48
FOCUS GROUP DISCUSSIONS
topics that were not outlined by the moderator in the introduction and participants need to be made aware of the fluid nature of the discussion. In addition, confidentiality can be difficult to maintain because researchers use quotations from study participants when reporting the study findings. This tradition underscores the need to ensure anonymity of data by removing any identifiable information from the transcript of the group discussion and ensure that quotations are reported in a way that does not disclose the identity of individual participants. The names of participants, if these were collected, should always be kept secure to protect the identities of participants and participants need to be informed how their identity will be protected.
Developing the Discussion Guide
A discussion guide is a pre-prepared list of topics or actual questions used by a moderator to guide the group discussion. A discussion guide can seem deceptively simple; however, it requires considerable forethought to develop an effective guide that fosters productive discussion and elicits useful information to meet the research objectives. Designing a discussion guide is one of the key tasks in preparing for focus group research.
A discussion guide has several functions. Its main purpose is to remind the moderator of the topics and questions that need to be covered to meet the research objectives. Although the questions on the discussion guide are structured in a logical sequence, the actual discussion may proceed in a more haphazard manner as the moderator follows issues raised spontaneously by participants. Therefore, "the moderator uses the guide as a resource to maintain the balance between the researchers focus and the group's discussion" (Morgan, 1997, p. 48). The discussion guide thus becomes a checklist to ensure that all topics were covered by the close of the discussion. The discussion guide also assists the moderator to manage the group discussion by providing effective "warm-up" questions to build rapport and "closing" questions to signal the discussion is coming to an end. It can also include prompts, instructions, and reminders for the moderator (e.g., a reminder to cover ethical issues, indicate where to do a group activity, or to thank participants and provide gifts at the end). A further function of the discussion guide is to introduce some consistency
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 49
in the data collected among different focus groups in the study, which is particularly beneficial if several moderators are used. This enables a comparison of issues across groups to be made during data analysis. This is not to suggest that each moderator needs to ask questions in the same order, but that the same topics or questioning strategies are used across each group so that the results are comparable.
A discussion guide may take different formats, ranging from a simple list of topics to more fully developed questions and probes. A topic guide consists of a list of topics to be covered during the group discussion from which the moderator formulates impromptu questions during the discussion itself. It is generally quick to develop and may lead to a more informal style of questioning. However, there is a great deal of pressure on the moderator to formulate questions on each topic as the discussion progresses, which requires an experienced moderator. A further limitation is the potential lack of consistency in questions asked between different groups particularly where several moderators are used, thereby reducing comparability in data analysis (Hennink, 2007). Furthermore, a topic guide also cannot be pilot tested to check the effectiveness of a questioning strategy, because the actual questions are developed spontaneously by the moderator during the discussion. In academic research the discussion guide is often more fully developed including pre-designed questions and probes. Although this type of discussion guide takes time to develop it overcomes the need for a moderator to formulate impromptu questions, and provides greater consistency in the questions asked across groups. These are important benefits for academic research, where the study population is often divided into subgroups and the issues compared between groups. One drawback may be the reduced spontaneity in the delivery of questions, although an experienced moderator can easily develop a more informal style of questioning using a pre-prepared discussion guide. Krueger (1998) provides a more detailed discussion of these different types of discussion guides. Some moderators may use no discussion guide at all, because they believe it imposes too much structure to the flow of a natural discussion (Greenbaum, 2000). However, "experienced moderators understand the benefits of using a discussion guide in providing structure and focus to the group discussion, while at the same time not feeling constrained by the guide when following
50
FOCUS GROUP DISCUSSIONS
new or interesting issues in the discussion" (Hennink, 2007, p. 49). Whichever type of discussion guide is used, it should be seen as a guide, not a rigid format, that helps the moderator focus the discussion while also maintaining the flexibility to change the topic order or explore new issues as they arise in the discussion.
A discussion guide is not a static research instrument. It often is moderately refined during the process of data collection as more is learned about the study topic. These modifications typically involve refining existing questions, adding more specific probes, or adding additional questions on new issues identified during data collection. The broad structure and focus of the discussion guide usually remains the same throughout the research process, but the refinements add focus, which allows an exploration of issues in greater depth in each subsequent group discussion. For example, a discussion guide may ask a question on the cost of a specific health treatment, whereby participants indicate that they can afford the treatment cost but the frequent clinic visits lead to high transport costs that are not manegable. In light of this information the discussion guide may be modified to add a specific probe about transport costs, thereby enabling researchers to use inductive leads from one group discussion to go deeper into the issues in subsequent groups and more fully explore the issues. Some consistency in the discussion guide is important because one needs to identify when saturation has been reached and data collection should cease, which is only possible with broad consistency in questioning.
Structure of the Discussion Guide
The structure of the discussion guide is important. It can assist the moderator to manage the group discussion by effectively opening the discussion, focusing on key topics, and bringing the discussion to a close (Hennink et al., 2011).
A discussion guide should have a clear and logical structure, even though the actual discussion may follow a different order. The sequence of questions and topics should make sense for the participants, so that a smooth discussion may follow with limited redundancy or repetition of issues. For example, if the discussion focuses on the process of migration, it makes sense to begin by asking about issues before migration, then ask about
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 51
the actual migration, followed by questions about the postmigra-tion experience. However, remember that once the group discussion is underway the order in which topics arise may differ from how they appear on the discussion guide. Nonetheless, a logically ordered guide can help the moderator to memorize the questioning structure, which can be useful during an active discussion to recall topics covered and those to still pursue.
An effective structure follows an hourglass design as depicted in Figure 2.2. The basic principle is to begin the discussion with broad questions to build rapport among group participants so that they begin to feel comfortable in the group environment before moving on to discuss the more specific topics that are critical to meet the research objectives, and then move again to broader summary issues to end the discussion. Each part of the hourglass structure has a specific purpose. The wide top of the hourglass depicts the broad introduction given to participants to provide cognition so they know what to expect during the group discussion. The opening questions begin to develop rapport with participants and make them feel comfortable in the group environment. As the hourglass narrows the discussion guide moves
Provides cognition Provides rapport
V Provides data
J
Provides closure Post-Discussion ^* Provides information
Figure 2.2. Hourglass design of the focus group discussion guide.
52
FOCUS GROUP DISCUSSIONS
toward more specific questions that provide data to respond to the research questions. A discussion guide may include several topics with a series of specific questions under each topic. These topics are placed in the center of the discussion guide so that there is sufficient time for participants to feel comfortable in contributing to the discussion before asking these questions, thereby improving data quality. These specific questions typically comprise one-half to two-thirds of the discussion guide. The hourglass then begins to widen again, depicting a move back toward broad questions that provide closure to the discussion. Closing questions indicate that the discussion is coming to an end and can be immensely helpful for wrapping up an active discussion. In some cases there may also be a post-discussion stage whereby participants are provided with information or resources or asked to complete a deomgraphic survey.
A typical discussion guide includes the following components: an introduction, an opening question, a series of short introductory questions, transitions questions or statements, key topics with specific questions, and closing questions. These components (based on Krueger & Casey, 2009) each have a different purpose as described next. An example discussion guide that includes these components is shown in Figure 2.3 from a study among adolescents attending a summer camp for overweight children.
Introduction
A focus group discussion typically begins with an introduction by the moderator, which may be included in the discussion guide as a written narrative or a series of bullet points. The moderators introduction begins the process of rapport development and should therefore be given in a friendly informal way to make participants feel at ease, rather than reading a formal statement. An introduction script is shown in the example discussion guide in Figure 2.3.
During the introduction the moderator conducts a number of tasks. The moderator should first welcome and thank participants for attending the discussion, and may ask participants to introduce themselves. A note-taker should also be introduced, so that participants are aware of their role. It can be helpful to indicate why participants were selected for the study (e.g., they may share
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 53
Introduction
Thank you all for coming today. My name is (moderator) and this is (note-taker). We are helping (health agency) to find out about your experiences of this summer camp and how it can be improved. The best way to do this is to talk to people who attended the camp, so we are holding these discussion groups with boys and girls at the camp this week. In our discussion today we just want to talk about your experiences at this camp, what you did here, what you learned, what you liked and didn't like - all to help improve the experience for the next camp.
We are not part of the camp itself and are just collecting the information, so we hope that you will feel comfortable to share with us what you really thought about this camp. Please don't feel shy, we want to hear from all of you about your time here. You are the experts because you have been at camp this week and we are here to learn from you. There are no right or wrong answers we simply want to hear your thoughts and suggestions. I have some questions for you but also feel free to add other things you feel are important as we go along.
During our discussion (note-taker) will be taking notes and reminding me if I forget to ask something, but s/he cannot write down every word we say so we would like to record the discussion so that we don't miss anything that is said. Please don't be concerned about this, our discussion will stay confidential and only the research team will listen to the recording. Camp leaders will not listen to the recording. Is it OK with everyone to record the discussion?
During our discussion please let everyone share their views, but only one person should speak at a time so that the recording will be clear. Just join in when you have something to say, we will not be going around the group for every question. Remember we want to hear all your views. It's OK to disagree with others if you have a different opinion but please also respect other people's views. Also, everything that you hear today should be confidential and not shared with people who are outside the group. This discussion will last about one hour, please help yourself to the refreshments. Are there any questions before we start?
Figure 2.3. Example focus group discussion guide.
54
FOCUS GROUP DISCUSSIONS
Let's start by introducing ourselves....
1. Let's each share our first names and where you are from.
2. What type of summer camps have you all been to before? (probe: activities, location)
EXPECTATIONS ABOUT CAMP
First, I would like to hear about your expectations before you came to this camp...
3. What made you all chose to come to this camp? (probe: parents, friends, activities, cost)
4. What did you expect this camp would be like before you came? (probe: people, activities, lessons)
a) What were you most excited about?
b) What were you most worried about?
5. How was the camp different to what you expected? HEALTHY LIFESTYLE LESSONS LEARNED
Now let's focus on what you did and learned at this summer camp...
6. What new things did you learn here about having a healthy lifestyle? (probe on four lifestyle lessons)
7. Which healthy habits from camp can you manage back at home? (probe: reasons)
8. Which healthy habits would be hard to keep at home? (probe: reasons)
ASSESSMENT OF CAMP
Now let's talk about all the things you liked and didn't like about this summer camp...
9. What did you like most about this camp? (probe: activities, people, camp leaders)
10. What were the things you didn't like about this camp? (probe: reasons why)
Figure 2.3. (Continued)
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 55
a) How could these things be changed for future camps?
11. What was the hardest thing you did this week? (probe: physical activity, restrict diet, no cellphones)
SUMMARY & CLOSING
I just have a few last questions....
12. How could more people be encouraged to come to this camp?
13. What would you tell your friends who are considering coming to this camp?
14. Of all the things we discussed today, which are the three most important things to improve this summer camp?
Are there any other things about this camp that you would like to share before we finish?
Thank you for sharing your thoughts with us today
Figure 2.3. (Continued)
a common experience or be residents of the same neighborhood). This reinforces homogeneity among participants, which makes people feel at ease and encourages participation in the discussion. A brief summary of the study is generally provided, as is an indication about how the information will be used. Ethical issues also need to be reviewed during the introduction. Although permissions may have been sought before participants joined the group, a moderator often reviews permission to record the discussion, emphasising that participation is voluntary, and ensuring participants of confidentiality of the discussion.
Finally, the moderator needs to outline how the group discussion will be conducted. This provides participants with cognition, so they know what to expect and feel confident to contribute to the discussion (Hennink & Diamond, 1999). Participants should be encouraged to speak at any time, but that only one person should speak at a time so all dialogue is clearly captured on the recording. It should be stressed that there are no right or wrong answers, that individual
5G
FOCUS GROUP DISCUSSIONS
views are important and different views are valuable, and that it is acceptable to disagree with others in the group who may have different opinions. The moderator can also indicate that they are not an expert in the topic and their role is to facilitate the discussion among the participants who are most knowledgeable about the issues. Participants should also be encouraged to share their comments with the whole group rather than the person seated next to them to discourage fragmentation of the discussion. The length of the session needs to be indicated at the beginning of the discussion. Many moderators also request that cell phones be turned to silent mode.
Opening Question
The first question on the discussion guide is the opening question. It is usually a simple question that all participants can respond to and begins to make participants feel comfortable to contribute to the discussion. The opening question is not intended to promote discussion, therefore it is often short and factual. For example, "Let's start with everyone telling the group which study program you are enrolled in and when you expect to graduate." Some moderators prefer to go around the group allowing each person to respond to the opening question. A subtle benefit of this strategy is that it allows each person to say something at the begining of the discussion, which can promote contribution throughout the discussion, because the longer one remains silent in a group the more reluctant they may be to contribute later in the discussion (Hennink, 2007; Ritchie & Lewis, 2003). The opening question is not intended to produce data and the information is rarely analyzed.
Another strategy for opening the discussion is to conduct a brief activity. For example, Kitzinger (1994) used pile-sorting as an opening activity in a focus group discussion about media images related to AIDS. Participants were asked to sort images into piles according to criteria given by the moderator, for example whether the group believed the images were effective or offensive. Similarly, Hesse-Biber and Leavy (2006) conducted group discussions among gay men, whereby the opening strategy involved the moderator reading a paragraph-long quotation to the group and asking for their reactions. This began an in-depth dialogue that led to many of the issues on the discussion guide. Using an activity can take time and reduces the actual discussion time; therefore, it needs to
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 57
be clearly related to the study topic whereby it can simultaneously build rapport and generate data, as in the examples described.
Introductory Questions
Introductory questions act as a "warm up" for the discussion and begin to focus participants' attention on the research issues. They typically comprise a series of questions with the moderator asking impromptu follow-up questions to encourage detailed responses from participants. It is useful to include several introductory questions, because it can take 10-15 minutes for participants to feel comfortable in a group discussion. An example of a series of introductory questions for a discussion on physical activity among the elderly may be as follows:
1. What types of physical activities are common among the elderly in this community?
2. Where do elderly people generally go to exercise?
3. With whom do elderly people usually exercise?
4. What type of elderly people exercise regularly?
An alternative introductory strategy is to ask the group to explain a term or phrase that is central to the research topic. For example, "We often hear the term 'glass ceiling' in relation to the promotion of women in the workforce, what does this term mean to you?" or "What does a 'balanced diet' mean?" These types of introductory questions provide the group with a common knowledge base about the term or phrase from which the moderator can then build the discussion.
Transition Statements
Transition statements signal a change between topics in the discussion guide. They may be used multiple times throughout the discussion, in particular when transitioning between the key topics and into the closing questions. Several transition statements are shown in the discussion guide in Figure 2.3. For example, the statement "First, I would like to hear about your expectations before you came to this camp" provides a transition between the introductory questions and the first key topic about participant's expectations of the camp. Similarly, the statement "Now let's talk about all the
58
FOCUS GROUP DISCUSSIONS
things you liked and didn't like about this summer camp" signals a move to a new topic about assessing the camp experience.
Key Questions
Key questions are the most important part of the discussion guide. They are the essential questions that generate data to meet the research objectives. This section may include a series of individual questions or several topics each containing a series of questions as shown in the example in Figure 2.3. Key questions include more probes than other parts of the discussion guide to generate detailed responses and promote a discussion on these critical issues. Key questions are placed about one-third to halfway through the discussion guide when participants are warmed up and comfortable in the group setting. At least half of the discussion time is allocated to the key questions. These questions generate the core study data and are therefore analyzed in the greatest depth. The section below on "question design" provides guidance on developing effective questions for this section of the discussion guide.
Closing Questions
Closing questions are designed to signal that the discussion is coming to an end. They help the moderator to effectively close the discussion, but can also provide valuable information that summarizes the issues discussed and can therefore be useful in data analysis (Krueger & Casey, 2000; Greenbaum, 2000). Different closing strategies can be used. A ranking strategy may be used, whereby participants are asked to rank all the issues discussed to identify those most important to them. For example, "Considering all the issues discussed this afternoon, which do you feel are the priority issues for government funding?" Alternatively, a summary strategy can be used, which involves the moderator giving a brief summary of the major issues discussed, then asking the group if this was an accurate reflection of the discussion. This ensures that important issues were not missed and allows participants to clarify points raised. Another closing strategy involves asking what message they would like to convey to a prominent person (e.g., a health minister, president, and policy-maker) about the issues discussed. For example, "Our discussion is almost over, but
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 59
if you had just one minute with the Minister of Health what message would you like to convey from our discussion today?" This type of question can synthesize critical issues from the discussion and provides important perspectives on the issues discussed. It is worth allocating some time for the closing questions because they may elicit fruitful data or stimulate further discussion.
Question Design
The questions in a focus group discussion guide often seem deceptively simple, but in reality they are carefully designed to promote an effective group discussion. Although the question guide for a focus group discussion may seem similar to that used in an in-depth interview there are several key differences (Hennink, 2007). First, there are fewer questions in a focus group discussion guide because the questions are being asked of a group of people and there needs to be sufficient time for participants to respond and engage in a discussion. Second, questions for a focus group are phrased to promote a discussion rather than to elicit a single response, as in an in-depth interview. Third, questions are phrased to seek broad perspectives rather than personal narratives as is sought in an in-depth interview. Suggestions on designing questions for a focus group discussion are highlighted next.
Use clear, short, and simple questions. This ensures questions are easily understood and participants can respond. Questions need to sound informal and conversational to create a comfortable, non-threatening environment for discussion. Therefore, use colloquial language appropriate for the type of study participants and avoid technical jargon or academic language. In general, the longer the question the greater the loss of clarity; therefore, identify the shortest way to ask each question. Use simple questions, and remember that a simple question does not mean a simple answer will be given. Using clear, short, and simple questions also helps participants to remember the question as the discussion proceeds, therefore helping to keep the discussion focused on each issue.
Design open and uni-directional questions. Open questions allow participants to respond from any perspective allowing them to freely share their own views. For example, "What is your opinion of the new course on maritime law?" is an open question because it invites comments on any aspect of the course including positive
60
FOCUS GROUP DISCUSSIONS
and negative issues. Questions need to be uni-dimensional, asking about only one issue at a time. Avoid "double-barrelled" questions with multiple parts because participants may respond to different parts of the question causing a confused discussion. For example, "How does the cost and quality of the new facility compare with others in the neighbourhood?" is a double-barrelled question asking about cost and quality at once. Separating this into two short questions on cost and then quality is more effective.
Reduce dichotomous questions. A dichotomous question elicits a "yes/no" response and therefore does not promote discussion. At times a dichotomous question is needed to clarify whether parti-pants have certain knowledge or experience before asking more detailed questions on the topic. In this case follow the dichotomous question with an open question. For example, ask "Has anyone used the new aquatic centre?" (a dichotomous question) and then follow with an open question, such as "What were your first impressions of this facility?"
Avoid direct personal questions. The lack of confidentiality in a focus group means that personal questions can feel confrontational and make participants feel uncomfortable. Try to phrase questions in a less personal way. For example, rather than asking "Have you ever used drugs?" the same issue may be rephrased as "What type of drug use is common among your peers?" Participants may still respond with their own experience, but they choose to do so rather than feeling compelled by the wording of the question.
Design questions that promote discussion. Consider different questioning strategies that promote an interactive discussion. A useful strategy involves using a statement to promote a discussion. For example, by stating "We have heard people refer to this place as the 'poor people's clinic'. How do you think this affects clinic use?" This statement provides a tangible situation for participants to discuss. Similarly, using a vingette followed by a series of specifc questions can prompt a discussion of specific issues highlighted in the scenario.
Probes
The discussion guide usually includes questions followed by a series of probes (see examples in Figure 2.3). Probes are reminders for the moderator to ask about certain topics related to a question if they are not raised spontaneously by the group.
For example, the question below is followed by a series of probes:
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 61
Who is least likely to be able to pay for health insurance in this country?
(Probe: unemployed/low income, elderly, migrant workers).
A motivated group of participants may highlight a range of issues in response to a question. In the previous example participants may have spontaneously remarked that those who are unemployed, elderly, or have low income are least able to pay for health insurance; therefore, only the probes on insurance regarding migrant workers needs to be asked. Thus, not all probes are needed in every group discussion because many issues are raised spontaneously in response to the question, so the use of probes is guided by the discussion itself. Probes are used most often with the key questions in the discussion guide, because this is where the greatest depth of information is required. Probes initially originate from topics identified in a literature review. However, they are only indications of important issues to probe about, and they may be refined as data collection begins and more is learnt about what is important to the study participants themselves. In the previous example one may learn that students also face difficulties getting health insurance, so "students" may be added as a probe in subsequent group discussions.
Number of Questions
A discussion guide has a limited number of questions. It should be remembered that in a focus group discussion each question is being asked of a group of people, so there needs to be sufficient time for multiple participants to respond to each question, discuss the issue, and raise new issues. A single question in the discussion guide may therefore lead to 5-10 minutes of discussion or longer if the topic is controversial or participants have diverse views. Therefore, most discussion guides have 12-15 questions for a 1-hour discussion.
Having too many questions puts enormous pressure on the moderator to cover all questions in the discussion period, which may lead to superficial coverage of the issues and reduced data quality. It also provides no flexibility for the moderator to follow up on new issues raised or to fully probe the discussion for depth and detail. A focus group discussion is a dynamic activity and time is needed to fully explore the issues, for impromptu questions to be added, and for some deviation from the discussion guide to explore new issues raised. This flexibility is critical to the success of a focus group discussion and to qualitative research in general.
62
FOCUS GROUP DISCUSSIONS
In addition, the topic of discussion and type of participants may influence the number of questions in the discussion guide. For example, a discussion among children or adolescents may include more questions because there may be less discussion than among a group of adults. Fewer questions may be included on a topic that is intense or emotional because more time may be spent discussing each issue. A discussion guide is typically designed for a group discussion of 1-2 hours. Longer sessions may lead to participant fatigue whereby the value of the information declines.
Data-Generating Activities
Conducting an activity in a focus group discussion can be an effective strategy to simultaneously build rapport, promote discussion, and generate data. An activity can immediately change a group dynamic, enabling participants to feel more at ease as they interact on a specific task. Therefore, activities can be effective in building rapport among participants. An activity can also effectively stimulate discussion. Even the simple task of ranking items in a list requires group interaction and discussion. In addition, an activity can provide an alternative means of collecting data on topics that may be difficult to discuss, by indirectly tapping into the thoughts and feelings of participants in a way that would otherwise be challenging through direct questioning. Therefore, an activity may elicit a different type of information that is complementary to the usual questioning strategies. One limitation of including an activity in a focus group discussion is that it reduces the time available for discussing other questions. Therefore, careful consideration of the value of an activity versus using simple questioning strategies is needed.
Many types of activities can be used in focus group discussions. Some activities involve writing, drawing, or responding to visual stimuli. For most activities the outcome of the activity provides data, such as the list developed, drawing created, or the verbal reactions to visual stimuli. However, where activities are intended to stimulate a discussion, it is often necessary to develop a set of questions related to the activity that tap into participants' thought processes and encourage them to critically reflect on issues raised by the activity. Therefore, many activities are followed by a set of discussion questions. When using activities, it is important to keep the focus group intact, rather than split participants into small groups. The intent of an activity is to promote discussion within
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 63
the focus group itself, therefore creating small groups fragments the focus group and creates difficulties in recording multiple discussions, thereby losing valuable data.
When planning to use an activity in focus group research consider the following:
• What is the purpose of the activity?
• How will the activity promote interaction and discussion between participants?
• What type of data will be generated?
• How will the data be recorded (i.e., written or verbal)?
• What is the moderator's role during the activity?
• How do moderators need to be briefed?
• At what point in the discussion should the activity be conducted? (e.g., beginning or middle)
• How much time is needed for the activity?
• What equipment is needed (i.e., marker pens, cards, paper, and so forth)?
• Can the activity be piloted?
• How will including an activity affect other topics to be discussed?
A wide range of activities are commonly used in focus group discussions. Selecting an appropriate activity is determined by the study topic and overall purpose of the activity. Some activities and examples are given in the list below.
Types of Group Activities
Participants are asked to make a list of items, either individually or as a group. This can be followed up by ranking, comparing, or discussing items listed. For example, ask participants to collectively "Make a list of all the water sources in this village" or "List the benefits and limitations of the health screening program." Participants may make the list as a whole group or write individual lists and compare them in the discussion. Krueger and Casey (2009) caution against the moderator standing at a whiteboard to generate a list, because this can change the group dynamic with the moderator dominating the activity.
64   :   FOCUS GROUP DISCUSSIONS
Ranking	Participants are asked to rank items by given criteria (e.g., cost, quality, effectiveness, convenience, and so forth). For example, a moderator may tell participants "Using the items on the list, rank them from the highest to lowest quality." They can be provided a pre-prepared list to rank or rank a list generated by the group itself.
Pile Sorts	Participants are asked to sort a range of images, photographs, or words into different piles by various criteria. They may be given criteria by the moderator or be asked to develop their own criteria of categories that distinguish items. For example, Quintiliani, Campbell, Haines, and Weber (2008) asked focus group participants to write down components of a healthy lifestyle (e.g., exercise, balanced diet, and so forth) on index cards and then sort these cards into piles according to their common characteristics. The piles were then discussed as a group guided by a moderator. In contrast, Kitzinger (1994, cited in Green & Thorogood, 2004) provided focus group participants with pre-prepared cards with statements about who was "at risk" from AIDS, which participants then sorted into groups of differing risk level. This activity prompted discussion among participants on the reasoning for the groupings, thereby providing valuable data on participants' assumptions and knowledge about AIDS. In addition, participants were provided a set of pictures from television and news reports and asked to construct a news report about AIDS.
Visual Props	Props are shown to the group to stimulate discussion, such as educational material (i.e., posters, video clips); current events (i.e., newspaper article, media clip); advertising material; and products. For example, in health research, props have included contraceptive devices, breast implants, and health promotion material, which are shown to the group for initial reactions followed by a series of questions for discussion.
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 65
Vignettes	A short scenario is read aloud to participants, followed by a series of specific questions. For example, participants may be asked what advice they would give to characters in the vignette, whether the scenario is common in their communities, how these issues are typically dealt with, and so on. Brondani et al (2008) describe using vignettes in focus group discussions with older adults to broach discussion on sensitive issues related to oral health. The vignettes included typical scenarios faced by older adults with dentures, which enabled participants to refer to situations in the vignette when discussing sensitive issues, such as personal hygiene, social expectations of facial appearance, and gender norms around oral hygiene.
Drawing	Participants are asked to make a drawing on a specific topic given by the moderator. Figure 2.4 shows a drawing developed in a discussion on stress among graduate students. Students were asked to draw a tree and then in the roots of the tree write all the causes of stress for graduate students. Examples in the drawing include "finances," "coursework," "thesis," "lack of sleep," and "future career." Students indicated that these stressors are compounded by time (as they depicted by the clock). They were then asked to write in the tree branches all the effects of stress. Some examples shown include "anxiety," "lack of exercise," "competitiveness," "exhaustion," and "isolation." After a drawing activity the moderator may lead a group discussion on the issues highlighted in the drawing, thereby providing valuable data that are inductively derived (from the drawing activity) and have depth and context (from the subsequent discussion).
Mapping	Participants are asked to sketch a map of their neighborhood and then mark certain items or places on the map as directed by the moderator. For example, participants may be asked to sketch a map of their neighborhood and then mark all the places they believe are safe or dangerous, hygienic or unhygienic, and so forth.
66
FOCUS GROUP DISCUSSIONS
Cartoons    Participants are asked to react to a cartoon or
complete an empty speech bubble with a word or thought, which can effectively lead to a discussion. A cartoon may be used to begin a discussion on a difficult topic. Barbour (2007, p. 84) states that cartoons "often tap into and express succinctly in an amusing way, difficult and keenly felt dilemmas, but take the sting out of thinking about these. They thus, simultaneously break the ice and give permission to raise difficult issues."
Figure 2.4. Example drawing activity depicting the root causes and outcome of stress.
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 67
Wherever possible it is important to pilot-test any stimulus material or activities to ensure they generate the intended type of discussion. For further reading on developing activity-oriented questions for focus group discussions see Colucci (2007). Many participatory tools, in particular community mapping, can also be effectively adapted for use in focus group discussions. For examples, see the selection of tools in International HIV/AIDS Alliance (2006).
Translating the discussion guide
When focus group discussions are conducted in a different language it is recommended to translate the discussion guide into the language of the discussion. Even when the moderator is bilingual a translated discussion guide is recommended. This takes a lot of pressure off the moderator who would otherwise need to translate each question spontaneously during the group discussion. This is described in an extract from a bi-lingual researcher in Zambia.
I thought I am competent in the language of my country and I can read a question in English and directly translate it, no problem. But it proved very, very difficult. Translating it correctly from English to the local language, right there and then, was very, very difficult. The easier way is to translate the whole instrument into the local language and use the local languge. (Hennink, 2007, p. 66)
Translating the discussion guide allows the most appropriate colloquial language to be used so that questions are understood by participants as intended, thereby increasing data quality. Using appropriate colloquial language can also aid rapport development with participants. It is important that the translation uses an informal style of language appropriate to the study participants, because too formal language can create unwanted distance between the moderator and participants (Hennink et al., 2011). This type of translation can often be achieved by asking bilingual professionals from the study setting (i.e., teachers, nurses, and so forth) to translate the guide, rather than a professional translation service. It may be necessary to translate the discussion guide into several languages, particularly if the study is conducted in several regions of a country where different languages are used. In some countries there may be a national lingua franca that can be used.
68
FOCUS GROUP DISCUSSIONS
The most important consideration is to identify the language in which participants would feel most comfortable to converse.
It is imperative to check the quality of the translation. Back-translation is a common strategy, whereby the text is translated back into the original language and checked for accuracy. What is most important is that the translation conveys the intended meaning of a question, rather than simply being an accurate translation of the words. Another strategy is to read translated questions to individuals familiar with the language and identify how each question is understood. For detailed guidance on translating a focus group discussion guide, selecting an appropriate language, and checking translations see Hennink (2007).
Piloting the discussion guide
It is good practice to pilot-test the discussion guide, because it can be difficult to predict how participants will interpret questions. Pilot-testing involves asking the discussion questions to a group of people with similar characteristics to the study population (if possible) or using the first focus group discussion as a pilot. After the pilot-test the questions and the overall structure of the discussion guide should be reviewed. Review questions to ensure they are clear, understood as intended, use appropriate language, are logically ordered, and there is no repetition or redundancy. Also, review the overall structure of the guide to assess the sequence of topics, number of questions, and the amount of time that participants discuss each issue. This may lead to reordering or removing questions to facilitate a smooth flow between topics and questions. The following questions (Hennink et al., 2011, p. 149) maybe asked during the pilot-test:
• Was sufficient information given in the introduction to the group discussion?
• Were all questions understood as intended?
• Do any questions need to be re-worded to improve clarity?
• Is the structure of the discussion guide working well?
• Does the topic order need to change?
• Will the information help to answer the research question?
• Is the discussion guide an appropriate length for a 60-90 minute discussion?
DESIGNING AND CONDUCTING  FOCUS  GROUP  RESEARCH   : 69
The discussion guide is not a stand-alone instrument. The moderator is an important component in the delivery of questions and clarifications, and in probing the discussion. Therefore, the questions and their delivery by the moderator need to be assessed in the pilot-test. Pilot testing is thus an opportunity to also provide feedback to the moderator. During the pilot-testing it is important to identify whether any problem identified is caused by the design of the guide, the skills of the moderator, or perhaps issues related to the participant group. The guide can then be redesigned, the moderator retrained, or issues related to participants themselves can be reviewed.
Conducting Focus Group Discussions
Roles of Moderator and Note-Taker
Conducting the group discussion is the central activity in focus group research. It generates the study data; therefore, an effective group discussion is critical. Managing an effective discussion can be challenging and rewarding. A focus group discussion is typically conducted by a moderator and note-taker team.
A note-taker has multiple roles. Primarily they are responsible for writing down the key points raised in the discussion in as much detail as possible. The note-takers summary is important because it is the only record of the discussion if the recording device fails or participants refuse permission to record the discussion. The note-taker's summary needs to include the main points discussed and if possible key phrases or short sentences that reflect participants' expressions on critical issues. The summary should separate facts from any interpretation of the issues by a note-taker. In addition, a note-taker can also attend to any disturbances to the group, such as latecomers, so that the moderator can focus on conducting the group discussion. A note-taker sits outside the actual discussion circle to attend to these issues and to take notes unobtrusively. The note-taker can also operate the recording device and assist the moderator with other tasks that arise. More details on recording the group discussion are provided later in this chapter.
The moderator has the critical role of conducting the group discussion. In many ways the role of the moderator is similar to that of an interviewer in an in-depth interview in that they are responsible