Synchronization between statistics, research hypotheses and research aims
-
Upload
uitmshahalam -
Category
Documents
-
view
5 -
download
0
Transcript of Synchronization between statistics, research hypotheses and research aims
13/04/2015
1
Dr. Geshina A Mat Saat 7 April 2015
Synchronization between statistics, research hypotheses and research aims
USM/GMS.biostats2015 1
Biostatistics Sharing Session@PPSK
School of Health Sciences, Universiti Sains Malaysia, Kubang Kerian, Kelantan
OBJECTIVES
• Link Bloom’s Revised Taxonomy with why, what, how, when
• To explain the relationship between research aims, research hypotheses & statistics in order to get the research method ‘right the first time’
• Synchronization between statistics, research hypotheses and research aims via experiential learning
OUTCOMES
• Be able to understand & link Bloom’s Revised Taxonomy with why, what, how, when
• Get the research method ‘right the first time’
• Able to synchronize between statistics, research hypotheses and research aims via eexperiential learning
USM/GMS.biostats2015 2
13/04/2015
2
PART ONE: BLOOM’S REVISED TAXONOMY
USM/GMS.biostats2015 3
Introduction: Why Bloom’s Revised Taxonomy?
• Graduate students are expected to: – improve their general intellectual skills
– attain proficiency in one or more academic disciplines of
their choice
– develop interpersonal and leadership skills needed for productive careers and effective citizenship
– other than the above, develop professional, research, and scholarship skills
USM/GMS.biostats2015 4
13/04/2015
3
Introduction: Why Bloom’s Revised Taxonomy?
• To what degree have students attained desired educational goals? – in other words, what are the evidences?
• Many universities around the world makes use of Bloom’s Revised Taxonomy to determine & measure desired educational goals.
• Therefore it is to the graduate student’s benefit to learn and understand how to use Bloom’s Revised Taxonomy in the research process and writing.
USM/GMS.biostats2015 5
Introduction: Why Bloom’s Revised Taxonomy?
• Bloom's taxonomy refers to a classification of the different objectives that educators set for students (learning objectives).
• It divides educational objectives into three domains: – Cognitive: Skills revolve around knowledge, comprehension, and
critical thinking on a particular topic.
– Affective: Skills describe the way people react emotionally and their ability to feel other living things' pain or joy. Affective objectives typically target the awareness and growth in attitudes, emotion, and feelings.
– Psychomotor: Skills describe the ability to physically manipulate a tool or instrument like a hand or a hammer. Psychomotor objectives usually focus on change and/or development in behavior and/or skills.
• Within the domains, learning at the higher levels is dependent on having attained prerequisite knowledge and skills at lower levels (Orlich et al, 2004)
USM/GMS.biostats2015 6
13/04/2015
4
• Typically, the top three levels within a domain are used as guides to formulate postgraduate level objectives.
• This is to draw attention to a researcher’s cognitive acumen, affective development and psychomotor skills.
• The following table presents examples or appropriate objectives at the higher ends of the three domains for research.
USM/GMS.biostats2015 7
Introduction: Why Bloom’s Revised Taxonomy?
USM/GMS.biostats2015 8
(Adapted from Anderson et al, 2000)
Bloom’s Revised Taxonomy: Cognitive domain
13/04/2015
5
USM/GMS.biostats2015 9
Category or 'level' Behavior descriptions Examples of experience, or demonstration and evidence to be measured
'Key words'
1. Receiving Open to experience, willing to hear
Listen to teacher or trainer, take interest in session or learning experience, take notes, turn up, make time for learning experience, participate passively
Ask, listen, focus, attend, take part, discuss, acknowledge, hear, be open to, retain, follow, concentrate, read, do, feel
2. Responding React and participate actively
Participate actively in group discussion, active participation in activity, interest in outcomes, enthusiasm for action, question and probe ideas, suggest interpretation
React, respond, seek clarification, interpret, clarify, provide other references and examples, contribute, question, present, cite, become animated or excited, help team, write, perform
3. Valuing Attach values and express personal opinions
Decide worth and relevance of ideas, experiences; accept or commit to particular stance or action
Argue, challenge, debate, refute, confront, justify, persuade, criticize,
4. Organizing or Conceptualizing Values
Reconcile internal conflicts; develop value system
Qualify and quantify personal views, state personal position and reasons, state beliefs
Build, develop, formulate, defend, modify, relate, prioritize, reconcile, contrast, arrange, compare
5. Internalizing Values
Adopt belief system and philosophy
Self-reliant; behave consistently with personal value set
Act, display, influence, solve, practice,
Source: www.d.umn.edu/vcaa/assessment/bloomoverviw.docx
Bloom’s Revised Taxonomy: Affective domain
USM/GMS.biostats2015 10
Category or 'level' Behavior descriptions Examples of activity or demonstration and evidence to be measured
'Key words'
1. Imitation Copy action of another; observe and replicate
Watch teacher or trainer and repeat action, process or activity
Copy, follow, replicate, repeat, adhere, attempt, reproduce, organize, sketch, duplicate
2. Manipulation Reproduce activity from instruction or memory
Carry out task from written or verbal instruction
Re-create, build, perform, execute, implement, acquire, conduct, operate
3. Precision Execute skill reliably, independent of help, activity is quick, smooth, and accurate
Perform a task or activity with expertise and to high quality without assistance or instruction; able to demonstrate an activity to other learners
Demonstrate, complete, show, perfect, calibrate, control, achieve, accomplish, master, refine
4. Articulation Adapt and integrate expertise to satisfy a new context or task
Relate and combine associated activities to develop methods to meet varying, novel requirements
Solve, adapt, combine, coordinate, revise, integrate, adapt, develop, formulate, modify, master
5. Naturalization Instinctive, effortless, unconscious mastery of activity and related skills at strategic level
Define aim, approach and strategy for use of activities to meet strategic need
Construct, compose, create, design, specify, manage, invent, project-manage, originate
Based on RH Dave's version of the Psychomotor Domain (Developing and Writing Behavioral Objectives, 1970). The theory was first presented at a Berlin conference 1967, hence you may see Dave's model attributed to 1967 or 1970). Source: www.d.umn.edu/vcaa/assessment/bloomoverviw.docx
Bloom’s Revised Taxonomy: Psychomotor domain
13/04/2015
6
USM/GMS.biostats2015 11
Objective Cognitive domain
Affective domain
Psychomotor domain
1 To construct a Standard Operation Procedure for nursing practices in ward hygiene management.
Construct Organization Articulation
2 To distinguish organizational ethics in five departments within Hospital ‘A’
Analysing Valuing Naturalization
3 To create a model linking patient support system and improved health status
Creating Internalising values
Precision
Bloom’s Revised Taxonomy and Research Objectives
Introduction: Why Bloom’s Revised Taxonomy?
• Application of Bloom’s Taxonomy in research provides indicators of professional, research, and scholarship skills.
USM/GMS.biostats2015 12
INTR
OD
UC
TIO
N Statement of the problem
Significance of the study Aims (Direction) Objectives Hypothesis
LITE
RAT
UR
E R
EVIE
W Determine what has already been
written on a topic
Construct an overview of key concepts
Identify major relationships or patterns
Identify strengths and weaknesses
Identify any gaps in the research
Identify any conflicting evidence
Construct a solid background to a research paper’s investigation
13/04/2015
7
Introduction: Why Bloom’s Revised Taxonomy?
Methodology chapter: why, what, how, when
USM/GMS.biostats2015 13
(Adapted from: Saunders et al, 2007)
Ch
apte
r su
b-h
ead
ings
Introduction: Why Bloom’s Revised Taxonomy?
USM/GMS.biostats2015 14
DIS
CU
SSIO
N What was done
What was found
What do the findings mean? To the best of your knowledge, why do you think that is? What accounts for these results?
Why are the findings significant/important/useful? How can they be used, and who can use them? What went wrong?
How do the limitations of your study affect the results?
13/04/2015
8
PART TWO: RELATIONSHIP BETWEEN RESEARCH AIMS,
RESEARCH HYPOTHESES & STATISTICS
USM/GMS.biostats2015 15
The relationship between practical problems and research problems
Practical problem
Research question
Research problem
Research answer
USM/GMS.biostats2015 16
13/04/2015
9
The relationship between research aims, research hypotheses & statistics
Research Aim
Research Hypotheses
Statistics
USM/GMS.biostats2015 17
The relationship between research aims, research hypotheses & statistics
• The development of the research aim, including a supportive hypothesis and objectives, is a necessary key step in producing relevant results to be used in evidence-based practice.
• A well-defined and specific research aim is more likely to help guide in making decisions about study design and population and subsequently what data will be collected and analysed (Brian, 2006).
• A good research question (O’Leary, 2004): – Defines the investigation – Sets boundaries – Provides direction
USM/GMS.biostats2015 18
13/04/2015
10
Research Aim
• An aim identifies the purpose and focus of the investigation.
• In other words, the aim answers the question ‘What is the point of the study?’
• The aim typically involves the word: investigate, investigation, explore, or analyse or other verbs.
• Examples: – Brewin et al (1999) investigated the ability of a diagnosis of
acute stress disorder and its component symptoms to PTSD.
– The aim of Tan et al’s (2014) research was to explore and evaluate previous work focusing on the relationship and links between regulation and PM.
USM/GMS.biostats2015 19
Research Aim
• Objectives specify the components of how the aim will be met: – Deconstruct a research aim into operationalised study
components – what you will do to achieve the aim.
– Primary (or main or general) and secondary objectives (or specific) translate directly into primary and secondary hypotheses which may be measured.
– Prioritise and limit objectives to avoid too many hypotheses and outcomes – what can you actually do given the time & resource limitations.
– Presented in active language, e.g. ‘to quantify’, ‘to determine’
USM/GMS.biostats2015 20
13/04/2015
11
Hypothesis
• A hypothesis (plural hypotheses) is a precise testable statement of what the researcher predict will be the outcome of the study.
• This usually involves proposing a possible relationship between two variables: the independent variable (what the researcher changes) and the dependant variable (what the research measures).
• The hypotheses can be expressed in the following ways: – The null hypothesis states that there is no relationship between
the two variables being studied (one variable does not affect the other). It states results are due to chance and are not significant in terms of supporting the idea being investigated.
– The alternative hypothesis states that there is a relationship between the two variables being studied (one variable has an effect on the other). It states that results are not due to chance and that they are significant in terms of supporting the theory being investigated.
USM/GMS.biostats2015 21
Adapted from: McLeod (2014)
A good hypothesis is short, clear and should include the operationalized variables being investigated.
• A one-tailed directional hypothesis predicts the nature of the
effect of the independent variable on the dependent variable. – E.g.: Children will recall less of what they ate then adults.
• A two-tailed non-directional hypothesis predicts that the
independent variable will have an effect on the dependent variable, but the direction of the effect is not specified. – E.g.: There will be a difference in how many food types are
correctly recalled by children and adults.
USM/GMS.biostats2015 22
Hypothesis
13/04/2015
12
Freeman and Tyrer (1992) offer an example:
• Vague hypothesis: ‘Does psychotherapy help patients with anorexia nervosa?’
• Precisely formulated question: Does psychotherapy, in the form of cognitive therapy, when given for ten weeks, lead to significantly greater gain in weight in anorectic patients than in those not receiving cognitive therapy?’
• Null hypothesis: There is no difference in the weight gain of patients with anorexia nervosa when treated with cognitive therapy compared with a control procedure.
Adapted from: McLeod (2014)
USM/GMS.biostats2015 23
Hypothesis
USM/GMS.biostats2015 24
The Hypothesis Dilemma
A hypothesis may not be appropriate if: • You do not have an educated guess about a particular
situation
• You do not have a set of defined variables
• Your question centres on phenomenological description
• Your question centres on an ethnographic study of a cultural group
• Your aim is to engage in, and research, the process of collaborative change
13/04/2015
13
USM/GMS.biostats2015 25
From Hypothesis to Statistics
• What you write as a hypothesis is directly linked to the statistic used later.
• The type of measurement (slide 27) will determine the type of statistical method used (slide 28).
• Statistical methodology (Lund Research LTD, 2014): – Descriptive statistics summarizes data from a sample using indexes
such as the mean or standard deviation
– Inferential statistics draws conclusions from data that are subject to random variation (e.g., observational errors, sampling variation)
• The next step is to determine whether interest is in relationship or
differences (see slide 28). – ‘Relationship’ will lead to the selection of multiple regression,
linear-regression, spearman’s r or product-moment correlation.
USM/GMS.biostats2015 26
From Hypothesis to Statistics
• Standard statistical procedure involve the development of a null hypothesis. – Working from a null hypothesis two basic forms of error are recognized:
• Type I errors (null hypothesis is falsely rejected giving a "false positive")
• Type II errors (null hypothesis fails to be rejected and an actual difference between populations is missed giving a "false negative").
• In other words, for example - if units of analysis are categorical for both independent and dependent variables, multiple regression can not be done.
• Therefore it is vital to be able to link aim to objective to hypothesis to statistical methodology (descriptive or inferential) to ensure your research methodology is correct and that the items in your quantitative survey are relevant and measurable for the statistical methodology selected.
13/04/2015
14
Scales of Measurement and their Statistics
USM/GMS.biostats2015 27
USM/GMS.biostats2015 28
13/04/2015
15
Selecting a Statistical Test
USM/GMS.biostats2015 29
Non Parametric Tests
USM/GMS.biostats2015 30
Anderson–Darling test Statistical Bootstrap Methods Cochran's Q Cohen's kappa Friedman two-way analysis of variance (by ranks) Kaplan–Meier Kendall's tau Kendall's W Kolmogorov–Smirnov test Kruskal-Wallis one-way analysis of variance (by ranks)
Kuiper's test Logrank Test Mann–Whitney U or Wilcoxon rank sum test McNemar's test median test Pitman's permutation test Rank products Siegel–Tukey test Spearman's rank correlation coefficient Wald–Wolfowitz runs test Wilcoxon signed-rank test
13/04/2015
16
Categorical Data Analysis
• Common tests: – Chi-square – Fisher’s Exact Test – McNemar test
• Reference for conducting categorical data analysis: http://courses.ncssm.edu/math/Stat_Inst/PDFS/Categorical%20Data%20Analysis.pdf
• E-book:
http://www.planta.cn/forum/files_planta/categorical_data_analysis_202.pdf
USM/GMS.biostats2015 31
PART THREE: SYNCHRONIZING
USM/GMS.biostats2015 32
13/04/2015
17
USM/GMS.biostats2015 33
It’s a morbid but necessary job in ageing Japan: to clean up apartments where elderly people have died alone. In March, the body of an elderly man was found on the floor of his apartment in downtown Tokyo. Neighbours hadn’t noticed the octogenarian’s absence. His bank made the rent payments on time, his family didn’t visit, and the only reason for the body’s discovery was a slight smell that troubled the tenant in the flat below. He had been dead for a month. In rapidly ageing Japan, more people are dying alone and unnoticed in a country where more than 5 million elderly people live alone and increasingly isolated lives. For these so-called “lonely deaths”, or kodokushi, families and landlords in Tokyo are increasingly turning to Hirotsugu Masuda and his clean-up crew to salvage apartments where the occupant’s body lay undiscovered for days or weeks. “This has started becoming a bit more common in the world and it’s become more recognised that there’s this sort of job,” says Masuda, whose services are required 3-4 times a week in summer when bodies decompose faster. When Masuda’s team turns up at the Tokyo apartment, police have taken away the corpse but body fluids have seeped into the floor. Workers wearing protective gear spray the apartment with insect repellent, using gloved hands to pack the trash in boxes. The six-hour exercise is conducted discreetly to avoid upsetting neighbours. The crew tells onlookers they are moving house. Masuda’s firm works almost exclusively with “lonely deaths”, charging between ¥81,000 (RM2,494) and ¥341,000 (RM10,497) depending on apartment size. When they are done, incense and flowers are placed where the body had been found, with the man’s photo put where his head had been. Victims forgotten by families are not given a funeral and their remains are interred in unmarked graves. It's a phenomenon that experts say will soon become the norm. "There's likely 40,000 of these cases and we think that in 10 years, it's likely to go over 100,000 cases," says Hideto Kone, an NGO official working on such cases. Data shows victims are more likely to be male. – Reuters Source: Meyers, C (April 3, 2015). Cleaning up Japan’s ‘lonely death’ apartments. The Star online
Synchronizing
USM/GMS.biostats2015 34
Aim Hypothesis Statistics
To investigate why ‘lonely deaths’ occur
There is no relationship between number of family members and ‘lonely deaths’
Linear regression
There is no association between living styles and ‘lonely deaths’
Spearman’s r
There is no difference between genders and ‘lonely deaths’
Chi-square
*Underlined texts = variables of interest. 1. Must first be defined in units of analysis: ratio, interval, nominal, or ordinal 2. Assumptions underlying the statistics must be met before conducting analysis.
A worked example based on the case
If unsure of how to write the aim; use what, where, when, how, or why.
Important words linked to statistical methods to be selected
13/04/2015
18
Partial references
Brewin, C. R., Andrews, B., Rose, S., & Kirk, M. (1999). Acute stress disorder and posttraumatic stress disorder in victims of violent crime. The American Journal of Psychiatry, 156(3): 360-366
Brian H. R. (2006). Forming research questions. J Clin Epidemiol. 59:881–6. McLeod, S. A. (2014). Aims and Hypotheses. Retrieved from http://www.simplypsychology.org/ aims-
hypotheses.html O'Leary, Z. (2004) The Essential Guide to Doing Research. London: Sage. Orlich, D., Harder, R., Callahan, R., Trevisan, M., Brown, A. (2004). Teaching strategies: a guide to effective
instruction (7th ed.). Houghton Mifflin. Tan, K.H., Shi, L., Tseng, M.L., Cui, W.-J. (2014). Managing the indirect effects of environmental regulation and
performance measurement, Industrial Engineering and Management Systems, 13 (2): 148-153. V. (July 17, 2013). How to write a Discussion chapter for your thesis or dissertation. Retrieved from:
http://prconnections.net/2013/07/17/how-to-write-a-discussion-chapter-for-your-thesis-or-dissertation/
Queensland University of Technology (Feb 12, 2015). Writing a literature review. Retrieved from:
http://www.citewrite.qut.edu.au/write/litreview.jsp
USM/GMS.biostats2015 35