DESIGNING QUALITATIVE RESEARCH
description
Transcript of DESIGNING QUALITATIVE RESEARCH
1
MELLISSA WITHERS, M.H.S., PH.D.CMORE SERIES
MARCH 19, 201312PM
DESIGNING QUALITATIVE RESEARCH
2
Not everything that can be counted counts and not everything that counts can be counted.
--Albert Einstein
3
OVERVIEW
Review What is Qualitative Research? When to use Qualitative vs. Quantitative
Methods
The Steps Sampling Methods Analysis
4
5
WHAT WENT WRONG?
6
A QUICK REVIEWFrom Harris, S. (1991). “You want proof? I’ll give you proof!”: More cartoons from Sidney Harris. New York: W. H. Freeman and Company.
7
WHAT IS QUALITATIVE RESEARCH?
Any type of research that produces findings not arrived by statistical procedures or other means of quantification.
Refers to research about:Person’s livesLived experiencesBehaviors, emotions, feelingsOrganizational functionSocial movementsCultural phenomena
Strauss & Corbin, 1998
8
QUALITATIVE RESEARCH
GOALS: Explore, discover, understand, describe
TYPES OF RESEARCH QUESTIONS:Why?How?
9
CHARACTERISTICS
Starts with general question or problem
Usually no pre-defined hypothesis
Uses a small, purposeful sample (not random)
Often done in naturalistic settings
Creswell, 2009
10
CHARACTERISTICS (cont.)
In-depth analysis
Present results descriptively
Focus on participants’ meanings; holistic account
Emphasize personal experiences and interpretation over quantification
Interpretive; Researcher as key instrument- awareness of own orientations, biases, experiences
11USE QUALITATIVE METHODS…
To explore a topic/population that has previously not been studied
To gain a more holistic, contextualized understanding
To gain insight into possible causal mechanisms
To study behavior in a natural setting (gangs, homelessness, drug use, etc.)
12
USE QUALITATIVE METHODS…
To understand local terms, meaning
To develop survey instruments
To explore quantitative findings
To test feasibility
To test standardized measurements or instruments on different cultures, populations
13
14
15
16
17
18
WHERE DO YOU GO FROM HERE?
19
THE PROCESS
Systematic process!
STEP 1: Decide on sampling procedure
STEP 2: Decide on the specific methods
STEP 3: Decide on how to analyze data
20
HOW TO DESIGN A STUDY…
STEP 1: DECIDE ON A SAMPLING PROCEDURE
How much is known about this topic and population
Clear eligibility/exclusion criteria
Homogeneity of the population
How to gain access to this population
Your resources
21
APPROPRIATE SAMPLE DESIGN DEPENDS ON…
Degree of accuracy required
Range of possible experiences (homogeneity of target group)
Need for statistical analysis
Difficulty reaching your population
Ethical issues
Resources (i.e. time / money)
22
SAMPLING
Adequacy of sample depends not so much on the number of cases
Need for smaller but focused samples rather than large, random samples
Who will you learn the most from?
Depends on the proper specification of the cases to be analyzed
23
SAMPLE SIZE
Required size (n) is often unknown
Recent guidance- n=30 is large enough sample (Dworkin, 2012)
Maximize possibility that all perspectives (positive and negative cases) have been explored
Redundancy in information is often a sign that the sample size is adequate -“saturation point” (Morse, 1995)
24
PURPOSEFUL SAMPLING
Non-probability or non-random sampling
Aim is not statistical representativeness but to gain access to the full range of views, perspectives, themes in the population
Used when there is a limited number of individuals who have the relevant information
Sometimes the only meaningful way to investigate
25
PURPOSEFUL SAMPLING
Select cases rich in information with in-depth understanding
Choose subjects who are in the best position to provide data
“Maximum Variation”
26
PURPOSEFUL SAMPLING
Examples: Snowball Quota Stratified Convenience Homogenous Typical case vs. extreme/deviant/critical case Maximum variation
Patton, 1990
27
PURPOSEFUL SAMPLING-EXAMPLES
Snowball sampling Starting with a small group and asking for further contacts Useful for sensitive topics
Quota sampling Population is stratified and numbers within strata are decided Contacts are made until quotas are full Quotas can be proportional or non-proportional to the population
Confirming or disconfirming cases Other examples to confirm research
Criterion sampling all meet pre-determined criterion--- ex: all dropped out of school,
all in Iraq war
28
PURPOSEFUL SAMPLING-EXAMPLES
Convenience sampling e.g., interviews on the street; simply asking for
volunteers; using clients in clinical or business setting quick, convenient, less expensive not generalizable at all Sometimes only way to reach population
Stratified purposeful sampling Need to stratify in order to make generalizations
about comparisons between groups Each strata will be homogeneous
29SAMPLING EXAMPLES
Sexual practices among men on the down low
Women who use cocaine and who have children under age 5
Why some immigrant Asian women do not seek pre-natal care
Age at first sex among Latino men
Stigma among Mexican parents of children who have cleft lip or palate
30
HOW TO DESIGN A STUDY…
STEP 2: DECIDE ON A METHOD
How detailed you need the results
Potential biases
Your time, $$ and resources
31
METHODS
Depend on research question & theoretical and philosophical framework
Examples of methods:Documentary ResearchParticipant ObservationInterviewsFocus GroupsCase StudiesOral HistoriesObservationsPhotovoice
32
SELF-REFLEXIVITY
Objectivity is not possible (not required)
Acknowledge your framework, experience, perspective
Researcher influences process from the very beginning all the way through (research question)
Requires self-awareness, transparency
Convince the reader that the researcher(s) is sufficiently knowledgeable and will produce trustworthy results
33
RELIABILITY & VALIDITY
Instead, think of:TrustworthinessApplicabilityRespectAuthenticityFairnessCredibilityMeaning in-contextConsistency
34
STRATEGIES TO INCREASE TRUSTWORTHINESS
Detailed description of methods
Continuous checking for representativeness of data and fit between coding categories and data
Multiple members of team for analysis
35
STRATEGIES TO INCREASE TRUSTWORTHINESS
Prolonged contact with informants
Continuous validation of data (member checks)
Self-reflexivity; transparency/competence of researcher
Triangulation
36
Multiple data sources
Multiple kinds of data
Multiple data collection strategies
Subjects(data sources)
Data collection strategies
Kinds of data
TRIANGULATION
37
A WORD ABOUT FOCUS GROUPS
Depends on research question and population
Focus groups not appropriate for sensitive topics
Group dynamics may influence process
Logistically difficult
Often no savings for time & $
38
IRB ISSUES
You should have a qualitative expert help you write the study protocol (methods) and the IRB application
You will need documents such as: Question guides Screening scripts Recruitment materials
39
40
HOW TO DESIGN A STUDY…
STEP 3: DECIDE ON DATA ANALYSIS PLAN
Analysis methods vary
Usually based on looking for patterns, themes, linkages between them
Represent people through and in their own words (Miles & Huberman, 1994)
Examples: content analysis, grounded theory
41
DATA ANALYSIS
Selecting, focusing, simplifying & transforming
Not linear; circular; iterative process
Often occurs simultaneously with data collection
Multiple readings of data
Examination of patterns/themes
42
DATA ANALYSIS
All analytical choices (which codes, quotations to use)
Use of quotes, examples; usually not quantifications
Coding, summaries, clusters: a final report
Miles & Huberman, 1994
43
44
DATA ANALYSIS SOFTWARE
Software such as Atlas.ti, Envivo, Nudist helps organize
Does not do the analysis for you
45
ESTIMATING TIME FOR ANALYSIS
Depends on if you are audio-taping, transcribing (and translating from another language) Getting the transcripts can take TIME
Takes much more time to do the analysis- at least 2 hours of analysis for every hour of interview
Should be analyzed by multiple members of team, and validated by member of the target group
46
TRANSCRIPTION?
The need to transcribe depends on how detailed you want the data to be
If rich, detailed quotes will be helpful to illustrate a complex process, consider transcription
Sometimes you can audiotape and review tapes in order to write notes and come up with major themes without nuanced, contextual quotes
47
TRANSCRIPTION SERVICES
Professional transcription services
Estimate about 60-90 mins for interviews & focus groups
$75-90 per hour of audiotape
Assumes you have clear audio and only two English speakers More speakers or foreign language= higher cost
48
REFERENCES
Creswell, J. (2009) Qualitative, Quantitative, and Mixed Methods Approaches. 3ed. Lincoln, NB. Sage.
Dworkin, S.L. (2012) Sample Size Policy for Qualitative Studies Using In-Depth Interviews. Arch Sex Behav ior, 41:1319–1320.
Merriam , S.B. (2002) Qualitative Research in Practice: Examples for Discussion and Analysis. San Francisco, CA. Jossey-Bass.
Miles, M.B. & Huberman, A.M. (1994) Qualitative Data Analysis: An Expanded Sourcebook. 2ed. Newbury Park, CA. Sage.
Morse, J. M. (1995). The significance of saturation. Qualitative Health Research, 5, 147–149
Strauss, A. & Corbin, J. (1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 2ed. Thousand Oaks, CA. Sage.
Patton, M.Q. (2002) Qualitative Research and Evaluation Methods. 3ed. Thousand Oaks, CA. Sage.
49
ACKNOWLEDGEMENTS
Thank you to my mentors at UCLA:
Dr. Kagawa-Singer Dr. Carole BrownerDr. Paula Tavrow