II –focus group DISCUSSIONS
2.1 - What is a focus group discussion?
A focus group discussion (FGD) is:
- a method of qualitative research;
- a situational method;
- a multi-dimensional communication process.
A focus group is an organised discussion – though structured in a flexible way – of between 6 and 12 participants. It usually lasts one or two hours and provides the opportunity for all the respondents to participate and to give their opinions. Dominant and submissive relationships which develop within the group, as well as side conversations, can be controlled. Smaller groups and those with a narrower range of characteristics tend to be more coherent and interactive.
Focus groups differ from informal group discussions in a number of aspects. First, specific, pre-determined criteria are used for recruiting focus group participants. Second, the topics to be discussed are decided beforehand, and the moderator uses a pre-determined list of open-ended questions arranged in a natural and logical sequence. The moderator may even memorise the questions beforehand.
Finally, focus groups rely on discussion between participants about the topics presented, and group members may influence each other by responding to ideas and comments that arise during the discussion. There is no pressure on the moderator, however, to have the group reach consensus.
Focus group discussions are not always easy to conduct and require planning and training.
2.2 - Aim of focus groups
Focus groups are used to:
- gain understanding of the subject being researched;
- provide an accurate picture of the subject’s experience of reality;
- evaluate and analyse needs;
- formulate interventions;
- test new ideas or programmes;
- improve existing programmes;
- obtain a wide range of information on a given topic in order to develop more structured questionnaires;
- inform policies.
These types of discussions allow the researcher to identify where participants’ points of view converge and diverge, and to investigate the whys and wherefores or certain phenomenon.
2.3 - Composition of groups and analysing data by group
Focus groups should be composed of homogeneous members of the target population. It is a good idea to form groups of respondents who are similar in terms of social class, age, level of knowledge, cultural/ethnic characteristics and sex (in addition to any other variables you may identify). This will to create an environment where participants are comfortable with each other and feel free to express their opinions.
Data should be analysed by group. This also allows for the results of different types of group to be compared.
It should be noted that the two paragraphs above describe what should happen in an ideal situation. The experience of the authors of this document in organising focus group discussions in various West African countries has been varied. In most cases, your experience will be different from what has been described above. Despite careful planning on the composition of the group, various cultural factors can influence group dynamics such as who gets to speak, etc.
2.4 - Strengths of focus group
The advantage of focus groups over individual interviews is that the comments of one participant can generate comments from other participants. Ideas and opinions can be developed and explored more so than in individual interviews. These types of discussions can be very productive. Researchers and interviewers can benefit from the ideas generated in these discussions. In a short amount of time, a large quantity of information can be collected – often more quickly and at a lesser cost than via individual interviews.
Focus groups are useful for obtaining preliminary information about beliefs, ideas, opinions, attitudes and behaviours in a community. They are particularly helpful in identifying pertinent and appropriate questions for individual interviews (semi-structured or structured).
2.5 - Weaknesses of focus group
Some weaknesses are:
- The researcher has less control over the flow of discussion in the group interview as compared to the individual interview;
- Focus groups cannot tell you about the frequency or distribution of beliefs in a population;
- Results are harder to analyse than individual interviews. Participants’ comments must be interpreted within the particular social environment created by the focus group (a discussion among strangers in a neutral place), and care must be taken to avoid lifting comments out of context or out of sequence;
- Because the amount of response time required for any given question is increased by having discussion among participants, the number of questions that can be addressed is smaller than in individual interviews;
- Facilitating and conducting a focus group interview requires considerable skill. It is important to know how to manage the interview so that one or two people do not dominate, and all participants are able to share their views.
Once the discussions are over, the analysis of the information can be long and costly. Very often, the researcher will have to listen to the tape recordings of the discussions again, transcribe the comments, code and analyse them etc. All of these steps are lengthy and require specific skills.
2.6 - Managing a focus group
An experienced moderator will be in control of the group, less experienced moderators are often afraid of losing control and this anxiety will make itself shown in several ways:
- Asking questions that show that the moderator already knows the answer;
- Asking questions in a mechanical way;
- Asking a question without giving the participants a chance to respond to the previous one;
- Interrupting when it is not necessary or not giving the participants a chance to finish what they were saying;
- Jumping to conclusions – the moderator makes a deductions not based on what the participants have said;
- Giving advice to participants about what they should do;
- Changing the subject too quickly and not giving participants enough time to express themselves before moving on.
Another problem with focus groups is the consensus effect. Depending on the composition of groups, certain participants may have a tendency to adopt the opinions of the group leader which makes the data generated in the focus group useless. An experienced moderator can avoid this.
Annex C gives information on specific techniques to avoid these types of problems.
2.7 - Other group interview techniques
Not all group interviews are focus groups. As we have seen, focus groups are special types of interviews where the researcher has control over the purpose, size, composition of the group, and procedures governing the discussion. Other less formal but equally useful group discussions exist. For example, interviewers can conduct discussions with naturally pre-existing groups such as families or teachers and students. These discussions can take place in a cafe or at a women’s health centre, for example. In these types of interviews, the researcher has less control over the sample size, group composition, etc., but the context is more natural.
Non-focus group discussions often involve unstructured or semi-structured interviewing techniques.
Other techniques resemble those employed in standard discussion groups and vary according to the number of participants and/or the length of the discussion.
2.8 – Dyads and triads
For several years now, researchers have been interviewing dyads (groups of two participants) and triads (groups of three participants). Smaller groups allow the moderator to avoid an individual’s response being influenced by the group, a problem often encountered in larger groups. Many qualitative research specialists highlight the value of these techniques (including individual interviews) when in comes to testing out creative concepts. Dyads and triads can replace traditional focus group discussions by making optimal use of research budgets where many segments of the target population have to be interviewed. They can also replace one-on-one interviews and, in this way, save time. These smaller groups are particularly useful with children and adolescents as age range can be restricted and turbulent behaviour minimised.
These methods facilitate the collection of information on questions, products and services when carried out by a qualified moderator. Small group interviews can last from 30 minutes to a whole day depending on the needs and type of research being conducted.
Many participatory data collection techniques exist that require group members to work together on a specified task. Participatory methods are particularly useful for action-based projects; researchers work with community members to identify appropriate intervention strategies.