediated moderation, a mediator variable accounts for the relation between the

moderator and outcome. Reusing the smoking cessation example where the moderator

remains the client’s quit-self-efficacy based on number of previously failed attempts,

another mediator in the group programme is targeted self-efficacy enhancement

techniques, which interacts with the moderator’s influence on smoking abstinence (de

Hoog, Bolman, Bernt, Kers, Mudde et.al., 2016).


The following chart shows illustrative examples of mediator relationships from actual

studies.

A.

Moderator

(pre-

existing

circumstan

ce that

facilitates

or impedes

B’s impact

on D and

thus E)

B.

Independe

nt variable

(input

element)

C.

Direction or

strength of

moderator

impact on

Mediator D

D.

Mediator (client

thoughts,

feelings and

behaviours

arising or

changing as a

result of therapy

that partially or

completely

cause E)

E.

Dependent

variable

(measured

outcome of

interest)

Example 1:

Dingemans,

Spinhoven

and van Furth

(2007)

Cognitive

behaviour

therapy for

binge

eating

[Therapy

factor]

Reduction in

weight concern

Abstinence

from binge

eating

Example 2:

Wilson,

Fairburn,

Agras, Walsh

and Kraemer

(2002)

CBT and

IPT for

bulimia

[Therapy

factor]

Reduction in

dietary restraint

Reduction

in binge

eating and

purging

241 of 271

Example 3:

Moscovich,

Hofmann,

Suvak and In-

Albon (2005)

Passage

of time

[Extrather

apeutic

factor]

Change in social

anxiety

Change in

depression

Example 4:

Teasdale

et.al. (2002)

Cognitive

therapy

and

mindfulnes

s cognitive

therapy

[Therapy

factor]

Change in

accessibility of

alternative

perspectives

Reduction

in risk of

relapse

In Example 4 (Teasdale, et.al. 2002) low meta-cognitive awareness is a moderator of

outcome which is directly addressed by the mediator, but the researchers do not

discuss or analyse its moderating impact on outcome, only the possible mediating

impact of improving this in therapy.

Appendix 8: Tests and causal syntax chart for predictors Tests

Predictors are calculated using multiple regression analyses. Predictors should be

measured at least at the outset and end of therapy.


The following chart shows illustrative examples of types of predictors from actual

studies.

Study Predictor

variable

Type of

therapy

Time

predictor

was

measured

Function

of

correlation

Outcome

Example 1:

Kindt,

Buck,

Arntz and

Conceptual

processing

[client skill

CBT

Imagery

Rescripting

for PTSD

Pre- and

post-

treatment

Positive

Decrease in

PTSD

symptoms

after therapy

242 of 271

Soeter

(2007)

developed

in therapy]

Example 2:

Dingemans

,

Spinhoven

and van

Furth

(2007)

Palliative

reacting

i.e.

eating/drin

king in

response

to stress

[client

factor]

CBT for

binge

eating

disorder

Pre-, mid-

and post-

treatment

Negative

Abstinence

from binge

eating at end

of therapy

Example 3:

McCrady,

Hayaki,

Epstein

and Hirsch

(2002)

Quality of

pre-

treatment

marital

relationship

[extrathera

peutic

factor]

Alcohol-

related

behavioural

couples

therapy for

men

Pre- and

monthly

post-

treatment

for 6

months

Positive

Abstinence

from drinking

alcohol at 6

month follow-

up

Example 4:

Rounsavill

e,

Weissman

and

Prusoff

(1981)

Psychother

apy

techniques

exploratory

, reflective

and

directive

[Therapy

factor]

Psychother

apy for

depression

Pre-therapy

and at 16

weeks after

commence

ment

Zero

Improvement

in depressive

symptoms

243 of 271

Appendix 9: Participant recruitment poster

Study on how people make counselling work

Are you aged 18+?

Will your counselling finish in May, June or July 2016?

After your counselling is finished, are you interested in confidentially sharing with a

researcher your experience of desired changes happening or not happening?

If you answered YES to these questions, you may be eligible to participate in a Counselling

Psychology doctoral research study.

The purpose of this research is to learn directly from counselling clients how and why

they were able (or not able) to make desired changes in life. And to explore personal and

social factors that can be helpful and unhelpful when trying to make desired changes in

life.

Research interviews take approximately 30-60 minutes of your time and will be held at

Oakwood House, Davenport. You will need to spend a further 15-30 minutes reviewing the

researcher’s transcript/analysis of your interview sometime after your interview.

If you are eligible to participate you will be given a participant information form and a list

of questions to think about in advance of your interview. Your informed consent will be

obtained before commencing. Some demographic details will be collected.

Contact: This study is being conducted by Caroline Vermes, MA, M.Ed, MBACP (Accred.)

who is undertaking the Doctorate in Counselling Psychology at University of Manchester.

You can contact her at [email protected]. Alternatively speak to

your therapist who will pass on your details to the researcher.

Ethical approval granted by University of Manchester Institute of Education

mailto:[email protected]

244 of 271

Appendix 10: Participant information sheet

Personal and social factors associated by clients

with change and non-change since the commencement of therapy

Participant Information Sheet

You are being invited to take part in a research study as part of a student project on the Doctorate in

Counselling Psychology. Before you decide to participate it is important for you to understand why the

research is being done and what it will involve. Please take time to read the following information carefully

and discuss it with others if you wish. Please ask if there is anything that is not clear or if you would like

more information. Take time to decide whether or not you wish to take part. Thank you for reading this.

Who will conduct the research? Caroline Vermes

Title of the Research: Personal and extratherapeutic factors associated by clients with change and non-

change since the commencement of therapy.

What is the aim of the research? This study is designed to explore and analyse how psychological therapy clients describe their own efforts to make desired changes. It aims to understand how and what personal and social resources clients use to help themselves; and what sorts of obstacles clients describe as getting in the way of making desired change. The study seeks a better understanding of how clients work towards therapeutic outcomes. The results may ultimately seek to extend theory about client agency. This will help educate practitioners so they can adapt therapy to identify, work with and support client self-healing capabilities; and understand and address barriers to well-being.

Why have I been chosen? You responded to an invitation to participate. You may have been selected

on the basis of demographic characteristics requested in the call for participants.

What would I be asked to do if I took part? You will take part in a confidential interview for 30 - 60

minutes with the researcher about recent desired changes you may and may not have made in your life.

About a week in advance of your interview, to help you prepare, you will be sent a short list of questions

to think about. Your thoughts about these questions will be discussed in the interview. Afterwards, within

12 months of your interview, you will also be asked to read and comment on a summary of the

researcher’s analysis of the anonymised information you and others have shared. This is likely to happen

over email so you will be asked to supply an email address that you access regularly. If you do not use

email, another communication method will be agreed with you. Reading and commenting on the research

summary may take 15 – 60 minutes depending on how much time you want to give it.

Are there any risks if I participate? There is a small chance some participants may become upset when

discussing their thoughts about the research questions. This could happen before, during and/or after the

interview. It may be related to thinking about or describing any difficulties encountered when trying to

make desired changes in life. The researcher will check this with you in the interview. You will be given

contact details for support services, along with the researcher and research supervisor’s contact details,

should you need to further discuss your reactions to participating in this research.

What happens to the data collected? The interview will be digitally audio-recorded by the interviewer.

The audio file will be stored on securely encrypted memory stick that can only be accessed by the

researcher. The recording will be transcribed and anonymised. Transcripts will be analysed by the

245 of 271

researcher for salient themes. The audio files will be kept securely in electronic storage for up to 5 years

by the researcher then will be permanently deleted by the researcher.

How is confidentiality maintained? Participant identifying information will be anonymised upon

transcription. This means any personal information you share in your interview that could identify you will

be removed before the data is analysed. Basic demographic information including gender, age range,

ethnicity and education will be attached to the transcript for differential analysis purposes. The information

you share with the researcher is confidential. The researcher cannot disclose to anyone that you

participated in the research. However, if you disclosed in your interview any information that suggested

that a child or adult was at risk of neglect or abuse, the researcher would have an ethical responsibility to

report some details to the appropriate authority. Quotes from your interview may be used in the final write

up to illustrate themes or other important findings. These quotes will not include any personal information

that could identify you.

What happens if I do not want to take part or if I change my mind? It is up to you to decide whether

or not to take part. If you do decide to take part you will be given this information sheet to keep and be

asked to sign a consent form. If you decide to take part you are still free to withdraw up to the date that

data from your interview transcript is entered into analysis software 30th September 2016.

Will I be paid for participating in the research? No

What is the duration of the research? 1 x 30-60 minute interview. Then, within 12 months of your

interview you will be sent one or two follow up emails asking for your comments and views on the work.

This may take 15-30 minutes depending on how much time you want to spend on it.

Where will the research be conducted? By prior arrangement at Oakwood House, 104 Kennerley

Road, Stockport, SK2 6EY

Will the outcomes of the research be published? This research may be submitted for publication. It

may also form background research for further research which may be submitted for publication.

Criminal Records Check (if applicable) DBS clearance has been obtained on behalf of the researcher

by the University of Manchester.

Contact for further information [email protected]

What if something goes wrong? If there are any issues regarding this research that you would prefer

not to discuss with Caroline Vermes, please contact the research supervisor Terry Hanley on 0161 275

8815 or [email protected]. If there are issues with the research you wish to discuss with

the researcher, please contact Caroline Vermes on 0786 333 7965 or

[email protected].



246 of 271

Appendix 11: Interview schedule Form A:


started therapy



started therapy


Form B:


started therapy



started therapy


247 of 271

Appendix 12: Participant consent form

CONSENT FORM

If you are happy to participate, please read, tick the boxes and sign the consent form below:

1. I confirm I have read and understand the attached information sheet about the above

study and have had the opportunity to consider the information. I have had the

opportunity to ask questions and have had these answered satisfactorily

2. I understand that my participation in the study is voluntary and that I am free to withdraw

at any time up to the stipulated deadline without giving a reason

3. I consent to my therapist passing my contact details to the researcher

4. I understand that the interviews will be audio-recorded

5. I agree to the researcher creating an anonymised transcript of the audio-recording

6. I agree to the transcript being subject to research analysis

7. I agree to anonymised quotes from my interview appearing in written versions of the

study

8. I agree that any data collected may be published in anonymous form in academic books

or journals

I agree to take part in this project

________________________ _______________ ______________________________

Name of participant Date Signature

________________________ _______________ ______________________________

Name of researcher Date Signature

248 of 271

Appendix 13: Transcript portion [blue highlighted text included in concept analysis]

CV: So to kick off the interview [P4] is there anything in particular that hasn’t changed as you hoped it

might, since you’ve done your therapy?

P4: So yeah, I’ve had a think about this, they’re quite probing questions actually. The main thing…is it

okay to start with a bit of background about me?

CV: It certainly is. And only go as far as you want to with anything that we’re talking about.

P4: Okay. So I was referred through to [service A] by my GP because for quite a long time I’ve had

probably underlying [D3] and [D2] issues. So, I’ve had quite a lot of change over the past few months. So

I decided at the start of the year that I would need...that I probably needed some additional support to

try and break the cycle for good. Because I’ve been through cycles of counselling etcetera before and

possibly not taken it a seriously as I should have done, or given as much into it as I should have done in

the past. So, (sigh) but off the back of that, having come for therapy or counselling to tackle [symptom

of D3] and the associated [D3], the main thing that hasn’t gone as well as I want was the overcoming

[D2].

So I still don’t think that I manage that particularly well. And I think I still have times when I feel quite

overwhelmed with that feeling of anxiousness. And I’m hoping with time, because I, trying to think when

I started with [therapist 4A], it was maybe back in March? And I think…at the time, yeah, it’s probably

got a little bit better. But not to the extent that I would have hoped it would have done. And I’m hoping

now my one-to-one sessions have stopped with [therapist 4A] I’ll go into group sessions for overcoming

[D2] and that will potentially help tackle that a little bit more head on. And also dealing with interacting

with other people that have similar issues might actually be quite useful for me because I’m a bit of a

“heart on my sleeve” type of person. So, I like to talk things through. So I think that…and bounce off

other people, so I do think potentially that will be maybe a more conducive environment to help me get

a better balance over those issues. But I think certainly that would be the one key thing for me, that.

CV: So reduce [D2], overall, that would be something you would hope to have gotten out of therapy, and

feel like you still got a ways to go with that.

P4: yeah, yeah, maybe I put too much of an expectation on what I’d hoped to achieve. And maybe it…I

think part of it is untangling the two issues, although they are very interlinked. I think that, yeah, the

249 of 271

[D2], I’d hoped that that might…because I’d always explain my [D2] on (clicking fingers) causing the [D3].

So, almost like the emotional crutch. So I’d have hoped, and it’s been weird in some sense because, I’ll

come onto why, the thing that has got better, that my [D3] is more under control, but my [D2] is still

quite high.

CV: Yes, yes. So you untangled them.

P4: So that’s a good thing, which I’ll talk a little bit more about later. So I think I’ve been able to separate

the two issues which have been good. But the one that’s benefitted the most over what, the past four-

five months, is the [D3] side of things. The bit that’s still quite bad for me is the [D2].

CV: Do you have any thoughts about why the [D2] is still as bad as it is at this point?

P4: yes. So I think it’s because I’ve had quite a lot of transition recently. I’ve been, I’ve changed roles

recently. I probably put myself under an awful lot of pressure both from an academic and from a

professional point of view. I’ve just moved into quite a senior role. And having left, I was with [previous

workplace], I left [previous workplace] after seven and a half years. So it’s where I grew up, [previous

workplace], from when I graduated. So it’s been quite a big step for me to leave that support network

behind. And moving somewhere that’s completely unknown, and operating at a higher professional

level. I think almost because of those pressures that have been put on, that’s probably kept my [D2] up

here (demonstrating with hand). And maybe had I not gone through that at the time I’ve been going

through therapy, at the moment I may not have had, it may have come down, but I don’t know.

CV: It may have come down but you don’t know. But you’re certainly thinking that the therapy

coinciding with this significant change in your workplace social support, and your workplace

expectations may have something to do with the [D2] being still quite profound?

P4: Yeah. I think so. I think, well, I haven’t learned yet to be able to drop the expectations that I have on

me. And again, that’s compounded by the fact that (clicking fingers) I want to be able to perform well

and create a good impression where I’ve gone. So that’s I think, part of the…it’s all tangled up, but yeah.

And I haven’t learn to re-programme yet. My, almost setting my expectations of myself lower. Not that

necessarily anyone else has that high expectation of me.

CV: Okay, yes. So you see this as coming actually from inside of you, that even though you’ve been

through these external changes, leaving [previous workplace], starting your new job, that there’s

something going on within you that you want to recalibrate.

250 of 271

P4: Yes. And I think I struggle with that, and I don’t know why. I think that I’m a quite practical person

generally. But it’s something that, I don’t know, I’m scared of failing but then I set myself up to fail.

Which is probably (laughing) where the [D2] comes from!

CV: (laughing) But you push yourself to, or your expectations are so high that then there is a chance that

you might not meet them. And then that causes more [D2].

Appendix 14: Transcription considerations I gave considerable thought ‘for and against’ replacing references to formal diagnostic

terms participants used to describe problems they sought counselling help for (such as

‘depression,’ ‘anxiety,’ ‘bulimia’ and so on). There were several reasons I wanted to

replace diagnostic terms with codes. A practical reason was to enhance participant

anonymity. An epistemological reason was that this research is expressly not about

understanding how therapists work with clients with particular diagnoses - a focus that I

believe serves the politics and economics of the psy-professions. I did not want readers,

likely to be psy-professionals, to be tempted to make assumptions about the

participants’ therapeutic journeys, based on particular knowledge and views about

specific diagnoses and their treatment. Ideological reasons to replace diagnostic terms

is that they may obscure the contexts in which my female participants described their

efforts to overcome health and emotional difficulties (Porter 2002); or may pathologise

common female responses to demanding maturational journeys (Chesler, 1972; Brown,

1994). Removing diagnostic terms could assist my intention to contribute to theory

about the complexities of people’s efforts at life improvement “beyond the individualizing

and pathologizing force of psychological categories” (Farley, 2017).

On the other hand, I also had two reasons not to replace participant references to

diagnostic labels. The first was that even with diagnosis terms removed, there was

sufficient description in most interviews of participants’ symptoms and therapeutic goals

that diagnoses might be guessed by a psychologically trained reader. Additionally, all

participants chose to name at least one diagnosis without any prompting from me, and

chose to describe their experience of its impact on their life, so I questioned my use of

authorial power to obscure terms that were clearly important to participants.

251 of 271

I ultimately opted, on the balance of arguments above, to replace diagnostic terms with

coded descriptors. No participant commented on this in their transcript check.

The final transcripts were not wholly verbatim documents. I made three significant

modifications from the spoken to written words of the participants:

A. Locally accented pronunciations were transcribed in standard written English, for

instance the spoken term, “It sounds good that, dunnit?” was transcribed as “It

sounds good that, doesn’t it?” On the other hand, local sentence structure and

word choice were unaltered, for instance, “It were wrote all over her face.”

B. Participant speech fillers were removed, including “um” and “you know.” “Sort of”

and “kind of” were removed where they were qualifiers that did not change the

meaning of the words they preceded; but they were kept where they indicated “a

type of something.” “Like” was removed where it acted as a filler, for example ‘Do

you mean you, like, didn’t stop at the red light?’ However, it was retained where it

was a preposition meaning “for example;” where it was an adjective meaning

“similar;” where it was a verb indicating “finds agreeable;” or where it indicated

the participant’s introduction to retelling someone else’s speech, for example, ‘My

sister was like, ‘Are you coming down the pub with us?’ Response tokens such

as “yes” and “mm-hmm” were also removed from my interjections where they

were simply given as encouragement to participants’ narrative, or indicated that I

was following what they were saying.

C. Punctuation was added. Where a participant spoke in run-on sentences, these

were broken down to briefer segments. For example, the following transcribed

text from Participant 4 was actually spoken as one sentence: ‘I probably put

myself under an awful lot of pressure both from an academic and from a

professional point of view. And I’ve just moved into quite a senior role. And

having left…I was with [previous workplace], I left [previous workplace] after

seven and a half years. So it’s where I grew up, [previous workplace], from when

I graduated. So it’s been quite a big step for me to leave that support network

behind. And moving somewhere that’s completely unknown, and operating at a

higher professional level.’ While adding punctuation I was careful to retain the

252 of 271

meaning and substance of what was said in the conversations, and to be

respectful of participants’ intentions at all times.

Interview texts were tidied or “denaturalised” (Oliver, Serovich and Mason, 2005,

p.1277) in this way for two reasons. First, I was interested in what people said about

their experiences, thoughts, beliefs, feelings and actions, rather than the mechanics of

how they said it. As I planned to include that a substantial number of participant quotes

in my findings and analysis chapters, the addition of punctuation and removal of filler

words improved the readability of the text (Kvale, 2007). Secondly, this editing was

intended to make reading their transcripts more agreeable to participants. My decision

to edit the text this way was reinforced when I sent the first transcribed interview to

Participant 1 as a completely unredacted verbatim document. In her reply, she

expressed disbelief that the transcript accurately replicated her speech patterns:

‘There's one thing that jumps out at me, which is all the ‘um’s’ and ‘kind of’s.’ At first I

thought maybe I just wasn’t aware of how much I said it.’ I did not want to inadvertently

adopt a position of ‘academic power’ over my participants by presenting them with

transcripts that could distract them from checking that I’d accurately transcribed their

narrative; or worse, that could cause them dismay about their conversational speech

patterns. “Our scholarly representations of those performances, if not sensitively

presented, may constitute an attack on our collaborators’ carefully constructed sense of

self” (Borland, 1991, p.71). I explained to P1 I would remove the filler words. After this, I

tidied each text as above before sending completed transcripts to the remaining

participants.

Addition of descriptive statistics and key attributes to transcripts

Date of interview

Length rounded to the closest minute

Total word count

Words per minute (combined speakers)

Number of participant mentions of different types of helping professionals and

services (to assist with analysis of how self-directed versus professional-directed

participants were)

253 of 271

Appendix 15: Interview summaries

1. Dana 6732 words 48 mins (140.25 wpm) 06.06.2016

The change in practical terms:

Sleeping well

What brought this about?

[Attended counselling, used self-help materials]

Sleep inducing medication

Not watching t.v. before bed

Trying not to worry

Confidence arising from landing the perfect job

Yoga

Homeopathy

Got to a really bad place – calling Samaritans

The non-change in practical terms:

Sexual intimacy with partner

Obstacles to change?

[relative]

History of health issues

Negative change in self-image

[relative]

2. Ella 8886 words 50 mins (177.72 wpm) 16.06.2016

The change in practical terms

No more D3 behaviour

Growing the angel, reducing the devil

Stopped dependence on others pushing me. Doing it for myself and others


[Attended CBT, service A programme, participating in a combination of modalities]

254 of 271

Got to a really bad place – coma

Not letting things bother me

Taking better care of self, thinking more clearly

Changed my goals from unrealistic to healthy

Journaling to understand personal susceptibilities

Change in ‘mindset’ (CBT language)

Pushing myself out of (dis)comfort zone to go out socially

Diversion from upsetting thoughts

Setting self-limits

Becoming part of the group

Being loved and needed

Family party

Awareness of ageing process, time ticking, need to be a grown up

Seeing others not getting better; not wanting to be like others

Creating a different version of me

I have to pick myself up rather than relying on others

Stopping avoiding taking care of myself by over-doing for others. Caring more for others by taking better

physical and emotional care of myself

The non-change in practical terms

Still experiencing symptom of D4

Obstacles to change?

This won’t change, I’m stuck with this (it’s not a treatment shortcoming)

Had [D2] since I was a child

Resignation/acceptance that I might be stuck with this

3. Luka 4113 words 30 mins (137.10 wpm) 14.07.2016


In remission from disease

Self-acceptance

255 of 271


[Attended CBT, analytic and online therapy, support group and used self-help materials]

Decision to just deal with it

Got to a really bad place – emergency GP

Asked sister if I was ill and she said yes

Sought help from a range of sources

It just happened naturally, subconsciously; but also through application of “techniques” e.g. goal

modification to more realistic self-care

Non-change in practical terms

Underlying issues unchanged

Obstacles to change

Sporadic attendance

I was distracted, not focussed in sessions, did not know what to say

No in-person relationship with therapist

4. Kyra 7885 words 48 mins (164.27 wpm) 14.07.2016


Separating D2 from D3 as different problems to work on

What has brought this about?

[Telephone and online counselling an advantage as not “fluffy,” helps contain emotion]

Realising I need to make life changes – frustrated with the unwell cycle

Wanting the option to have children

Not wanting to lose relationships with loved ones especially partner

I don’t want to be defined by a problem that has caused multiple emergency hospitalisations (don’t

want to seem vulnerable)

New job, need to be my best

Seeing the problems as a “snappable cycle”

Going to the GP as an entry point

Self-monitoring and reflective learning


256 of 271

D2 not reduced enough

Obstacles to change

Difficult to change “habits of a lifetime”

Still need to understand and explore causes of D2

Think a group might help better than 1:1

Higher professional expectations in new job

High personal expectations

I’m scared to fail but then set myself up to fail

5. Lucy 7654 words 51 mins (151 wpm) 17.08.2016


Getting passion for life back

Getting relationships with friends, family and husband back.

What helped bring this about

Realising I’d become isolated and lonely due to D1

Fear of losing relationships/wanting to preserve relationships

Sharing problems with others; allowing others to help

Seeing I have a choice how to behave, how to look after myself

Physical health deterioration – health crisis

Friends interested in psychology

Self-reflection

Taking ownership of recovery

Reading recovery stories, other people’s experiences in print

Other people’s experiences – known friends

Use of social media support groups

Switching/converting drive for illness to drive for wellness

Using certain character traits as a force for good rather than a force for self-destruct e.g. goal-oriented;

perfectionistic; stubborn

Facing the problem; stopping denial; challenging myself to change old habits and beliefs

257 of 271

Turning point: Friend wanting to talk to me about the problem. Going to the GP


Not overcome D4

Obstacles

Complacency

Denial

False sense of security due to some improvement

The D4 still is strong in my thoughts and feelings

Fear of the unknown

6. Zoey 6609 words 51 mins (129.59 wpm) 22.08.2016


More self-awareness. Understanding why I react certain ways and why I have behaved certain ways in

the past. Understanding how D2 has been present in my life. Seeing D2 as a problem and isolating it

from other issues


[Opening up to therapist and trusting them after a difficult start; using CBT techniques]

Confiding in friends.

Reading The Empathy Trap.

Put things in boxes and label them

Online support group

Reflective/autobiographical writing in blog


How people hurt me and how I react to this in a childlike way – I just put it to one side and keep going.

Obstacles/perpetuators

Friends hurt me. They don’t know how to react to my disclosure about abuse.

Negative reaction to blog

Not being able to trust my friends

Being lonely

258 of 271

7. Thea 5656 words 38 mins (148.84 wpm) 31.08.2016

Change in practical terms

Happier

Going to Uni

Self-worth improved


[Got on well with therapy team and TS2 therapist]

Met gf on Tinder

She made me happier

I trusted her so could accept her compliments which improved how I felt about myself

Physical activity club made me feel strong and capable, sense of mastery

Extended social life though physical activity club

I took therapy more seriously – I was ready – I asked to practice known CBT techniques


Motivation to do routine self-care activities and domestic maintenance (making bed, drawing curtains)


This goal wasn’t the focus of therapy, not what I was referred for, not a pressing health issue

Family did not acknowledge D1 as a problem for me

8. Lily 8037 42 mins (191.35 wpm) 06.09.2016


[Improved D3 through CBT techniques and therapist instructions]

Realising relationship with [exercise] isn’t healthy. Realised there are other things in life. Don’t want to

worry about it any more

More confident

Moving to [English city 1]: fresh start, getting away from environment that reminds me of being ill. Hope

of establishing new routine


Planning ahead. Weekly lists. Awareness of perfectionism and how it works. Internalised therapist tells

me how to respond

259 of 271

Pie chart of self-evaluation

New job


Still bothered by D3 and feelings. Hoped they would go away completely.


2 months of therapy isn’t enough to help me change thoughts and feelings. All it can do is help me

change how I can react. I have innate traits i.e. perfectionism and inner competitiveness that I have to

learn to channel the right way. D3 voice – how to disentangle that from an activity I love and that

defines me.

9. Esme 8730 62 mins (140.84 wpm) 07.02.2017


Relaxing over-self-control without letting go completely


[therapist helped me realise I was too controlling. She didn’t mollycoddle me. She pushed me]

Giving some control over housework to bf and generally being less controlling with him

Pushing myself out of my comfort zone

Getting impatient with myself

Trusting bf’s view of the problem and its effect on our relationship

Reducing self-absorption, improving empathy

Seeing myself through others eyes – adjusting self-perception

Not being able to deny impact of behaviour on health

Aiming to help a friend then realising I should take my own advice


Nothing. But I regret having gone through this experience. It wasted part of my life and didn’t make

anything any better


Culture makes women discontented with themselves

260 of 271

Appendix 16: Entire interview data spreadsheet

Appendix 17: Example ephemeral reflective notes page

261 of 271

Appendix 18: Example of peer debrief meeting minutes Thesis progress meeting

Online meeting minutes Wed 26/09/17

Chair: [redacted]

Minutes: [redacted]

Attended: Caroline, [redacted], [redacted]

Apologies: [redacted]

Caroline:

- 'Old' intro chapter now split into more usual format intro + short lit. review.

- Meeting with Terry on Tuesday 03/10

- Methodology needs some more work, but not re-opened yet.

- Submission planned in May

Aims: next week 04/10/17

Complete both intro and lit review complete for the meeting with Terry (by Monday)

Restart the methods chapter

[remainder of minutes redacted]

Next meeting

Proposed time Weds 04/10/17, 10:30am

262 of 271

Appendix 19: MIE Approval emails

Ethics Education

To:

Caroline Vermes

Cc:

Terry Hanley

Inbox

Tuesday, November 24, 2015 10:28 AM

Dear Caroline Ref: PGR-7232175 Project Title: Personal and extratherapeutic factors associated by clients with change and non-change since the commencement of therapy. I am pleased to confirm that your ethics application has now been approved by the School Research Integrity Committee (RIC) against a pre-approved UREC template. If anything untoward happens during your research then please ensure you make your supervisor aware who can then raise it with the RIC on your behalf This approval is confirmation only for the Ethical Approval application. Regards Georgia Irving

minor MIE ethics change

Terry Hanley

Actions

To:

Caroline Vermes

Tuesday, March 08, 2016 11:58 AM

For info – your thesis change is approved

Terry

Dr Terry Hanley CPsychol AFBPsS | Senior Lecturer in Counselling Psychology | Ellen Wilkinson Building, A6.15 | University of Manchester |

Oxford Road | M13 9PL | tel: +44 (0)161 275 8815 | email: [email protected] (preferred contact) | skype: terry.s.hanley | new

twitter accounts @DrTerryHanley [personal twitter] @UoMCounsPsych [programme twitter]

https://outlook.manchester.ac.uk/owa/redir.aspx?C=PLTrvOipCmG7_OXbRCZXZTGaC-afzkBHh5Xjn--uwp7YzoI9L8XVCA..&URL=mailto%3aterry.hanley%40manchester.ac.uk

263 of 271

Appendix 20: Adverse event flow chart

264 of 271

Appendix 21: Client factors in Castonguay and Beutler (2006) Dysphoria

disorders

Anxiety disorders Substance misuse Personality

disorders

Age Age of adult clients

is inversely related

to the quality of

outcome

Evidence that age

has no prognostic

value

Better outcome

associated with

abstinence-based

social support for

adolescents

Not assessed in

Castonguay and

Beutler (2006)

Gender Young male clients

more likely to drop

out

Mixed evidence

favouring the

conclusion that

gender has no

prognostic value

Mixed and

contradictory

evidence

Insufficient evidence

Ethnicity Insufficient evidence Evidence that

ethnicity has no

prognostic value

(further research

recommended)

Insufficient evidence Not assessed in

Castonguay and

Beutler (2006)

Socioeconomic status Clients from lower

SES more likely to

drop out

Clients from lower

SES more likely to

drop out

Employed clients

have better

outcomes

Clients from lower

SES more likely to

drop out

Religion Insufficient evidence Insufficient

evidence

Not assessed in

Castonguay and

Beutler (2006)

Not assessed in

Castonguay and

Beutler (2006)

265 of 271

Appendix 22: Demographic information sheet Research participant information sheet

Participant number________________________ Counselling at _____________________

Gender (please circle the best option)

Female Male Trans Other Prefer not to say

Age (please circle one)

18-25 26-30 31-40 41-50 51-60 61+ Prefer not to say

Relationship status (circle all that apply)

Single Boyfriend/girlfriend Married (heterosexual) Married/civil partnership (same-sex)

Same-sex long term relationship Heterosexual long-term relationship Divorced

Widowed/Widower Other Prefer not to say

Ethnic group (please circle the best option)

British White Irish White Other White (state)__________________________

British Black (Caribbean) British Black (African) Other Black (state)_________________

Mixed Black Mixed heritage Black and White

Indian (Asian or British Asian) Pakistani (Asian or British Asian) Bangladeshi (Asian or British Asian)

Other Asian Mixed Asian Mixed heritage Asian and White

Other ethnicity (state)_________________________ Prefer not to say

Last level of education completed (circle best option)

Primary school Secondary GCSE A Level/ A Level equivalent

Apprenticeship Vocational training

University Bachelors

University Masters

University Doctorate

Other (state)___________________________

Prefer not to say

Current employment (circle all that apply)

Homemaker Self-employed Full-time employed Part-time employed

Full-time student Part-time student Jobseeker Apprentice Volunteer

Other (state) ______________________________________________

Email address: _________________________________________________________________

266 of 271

Appendix 23: Definitions of causal relationships in research

Definition

Negative predictor The presence of this factor during therapy predicts a poorer

outcome at end of therapy or at follow-up. Statistical analyses

include variants of hierarchical multiple regression

Positive predictor The presence of this factor during therapy predicts a better

outcome at end of therapy or at follow-up.

Negative mediator This factor arises during therapy and mediates between therapy

factors (or other client factors) and a poorer outcome at end of

therapy

Positive mediator This factor arises during therapy and mediates between therapy

factors (or other client factors) and a poorer outcome at end of

therapy

Negative moderator Client characteristic present before the commencement of therapy

that negatively effects outcome

Positive moderator Client characteristic present before the commencement of therapy

that positively effects outcome

Other association with negative

outcome

Non-statistically derived client factors that negatively effect

outcome

Other association with positive

outcome

Non-statistically derived client factors that positively effect

outcome

Appendix 24: Included studies with social factors findings

A. Negative prediction (i.e. adverse social circumstance predicts worse outcome)

*Hartmann, Orlinsky, Weber, Sandholz, Zeeck, (2010)

* Nakano, Lee, Noda, Ogawa, Kinoshita, Funayama, Watanabe, Chen, Noguchi and Furukawa

(2008)

* Dow, Kenardy, Johnston, Newman, Taylor and Thompson (2007)

*Ritsher and Phelan (2004)

Hooley and Teasdale (1999)

Hooley and Teasdale (1989)

B. Positive prediction (i.e. supportive social circumstance predicts better outcome)

*Binford, Mussell, Crosby, Peterson, Crow and Mitchell, J.E. (2005)

267 of 271

* McCrady, Hayaki, Epstein, and Hirsch (2002)

*Zywiak, Longabaugh and Wirtz (2002)

*Steinmetz, Lewinsohn and Antonuccio, D.O. (1983)

* Toth, Rogosch, Oshri, Gravener-Davis, Sturm and Morgan-Lopez (2013)

Mallinckrodt (1996)

Renneberg, Chambless, Fydich and Goldstein (2002)

Jarrett, Eaves, Grannemann and Rush, A.J. (1991)

Fournier, De Rubeis, Shelton, Hollon, Amsterdam and Gallop (2009)

C. Positive association (i.e. supportive social circumstance associated with better outcome)

* Khattra, Angus, Westra, Macaulay, Moertl and Constantino (2017)

Davidson, Borg, Marin, Topor, and Sells (2005)

Constantino, Morrison, Coyne, Goodwin, Santorelli and Angus (2017)

268 of 271

Appendix 25: Helpful elements density table Element Number of

participants (out of 9)

Number of studies

(57 out of 68)

1Positive: Trusting family and friends helped with problem recognition

1Positive A: Partnership quality 5 5

1Positve B: Speaking with family and friends 9 0

1Positive C: Family/friends raised the problem 6 0

2Positive: Hitting rock bottom

2Positive A: Accessing crisis services 4 0

2Positive B: Accessing GP services 7 0

3Positive: Precluding relational loss

3Positive A: Preserving current relationships 6 0

3Positive B: Protecting future relationships 4 0

4Positive: Self-reclamation

4Positive A: Autonomous responsibility 3 2

4Positive B: Intention 6 6

4Positive C: Maturity 7 0

5Positive: Persistently accessing help

5Positive A: Help from family/friends 5 6

5Positive B: Help from staff at school/work 3 0

5Positive C: Self-help and social media 7 0*

5Positive D: Support groups 5 0*

5Positive E: Medication 3 0*

5Positive F: Persistence with counselling 9 0

6Positive: Improving social activity

6Positive A: 6 3

6Positive B: Re-engagement 4 2

6Positive C: Secure private home 2 1

6Positive D: New job/leaving bad job 6 0

6Positive E: Proximalising/distalising 3 0

269 of 271

6Positive F: Role models 4 0

7Positive: Self-helping behaviours

7Positive A: Journalling 5 4

7Positive B: Insight-led behavioural change 8 4

7Positive C: Exposure and flexibility 5 4

7Positive D: Exercise 3 0

8Positive: Helpful thinking

8Positive A: Helping self-talk 8 2

8Positive B: Thought-stopping and restructuring 7 1

8Positive C: Recognising accomplishments and self-efficacy

8 4

8Positive D: Hope for future 7 1

8Positive E: Spirituality 0 3

8Positive F: Gratitude 0 1

9Positive: Reintegrating

9Positive A: Owning evolution 7 1

9Positive B: Self-acceptance 4 7

9Positive C: Life on life’s terms 3 0

9Positive D: Minding regret 2 0

9Positive E: Rejecting unhelpful stereotypes 2 0

*study exclusion criteria precluded possible entries

270 of 271

Appendix 26: Unhelpful elements density table Element Number of

participants Number of

studies (29 out of 68)

1Negative: Compromised support

1Negative A: Difficult family relationships 4 4

1Negative B: Family mental health issues 3 0

2Negative: Ignoring problem

2Negative A: Unrecognised crisis 1* 1

3Negative: Socially disconnected

3Negative A: Alienation 4 2

3Negative B: Guardedness 6 1

3Negative C: Social difficulty 6 4

4Negative: Unready to change

4Negative A: Fear 3 0

4Negative B: Denial 6 0

5Negative: Unable to use help

5Negative A: Find counselling fruitless 4 1

6Negative: Feeling oppressed

6Negative A: Adverse work 7 0

6Negative B: Adverse partnership 0 1

7Negative: Retained unhelpful behaviours

7Negative A: Unhelpful reactions 5 1

7Negative B: Avoidance 6 3

8Negative: Retained unhelpful thinking

8Negative A: Perfectionism 5 3

8Negative B: Unhelpful self-talk 6 2

8Negative C: Obsessions 0 1

8Negative D: Low meta-cognition 0 1

8Negative E: Dissociation 1* 1

9Negative: Unwell identity

271 of 271

9Negative A: Endorsed unhelpful stereotype 5 1

9Negative B: Problem enmeshment 8 0

9negative C: Long illness 6 2

*1 fell beneath threshold of 2 to be considered a participant-informed element

Client factors associated with psychotherapeutic process and ...

Documents

Transcript of Client factors associated with psychotherapeutic process and ...