L20 Qualitative Research Flashcards
Is qualitative research interpretivism or Positivism
Interpretivism
• Researcher & social world interact
• Characteristied by interpretivist theory of knowledge
• Facts & values not distinct
• Affected by researcher values & perspectives
• Not possible to do objective, value-free research
what is positivism ?
Quantitative
• World unaffected by the researcher • Facts and values distinct, possible to do objective, valuefree research that is generalisable • Natural science method (hypothesis testing, causal explanations appropriate to understand social phenomena)
When to use qualitative research
When there is not much to go on
• Little known about the topic, limited theory to guide topic
• When you want a particular perspective on the topic
• The perspective of a specific person/group, or a more holistic perspective
• When you want to explain quantitative results
• When you want to identify relevant questionnaire items
Key Features in Qualitative Research:
Collect and assess data Abstract and reduce data Explore key themes and patterns Evaluate alternative explanations Formulate and verify conclusions -NON LINEAR, ITERATIVE
- Uses INDUCTIVE reasoning
- using observations to formulate an idea or theory
- exploring new phenomena
- Starting with observations»_space;» theory
- Doesn’t typically use DEDUCTIVE reasoning (
- Starting with theory»_space;» confirmation)
does qual research have high validity and reliability?
Has high VALIDITY
• measuring what we are supposed to measure
• e.g. describe your perception of the experience of chemotherapy in your own words
Has low RELIABILITY
• inconsistent response given same conditions
• e.g. describing the same experience of chemotherapy in a different way on two separate occasion
Methods: Data collection
- Interviews
- Open ended questionnaires
- Observation
- Participant / non-participant
- Overt / covert observation
- Document analysis (written/audio/video)
- Oral histories
Methods: Interview types
- Structured interview
- Set questions, set order, usually with a limited range of responses
- Semi-structured interview
- Open-ended questions, with prompts to gain further insight
- In-depth interview
- One or two issues covered in detail
- Questions follow what the interviewee says
- Focus groups
- A form of group interview that capitalises on communication between group participants
9 sampling methods
- Confirmatory/discriminatory cases
- Elaborates on quant. Findings, looks for typical/variant examples providing deeper understanding
- Confirming patient satisfaction with consultation results - Convenience
- Faster & less expensive; possibly less rigorous
- Asking doctors in a single cancer centre about family involvement in consultations - Criterion
- Sample all cases which meet criteria
- Interviewing all families taking part in consultations in 1 week - Extreme (deviant) cases
- Describes something highly unusual
- Interviewing any patients who have lodged complaint - Homogenous
- Focus on singular phenomenon, simplifies research & describes in depth
- Conducting focus groups with support people who attended appointments with men making decisions about chemotherapy - Maximum variation
- Shows range of behavior & identify commonalities across cases
- Asking as many people as possible about their experience with doctors - Opportunistic
- Captures ‘the moment’ when exemplary data can be collected
- In-depth interviews with people who have made a complaint - Politically important cases
- Highlights examples which can be linked to policy change
- Hosting focus group with patients who received access to a new drug - Random purposive
- Adds external validity to qualitative process
- Using randomization to determine who you will contact
5 most common sampling methods
1- Snowball - current participants knowledgeable about/invested in research to find others
- Asking participants ‘who else’s opinion is important for us to understand’
2. Stratified purposive - Facilitates data comparison on one or more variables of interest
- Asking participants to recommend other participants of the same sex or diagnosis
3. Theory-based - Illuminates theoretical concepts with intent to elaborate or expand
- Observing whether matching patients preferred style of decision-making with decisions increases adherence to treatment
4. Typical case - Elaborates on average experience
- Observing how patients interact with their doctors
5. Mixed
Theoretical method
o Used in grounded theory studies
o Research starts from homogenous (small) and moves to a more heterogeneous (larger) sample
o Occurs sequentially and alongside data analysis
o Previously analysed data guides what data needs to be collected next
Methods: Data Analysis
• Data usually presented in written form (observations, transcribed audio-taped interviews)
• Data is transcribed, coded, categorised
o Code and categorise
o Present themes/patterns from data
Data saturation = How do you know when you have enough
– Phase of analysis in which the researcher has continued sampling and analysing data until no new data/themes appear
– When you have learnt all you can about this phenomenon
– You are getting the same answers from informants
– Interview until 3 consecutive interviews with no new concepts (after no new info, you have reached data saturation)
– Member checking (ask the participants if what you found is what they think they understand to be the case)
Rigour in qualitative research
/Problems with qualitative research
• Ensuring rigour
• Use appropriate methods to maximise validity and reliability
• Results may be influenced by personal bias & idiosyncrasies
• Generalising beyond the sample (you should not say this data could reflect the broader community)
• Aims to capture the range of views, not representative views
• Time consuming
• Smaller samples based on theoretical saturation are achievable
Small selective samples - not generalisable
How to do Rigorous data collection
• Interview / focus groups:
o Audio or video-recorded
o Transcribed in full
o Participant confidentiality assured (removing names and places from the interview)
• Interviewer:
o Suitably trained in qualitative methods
o Accepted by participants as trustworthy
o Familiar with the context that people are coming
Managing personal bias & idiosyncrasies
- Investigators own personal bias and idiosyncrasies can influence the qualitative research process» need to be reflexive
- Reflexivity - awareness of the way own beliefs/attitudes affect the research process and outcomes
- One technique— use of a journal throughout research process which reflects researcher’s thoughts, feelings, questions, frustrations.
- Researcher may become aware of biases/assumptions»_space; may alter the way they collect data or approach analysis (keep personal journal so you can see your potentially biased thoughts written down)
Rigorous coding
- Development and application of coding frame (with themes/concepts) can improve rigour
- Immersion in the data (e.g. read transcripts multiple times)
- Multiple researchers should code some of the same transcripts and compare results to improve the coding frame
- Apply final coding frame systematically to all the data by annotating the transcripts with codes
Rigorous analysis
- Agreement between the researchers on the categorisation of data can be calculated
- Map the relationships between themes to explain findings
- Make sure the themes/concepts are supported by the data rather than based on your assumptions
- There are qualitative data programs that can help!
When to use software? Software is hard to learn how to use!
- When the size of the study team is very high 4+ required
- When the size of the data set is over 100 files
- When analysis goal is more theory building rather than exploratory
- When the study is highly complex
e. g. NVivo, MaxQDA, Dedoose, Atlasti
Rigorous validity checking
• Triangulation
o Compare same issue from different sources, methods, points (e.g. researcher notes, interview transcripts, audio-tapes)
• Provide evidence for themes/concepts
o Use participants’ own language to demonstrate the source of the theme or concept (e.g. illustrative quotes)
• Participant feedback/validation/verification
o Member checking
o Ask participants to review your interpretation of their data
Qualitative approaches used in health research
Grounded theory
• Strong tradition grounded theory in health research
Aims
• to collect and analyse qualitative data to
• describe components of a phenomenon
• the relationships between them
• generate a theory of the phenomena that is ‘grounded’ in the data (from the ground up)
Methods
• Complex iterative process, no distinct end point
• Coding: identifying/describing codes, relating codes to one another
• Memoing: recording thoughts/ideas as they evolve
• Diagrams: make sense of the data with respect to the emerging theory
Interpretive Phenomenological Analysis (IPA)
• Aims
• To gain a detailed examination of the participant’s lifeworld, personal experience/perceptions of an event
• To gain understanding of the lived experience through its meaning to individuals
Methods
• Data collection: open questions, look for meaning of the experience for that person, check interpretation with the person
• Data analysis: Goal is to describe phenomena and improve understanding; Might not have obvious ‘conclusions (may tell you how someone will interact with an intervention you’re developing)
Ethnography
Aims
• Seeks to understand human behaviour and individuals’ experiences within a group culture
• What is the experience of a breast cancer support group participant?
Methods
• Most common approach is participant observation (e.g. field research)
• Ethnographer becomes immersed in the culture and records extensive field notes
Thematic analysis
Useful approach for doing applied qualitative research in health
• Method for identifying, analysing, and reporting patterns (themes) within data.
• Highly flexible- not bound to a particular theoretical approach
• Progresses from description of data»_space;> interpretation of the significance of patterns and their broader meanings and implications
• Authors provide step-by-step guide to thematic analysis
Phases of Thematic Analysis
- Familiarising yourself with data
- Generating initial codes
- Searching for themes
- Reviewing themes
- Defining and naming themes
- Producing the report
Framework method • Ritchie and Spencer (2002)
- Analytic approach which sits within the broad family of thematic analysis
- Tool for supporting thematic analysis because it provides a systematic model for managing and mapping data
- 7 key steps (Gale et al., 2013)
- Transcription
- Familiarisation with the interview
- Coding
- developing a working analytical framework
- applying an analytical framework
- charting the data into the framework matrix
- interpreting the data
Specific strategies to ensure methodological rigour- 6
- Immersion: Steps 1 and 2 focus on researcher immersion with the data
- Team-based coding: Advocates for several researchers to engage in coding and for all team members to read summaries to offer perspectives during analysis
- Close attention to participant experiences: Through charting researchers pay close attention to describing the data using participant’s own expressions, before moving onto interpretation
- Systematic, thorough analysis:
- Inclusion of contextual data: Flexible enough that non-interview data (e.g. field notes) can be included in matrix
- Audit-trail: Framework method ensures there is a clear audit trail from original raw data to final themes, including illustrative quotes.
Framework method- Pitfalls Potential pitfalls of this approach
- Temptation to quantify: Systematic approach, matrix format, spreadsheet-look»> can be appealing to those trained quantitatively
- Resource intensive: Framework method is time consuming and resource-intensive
- High training needs: High training component to successfully use Framework Method
