Using Focus Groups in research: more than just a fashion statement?

1
Using Focus Groups in research more than just a
fashion statement?

• A/Prof Paul Ward
• Discipline of Public Health
• paul.ward_at_flinders.edu.au

2
Plan of Session

• Introductions expectations, needs/wants
• What are focus groups and why/when might you use
  them?
• Epistemological and methodological issues
• Critique - advantages and disadvantages
• Research Ethics
• Practicalities
• Thinking about analysis

3
Introductions

• What are your knowledges and experiences of focus
  groups?
• Why are you here? (and why now?)
• What do you want out of this session?

4
What are focus groups and why/when might we use
them?
5
Remember

• As with all research methods, let your research
  question determine your methodology and methods
  (NOT the other way around!!!)
• Just because they may be seen to be trendy (Tony
  Blair seemed to lead the UK on the basis of them)
  dont be fooled into just using them
  uncritically

6
What are Focus Groups?

• Group discussions/ group interviews
• Facilitated by researcher (often helped by
  another person)
• Getting people to think about, discuss/debate an
  issue (or set of issues) related to your
  research question
• More than just a bunch of individuals responding
  to your questions
• Allows for interaction (new data, conflicting
  data)
• Allows participants to interpret each others
  responses
• Not just a series of separate conversations with
  you!!
• The sum should be more than its individual
  parts!!!!

7
Why/when might you use them?

• MAIN REASON when you want to generate
  discussion/debate about your research question/
  area of research
• You do not think that individual responses will
  yield the kinds of data you want
• When you think that individuals may not be able
  to provide adequate responses on your research
  question
• Havent thought about the issues
• The issue requires discussions
• Might be normal/ everyday/ taken for granted
  issues
• When people may not feel safe talking in
  individual interviews
• Young people?
• Groups that may work on basis of collective
  discourse

8
Why/when might you use them?

• Other reasons are the same as all qualitative
  methods
• Lack of research in area
• Wanting to develop theory
• Wanting to develop a quantitative stage of study
• Etc, etc
• REMEMBER this is not a group counselling
  session there is always a possibility of a
  research relationship to stray into a
  therapeutic relationship but even more so in
  focus groups

9
Situating focus groups within an epistemological
and methodological framework
10
Rationale for talking about epistemology and
methodology here

• Highlight the philosophical bases of ALL research
• Explore the point or rationale of focus groups
• Situate them within epistemological and
  methodological framework
• There is a tendency for cookbook methods
• Add 1 RCT, wait for 18 months, then sprinkle with
  some focus groups.
• Add 4 focus groups and 15 interviews, wait for 1
  year.
• However this divorces methods from their
  philosophical bases
• Therefore quick re-cap on epistemology,
  methodology and method

11
Epistemology

• Concern the questions of what is (or what should
  be) acceptable forms of knowledge
• Central to this is whether the social world (i.e.
  that involving people and structures) can or
  should be studied according to the same
  principles and procedures as the natural sciences
• Positivism
• objectivity, generalisation, the development of
  general laws/ truths and testing of theories
  (deductivism)
• Interpretivism -
• social reality can only be understood through
  social constructions such as language,
  consciousness and shared meanings.
• Does not predefine variables, but explores human
  sense-making in naturalistic settings.

12
In essence..

• Positivism is about explanation (need for
  statistical generalisation etc)
• Interpretivism is about understanding (need for
  depth and context etc)

13
Methodology

• Method and methodology are often used as though
  they were synonyms - they arent
• Methodology is not just a posh word for method.
• Methodology is the study of methods and refers to
  the strategy or approach to research.
• Very much related to the epistemological position

14
Method

• Method is a specific technique (or set of
  techniques) for data collection
• Informed by methodology (which is shaped by
  epistemology)
• Within quantitative methodology questionnaire
  survey, experiment etc
• Within qualitative methodology focus groups,
  individual interviews, observation, documentary
  analysis

15
Links
Epistemology
Positivism
Interpretivism
Methodology
Quantitative
Qualitative
Experiment
Method
Interview
Focus group
Technique
RCT
16
Types of research questions

• Describing or answering questions about a
  particular, localised occurrence or context
• Understanding the perspectives of particular
  groups towards events, beliefs or practices
  (dentists, GPs, nurses, patients, homeless etc)
• Exploring complex research area where little is
  known (theory generation)
• .. Answering the WHY question..
• Why do young girls continue to smoke?
• Why do young men engage in many high-risk
  activities?
• Why are some health care services inequitable
  provided?
• providing CONTEXT and UNDERSTANDING

17
Some benefits of qualitative research

• The potential to illuminate everyday life to
  better understand the familiar and strange
• Can provide specific, concrete details to guide
  an understanding of a particular setting
• Can provide interpretation of local meanings that
  activities and practices have for a group engaged
  in them
• To illuminate differences across settings how
  different GPs interpret guidelines/ diagnose flu/
  prescribe antibiotics etc

18
Characteristics of qualitative research (1)

• Natural context occur in natural settings
• Places where people interact (GP/dental surgery,
  classroom, street corner etc)
• Study of inanimate objects (how health
  care/policy is developed or organised)
• Non-manipulative study situations/objects
  intact
• Researcher observes, interviews, records,
  describes settings as they are
• Researcher as instrument researchers engages
  in a situation and attempts to make sense of it
• Data collected through human observation
• Data interpretation through human perceptions
• Subjectivity of researcher insights,
  experiences, perceptions of researcher are
  important part of the study

19
Characteristics of qualitative research (2)

• Interpretive character researcher attempts to
  explain why and how something is happening -
  focus is on meaning rather than specific
  behaviours
• Focus on process, rather than outcome how and
  why things happen
• Depth (rather than breadth) of understanding
  hence the smaller number of cases
• Inductive analysis research begins with
  open-ended questions rather than attempting to
  test a priori hypotheses.
• Context sensitivity findings are placed in
  social, historical and spatial context limiting
  generalisations

20
Characteristics of qualitative research (3)

• Empathic neutrality complete objectivity is not
  possible but pure subjectivity undermines
  credibility
• Attempt to understand (not prove) something
• Goal is not to advocate or to advance personal
  agendas
• Personal experience is included as part of
  relevant data, although non-judgemental stance
  towards whatever content may emerge from data
• Reflexivity attempting to include the role of
  the researcher in the whole process (question
  construction, design, analysis etc)
• Flexibility of design research is open to
  adaptation as understanding deepens or situations
  change
• Focus on emic perspectives write from the
  perspectives of the participants (emic) rather
  from the researchers own perspective (etic).
• Perspectives of the participants in the study
• What are they thinking, why are they thinking it,
  what are their assumptions, motives, goals values
  etc.

21
Some limitations

• Subjectivity is inherent
• In ALL research (not just qualitative)
• Reflexivity is championed in qualitative
• Labour/ time intensive
• Underestimation of the vast amount of time it
  takes to undertake the data collection, arrange
  the interviews/ focus groups, transcribe the
  data, undertake data analysis etc
• Misunderstanding of novice researchers
• Many researchers think it is a soft option, but
  have little understanding of the complexities
  involved
• Quality and trustworthiness of studies are then
  compromised
• Limited generalisability moderatum
  generalisability
• However, this is not the point of qualitative
  studies, therefore seems an inappropriate
  standard by which to judge

22
Critique of focus groups
23
Some Advantages

• Has many of advantages of many qualitative
  methods, but in addition.
• Group dynamics more than the sum of its parts
• Generating new data that wouldnt have been
  gained through individual interviews
• Participants bouncing off each other
• In Literature - Time efficient
• Not a good methodological justification
• Not wholly convincing either (set up time,
  arranging lots of people for same time,
  developing appropriate themes, transcribing,
  analysis etc etc)

24
Some Limitations

• Maybe less detailed or in-depth than some
  interviews
• Potential for group speak
• Some people going with consensus rather than
  offering an alternative viewpoint
• Potential for power dynamics
• Different levels of involvement
• In Literature, issue of Public vs Private views
• May only get public views in focus group
• Holds onto positivistic notion of a truth
  which is private
• IF there is a private view, hidden from public
  view, can researchers ever get to it???????

25
Research ethics
26
Research ethics

• ALL research needs to consider and respond to the
  following ethical principles
• Research Merit and Integrity
• Respect
• Beneficence
• Justice
• However, when thinking about using focus groups,
  we have some additional ethical issues to consider

27
Research ethics

• Data control if 1 person wants their data to be
  removed, how does one do this within the context
  of a discussion - and make it amenable to
  analysis?
• Confidentiality and anonymity a variety of
  things may be talked about within the group, how
  do you make sure none of that goes outside the
  group?
• Power there may be distinct power dynamics
  how do you make sure each person has an equitable
  voice?

28
Practicalities of focus groups
29
Practicalities

• Setting it up
• Deciding on composition and size
• Running focus group
• Transcribing
• Potential pitfalls

30
Setting it up

• Adequate physical facilities
• Big enough room
• friendly atmosphere
• Child-care facilities?
• Disabled access?
• Car parking or public transport routes?
• Relevant recording equipment
• Good quality audio recorder to pick up a
  variety of voices at different distances from
  microphones
• Make sure you have pre-tested it and know how it
  works
• Refreshments
• Incentives?
• Okay consistent and in-line with what youre
  asking them to do

31
Deciding on composition and size

• Heterogenous vs homogenous?
• Heterogeneous
• diversity of perspectives on central research
  question
• Homogeneous
• May be good for sensitive issues Kitzingers
  work on HIV
• shared experiences good for bouncing off each
  other and a sense of safety
• May be good for marginalised groups for whom
  individuals may feel powerless within a
  heterogeneous group
• Whats a good sized group?
• Ideal size is 6 to 10 participants
• Not enough discussion vs alienation of some
  people or too rowdy
• Over-recruit!!! invariably get last-minute
  cancellations

32
Running focus group

• Role of facilitator (YOU) - success depends
  heavily on skills of the facilitator
• keeping discussion going
• keeping discussion within the research framework
• making sure people have a say fairness and
  respect
• trying to involve quiet participants
• trying to make ethnographic notes on the
  non-verbals (HARD so maybe take an observer
  along too)

33
Transcribing

• Takes MUCH longer than transcribing individual
  interviews
• Help the transcriber ask people to go around
  room and either say their name or pick a
  pseudonym
• Level of voice is really important ask people
  to speak up
• Also civility rules ask people (wherever
  possible) not to talk over one another if this
  happens its not transcribe-able so ask to
  repeat maybe?

34
Some Potential Pitfalls

• Dominance by one person
• Going off topic
• Writing up sometimes its hard to see that
  there was more than 1 person in the room at the
  time!!!! analysis and writing is on the basis
  of group discourse, not individual discourse

35
Introduction to Qualitative Analysis

• .. Not a workshop on analysis. But you need to
  think about it BEFORE collecting data.

36
Transcript provided

• Example of part of focus groups transcript
• THIS is what youll have to analyse
• Have a quick look through, then well talk about
  methods of analysis.

37
Background to qualitative analysis

• Inductive theory development
• Looking for patterns in the data emergent
  themes
• No pre-defined variables to focus on
  variables/themes/issues are defined as a result
  of the analysis
• Data collection and analysis occur simultaneously
• No single agreed approach to analysis depends
  on theoretical orientation, type of data,
  research question, research setting etc
• Involves reading, re-reading, and re-reading
  piles and piles of notes, transcripts, initial
  analyses until it makes sense

38
Useful quote to keep in mind..

• Plummer (quoted in Chapple Rogers (1998,
  p.559)) suggests that
• analysis is the truly creative part of the
  work it entails brooding and reflecting upon
  mounds of data for long periods of time until it
  makes sense and feels right, and key ideas
  and themes flow from it. It is also the hardest
  part to describe.

Chapple A Rogers A. Explicit guidelines for
qualitative research a step in the right
direction, a defence of the soft option, or a
form of sociological imperialism? Family
Practice 1998 15 556-561.
39
Stages involved in ALL analytical approaches

• Organising the data
• Developing categories, themes, and patterns from
  the data
• Testing emergent theories/hypotheses against the
  data
• Searching for alternative explanations in the
  data
• Writing the report

40
Some issues to bear in mind

• The process is not linear
• Iterative stages build upon each other, and
  inform each other
• Cyclical interpretation may lead to
  re-classification
• Process does not have the same clear structure
• Sometimes unclear which stage you are at
• May go from description to interpretation
• Unpredictability
• The process of analysis
• When is analysis finished?
• How long will it take?

41
(1) Organising the data

• What are data?
• Transcripts of focus groups
• Field notes
• Memos
• Reflections or comments by researcher (and
  recorder)
• Develop a data management system
• All data are organised, dated and sequenced
• Manual or computer? - for large datasets,
  computers make life easier
• WARNING although they are called Qualitative
  analysis software packages they are glorified
  filing cabinets
• YOU do the analysis, the software package helps
  you to organise the data, to sort the data, and
  to find the data at a later stage.

42
(2) Describing the data

• Provide a picture of the setting, people and
  events
• Setting where and when did research take place
• People describe who participated
• Events what happened throughout the research
  process
• Provides a context, within which to situate the
  analysis of verbal/textual data
• Called a thick or rich description
• Also consider issue of reflexivity
• Why that setting and those people?
• Describe inter-personal relationships?
• Describe what it felt like in each encounter

43
(3) Searching and coding the data

• SEARCHING
• Involves reading and re-reading the data to get
  a sense of it
• Reading ALL data (memos, transcripts,
  reflections)
• What seem to be the important issues emerging
  from the data?
• As you read make notes of the sections, themes,
  issues that initially seem important
• Coloured pens
• Notes in margins
• Underlining
• Make notes of particular themes that seem to
  re-occur
• Keep searching until you feel that you have
  documented the main issues emerging
• And have highlighted where these occur in the data

44
(3) Searching and coding the data

• CODING
• Breaking data into smaller units to make sense
  of the data
• Naming and labelling of phenomena through close
  examination of the data.
• Strauss Corbin (2004, p.303) state that
• During open coding the data are broken down into
  discrete parts, closely examined, compared for
  similarities and differences, and questions are
  asked about the phenomena as reflected in the
  data. Through this process, ones own and
  others assumptions about phenomena are
  questioned or explored, leading to new
  discoveries.
• The transcripts are read, and re-read, and each
  discrete incident, idea, event, concept etc are
  labelled (coded).
• Similar incidents or ideas are given the same
  label (code), thereby allowing comparison both
  within and between transcripts

45
(4) Categorising data

• Process of coding may have developed hundreds of
  codes
• Although these will be discrete, there will be
  similarities between some codes
• Also, need to have a smaller number of categories
  to make the analysis manageable, and the findings
  readable
• Therefore collapse codes into meaningful
  categories
• Category is a group or classification of
  individual codes/ labels/ themes
• One method of coding and categorising data is
  Constant Comparison

46
Constant comparison

• Comparing newly identified topics/themes to
  determine if they represent NEW categories or
  should be placed in existing categories
• is this theme similar to or different from
  existing categories?
• If similar these are added to the existing
  categories which are changed accordingly
• If not similar new categories are created
• Continues until data saturation
• No new categories emerging
• All categories have been exhausted
• How do we know when weve reached
  saturation??????
• Data saturation
• When no new themes/concepts emerge from the data
• When weve looked for negative cases (data that
  contradicts the emerging category) and discrepant
  cases (data that adds a different perspective to
  the category)
• Negative/discrepant cases may be sought within
  the other transcripts, although also may involve
  recruiting new research participants

47
Constant comparison

• Some steps in constant comparison
• Collect data from several cases
• Identify important issues or themes
• Identify recurrent themes and use them to create
  categories
• Collect additional data to provide more examples
  for each category and to elaborate on dimensions
  of each category
• Describe how categories account for documented
  events
• Reformulate some categories and delete others as
  dictated by data
• Identify patterns and relationships between
  categories
• Develop a theory by continuing to collect and
  compare data and refining categories and
  relationships

48
(5) Synthesising the data

• Involves looking for patterns and links between
  categories
• Need to search for, and determine the links
  between categories
• How do they all fit together?
• A process of re-assembling the data
• bring back the complexity of the data
• links between categories
• context that is lost in coding and categorising
  (sometimes called fracturing)
• Essential for integrating and understanding the
  holistic nature of the data
• Really useful for reporting the data
• Provides a framework for reporting
• Can report each category, and also how they all
  fit together

49
Web of qualitative data analysis
Organise
Describe
Synthesise
Search
Categorise
Code
50
Summary

• Focus groups can be very good for generating
  discussion/debate about your research questions
• Main Adv more than the sum of its parts the
  added value that comes from debate/discussion
• Potential pitfalls power, voice and group speak
• Need skillful moderator to minimise these
• If thinking about using focus groups make an
  informed choice based on your RESEARCH QUESTION,
  rather than because they seem to be fashionable
  in research!!!