© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

189

Original A

rticle

Blackwell Science, LtdOxford, UKNUPNursing Philosophy1466-769XBlackwell Publishing Ltd 20034Original Article

Truth and Validity in Grounded Theory

Kirsten Lomborg and Marit Kirkevold

Correspondence: Kirsten Lomborg, Department of Respiratory

Diseases, University Hospital of Aarhus, DK – 8000 Aarhus C,

Denmark. Tel.: 45 89492113; fax: 45 89492110;

e-mail:k-lomborg@akh.aaa.dk

Truth and validity in grounded theory – a reconsidered realist interpretation of the criteria:

fit

,

work

,

relevance

and

modifiability

Kirsten Lomborg

*

,†

RN

BA MSN

and Marit Kirkevold

*

,‡

RN EdD

*

Department of Nursing Science, Faculty of Health Sciences, University of Aarhus, Denmark,

†

PhD Student, Department of Respiratory Diseases, University Hospital of Aarhus, Denmark, and

‡

Professor, Institute of Nursing Science, University of Oslo, Norway

Abstract

Grounded theory is a frequently used approach in nursing research.Over the years the methodology has developed in different directionswith ambiguous answers to questions of truth and validity. This ambi-guity influences the interpretation of the criteria for quality judgementof grounded theories:

fit

,

work

,

relevance

and

modifiability

. In particular,the criterion

fit

seems to be caught in a vacuum between differentepistemological and ontological positions.

Fit

can be interpreted eitherfrom a realist or from a nonrealist position but both present problems.A realist position is problematic if it insists on an immutable empiricalworld and ignores the social and psychological aspects of human life. Anonrealist position can either be argued to rely on hidden realistassumptions and therefore to be inconsistent, or it can be relativistic,opening up the possibility of ‘anything goes’ attitudes in research andsolipsistic confirmations of the world view of researchers with little ormisleading practical impact. A reconsideration of the realist position issuggested, in which validity is regulated by the social constructed reality‘as it really is’. From this position

fit

is a matter of correspondence tofacts in social reality. The criteria

work

,

relevance

and

modifiability

areargued to support the fitness of a theory, and to be useful in the broaderevaluation of the quality of grounded theories.

Keywords:

grounded theory, methodology, epistemology, realism,

truth, validity.

190

Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

Introduction

Thirty-six years have passed since grounded theorywas first introduced as a sociological method forempirical research (Glaser & Strauss, 1967). Sincethen the methodology has been frequently used notonly in sociology, but also in other disciplines andprofessions and grounded theory has become one ofthe most common methodological approaches innursing research (Morse, 2001). As nursing is apractical discipline, the pragmatic position ofgrounded theory is appealing, and with its embed-ded focus on human action and interaction, themethodology seems relevant for studies of people’sresponses to illness and disease and patient–nurseinteractions.

Originally the sociologists Glaser (1930–) andStrauss (1916–96) introduced grounded theory as amethodology that emphasized the practical useful-ness of the theories in the social area under study, andthe concepts

fit

,

work

,

relevance

and

modifiability

were introduced as criteria for quality judgement(Glaser & Strauss, 1967). Assumptions about truthand validity were only a secondary issue, resting onepistemological and ontological premises that weremore or less taken for granted. Over the years themethodology has developed in different directions(Glaser, 1992; Stern, 1994; Annells, 1997b; Strauss &Corbin, 1998b; Charmaz, 2000), and postmodern andconstructivist versions of the methodology withinherent nonrealist interpretations of truth and real-ity are now being advanced (Charmaz, 2000; Mac-Donald & Schrieber, 2001). Considerations of theimplications of these nonrealist interpretations inrelation to the validity and the quality judgement ofgrounded theories are, however, scarce.

Ambiguities in the different interpretations ofgrounded theory call for a clarification of the episte-mological and ontological premises. When consider-ing truth and validity,

fit

– the extent to which a theoryfits the empirical situations in the social area understudy – deserves particular attention.

Fit

can either beinterpreted from a realist position as a matter of cor-respondence to reality or it can be interpreted fromone of the various nonrealist positions as a matter ofcoherence, consensus or pragmatic usefulness,

depending on the theory of truth inherent to the spe-cific nonrealist position. Both the realist and thenonrealist positions present problems. Realism isproblematic if it insists on a fixed or immutableempirical world and ignores the social and psycholog-ical aspects of human life. These ideas have beenwidespread until now, inherited from the logical pos-itivistic and postpositivistic traditions. Nonrealismcan either be argued to rely on hidden realist assump-tions and therefore to be inconsistent, or it can berelativistic. With a radical nonrealist position, the sci-entific reasoning during the research process fallsapart, and everyone (individual researchers and aca-demic societies) can maintain whatever theory theyprefer, based on their own construction of reality.Such relativist theoretical constructions tend to havelittle or, in the worst case, a misleading practicalimpact.

The endeavour of this paper is to reconsider a real-ist position that acknowledges the human constructedsocial reality ‘as it really is’ as an ideal against whichscientific theories can be assessed, and in additionjustifies the Correspondence Theory as a method-ological tool for the investigation of social facts. Forthis purpose, the American philosopher John R.Searle’s (1995) analysis of the construction of socialreality is introduced. His realist position is argued tobe consistent with grounded theory. His insistence onan external social reality means that

fit

can be definedas a matter of correspondence to facts in this socialreality. Consequently, the validity of a grounded the-ory is regulated by empirical facts and reasoning, andcan be assessed throughout the whole research pro-cess from data collection and analysis to theorydevelopment. The criteria

work

,

relevance

and

modi-fiability

are argued to support the criterion

fit

and tobe practically useful in the broader evaluation of thequality of grounded theories.

Grounded theory

Before going into details of the concepts of truth andvalidity, some background knowledge of groundedtheory is needed. In the following section key aspectsof the methodology will be summarized. First, theintentions and characteristics of the methodology are

Truth and Validity in Grounded Theory

191

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

presented in brief. Secondly, the original criteria forquality judgement are outlined.

Intentions and characteristics

The methodology was introduced in

The Discovery ofGrounded Theory

(Glaser & Strauss, 1967) as amethod to explore social processes and reveal thehuman characteristic of anticipating and respondingto various life circumstances. The methodologyattempted to close ‘the embarrassing gap betweentheory and empirical research’ (Glaser & Strauss,1967, p. vii). From a symbolic interactionist perspec-tive, Glaser and Strauss provided a critique of theoverspecialized deductive use of pre-establishedgrand theories in sociology, and from this pragmaticposition, combined with a rigorous, systematicapproach, they developed their constant comparativemethod.

1

This approach was intended to ‘ground’ thetheory, which Glaser and Strauss thought was possiblewhen generating theory from empirical data and witha new type of systematic comparative analysis. Theydid not claim to present a cut-and-dried method thatshould once and for all prescribe to sociologists howto carry out their research. What they did was to stressthat the constant comparative method should address‘the context of discovery’ which had been repressedby the exclusive focus on ‘the context of justification’,which – not least because of ever more refined meth-ods – was being applied within social science. Due tothis approach, grounded theory attempted to empha-size the usefulness of social theories in practice. Themethodological pivot was the inductive process ofdata collection and analysis in which the generationor development of hypotheses or theories about socialprocesses was carried out systematically.

Criteria for quality judgement

The concepts

fit

,

work

,

relevance

and

modifiability

were introduced as criteria for judging the quality ofa theory. By

fit

was meant that categories shouldemerge from data and should not be selected from apre-established theoretical perspective. ‘Discovered’like this, a grounded theory would

fit

empirical situa-tions. By

work

was meant that theories should pro-vide predictions, explanations and interpretations ofwhat was going on in the area under study. By

rele-vance

was meant that theories should be relevant toaction in the area it purported to explain, focusing onthe emerged core problems and processes. By

modi-fiability

was meant that a grounded theory might gothrough changes when new data emerge, generatingqualifications to the theory (Glaser & Strauss, 1967;Glaser, 1978).

These criteria are closely internally related. Itseems reasonable to believe that, all being equal, arelevant theory should work in practice, and that atheory, generated from empirical data, should be rel-evant. If one recognizes that social reality is dynamicand therefore continuously goes through slightchanges, then the fitness of a theory must be under-stood as being situated in a context with possibilitiesfor modification as the theory is brought into newcontexts where new data are available.

Glaser and Strauss introduced the criteria as beingon the same epistemological level, but a closer lookat the meaning of the four criteria demonstrates thatthe criterion

fit

forms the basis for the other three.The exact meaning of

fit

has, however, become con-fused due to contradictory understandings of theepistemological and ontological foundation ofgrounded theory.

Fit

, therefore, deserves specialattention when considering truth and validity ingrounded theory.

Truth and validity in grounded theory

Is it possible to ‘discover’ a theory? If the answer isyes, then what are we doing when we discover a the-ory, and where should we look for such discoveries?When doing grounded theory we must obviously goto everyday life in the social area under study and

1

In the literature the term ‘grounded theory’ is used for the

methodology and method as well as for the resulting product of

the research, that is, the theory. However, some make a distinc-

tion and label the method ‘the constant comparative method of

qualitative analysis’, and reserve the term ‘grounded theory’ for

the developed theory. In this paper the term ‘grounded theory’

is used in the broad sense, as the originators used it from the

beginning (Glaser & Strauss, 1967).

192

Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

collect data from field (observation and interview)and documentary materials (Glaser & Strauss,1967). However, the epistemological and ontologicalstatus of everyday life is ambiguous. This status hasfundamental implications for the concepts of truthand validity and the criteria for quality judgement.Before we examine this status and elaborate on itsimplications for grounded theory, we should brieflycomment on the realist and the nonrealist philo-sophical stances regarding the nature of truth andvalidity.

Truth and validity

The realist ontological position assumes that anobjective world exists independently of our knowl-edge, beliefs, theories or descriptions about it. Thisreality exists whether or not we can experience it orhave conceptions of its nature. A realist concept oftruth maintains an epistemological dimension, con-tending that human beings have – to a certain extent– access to reality, and that in principle, we have themeans to discover whether a given statement or the-ory is true or false. We say that scientific theories aretrue if they correctly describe or correspond to thereal world (Hussey, 2000). So, from the realist posi-tion it is possible to obtain a substantial amount ofreliable and relatively observer-independent infor-mation about the world. The overall scientific idealaims to access this reality by using disciplined reason-ing and observation to discover and explain regular-ities in the world. Thus the realists claim that althoughthe final truth about reality can never be establisheddue to the limitations of all human knowledge, theidea of truth does not need to be abandoned. Accord-ing to this position theories about the world are dis-covered and the world ‘as it really is’ is the regulatoryideal. In principle, we can therefore have warrantedbeliefs in theories (Packard & Polifroni, 1992; Schu-macher & Gortner, 1992).

In contrast, several nonrealist positions have alsobeen advanced, incorporating a wide variety of philo-sophical views pertaining to truth. These positionsreject ontological and/or epistemological realism, andtherefore truth cannot be related to an external real-ity (Kirkham, 1995). It is beyond the scope of this

paper to elaborate on the diversity of such nonrealistpositions. What is needed is an examination of thefoundational assumptions that are embedded in thedifferent interpretations of grounded theory and aconsideration of the recent developments of themethodology.

The following section will therefore examine: (1)the original version of grounded theory; (2) thedifferent versions of grounded theory that Glaserand Strauss have subsequently developed, and (3)recent postmodern developments of grounded the-ory. The purpose of this examination is to addressthe ambiguities embedded in the grounded theoryapproach.

The original version of grounded theory

In the original introduction of grounded theory, theconcept of truth is only indirectly addressed. Glaserand Strauss seem to hesitate when using the term‘real’, and to avoid using the term ‘validity’. Instead,they use the terms ‘credibility’, ‘trustworthiness’ and‘faithfulness’, which, they state, can be ensured bytheir systematic comparative method:

Paying heed to these strictures on emergence and the appli-

cation of integrative schemes, as well as to strictures on the

emergence of concepts can insure that substantive and for-

mal theories will correspond closely to the ‘real’ world.

(Glaser & Strauss, 1967, p. 42)

The inductive method is supposed to ensure thattheories are ‘closely related’ to the daily realities ofthe substantive area under study:

Clearly, a grounded theory that is faithful to the everyday

realities of a substantive area is one that has been carefully

induced from diverse data, as we have described the process.

Only in this way will the theory be closely related to the

daily realities (what is actually going on) of substantive

areas, and so be highly applicable to dealing with them.

(Glaser & Strauss, 1967, p. 239)

By stating this, Glaser and Strauss simultaneouslyindicate a reservation about the use of ‘real’, and anunderstanding of validity related to truth as corre-spondence to reality, meaning ‘what is actually goingon’. This demonstrates an epistemological ambiguity.

Truth and Validity in Grounded Theory

193

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

The ambiguity may, however, be the result of Glaserand Strauss being particularly cautious in theirapproach to reality and validity, as they were attempt-ing to distance grounded theory from the dominatinguse of grand theories and the prevailing tradition ofsociological research:

This criticism [of grounded theory] stems from sociologists’

taking as their guide to credibility the canons of rigorous

quantitative verification on such issues as sampling, coding,

reliability, validity, indicators, frequency distributions, con-

ceptual formulation, hypothesis construction, and presenta-

tion of evidence. But in this book we have raised doubts

about the applicability of these canons of rigor as proper

criteria for judging the credibility of theories based on flex-

ible research. (Glaser & Strauss, 1967, p. 224)

Still, the original grounded theory approach leavesresearchers and readers with ambiguous understand-ings of the epistemology and the related ontologicalpremises. It is an open question whether the concept

fit

relates to a reality and, if it does, how. The origina-tors each later developed their own direction ofgrounded theory and Strauss was influenced by hiscollaboration with nurse researcher Juliet Corbin(Strauss & Corbin, 1998a). These developmentsincrease the ontological and epistemological ambigu-ities of Glaser and Strauss’s original version.

The two directions of grounded theory

Annells (1997a) distinguishes between Glaser’sclassic mode, characterized as ‘critical realist’ and‘modified objectivist’ and Corbin and Strauss’s refor-mulation, characterized as ‘relativist’ and ‘subjectiv-ist’. Charmaz (2000) recognizes that the twodirections are to a degree conflicting, but states thatboth have positivistic underpinnings. She states thatboth Glaser’s position and Strauss and Corbin’s posi-tion ‘endorse a realist ontology and positivist episte-mology, albeit with some sharp differences’ (p. 513).Where Glaser remains positivistic, Strauss andCorbin move into postpositivism, she states, becauseStrauss and Corbin give voice to their respondents,acknowledging their view of reality. Charmaz findsthe position of Strauss and Corbin especially ambig-uous, as they move between objectivist and construc-

tivist assumptions. This is consistent with MacDonaldand Schreiber who state that with Strauss and Corbinpeople can find support for any ontology they wish(MacDonald & Schrieber, 2001).

The above interpretations of the two directions arenot identical. We need to look at Glaser’s and Corbinand Strauss’s original writings. A review of Glaser’sposition (Glaser, 1978; Glaser, 1992; Glaser, 1999;Glaser, 2001) shows that for him the inductive methodfor generating theory and the analytical process ofabstraction and conceptualization are the most impor-tant aspects of grounded theory. In fact, Glaser payslittle attention to communicating the underlying epis-temology, and he warns researchers not to becomeparalysed when trying to figure out how to get ‘bed-rock objectivity’ and stay conceptual (Glaser, 2001, p.48). Still Glaser stresses that

fit

is the strength ofgrounded theory, both with the substantive researchdata from which it is generated and with generalizingit to other data by constant comparison. Glaseremphasizes that grounded theory is ‘what is’, and notwhat should, could or ought to be (Glaser, 1999).

Strauss and Corbin explicitly state their foundationin American pragmatism with Dewey and Mead astheir epistemological sources (Strauss & Corbin,1998a). In an overview of grounded theory, theymake clear that ‘a theory is not a formulation of somediscovered aspects of a preexisting reality “outthere”. To think otherwise is to take a positivisticposition that, as we have said above, we reject, as domost qualitative researchers’ (Strauss & Corbin,1998b, p. 171).

2

Instead, Strauss and Corbin imply anonrealist position, stating that truth is enacted and

2

In adopting this argument, Strauss and Corbin imply that a

positivist position is undesirable, but they fail to acknowledge

that some of the most conspicuous directions of positivism –

phenomenalism, empiricism and instrumentalism – are in fact

ontologically nonrealist (Schumacher & Gortner, 1992). The

phenomenalistic version of empiricism, for instance, stresses

that experiences have a subjective character, and in instrumen-

talism scientific theories are nothing more than sets of rules for

connecting one set of observational phenomena with another,

and the existence of things in the world is not a scientific concern

(Chalmers, 1978, p. 116).

194

Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

theories are interpretations made up of different per-spectives at different times. Without further definitionthey claim that it is possible to make judgementsabout the soundness of theories even though they aretemporally limited and never fully established. Theo-ries will always be provisional and become outdatedbecause they are embedded in historical epochs anderas (Strauss & Corbin, 1998b).

Neither the recognition that the truth is never fullyestablished nor that theories are historically embed-ded is, however, contradictory to a realist positionthat assumes the existence of a reality that is – at leastpartly – independent of human consciousness. Itseems unclear whether ‘soundness’ is to be equatedwith validity or not. Like Glaser, Strauss and Corbinmaintain

fit

as a criterion for judgement. A theoryshould be faithful to the substantive data, they claim,repeating from Glaser and Strauss’s earlier work that

fit

indicates some kind of close correspondence.

Fit

,as they further explain, is a powerful condition forunderstanding and usefulness (Strauss & Corbin,1998b), thus leaving the reader with a vague intuitionthat their position is realistic, because correspon-dence normally refers to a realist position.

In summary, the concepts of reality, truth and valid-ity in grounded theory are certainly suggested, butnot sufficiently elaborated in the originators’ writing.Both of the two directions maintain

fit

as a centralcriterion for the judgement of a grounded theory.Glaser does not participate in the discussion on theepistemological and ontological level, and his posi-tion and interpretation of

fit

is therefore vague.Corbin and Strauss indicate a nonrealist position, buttheir argumentation is inconsistent and their use ofthe concept

fit

is ambiguous. Consequently it isunclear both how the researcher should ensure thequality of a theory and how a given grounded theorycan be critiqued. This lack of clarity is also evidentwhen looking more closely at the recent develop-ments of grounded theory.

Recent developments of grounded theory

Annells (1997a) challenges users of grounded theoryto contribute to the development of the method,while justifying their methodological modifications.

She recommends researchers to consider not only theinquiry focus, practical issues, inquiry paradigm andthe intended research product, but also the philo-sophical inquiry regarding concepts of reality, truthand validity (Annells, 1997b). However, mostresearchers using grounded theory do not discussepistemological and ontological matters of truth andreality and some even fail to reflect explicitly on thequality or validity of their research. Still there areexceptions. Recent developments consider groundedtheory in the light of postmodernism and poststruc-turalism, inspired by social constructivism.

Charmaz (2000) advocates a constructivistgrounded theory that assumes relativism and recog-nizes knowledge as a mutual creation between theviewer and the viewed, aiming towards an interpretiveunderstanding of subjects’ meanings. She emphasizesthe distinction between reality and truth like this:

A constructivist grounded theory distinguishes between the

real and the true. The constructivist approach does not seek

truth – single, universal, and lasting. Still, it remains realist

because it addresses human realities and assumes the exist-

ence of real worlds. (Charmaz, 2000, p. 523)

Charmaz rejects the existence of a single unidi-mensional world, stressing that our knowledge isbased upon our perspective. In relation to symbolicinteractionism, Charmaz states that research prod-ucts do not constitute the reality of the respondents’reality (Charmaz, 2000). In other words theresearch product is influenced by the world view ofthe researcher and the definitions of the situationthat the researcher relies on. It will always be a con-struction built upon dialectic relations among therespondents and the researcher. Due to the impor-tance of applicability and usefulness, Charmazacknowledges pragmatism in grounded theory, butrejects an aspect of the symbolic interactionistextension, namely the obdurate character of sociallife that is embedded in Blumer’s symbolic interac-tionism. According to Charmaz (2000, p. 523), real-ity is a construction that is made by human beings:‘Thus a grounded theorist constructs an image of

a

reality, not

the

reality – that is, objective, true, andexternal.’ Charmaz states that the developed catego-ries should be ‘consistent’ with the studied life, and

Truth and Validity in Grounded Theory

195

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

codes and categories should ‘preserve’ images ofexperience, composing a story.

MacDonald & Schrieber (2001) plead for an inter-pretation of grounded theory in which evolutionalchanges are legitimate. They state that few writingson grounded theory reflect how researchers actuallyuse the method. So grounded theory should be con-structed and reconstructed with a critical stance tomodernist and postpositivist views of the methodol-ogy. MacDonald and Schrieber position themselvesin pragmatism and antifoundationalism with a criticalfeminist perspective, and believe that truth and real-ity are locally constructed, bounded to situation, timeand space. Therefore, truth is not ‘discovered’, but‘constructed’ and ‘reconstructed’, and truth and real-ity are to be judged by those who are in the situations,rather than any externally imposed authority. Thecriteria for judgement of a grounded theory are con-sidered infinite and without firm foundation. Mac-Donald and Schrieber use the term ‘congruent’ forjudging the relation between research results and theparticipants’ experiences and understandings.

In summary, scholars are inviting contributions tothe discussion of the development of new versions ofgrounded theory. It appears that grounded theory ismoving from realism towards relativism. The status ofeveryday life is considered a human construction, andfrom this consideration both Charmaz and Mac-Donald and Schreiber draw the conclusion thatgrounded theories are constructions without externalreferences. From such a relativist position,

fit

cannotbe interpreted as correspondence to reality. If weaccept this rejection of truth as correspondence, howshould

fit

be understood? Alternative terms like‘congruence’, ‘consistency’ and ‘preservation’ aresuggested, but without further elaboration the alter-natives do not clarify a great deal.

The influence from pragmatism, symbolic interactionism and social constructivism

For the purpose of suggesting a clarification of theambiguities, a study of the sources that have inspiredthe diverse approaches in doing grounded theory willenable us to locate some fundamental problems. In

the following section the theoretical traditions of prag-matism, symbolic interactionism and social construc-tivism will be addressed, focusing on their embeddedepistemological and ontological assumptions.

Pragmatism

The emphasis on the practical usefulness of agrounded theory indicates its roots in pragmatism(Mounce, 2000) which is consistent with Glaser andStrauss’s references to Mead, Cooley and Dewey.This pragmatic influence on grounded theory meansthat the epistemological and ontological ambiguitiesare not surprising, because different pragmatic posi-tions have been accused of being inconsistent in theirremarks about truth and reality (Kirkham, 1995).Pierce, the originator of American pragmatism, andother philosophers belonging to the tradition such asJames, Cooley and Dewey can be interpreted in dif-ferent ways. Kirkham (1995) states that all the prag-matic positions are nonrealistic or quasi-realistic,meaning that they do not recognize truth as corre-spondence to an external reality. Kirkham demon-strates that although Pierce claims that truth isdetermined by consensus (in the scientific society)and not by correspondence to an external reality, herecognizes an objective reality, stating that ‘is true’ isequivalent to ‘accurately reflects objective reality’.How can this be? According to Kirkham, this isbecause ‘reality’ in Pierce’s view is not mind-inde-pendent after all. It is independent of individualminds (reality affects us through perceptions that wecannot control), but it is not independent of theminds in the community (the real is the idea on whichthe community ultimately settles). So Pierce’s theoryof truth is circular: ‘is true’ is equivalent to ‘accuratelyreflects objective reality’, ‘accurately reflects objec-tive reality’ is equivalent to ‘consensus in the commu-nity’, ‘consensus in the community’ is equivalent to ‘istrue’. His theory only seems plausible because it isparasitic on the Correspondence Theory of truth, butthis contradicts his understanding of objectivity asnot being independent of the human community’smind. Kirkham (1995) also demonstrates that James’theory seems most viable when treated as a theoryabout what makes a belief valuable (relative to indi-

196

Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

viduals), and not as a theory of truth. These nonreal-ist theories are not necessarily antirealist, but they doexplicitly reject truth as correspondence an externalreality and can be understood as forerunners ofsocial constructivism.

Symbolic interactionism

Grounded theory relies heavily on Mead’s and Blum-ers’s symbolic interactionism (Milliken & Schreiber,2001). Inspired from symbolic interactionism,grounded theory is concerned with the dynamic rela-tionship between persons (individual or collective)and society.

Symbolic interactionism (Blumer, 1986) is a socio-logical extension of a pragmatic position that assumeshuman beings construct and reconstruct the meaningof reality in a constant interaction with the self andothers. Due to this assumption, human action andinteraction and the construction and reconstructionof meaning in everyday life are central phenomena ofinterest for theory development. Symbolic interac-tionism strives, as Blumer states, to acknowledge thenature of the empirical world by organizing a respect-ful methodological stance. Symbolic interactionismrests on three premises: (1) human beings act towardsthings on the basis of the meanings that the thingshave for them; (2) the meaning of things is derivedfrom, or arises out of, the social interaction betweenpeople, and (3) as people deal with the things theyencounter, meanings are modified through an inter-pretative process (Blumer, 1986). So the dynamicprocess of action and interaction constantly changessituations and contexts.

At first sight, the three premises seem to contradictthe idea that there can be some kind of direct objec-tive grasp of the empirical world and that knowledgemirrors what is ‘out there’. However, Blumer posi-tions himself between idealism and a certain type ofrealism. The former he finds solipsistic and untenable,while he finds the latter, which he labels ‘traditionalrealism’, seriously impaired. This is due firstly to itsinsistence on a fixed or immutable empirical worldand secondly to the philosophical doctrine that theempirical world has to be seen and cast in terms ofthe findings of advanced physical science with a per-

nicious impact on social and psychological science.Blumer insists that an empirical science presupposesthe existence of an empirical world, which exists assomething available for observation, study and anal-ysis. This empirical world is the point of departureand the point of return in any case of empirical sci-ence. It has an obdurate character, meaning that it isable to resist and ‘talk back’:

Fundamentally, empirical science is an enterprise that seeks

to develop images and conceptions that can successfully

handle and accommodate the resistance offered by the

empirical world under study. (Blumer, 1986, pp. 22–23)

Reality, he goes on, exists and can be discovered inthe empirical world. The key point is whether what issignified or implied to be the nature of the empiricalworld in a given study is actually the case

.

Social constructivism

As we have seen, recent interpretations of groundedtheory question realism and position themselves withcritical perspectives that advance constructivism andrelativism.

Social constructivism is a movement that has arisenfrom, and is influenced by, a variety of disciplines andintellectual traditions. One of the sources is symbolicinteractionism. There is no single description of socialconstructivism, or social constructionism as it is some-times labelled. Social constructivism ranges betweenso-called ‘weak’ or ‘moderate’, and ‘strong’ or ‘radi-cal’ positions (Schwandt, 2000). Burr (1995) charac-terizes all social constructivist positions as having: a)a critical stance towards taken-for-granted knowl-edge; b) historical and cultural specificity; c) believingthat knowledge is sustained by social processes, andd) believing that knowledge and social action gotogether. However, these characteristics are notexclusively related to social constructivism. The lineof demarcation from other traditions in social andpsychological science is the antirealist doctrine, her-alding a radically different model of what it means todo social science (Burr, 1995). Social constructivismdenies that our knowledge is a direct perception ofreality and offers a view of sociological research thatmaintains that truth is constructed, both individually

Truth and Validity in Grounded Theory

197

© Blackwell Publishing Ltd 2003

Nursing Philosophy

,

4

, pp. 189–200

and collectively, is multiple and shifting, and thatthere are no such things as objective facts.

The problem is, however, that the social construc-tivist position leaves nothing to replace the under-mined epistemological ground. If judgements acrossdifferent individual or cultural theoretical frame-works are not possible, we end up in a solipsisticposition where truth can only be determined withina particular conceptual scheme (Schwandt, 2000).Consequently, people are left with a dilemma of howto construct meaning and make judgements aboutgood and bad, justice and injustice, truth and false-hood in a world without external signposts (Mac-Donald & Schrieber, 2001). Without truth as aregulatory ideal, there is no set of standards againstwhich research results, like grounded theories, can bevalidated and scientific advances can be recognizedand used for practical interventions (Schumacher &Gortner, 1992; Stajduhar

et al.

, 2001).

The implications for

fit

The examination of the pragmatic roots of groundedtheory can explain the embedded ontological and epis-temological ambiguities in grounded theory. Theseambiguities make possible diverse interpretations ofthe methodology, including the recent constructivistmovement, which rejects the objective character ofthe empirical world implied by symbolic interaction-ism. But this nonrealist stance leaves us with seriousrelativist problems. Without possibilities of externaljudgement of the validity of a given theory, the concept

fit

appears vague. We should therefore carefully recon-sider the realistic position to be sure that we are not‘throwing out the baby with the bathwater’. What weare looking for is a position that simultaneouslyacknowledges the human constructed nature of socialreality and retains the objective character of this verysame social area. For this purpose, we may profit fromrecent philosophical considerations of truth and real-ity related to social aspects of human life.

A reconsideration of realism

Over the last decades, realism has been advanced inphilosophical branches of natural and social sciences.

Among others, the philosophers Hacking, Harré andSearle have each contributed with their perspectiveon realism (Hussey, 2000). In the following subsec-tion, Searle’s position will be briefly introduced as aphilosophical foundation that underscores and clari-fies the notion implied by Blumer’s symbolic interac-tionism, thereby supporting the continued relevanceof the original criteria for quality judgement ofgrounded theories.

Realism and the human constructed social reality

Searle’s book The Construction of Social Reality(Searle, 1995) is a treatise on social ontology inwhich Searle considers the preconditions for the pos-sibility of objective judgement about social facts(Hacking, 1997). According to Searle, social realityexists against a background of human beings withminds, intentions and language, and with innate andacquired abilities to understand, interact and com-municate with others. Human beings have a collec-tive intentionality and a capacity to represent objectsand states of affairs. These representations give riseto ‘social facts’, meaning social conditions in every-day life that we constantly encounter. ‘Fact’ is a gen-eral term to designate how-things-are-in-the-world.Some social facts are institutions like ‘hospitals’ and‘nursing’. Concepts that name social facts are self-referential. For instance, when everybody alwaysthinks that certain types of activities in a certain con-text are ‘nursing’ and they use and treat these activi-ties as nursing, then such activities are nursing.Collective imposition of meaning on things and phe-nomena is not necessarily a matter of a deliberateact or set of actions: ‘The creation of institutionalfacts is typically a matter of natural evolution’(Searle, 1995, p. 125).

Searle’s position is ontologically realist. When wemake theories about the world, we use the languageand the concepts at our disposal, but this gives us noreason to abandon the claim that there is a reality thatis independent of our descriptions. Neither does itjustify idealism, which claims that our descriptionsare all that exist.

Searle also defends epistemological realism,advancing a version of the Correspondence Theory

198 Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003 Nursing Philosophy, 4, pp. 189–200

that is immune to claims of not bridging the gapbetween our concepts, statements and theories onone side, and the world on the other side. Searlestates that there has been a misleading tendency tothink that facts, if they really exist, must be a strangekind of complex object and that correspondencemust signify a special type of picturing or isomor-phism. However, there is no gap to close because ourconcepts, statements and theories are used, and canonly be used, against the background of humanexperiences in the world. The world is alreadyembodied in our statements and concepts because itis through our human experiences in the world thatwe learn language and linguistic meaning. Therefore,a statement that we understand cannot have anasserting content of which we do not already knowthe truth conditions.

The statement that ‘true statements are true,because they correspond to the facts’ seems self-evident. Nonetheless, we can add nothing more.‘True’ is just the adjective used for assessing the accu-racy of statements. Statements are assessed as beingtrue when they represent things as they really are.‘Fact’ refers to ‘that in virtue of which a statement istrue’. Facts do not need statements in order to exist,but statements need facts in order to be true. Finally,‘correspondence’ means a general characterization ofthe variety of ways in which statements can accuratelyrepresent how things are.

In summary, Searle’s analysis offers an ontologicaland epistemological realism that acknowledges thehuman constructed nature of social reality and retainsthe objective character of this very same social area.The Correspondence Theory is the methodologicaltool for the investigation of social facts.

Symbolic interactionism in Searle’s realist perspective

From Searle’s position it follows that much of ourhuman life depends on interpreting rightly the mean-ings of symbolic performances by grasping andaccepting social facts as facts. This is consistent withBlumer’s statements about human beings and theiractions, interactions and interpreting processes.Searle’s notion of collective intentionality seems con-

sistent with the symbolic interactionist idea of ‘jointaction’, which refers to a larger form of collectiveaction:

Thus, to act appropriately, the participant has to identify a

marriage ceremony as a marriage ceremony, a hold up as a

hold up, a debate as a debate, a war as a war and so forth.

(Blumer, 1986, p. 71)

Blumer emphasizes that joint action must be con-sidered as built up over time, having a career or ahistory. Such careers of joint action may be under-stood as collective impositions of meaning on thingsand phenomena. Although Searle does not intendto discuss social science and methodology, he sug-gests that working social scientists might find someof his ideas useful in that they should not take insti-tutions for granted (Searle, 1997). Social facts arereal but, given that they depend on somebody rec-ognizing them, they will vanish if people stop takingthem into account. Therefore, social reality isdynamic but that does not contradict the factualityof social reality. The factuality of social reality isprecisely what Blumer is defending when he statesthat the social reality is resistant and able to ‘talkback’: ‘It is the obdurate character of the empiricalworld – its ability to resist and talk back – that bothcalls for and justifies empirical science’ (Blumer,1986, p. 22).

In summary, it seems reasonable to believe thatBlumer’s thoughts are consistent with the philosoph-ical perspective of Searle. His reconceptualization ofrealism may serve as a philosophical foundation forreconsidering the implicit realist stance of symbolicinteractionism, thereby salvaging grounded theoryfrom the nonrealist social constructivist erosion of itsepistemological foundations. Given this reconsidera-tion of realism, the concepts of truth and validity arerelevant aspects of doing grounded theory.

The reconsidered realist version of grounded theory

From the above reconsiderations of a realist positionfollows that a valid theory is a theory that corre-sponds to the phenomena in the area under study. Theconcept fit is a simple way to express the correspon-

Truth and Validity in Grounded Theory 199

© Blackwell Publishing Ltd 2003 Nursing Philosophy, 4, pp. 189–200

dence to social reality, and fit serves the central func-tion of enabling external validation of research insocial processes to take place. Even though infinitevarieties of theoretical interpretations of a givensocial subject can in principle be established, notevery theory fits the subject under study.

Still, it is also important to stress the criteria work,relevance and modifiability for two reasons. Firstly, allbeing equal, workability, relevance and modifiabilityindicate and support the validity of a theory. Sec-ondly, validity is not the only aspect of interest whenconsidering the quality of theoretical findings. A truetheory can be more or less relevant to practice. Ifresearch is to contribute to the development of apractical discipline such as nursing, relevance andworkability are certainly important criteria in theorybuilding, and because of the dynamic nature of socialreality, theories must be modifiable.

Acknowledgements

This paper is part of a Danish research project, sup-ported by grants from Danish Nurses Organization,JCVU Centre for Higher Education, Department ofRespiratory Diseases, Aarhus University Hospitaland the Novo Nordisk Foundation.

We thank Kari Martinsen, Professor at Section ofNursing Science, Bergen University and Steen Wack-erhausen, Associate Professor at Department of Phi-losophy, Aarhus University for encouraging andhelpful comments.

References

Annells M. (1997a) Grounded theory method, Part I: within the five moments of qualitative research. Nursing Inquiry, 4(2), 120–129.

Annells M. (1997b) Grounded theory method, Part II: options for users of the method. Nursing Inquiry, 4(3), 176–180.

Blumer H. (1986) Symbolic Interactionism. University of California, Berkeley, Los Angeles, London.

Burr V. (1995) An Introduction to Social Constructionism. Routledge, London, New York.

Chalmers A.F. (1978) What Is This Thing Called Science? The Open University Press, Milton Keynes.

Charmaz K. (2000) Grounded theory. Objectivist and con-structivist methods. In: Handbook in Qualitative

Research, 2nd edn (eds N.K. Denzin & Y.S. Lincoln), pp. 509–535. Sage, Thousand Oaks, London, New Delhi.

Glaser B.G. (1978) Advances in the Methodology of Grounded Theory: Theoretical Sensitivity. The Sociology Press, Mill Valley, CA.

Glaser B.G. (1992) Basics of Grounded Theory Analyses: Emergence Versus Forcing. The Sociology Press, Mill Val-ley, CA.

Glaser B.G. (1999) The future of grounded theory. Qualita-tive Health Research, 9(6), 836–845.

Glaser B.G. (2001) The Grounded Theory Perspective: Con-ceptualization Contrasted with Description. The Sociology Press, Mill Valley, CA.

Glaser B.G. & Strauss A. (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research. Aldine De Gruyter, New York.

Hacking I. (1997) Searle, reality and the social. History of the Human Sciences, 10(4), 83–92.

Hussey T. (2000) Realism and nursing. Nursing Philosophy, 1(2), 98–108.

Kirkham R.L. (1995) Theories of Truth: A Critical Introduc-tion. Massachusetts Institute of Technology, MA.

MacDonald M. & Schrieber R.S. (2001) Constructing and deconstructing: grounded theory in postmodern world. In: Using Grounded Theory in Nursing (eds S.R. Schreiber & P.N. Stern), pp. 35–53. Springer Publishing Company, New York.

Milliken P.J. & Schreiber S.R. (2001) Can you ‘do’ grounded theory without symbolic interactionism? In: Using Grounded Theory in Nursing (eds S.R. Schreiber & P.N. Stern), pp. 177–190. Springer Publishing Company, New York.

Morse J.M. (2001) Situating grounded theory within quali-tative inquiry. In: Using Grounded Theory in Nursing (eds S.R. Schreiber & P.N. Stern), pp. 1–15. Springer Publish-ing Company, New York.

Mounce H.O. (2000) Pragmatism. Nursing Philosophy, 1(1), 80–81.

Packard S.A. & Polifroni C.E. (1992) The nature of scientific truth. Nursing Science Quarterly, 5(4), 158–163.

Schumacher K.L. & Gortner S.R. (1992) (Mis)conceptions and reconceptions about traditional science. Advances in Nursing Science, 14(4), 1–11.

Schwandt T.A. (2000) Three epistemological stances for qualitative inquiry. Interpretivism, hermeneutics, and social constructionism. In: Handbook of Qualitative Research, 2nd edn (eds N.K. Denzin & Y.S. Lincoln), pp. 189–213. Sage Publication, Thousand Oaks, London, New Delhi.

Searle J.R. (1995) The Construction of Social Reality. Penguin, London.

Searle J.R. (1997) Replies to critics of the construction of social reality. History of the Human Sciences, 10(4), 103–110.

200 Kirsten Lomborg and Marit Kirkevold

© Blackwell Publishing Ltd 2003 Nursing Philosophy, 4, pp. 189–200

Stajduhar K.I., Balneaves L. & Thorne S.E. (2001) A case for the ‘middle ground’: exploring the tensions of post-modern thought in nursing. Nursing Philosophy, 2(1), 72–82.

Stern P.N. (1994) Eroding grounded theory. In: Critical Issues in Qualitative Research Methods (ed. J.M. Morse), pp. 212–223. Sage Publications, Thousand Oaks.

Strauss A. & Corbin J. (1998a) Basics of Qualitative Research, 2nd edn. Sage Publications, Thousand Oaks, London, New Delhi.

Strauss A. & Corbin J. (1998b) Grounded theory method-ology. An overview. In: Strategies of Qualitative Inquiry (eds N. Denzin & Y. Lincoln), pp. 158–183. Sage Publica-tions, Thousand Oaks, London, New Delhi.