Medical Education Singapore Med J 2008; 49(3) : 256

CME Article
Focus group discussion: a tool for health
and medical research
Wong L P

ABSTRACT discussions help researchers tap into many different forms

Focus group discussion is a research methodology of communication that people use in day-to-day interaction,
in which a small group of participants gather to including jokes, anecdotes, teasing, and arguments. There
discuss a specified topic or an issue to generate is a tendency that during group discussions, attitudes
data. The main characteristic of a focus group is and perceptions are developed through interaction with
the interaction between the moderator and the others in the groups.(2) In this sense, focus groups showed
group, as well as the interaction between group dimensions of understanding that often remain untapped
members. The objective is to give the researcher or inaccessible by other forms of data collection.(3)
an understanding of the participants' perspective In the context of healthcare and medical research, focus
on the topic in discussion. Focus groups are rapidly groups are particularly apt due to the fact that most health -
gaining popularity in health and medical research. related conditions are created by social environments and
This paper presents a general introduction of the made within the social context.(4) Thus, focus groups are
use of focus groups as a research tool within the a popular method for assessing public experience and
context of health research, with the intention understanding of illness,(5'6) identifying ideas concerning
of promoting its use among researchers in health -risk behaviours and danger,(6'7) and discovering the
healthcare. A detailed methodology for the public's perception of causes of diseases.(8)Focus groups
conduct of focus groups and analysis of focus group are also exceptionally effective for study of sensitive issues
data are discussed. The potentials and limitations as well as issues that are difficult to access, such as acute
of this qualitative research technique are also mental distress, HIV/AIDS, or sexual health issues. They
highlighted. can also be used to gain insights into people's experiences
of ill health and health services,") and explore the
Keywords: focus group discussion, healthcare attitudes and needs of healthcare providers."
research, medical research, qualitative research In the literature, focus groups have been used to
technique examine a wide range of health- and medical -related
Singapore Med J 2008;49(3):2S6-26 I
issues, including, for example:
People's attitudes toward smoking and second-hand
INTRODUCTION smoke;('2
Focus group discussion is a form of qualitative research Identification of commonly -used local terminology
method in which the interviewer (also called the of symptoms or diseases;(13)
moderator) asks research participants specific questions Understanding of sexual abuse and associated
about a topic or an issue in a group discussion. Focus factors;("
groups, unlike individual interviews, provide the Health needs of gays and lesbians;(16)
added dimension of the interactions among members. Identification of psychosocial issues of patients.(16)
In conducting the focus group, the emphasis should be In relation to health services, focus groups have also been
placed on the interaction among group members. Instead used to explore issues, such as: (3)
of the moderator asking questions, the group members are Professional responses to changing management Health Research
encouraged to communicate with one another, exchanging arrangements; (17) DevelopmentUnit,
Faculty of Medicine,
ideas and comments on each other's experiences or points Developing ways to improve medical education and University of Malaya,
Kuala Lumpur 50603,
of view.' professional development." Malaysia

Wong LP, MSc, PhD

PURPOSE PROCEDURE Lecturer
Focus group discussions are frequently used to obtain Conducting a focus group requires a high level of Correspondence to:
knowledge, perspectives and attitudes of people about resources. The process involves formulating research Dr Wong Li Ping
Tel: (60) 3 7967 5748/
issues, and seek explanations for behaviours in a way questions, developing protocols, soliciting participants, 5738
that would be less easily accessible in responses to arranging venues, facilitating focus groups, transcribing, Fax: (60) 3 7967 5769
Email: wonglp@
direct questions, as in one-to-one interviews.(2'3) Group analysing data, and reporting the findings. ummc.edu.my
 Singapore Med J 2008; 49(3) : 257

Formulate research questions The group should not be so large as to preclude adequate
A clear and specific purpose statement is needed in order to participation by most members nor should it be so small
develop the right questions and elicit the best information that it fails to provide significant greater coverage than
from focus group discussions. For example, "To find out that of an individual interview.'21 Small groups are easily
how people define a healthy interpersonal relationship", dominated by one or two members, or discussion may fall
"To identify the healthcare needs of children living with silent if too few people contribute. On the other hand, a
HIV/AIDS", or "To assess the impact of sexual harassment large group lacks cohesion and may possibly break into
on individuals' lives at work and at home". Focus groups side conversations, or participants may become frustrated
are better used to explore specific or narrowly -focused if they have to wait for their turn to respond or get
topics; otherwise, the data obtained is likely to be diffused, involved. However, the number of participants depends
thus making data analysis a difficult task.(2,3) on the objective of the research; for example, smaller
groups (four to six participants) are preferred when the
Discussion guide participants have an intensive experience to share about
The purpose of the discussion guide in a focus group the topici2' or when the researcher wants participation
discussion is to provide a framework for the moderator from each subject.
to ask and probe questions. Using a discussion guide
increases the comprehensiveness of data collection and Number of focus group sessions
makes data collection more efficient. The guide merely The number of focus group sessions to be conducted will
provides the moderator with topics and issues that are to be mediated by factors such as the purpose and scale of the
be covered at some point during the group discussion. It is research, as well as the heterogeneity of the participants. (22)
not the equivalent of a survey instrument, and is not to be Often, a diverse range or participants is likely to necessitate
followed in detail or even necessarily in order. The probing a large number of sessions. Time, cost and availability of
questions in the discussion guide is to stimulate further participants may also limit the number of sessions that can
discussion, but not direct the discussion too much.(2) be held. Another guideline to the number of focus group
The guide should proceed logically from one topic to sessions is to use the concept of saturation,(20) to continue
another, and flow from the general to the specific. In a conducting focus group sessions until it seems to reach a
particular topic of discussion, the initial questions should saturation point, where there is repetition of themes and
be general, and as the discussion continues, the questions no new information is shared.
should become more specific and focused. Questions
should be open-ended, simple, unbiased and non- Group composition
threatening. Pre -testing the guide with several "mock" Group members in a focus group may be homogeneous
focus groups is essential. A well -designed focus group along some dimensions, and heterogeneous along others.
guide should allow the flexibility to pursue unanticipated The decision is also largely determined by the purpose
yet relevant issues that may be generated during the of the research. Some diversity in the composition of
discussion. the group may enhance discussion. However, a very
heterogeneous group can be threatening to participants
Recruiting participants and can inhibit disclosure. This is particularly evident
Generally, participants are recruited on the basis of their in the discussions about sensitive issues, such as factors
experience or involvement related to the research topic, associated with child sexual abuse, the sharing of
and whose opinions the researchers are interested in experience, and the sense of "everyone in the same boat"
hearing; for example, people with a particular disease, is particularly important to facilitate disclosure.' 23'
caretaker who is familiar with the patient, healthcare On the other hand, homogeneity within the group may
providers or government officials. In identifying the help to capitalise on the participants' shared experiences,
person to represent the group, it has been suggested that as they are more likely to talk freely and share experiences
purposive sampling can be employed.i19i Participants if they feel they have a lot in common. It is particularly
can be recruited from hospitals, community centres, via essential that some issues are better discussed by people of
advertising in the local newspaper or by writing letters to similar experiences or in the same situations, where their
local organisations. Snowballing recruitment technique is disclosures are encouraged in a nurturing environment.
also favourable, where initial contacts are asked to suggest For example, in a discussion on sexual behaviours,
people who would make interesting contributions to the younger and older women should participate in separate
discussion.'20' groups. Younger women may be reluctant to discuss sexual
behaviours in the presence of the elderly.
Size of the group The ideal is, therefore, a point of balance between
Most focus groups consist of between six and 12 people. the two extremes of heterogeneity and homogeneity.i24)
 Singapore Med J 2008; 49(3) : 258

When it is desirable to obtain data from different groups,

Participant I Participant 2
a general recommendation is to conduct a series of focus
groups using homogenous participants.'24' For instance, if
Participant 3
a study's aim is to explore public's view on HIV/AIDS,
one might conduct separate focus groups with healthy
people, people living with HIV and AIDS (PLHIV), and
family members of PLHIV.

Setting Participant 4
When setting up a focus group session, it is important to Notetaker
Participant 6
give careful consideration to the physical setting. The Participant 5
venue has to be comfortable and conducive to discussion.
Fig. I Sketch of a focus group seating plan.
In selecting a focus group site, it is important to make it
geographically convenient for the participants. Locations
that are hard to find may cause delays and scheduling valued and will remain confidential and anonymous, and
problems. Often the tape -recorder will pick up background why and who will have the access to the information.
noise; thus, the discussion room should be free from After the introduction, the moderator typically have
outside distractions. Participants should be arranged group members introduce themselves. This can help to
to sit around a table to enable them to see and hear one "break the ice" and build rapport among group members.
another. To preserve confidentiality, the moderator may ask
participants to introduce themselves by a pseudonym.
Role of moderator and note -taker The moderator must brief the participants to speak one
Focus group discussions require a small team. Ideally, two at a time to avoid garbling the tape-recording. It is also
people will be needed to conduct each focus group, one as essential for the moderator to assure participants that
the moderator and the other as the note -taker. Focus group all their contributions are valuable and important, and
moderators serve as discussion leaders. The moderator is to emphasise that there are no "correct" or "wrong"
responsible not only for guiding the participants through answers.
the discussion, but also for looking after the group In the process of discussion, the moderator should
dynamics to ensure all participants join in the discussion. use phrases such as, "Could you further elaborate what
When some participants dominate the discussion, the you have said?" "Can you tell me more?" or "Would you
moderator should address questions to individuals who are give me an example?" to obtain additional information.
reluctant to talk, in order to balance out participation.'25' To encourage in-depth exploration of a particular issue,
The note -taker will have to capture what was said the moderator may provide some ideas by probing (refer
and expressed, noting the tone of discussion, the order in to focus group guide prepared prior to the discussion).
which people spoke (by participant number or name), as The following is the example of a core question and the
well as phrases or statements made by each participant. probes:
It is extremely important for the note -taker to capture the Core question: "Does your hospital use men to be involved
information from the discussion as accurately as possible. in Pap smear screening?"
Non-verbal expressions, such as facial expressions or hand Probes:
movements, should also be noted. The note -taker should What are the strategies used to get men involved?
include a sketch of the seating arrangements, writing the What were the responses received?
name or the assigned number (Fig. 1). The notes will What are the barriers of men involvement in the
help greatly in transcribing the discussion to ensure the programme?
information is matched correctly.
It is important for the moderator to summarise
Conduct discussion each time before moving from one topic to another; for
The focus group should begin with a welcome. The example, "Before moving on to the next topic, let me see
moderator then introduces himself/herself and the note - if I have understood your point-of -view correctly, that is,
taker, and gives a brief overview of the topic of discussion that men's involvement greatly enhances cervical cancer
and objective of research, ground rules, and duration screening among women?"To curb a dominant participant,
of discussion (usually 45-90 minutes). The moderator the following phrase is helpful: "There are a few people
must also explain how the session is being recorded (in who wish to add to this point, we will come back to that
writing and/or tape-recorded), convey the expectation idea if we have time", and to encourage a quiet participant:
that everyone should contribute, that all contributions are "Do you have anything you would like to add to on this
 Singapore Med J 2008; 49(3) : 259

Belief about risk factors t'or cervical cancer [Qualis Research Associates, Colorado, USA]) have been
developed to make the task relatively easier. Nevertheless,
the researcher remains responsible for the interpretive
process of the analysis.
Above all, the coding merely allows the researcher to
.Se. rrulfr
trrrnNn,itted establish a connection of the data to facilitate data analysis.
irrfeclrou
The actual data analysis process can be classified into two
levels. The basic level of analysis is merely a descriptive
account of the data: explanation of what was said and
no assumption is made. The second level of analysis is
interpretative, which involves the comprehension of the
themes (or perspectives), creates links between the themes,
demonstrates how those themes emerged and generates
a theory grounded in the data.(28'29) Using a model to
illustrate the relationship and reciprocal influences of each
of the categories and themes is encouraged (Fig. 2).(30)
In reporting focus group findings, the results should
be presented in the context of the discussion within
the groups, from a series of sessions rather than from a
single focus group session. Researchers must consider
Fig. 2 Illustration of relationship and linkage between themes
and categories. the intensity of respondents' comments, as well as the
specificity of probe responses.(") Focus group results
issue?" are often expressed in impressionistic terms, and should
Moderators must be unbiased, respectful and able to be replete with statements, such as "many patients
listen. It is important that the moderator must not pass mentioned... ,", "several disagree... " and "almost none of
judgment, but should be a listener and should not be too the patients had ever... " Support findings by using direct
actively involved in the discussion except to guide it and to quotes to illustrate the different ways responses were
keep it focused. Biased responses, such as "That's a very expressed, and these should be reported anonymously.
good point! ", "Exactly!" or "Correct!" may introduce bias For example: When asked to define their understanding
to participants' responses. At the end of the discussion, about HIV, one participant noted, "I don' t know exactly
overall summarisation of important points should be made what HIV means, but it's something to do with soldiers
by the moderator to ensure correct interpretation as well as protecting the body. It kills the soldiers of the body." (13)
to allow the participants to elaborate their points further, if Although it has been suggested that numerical terms
any have been are left out. is inappropriate in reporting results of focus groups,(29s1)
it has been argued that some qualitative data can be dealt
Analysing the data and reporting findings with in a quantitative way. If a theme repeatedly appears
Basically, analysing focus group data is similar to in the data, it is alright to quantify how often it appears.
analysing other qualitative data. The actual words and Simple statistical frequencies can be used to describe
behaviours of the participants are the basis to the answer the important characteristic of the themes, although a
of a research question.(") The first step is to produce a generalisation is not possible. It should be noted that the
verbatim transcript of the entire discussion. If the focus sampling method and the number of members of a focus
group has been recorded in a language different from group(2) is usually not large enough to be a representative
the language in which the analysis takes place, the sample of a population. Thus, the data obtained is not
transcription must be translated. The complete transcript necessarily representative of the general population,
should then be compared with the handwritten notes taken unlike in a survey.
by the note -taker to fill in the gaps.
Once the transcribing is done, the next step is coding POTENTIALS AND LIMITATIONS
the data in the transcripts, which involves sorting the Focus group discussions have several advantages. It is
data and assigning them to categories.(27) Coding can be an excellent method for collecting qualitative data where
done manually, by "cutting and pasting" and using of participants are able to build upon one another's comments,
coloured pens to categorise data. More recently, a number stimulate thinking and discussion, thus generate ideas
of computer software packages (NUDIST, NVivo, Atlas/ and breadth of discussion. (1,2) It can produce high quality
ti, [QRS Int, Melbourne, Australia], and Ethnograph data because the focus group moderator can respond to
 Singapore Med J 2008; 49(3) : 260

questions, probe for clarification and solicit more detailed cancer detection practices and beliefs in black women attending public
health clinics. Health Edu Res 1994; 9:331-42.
responses. (24) Focus groups may aid in conceptualisation 8. Trilling JS. Selections from current literature: focus group technique in
and generate hypotheses, if the researcher is exploring chronic illness. Fam Pract 1999; 16:539-41.
a new area. Information told by group members can be 9. Naish J, Brown J, Denton B. Intercultural consultations: investigation
of factors that deter non-English speaking women from attending their
turned into hypothetical -type of questions used in surveys general practitioners for cervical screening. BMJ 1994; 309:1126-8.
(inductive method). (32) 10. Murray SA, Tapson J, Turnbull L, McCallum J, Little A. Listening to
local voices: adapting rapid appraisal to assess health and social needs
However, focus groups have some limitations. in general practice. BMJ 1994; 308:698-700.
A fundamental disadvantage of focus groups is its 11. Denning JD, Verschelden C. Using focus group in assessing training
susceptibility to bias, because group and individual needs: empowering child welfare workers. Child Welfare 1993;
72:569-79.
opinions can be swayed by dominant participants or by 12. Glantz SA, Jamieson P. Attitudes toward secondhand smoke, smoking,
the moderator.' 1'3' In addition, control over the group and quitting among young people. Pediatrics 2000; 106:E82.
13. Sifunda S, Reddy PS, Braithwaite RB, et al. Social construction and
discussion could be a problem and time can be lost on
cultural meanings of STI/HIV-related terminology among Nguni-
issues irrelevant to the topic if a discussion digresses from speaking inmates and warders in four South African correctional
the original topic. In such situations, the data could be facilities. Health Educ Res 2007; 22:805-14.
14. Meursing K, Vos T, Coutinho O, et al. Child sexual abuse in
"messy"; therefore, it is imperative that moderators need Matabeleland, Zimbabwe. Soc Sci Med 1995; 41:1693-704.
to have facilitator skills to overcome this potential setback. 15. Brotman S, Ryan B, Cormier R. The health and social service needs
of gay and lesbian elders and their families in Canada. Gerontologist
Groups are often difficult to assemble and response rate
2003; 43:192-202.
could be a problem. A telephone or mail reminder to the 16. Anderson RM, Barr PA, Edwards GJ, et al. Using focus groups to
participants of the time and place of the setting is helpful. identify psychosocial issues facing black individuals with diabetes.
Diabetes Educ 1996; 22:28-33.
It is advisable to over recruit by 20%, as some people may 17. Barbour R. Are focus groups an appropriate tool for studying
change their minds about participating or fail to turn up on organizational change? In: Barbour R, Kitzinger J, eds. Developing
Focus Group Research: Politics, Theory and Practice. London: Sage,
the day of the discussion.'24'
1998: 113-26.
18. Ivanoff SD, Sjöstrand J, Klepp KI, Lind LA, Lindqvist BL. Planning a
CONCLUSION health education programme for the elderly visually impaired person:
a focus group study. Disabil Rehabil 1996; 18:515-22.
This article has outlined the main features of focus 19. Stewart DW, Shamdasani PN. Focus Groups: Theory and Practice.
groups and examined methods for conducting a focus London: Sage, 1990.
20. Cameron J. Focusing on the focus group. In: Hay I, ed. Qualitative
group discussion. The interactive element makes focus
Research Methods in Human Geography. 2nd ed. Melbourne: Oxford
groups ideally suited to explore issues related to medical University Press, 2005: 156-82.
and health research. Nevertheless, if used appropriately 21. Merton RK, Fiske M, Kendall PL. The Focused Interview: A Manual of
Problems and Procedures. 2nd ed. London: Collier MacMillan, 1990.
and judiciously, focus groups may produce high quality
22. Morgan DL. Successful Focus Groups: Advancing the State of the Art.
data. Although they require careful and tedious planning, Newbury Park: Sage Publications, 1993.
followed by intricate analyses, focus groups remain 23. Finch H, Lewis J. Focus groups. In: Ritchie J, Lewis J, eds. Qualitative
Research Practice: A Guide for Social Research Students and
an invaluable research tool in health and medical Researchers. Thousand Oaks, CA: Sage, 2003: 172-98.
perspectives. 24. Morgan DL. Focus Groups as Qualitative Research. London: Sage,
1998.
25. Smithson J. Using and analyzing focus groups: limitations and
REFERENCES possibilities. Soc Res Meth 2000; 3:103-19.
1. Kitzinger J. The methodology of focus groups: the importance of 26. Mays N, Pope C. Rigour and qualitative research. BMJ 1995;
interactions between research participants. Sociol Health Illness 1994; 311:109-12.
16:103-21. 27. Dey I. Qualitative Data Analysis: A User-friendly Guide for Social
2. Kreuger RA. Focus Groups: A Practical Guide for Applied Research. Scientists. London: Routledge, 1993.
1st ed. London: Sage, 1988. 28. Basit TN. Manual or electronic? The role of coding in qualitative data
3. Kitzinger J. Qualitative research: introducing focus groups. BMJ 1995; analysis. Edu Res 2003; 45: 143-54.
311:299-302. 29. Hancock B. Trent focus for research and development in primary
4. Carter S, Henderson L. Approaches to qualitative data collection in health care: an introduction to qualitative research [online]. Available
social science. In: Bowling A, Ebrahim S, eds. Handbook of Health www.trentrdsu.org.uk/cros/uploads/Ilealth%20Needs%20Assessm
at:
Research Methods: Investigation, Measurement and Analysis. ent.pdf. Accessed May 8, 2007.
Buckingham: Open University Press, 2005: 219-29. 30. Wong LP, Wong YL, Low WY, Khoo EM, Shuib R. Knowledge and
5. Kitzinger J. Understanding AIDS: Researching audience perceptions beliefs of risk factors of cervical cancer among Malaysia women. Int J
of acquired immune deficiency syndrome. In: Eldridge J, ed. Getting Behav Med. In press, 2008.
the Message: News, Truth And Power. London: Routledge, 1993: 31. Fern EF. Advanced Focus Group Research. Thousand Oaks, California:
271-305. Sage, 2001.
6. Ritchie JE, Herscovitch F, Norfor JB. Beliefs of blue collar workers 32. Steckler A, McLeroy KR, Goodman RM, Bird ST, McCormick
regarding coronary risk behaviours. Health Edu Res 1994; 9:95-103. L. Toward integrating qualitative and quantitative methods: an
7. Duke SS, Gordon-Sosby K, Reynolds KD, Gram IT. A study of breast introduction. Health Educ Q 1992; 19:1-8.
 Singapore Med J 2008; 49(3) : 261

SINGAPORE MEDICAL COUNCIL CATEGORY 3B CME PROGRAMME

Multiple Choice Questions (Code SMJ 20083A)

True False
Question 1. How have focus groups been used in health research?
(a) To develop a testable hypothesis.
(b) To assess the effectiveness of health promotion programmes.
(c) To test the efficacy of drugs.
(d) To study the needs of marginalised groups.

Question 2. State whether the following statements are true or false:

(a) The moderator's role is to generate as many different views as possible from the participants.
(b) The moderator need not have adequate knowledge about the topic.
(c) In an ideal focus group, all the participants are very comfortable with each other, but none
of them know each other.
(d) Heterogeneity is the key to maximising disclosure among focus group participants.

Question 3. When is it important for the focus group moderator to involve himself/herself
during discussion?
(a) When the participants begin to discuss a completely irrelevant topic.
(b) When the participants become passionate about the subject.
(c) When the moderator wants to share his/her own experiences and perceptions.
(d) When the moderator disagrees with a wrong view made by participants.

Question 4. Focus group sessions should end when:

(a) The researcher is able to present information and draw linkage of the themes or views.
(b) Groups eventually give similar views.
(c) When issues discussed contradict each other, and linkage or pattern of themes fail to be identified.
(d) No more new ideas emerge.

Question 5. The following are limitations of a focus group:

(a) The possibility for the group to conform to one dominant opinion.
(b) The focus group does not allow for large volumes of data to be collected.
(c) The findings cannot be projected onto the entire population.
(d) Unprofessional moderating can lead to inaccurate conclusions.

Doctor's particulars:
Name in full:
MCR number: Specialty:
Email address:

SUBMISSION INSTRUCTIONS:
(1) Log on at the SMI website: http://www.sma.org.sg/cme/smj and select the appropriate set of questions. (2) Select your answers and provide your name, email
address and MCR number. Click on "Submit answers" to submit.

RESULTS:
(1)Answers will be published in the SMJ May 2008 issue. (2) The MCR numbers of successful candidates will be posted online at www.sma.org.sg/cme/smj by
15 May 2008. (3) All online submissions will receive an automatic email acknowledgment (4) Passing mark is 60%. No mark will be deducted for incorrect
answers. (5) The SMI editorial office will submit the list of successful candidates to the Singapore Medical Council.

Deadline for submission: (March 2008 SMJ 3B CME programme): 12 noon, 25 April 2008.