lecture 2 - qualitative data collection methods and approaches Flashcards
What is qualitative data?
Qualitative data is in the form of words rather than numbers.
Patient description of health
“You’re just constantly having to get through obstacles to get to the right doctor”
Self-reported ethnicity - nominal
1=Indian
2=Gypsy or Irish Traveller 3=Arab … etc…
Ranking of London Marathon runners - ordinal
First, second, third
Likert rating scale - interval
1 = not at all
2 = Slightly
3 = Moderately 4 = Very
5 = Extremely
How do you generate qualitative (word) data? - narrative/ spoken word
̈ Most common data collection methods are: ¤ individual interviews
¤ group discussions (focus groups)
¤ non-interactive methods (textual data from qualitative survey, diaries, media content, narrated stories using images)
Individual interviews
̈ Like everyday conversations between researcher and interviewee
¤ Focused on researchers need for data
¤ Conducted to ensure trustworthiness of data (reliability & validity) nReproducible: Someone else with same topic guide can generate similar
findings
nSystematic: No cherry-picking participants or observations from the interview nCredible: Reasonable design (questions and methods) to yield truthful accounts
nTransparent: Method used reported and accessible to others, so all know how data was generated
collected to ensure trustworthiness of data
Interview
̈ Illustrative example of interview ¤ Experiences of driving and
learning how to drive
what are key features of interviews?
knowledgeable - Familiar with topic, interview topic guide, has piloted and practiced
structure - Starts off with explanation of study and rounds off at the end
clear - Simple, easy, short questions without jargon, open-ended
gentle - Lets interviewee finish, tolerates pauses, gives people time to think and to finish, is slow to interrupt
sensitive - Listens attentively, empathises with what participant is presenting
open - Responds to what is important to interviewee, flexible questioning to respond to issues raised by interviewee
steering - Uses prompts, probes and sub-questions to get information needed for their research
critical - Prepared to challenge sensitively inconsistencies and ambiguities
remembers - Remembers what has been previous said, relates back currently said to past information
interprets - Can clarify and extend meaning of what interviewee has said (e.g., summarises )
balanced - Doesn’t talk too much or too little
ethically sensitive - Makes interviewee aware of ethical context (e.g., purpose of study, confidentiality of responses)
- setting for interviews
needs to be formal but not overly formal so interviewee feels comfortable. opposite seating
- Building and maintaining rapport
̈ Interviewees need to trust interviewer
¤ Interviewer interested in them and what they have to say ¤ Interviewer will not judge interviewee for their answers
̈ Interviewer should set the interview context
¤ Introduce yourself and give your name, aims of interview, reminder that interview can stop at any time and possibility of asking any questions; be sensitive to needs of interviewees (e.g., hard of hearing, frail, embarrassed, literacy level for topic)
̈ Keep in mind effects of interviewer personal characteristics (ethnicity, status, gender, social distance, clothing)
illustrative example - visual metaphor - using something visible to show something invisible
top tips for interviewing
People interviewed often in
vulnerable state
* Do not create expectations you
cannot fulfil
* Tell the person what you plan to do
with the information and then do it
more info in notes
- Types of interviews & interview questions
̈ Semi-structured [open-ended] interviews
¤ topic guide comprises open-ended questions for a defined area of
interest (e.g., quality of life and arthritis)
̈ In-depth qualitative or unstructured interviews
¤ Very broad open-ended questions
¤‘Tell me about your experiences of arthritis’
¤ Goal is to elicit information as is relevant to the interviewee
nquality of life may or may not be a topic interviewee brings up ̈ Topic Guides can enhance reproducibility
guide in notes
Topic guides to guide interview
̈ Used mostly in semi-structured interviews
¤ Key questions interviewer would like to cover
¤ Useful prompts to elicit discussion of topics if specific issues do not come up
̈ How to formulate questions?
¤ Must be as non-judgmental as possible
¤ Avoid leading informant to particular answers
̈ Use audio-recorders (for full transcripts)
how to formulate questions
1 - start with general qus
2 - type of language, emotion used
3 - use everyday vocab
4 - more sensitive qus towards the end
5 - ask open qus
6 - ask neutral qus
7 - use concrete rather than abstract qus
8 - use concrete events to help people remember
Individual interviews – doing better
̈ Most common data collection methods are: ¤ individual interviews
¤ group discussions
¤ non-interactive methods
̈ Illustrative example Experiences of driving and
learning to drive
Group discussions (focus groups)
̈ May be more useful than individual interviews for some topics
* ¤ Ease of disclosure in groups
- Expressing dissatisfaction with healthcare
- Sensitive topics that all group members share
* ¤ Group processes can help people to explore and clarify their views in ways
that would be less easily accessible in a one-to-one interview
* ¤ Groups give access to how people talk to each other about topic
- Informative about social structure (context, fabric) of the community and how opinions
¤ Informs how knowledge is formed in social contexts (behaviour in action) nMay not be appropriate for marginalised or vulnerable group members
characteristics of group interviews
Discussion with a group of people (formal, informal)
¨ Formal (focus group, natural group)
¤ Natural groups à interested in group norms;
¤ Focus groups à range of views
Practical aspects of focus groups* to stimulate discussion of target topic
̈ Participant recruitment
¤ Best is to have people willing to discuss and this is determined by:
- Rapport (as discussed)
- Awareness, understanding, interest and perceived gain from discussing research
topic X
- Compatibility in social structure (e.g., power, influence, etc)
¤ 6 to 10 people for sustainable discussion; over recruit 25% for attrition ¤ Ask for volunteers, or help of gate-keepers (tribal leaders, directors)
¤ Natural groups depend on access to them
Running a group (in brief)
From: Brikci, N., & Green, J. (2007). A guide to using qualitative research methodology. Field Research Médecins sans Frontières.
̈ Capable moderator(s) [usually 2 for leading discussion & operations]
¤ provide a clear explanation of the purpose of the group (e.g., duration, topic) ¤ help people feel at ease
¤ facilitate interaction between group members using topic guide
̈ Number of groups will depend on expected saturation
¤ Saturation = when additional data collection no longer generates new understanding
̈ Select appropriate setting
̈ Use tape/Zoom/Teams recorders (for full transcripts)/ dedicated note taker
̈ Use appropriate topic guide
̈ Start group with icebreaker
̈ Discussion starter to set initial positions (“How is treatment planned at your clinic?”)
̈ Final summary statement from each member to signal the approaching end of the group
Challenges in focus group interviews
̈ Balance between homogeneity (similar) and heterogeneity (dissimilar)
¤ Viewpoints too extreme (anti-vaxer Vs virologist)
¤ Social divisions too great (Company CEO vs shop floor assistant) ¤ Dominance of one voice (Loudest, most informed or confident)
̈ Participant interested in topic but not personally invested ¤ No motivation to engage fully
¤ Group context makes it easier to not engage
Collecting qualitative data – Doing better
̈ Commonalities between individual and focus group interviews
̈ Important to keep focused on eliciting data needed for the research topic
̈ Building rapport and engaging participant collaboration in providing the needed information
Experiences of driving and learning to drive
¤ How can it be done better?
non - interactive qualitative data collection
surveys - open text boxes in online surveys
observations - context observations, media content, diaries, medical records
quantitative and qualitative survey data
diagram in notes
users perception of website quality
questionnaire items have quantitative analysis
open comments have qualitative coding
these can both go through triangulation for internal validity
this is how new concepts are identified
thematic analysis of qualitative data
analyse the words
looks at ptps, quotations, codes, themselves and broad themes
observations and field notes
- record notes as soon as possible
- begin the record of each field on a new page
- use wide margins to add notes at any time
- record events in order of which they occur and note how long they last
- make notes as concrete, complete and comprehensive as possible
- record small talk or routines that dont appear too be significant at the time - may be important later
- let your feelings flow and write quickly without worrying about spelling - assume no one else will use the notes
- include diagrams or maps of the setting and outline your own movements and those of other during the period of observation
- record emotional feelings and private thoughts in a separate section
- avoid Evaluative summarising words
- reread notes periodically and record ideas generated by the rereading
Other sources of contextual data
̈ Reports of previous research (e.g., from literature search) ̈ Country reports about phenomena
̈ Clinic and service records
̈ Policy reports (e.g., national, international, NGOs, WHO)
̈ Oral data
¤ Informal conversations
Conclusions and re-cap
- ̈ Different strategies to collect qualitative data, which is data expressed in the form of words rather than numbers
- ̈ Data collection approach must ensure “trustworthiness” of data
- ̈ Individual and group interviews (focus groups) have pros and
cons; which is best depends on research goals - ̈ Non-interactive methods are also possible, especially now with online open surveys
̈ Contextual data is also collected to help in interpretation process
