Objective: To review and synethize the research material on focus groups with children and adolescents and to provide guidelines for future development.
Methods: Psychlit, Medline, and Cinahl electronic databases, as well as the reference lists of those articles consulted, were reviewed for information regarding focus groups with participants under the age of 18 years. Both empirical and methodological articles were part of this review.
Results: We review the utility of focus groups for exploratory research, program evaluation, program development, and questionnaire construction or adaptation. Based on previous research, we provide guidelines for focus groups with children and adolescents and outline suggestions for future development.
Conclusions: There is evidence to suggest that focus groups are a valuable means of eliciting children's views on health-related matters, given an appropriate research question. However, empirical research is required in order to investigate systematically the effect of different processes and variables on the final outcome of focus group interviews.
Increasing recognition of the benefits of the qualitative research paradigm has opened up new means of exploration and investigation. As part of this trend, there has been an increase in the use of focus groups as a viable alternative to traditional one-to-one interviews. Krueger (1994) describes focus groups as “a carefully planned discussion designed to obtain perceptions on a defined area of interest in a permissive, nonthreatening environment” (p. 6). The goal of the group is to elicit a discussion that allows the researcher to see the world from the participants' perspectives. For example, focus groups may be used to understand people's opinions of a program, event, or service, to explore the rationale behind people's thoughts or behaviors, or to facilitate the expression of people's perceptions of a certain phenomenon in an open, free, relaxed format. Focus group discussions allow the researcher to probe both the cognitive and emotional responses of participants while observing the underlying group dynamic.
The principles of focus groups set out by Merton, Fiske, and Kendall (1956) remain in common practice (Krueger, 1994). Vaughn, Schumm, and Sinagub (1996, p. 5) outline the core elements as follows: (1) the group is an informal assembly of target persons whose points of view are requested to address a single topic; (2) the group is small, 6-12 members, and it is relatively homogeneous; (3) a trained moderator with prepared questions and probes sets the stage and induces participants' responses; (4) the goal is to elicit the perceptions, feelings, attitudes, and ideas of participants about a selected topic; and (5) focus groups do not generate quantitative information that can be projected to a larger population.
Until the mid-1980s, focus group methods seem to have been used almost exclusively with adult research participants. Then, for the first time, a small number of publications appeared referring to their use with children. This work arose in fields as diverse as market research (McDonald & Topper, 1988), health (Heimann-Ratain, Hanson, & Peregoy, 1985), education (Vaughn, Schumm, Niarhos, & Gordon, 1993), and child welfare (Hill, Laybourn, & Borland, 1996). Despite this body of empirical studies, we could find no publications synthesizing the research in which focus groups have been used with children or providing interested researchers with guidelines on how to conduct them. This article attempts to fill this gap in the literature.
In order to identify articles that reported on the use of focus groups, searches were conducted using the Psychlit, Medline, and Cinahl electronic databases, as well as the reference lists of the relevant articles. Empirical and methodological articles were included. Using the search words “focus group” and “children,” we identified a total of 608 articles in the above databases from 1982 to the present time. Ninety-three empirical articles were identified as relevant to this review. The criteria used to reduce the number of empirical articles for inclusion in the review were (1) the abstract indicated that focus groups had been one of the primary methods of empirical data collection, (2) children under 18 years were one of the main target groups of the research, and (3) the research was related to pediatric health care. To obtain further information, we conducted searches within PsychLit using the following key words: “focus groups,” “ children and group work,” “children and group format,” and “children and group processes.”
Throughout this review, the main aim is to provide interested researchers with a means of determining whether focus groups would be appropriate for their purposes rather than to provide an exhaustive review of all relevant articles. Two further aims are to discuss current practices in focus groups with children and to provide suggestions for future research in the area.
Advantages of Focus Groups
Basch (1987) has suggested that focus groups with adults are a relatively easy way to learn about the ideas and opinions of homogeneous groups and that they produce a diverse range of information. The same advantages apply to the use of focus groups with children. The fact that children may respond in ways that they believe the researcher desires has long been seen as a threat to validity in one-to-one interviews between adults and children (Donaldson, 1978). By removing the emphasis on the adult-child relationship, the focus group may help to overcome some of these concerns. Lewis (1992) also points out that, unlike an interview, the focus group does not have to end when an individual does not respond, thus removing the pressure from a child who might otherwise be tempted to respond to a question that was not fully understood or outside his or her realm of experience.
Levine and Zimmerman (1996) suggest that a further important advantage of using focus groups with children is that the method acknowledges the participants as experts. The aim is essentially to discover children's view of their world. Because of this, the results are likely to have high face validity and can be useful in the development of programs, services, or conceptual models.
A final advantage of focus groups is their flexibility and ability to combine well with other qualitative and quantitative methods (Morgan, 1997; Vaughn et al., 1996). Focus groups can be used to provide meaning to a forced choice format questionnaire by examining the reasons and motives behind people's behavior, or they can be used to inform the design of an instrument.
Limitations of Focus Groups
Basch (1987) argues that one of the major limitations of focus groups is that they are not useful for testing hypotheses in the traditional experimental design. Nor does he believe that they are appropriate for drawing inferences about larger populations or for statistical testing and interval estimation, which require quantitative findings. Although the group interaction is generally seen as an advantage of focus groups, there is always the possibility that intimidation within the group setting may inhibit interaction (Lewis, 1992). Lewis points out that children in focus groups may also adopt themes previously raised by other children rather than offering their own opinion. However, she also acknowledges that this tagging on may indicate the salience of these ideas within the group.
There have also been challenges to the claims made about openness in responding in focus groups. For example, Kitzinger (1994) found that, when sensitive personal issues were being discussed, adults sometimes confided information to the researchers that they were not prepared to share with the group. Fern's (1982) research found that individuals who were interviewed on their own reported feeling more anonymous than individuals who participated in focus groups.
Another difficulty associated with the use of focus groups, which is not encountered in individual interviews, is scheduling a time and location convenient to all participants. Caution must be also be exercised, as the quality of the findings is inextricably tied to the skills of the moderator (Festervand, 1984-1985). Potential misuses include the use of focus groups for unsuitable topics (for example, to obtain biographical information about a person's life) and the use of focus groups with a small and unrepresentative sample as the only empirical evidence to support a decision (Morgan, 1997).
Review of Research
Focus groups can be used in a variety of different types of research studies. The examples chosen in this section are designed to reflect this diversity so that researchers can determine whether focus groups might be useful to their work.
Many aspects of health-related behavior that are well understood in adults have not been researched with children (e.g., Green & Hart, 1998). In these circumstances, focus groups can be a useful means of initial data collection, as they allow exploration of children's knowledge, perceptions, and experiences. In this section, examples of exploratory studies are taken from research on a range of children's health-related behaviors.
One widely studied issue using focus groups is the broad area of sexuality, including topics such as sexual activity, contraception, abortion, STDs, and parenting. One of the first such studies was conducted by Kisker (1985), who explored the general question of why contraceptive use was so poor among teenagers in the United States. Barker and Rich (1992) used focus groups to explore factors influencing adolescent sexual behavior with young people in Nigeria and Kenya. The focus groups were particularly appropriate because the authors were interested in peer interaction as a factor in decision making about sexual activity. Although most of the studies of sexuality have involved adolescents, such research has also been conducted with younger children. For example, focus groups have been conducted to determine children's understanding of AIDS and their affective responses to people with AIDS (Brown, Reynolds, & Brenman, 1994; Hoppe, Wells, Wilsdon, Gillmore, & Morrison, 1994).
In addition to sexuality, a number of other issues have been investigated using focus groups. One such issue is risk taking and injury prevention. Green and Hart (1998) used focus groups to investigate children's (7 to 11 years) understanding of accidents, accident prevention, and advice about avoiding accidents. Such research can provide educators with an excellent basis for developing a range of educational programs.
Of the 93 articles relevant to this review, 51 could be broadly classified as exploratory studies. Most of those studies reviewed were intended to contribute to the development and application of health education programs or services, rather than theory building. Among the commonly mentioned aims were informing health promotion programs (e.g., Beck, Summons, & Hanson-Matthews, 1987) and the development or expansion of health services (e.g., Barker & Rich, 1992). Some studies intended using the results of their focus group discussion to generate hypotheses (e.g., Bull & Hogue, 1998). Hockenberry-Eaton et al. (1999) developed a conceptual model of fatigue, and Hauser and Dorn (1999) wished to develop a service model based on the focus group data. Stanton, Black, Kaljee, and Ricardo (1993) used the focus groups in their study to guide the application of a theoretical model of adolescent sexual behavior. Other less commonly cited uses of focus group research include informing professional practice and as a teaching method (Brown et al., 1994; Murphy, Youatt, Hoerr, Sawyer, & Andrews, 1995). Of course, the information generated by these exploratory studies may be put to a much greater range of uses than was originally intended by the investigators.
Focus groups have increasingly been used in an effort to assess the needs of target groups and assist in the development of meaningful and effective health promotion programs. Such efforts span areas as diverse as general health promotion programs for adolescents (Deering, 1993), educational interventions to enhance interactions between health professionals and families (e.g., Croft & Asmussen, 1993), health enhancement interventions for chronically ill populations (Barlow, Shaw, & Harrison, 1999), and contextually relevant programs for specific sociocultural regions (Gazzinelli, Gazzinelli, Filho, Sa, & Kloos, 1998).
Examples of the use of focus groups in program development include, for example, Heimann-Ratain et al.'s (1985) smoking prevention program which was developed and pretested using focus group interviews. In common with other studies, they used focus groups as a means of obtaining information on the attractiveness and appropriateness of program components. Stevenson and Lennie's (1992) research is another good example of the use of focus groups in program development. They used focus groups at two separate stages in the development and implementation of a bicycle helmet program in Australia. The initial round of focus groups was used to develop a questionnaire that could provide baseline data on patterns of usage. A bicycle helmet program was subsequently developed based on preliminary research, and a helmet trial was conducted. After the trial, focus groups were conducted to identify the most popular helmets for each age group. Thus, focus groups can be usefully employed during different phases in the development of a program.
Overall, the findings of these studies suggest that focus groups can make a valuable contribution to the design and pretesting of educational programs and services for young people. Within this context, focus groups can fulfill important functions in terms of needs assessment and ensuring the feasibility of program components.
A small number of studies have used focus groups to evaluate an intervention program (e.g., Boaz, Ziebland, Wyke, & Walker, 1998; Watson & Robertson, 1996). Elliott, Gruer, Farrow, Henderson, and Cowan (1996) used focus groups and questionnaires as part of their evaluation of a theater production on HIV. Quantitative data were collected on knowledge, attitudes, and risk-taking behavior, and the focus groups were used to achieve a deeper level of understanding of the issues. The transcripts from the focus groups reflected and illustrated the findings from the quantitative parts of the research. The focus groups also provided a good account of how enjoyable the theater production had been.
Although the total number of studies in this section is small, results show that focus groups can provide important information on the utility of the intervention and the meaning of the messages conveyed.
Focus groups have numerous applications in the development and adaptation of questionnaires. For example, they can be used to determine respondents' ideas regarding the content of the questionnaire, the language used, and the question and response format. This could include the adaptation of questionnaires from one cultural setting to another. When the initial instrument has been devised, focus groups can also function as a pretesting tool by assessing the acceptability of the instrument to potential respondents, their comprehension of the language and layout, and the time taken for completion.
Many researchers have referred to the use of focus groups to generate items for inclusion in questionnaires (e.g., Amos, Gray, Currie, & Elton; 1997; Mwanga, Mugashe, Magnussen, Gabone, & Aagaard-Hansen, 1998). However, few make the links between the focus group data and the questionnaire items explicit. One exception is French, Christie, and West's (1994) research on the development of the Childhood Asthma Questionnaires (CAQs). The authors involved mixed groups of children with and without asthma in discussions about various aspects of their lives (e.g., activities after school, games, and physical education lessons). The factors of greatest importance generated by the children were subsequently turned into questionnaire items and combined with other items based on clinical literature on childhood asthma. The discussions not only yielded information important for the development of items for inclusion in the questionnaire but also influenced its overall format. The children's enthusiasm for “smiley” faces to depict how they felt during the group discussions prompted the adoption of the faces to represent the five points of a Likert-type rating scale in the completed questionnaire. French, Carroll, and Christie (1998) subsequently used focus groups as part of the process of modification of their CAQ for another culture.
To date, the potential contribution of qualitative data to questionnaire construction has not yet been fully exploited. In this respect, focus groups represent an economical means of obtaining large amounts of qualitative data in short periods of time.
Successful focus groups with children require attention to the developmental abilities and needs of the participants. Engaging children's interest and tailoring the interview to their level of comprehension are necessary prerequisites for maximum participation. This implies that the moderator must be familiar with the cognitive and social capacities of children of different ages. He or she must also be sensitive to the children's communication abilities and attention span. Within health services research, the moderator also needs to be familiar with the likely developmental changes in children's understanding of health and illness (Bearison, 1998). Failure to do so could result in inappropriate questions or inappropriate interpretation of children's responses. In the following section, we consider how the challenges of conducting focus groups with children as participants have been dealt with in previous research.
The optimum size of a focus group with children is generally smaller than that of adults. According to Hoppe, Wells, Morrison, Gillmore, and Wilsdon (1995), it is preferable to have four or five participants in the group to ensure at least three “ talkers.” The risk with smaller groups is that the group discussion will resemble parallel interviews as opposed to a dynamic group interaction, whereas larger groups may become difficult to control. In general, researchers tend to recommend somewhere between four and six participants per group (Greenbaum, 1988; Hoppe et al., 1995; Vaughn et al., 1996). These suggestions represent guidelines based on the views of experienced moderators and researchers. Searches of the literature found no empirical studies investigating the role of group size in focus groups with children of any age. Research on the effect of focus group size with adults (Fern, 1982) found that focus groups with eight members produced more ideas than focus groups with four members. However, the larger group size may be counterproductive with young children, who may have more difficulty keeping on task and a greater tendency to talk simultaneously and interrupt each other. Based on this review of 93 empirical studies, 33 studies involved focus groups with four to eight participants, and 30 studies had at least one focus group with more than eight participants. Data on group size were not presented in 25 studies.
Length of Interview
The appropriate length of the interview is shorter for children than for adults. Vaughn et al. (1996) recommend that focus groups should be approximately 45 minutes for children under 10 years and around 60 minutes for children between 10 and 14 years. Greenbaum (1988) recommends that 90 minutes is the maximum time period for group discussions with children; however, younger children will require shorter time periods. Searches of the literature found no empirical studies on the optimal length of a group discussion with children of different ages. The overall length of the group session will depend on many variables including the subject chosen, the skill of the moderator in eliciting children's views, and the time of day. In the 93 empirical studies considered in this review, the majority of focus groups with children and adolescents lasted between 30 and 90 minutes.
Age of Participants
Most researchers conclude that focus groups are not suitable for children under 6 years because they do not have the social or language skills to be effective participants in group discussions (Greenbaum, 1988; McDonald & Topper, 1988; Vaughn et al., 1996). In contrast, Vaughn et al. (1996) argue that children over 6 years are likely to be very effective participants in focus groups as they are likely to provide more spontaneous and fewer socially desirable responses than some adult participants.
Only four studies were identified in this review that utilized focus groups with children younger than 6 years. One of these is a study by Klein et al. (1992), which examined the influence of candy cigarettes on smoking behavior and attitudes among children in three age groups including one group of 4- and 5- year-olds. Limited information is provided on the success of the groups; however, the authors conclude that focus groups are unreliable with very young children.
In contrast, Charlesworth and Rodwell (1997) had more positive views about the use of focus groups with this age group. In their study, focus groups were used as a tool for program evaluation among children from kindergarten to fifth grade. They concluded that focus groups can be used effectively with very young children. Turner, Mayall, and Mauthner (1995) included a variety of creative activities in their focus groups with young children (5 to 6 years old) including writing, reading, and card sorting. Activities such as these are likely to facilitate children's participation in group discussion. However, more extensive research needs to be conducted regarding the value of focus group interviews with such young children.
The importance of ensuring that focus groups contain children who are broadly similar in age is emphasized by the experiences of a number of researchers. The reasons given for restricting the age range are typically based on the differing cognitive abilities and sensitivities of children of different ages. For example, Hoppe et al. (1995) worked with children from third to sixth grade and found that younger children were less reticent in discussing sexual topics than older children. The older children also appeared to have higher levels of anxiety about peer reactions. Thus, restricting the age range of focus group participants should make it easier for the moderator to compose questions appropriate for the participants and to ensure that topics are approached sensitively.
Gender of Participants
Homogeneity with respect to gender is frequently recommended when conducting focus groups with children (Greenbaum, 1988; Hoppe et al., 1995; Spethmann, 1992; Vaughn et al., 1996). According to Greenbaum (1988), younger children often dislike the opposite sex in a way that may hinder group productivity. For older children and teenagers, there is high interest in the opposite sex that can also negatively affect group productivity. Different interests and attitudes can also produce conflicting viewpoints that may not be conducive to a trouble-free environment. Davis and Jones (1996) reported that the mixed gender focus group approach worked well with 9- to 11-year-olds but had to be adapted for the 13- to 14-year-olds, where the agendas of teenage girls and boys turned out to be not only different but conflicting. The authors found that the formation of eight single-sex groups in the secondary schools resulted in some of the richest material in the study. In this review of 93 empirical studies, 28 used single-sex groups only. The remaining studies either used a combination of homogeneous and mixed gender groups or made no reference to the gender composition of the groups.
Ensuring that children feel comfortable and confident as members of the group discussion is a challenge to all moderators. Some researchers opt to include groups of friends as a means of ensuring productive groups of participants. Lewis (1992) argues that friendship groups have already passed through the early stages of group behavior, thus facilitating the free expression of ideas. Davis and Jones (1996) included both friendship groups and “representative” groups in their research, and they concluded that the latter groups required much more intervention and direction from the moderator than the former. Watson and Robertson (1996) likewise reported that friendship groups of teenagers facilitated group discourse in their evaluation of an HIV/AIDS education program. Not all researchers agree, however, on the value of friendship groups. Based on experience in advertising and marketing, Spethmann (1992) argues that peer pressure in group settings is diminished when children do not know one another. Only a minority of those studies reviewed selected groups on the basis of friendship links; therefore, the precise benefits of composing groups in this manner is unknown.
Overall, to obtain maximum output from focus group discussions with children, the composition of the group must be planned in advance, if at all possible. In addition to the above factors, ethnicity and social class may be other potential intervening variables. In general, the rule of thumb appears to be that homogeneity is best.
Focus Groups Versus Individual Interviews
Whether to choose a focus group or a series of individual interviews must depend on the research question and the time and expertise available to the researchers. Whereas the advantages and limitations of focus groups have already been discussed, it is worth considering the specific factors that may influence the choice of one method over the other. Lewis (1992) claims that a focus group can generate a greater range of responses than an individual interview; however, this claim is challenged by the findings of Fern (1982). In his research, Fern systematically compared the number of unique relevant ideas from focus groups and interviews with an equal number of adult individuals. His findings suggest that interviews produce a larger number of ideas. However, the number of ideas generated may not represent the most appropriate measure of the value of a focus group discussion.
In contrast to interviews, focus groups allow researchers to observe the dynamic interchange in group settings (Morgan, 1997) and are well suited to understanding shared meanings (e.g., Hockenberry-Eaton et al., 1999). However, O' Brien (1993) argues that individual interviews may be the preferred method for particularly complex topics so as to allow the interviewer to further explore the participant's experience in a manner that could not be done in a group setting (e.g., biographical personal details) (Morgan, 1997).
Another reason for choosing to use focus groups relates to the importance of the peer group in children's lives. Children are essentially social beings and spend much of their lives in groups (Dwivedi, 1993). Many of the behaviors of children are enacted within the context of groups; as such, the group setting represents a familiar and reassuring environment for children.
Very little has been published about the ethical issues involved in focus groups (Morgan, 1993). However, a number of principles relevant to all research with children are applicable. In line with the U.S. Department of Health and Human Services (1991) Policy for Protection of Human Research Subjects, the informed consent of parents and the assent of children would be required for participation in focus groups. Thus, parents need to be informed about the nature of the research and should be asked for written permission for their child to take part. The age at which children may legally consent to participate in research varies from jurisdiction to jurisdiction, but even when a child may not legally consent, his or her assent should be sought. To assent, children need to have the aims of the research and their role in it explained in developmentally appropriate language. The researcher should also explain how the information that the children provide would be used. Children lack social power (Hill, 1998; Vaughn et al., 1996), so the moderator must ensure that the children know that they are free to decline to participate even if their parents or guardians have already given consent. They should also be made aware that they can withdraw at any time and that withdrawal will have no negative consequences for them.
Once parental consent and children's assent has been freely given, the focus group moderator needs to consider ethical issues specific to focus groups. These issues arise from two aspects of the focus group discussion: (1) the fact that disclosures by participants are shared with all group members and not just the researcher, and (2) intense group discussion may give rise to stress or distress in individual participants. Explaining the meaning of and need for confidentiality should form part of the introduction to the focus group and could also include information on what will be done with the researcher's notes and tape recordings of proceedings.
Given the group context, it will not be possible for the moderator to guarantee that participants will not be upset or offended by one another's comments. Smith (1995) offers a number of suggestions for focus groups on sensitive topics. These include the need for the moderator to monitor stress levels of participants and to be prepared to intervene when necessary, the importance of having small groups, and the value of debriefing sessions in which participants can discuss their reactions to the discussion. He also emphasizes the importance of having a co-leader with clinical experience present in order to adequately monitor the “ comfort level” of the participants. Although Smith's suggestions relate to focus groups with adult participants, they are equally applicable to research with children.
Ethical concerns are relevant to all research in which focus groups are used; however, they are particularly pertinent when the topics discussed are sensitive. Thus, researchers organizing focus groups to discuss topics such as AIDS or sexual behavior would need to pay particular attention to these issues. Researchers working on such sensitive topics also need to inform children that confidentiality cannot be absolute and that a disclosure of abuse or neglect would have to be reported to the appropriate authorities (Hill, 1998; Koocher & Keith-Spiegal, 1998). In such circumstances, it would also be important to develop a policy on how such disclosures would be handled (Charlesworth & Rodwell, 1997).
Triangulation and Credibility
Approaches to establishing reliability and validity of qualitative data differ from those used with quantitative data (Merrick, 1999). Two approaches mentioned with the focus group literature are triangulation and credibility. Triangulation refers to the overlap between data produced using different methods. Potential sources of data for triangulation with focus groups include interviews and surveys (Morgan, 1993) and a range of other qualitative techniques. For example, Stanton, Aronson, et al. (1993) compared data obtained in focus groups with data obtained from individuals using a pile-sorting technique. Pile-sorting requires individuals to organize cards with phrases or pictures into clusters or categories based on a perceived shared dimension. The authors found that the pile-sorts revealed support for most of the views expressed in the group discussions. However, the authors also concluded that pile-sorts were useful in exposing divergent viewpoints, as some interpretations emerged that were not revealed in focus group discussions. For example, age and gender differences emerged in the analysis of the pile-sorting that were not apparent in the interpretation of the focus group discussion. This finding suggests that focus groups may mask some individual differences in interpretation that may be revealed by other methods of data collection.
Unfortunately, few studies have reported a comparison of data obtained from focus groups with data from surveys. Literature searches primarily produced studies involving adult participants (Folch-Lyon, de la Macorra, & Schearer, 1981; Saint-Germain, Bassford, & Montano, 1993; Ward, Bertrand, & Brown, 1991). All of these studies found an overlap between the data produced using the two methods, even though the questions asked in focus groups typically involved open-ended questions, whereas the surveys more commonly involved fixed response options. Two further studies evaluated the use of different methods among young people (Mitchell & West, 1996; Sussman, Burton, Dent, Stacy, & Flay, 1991). Mitchell and West concluded that there was a greater similarity between the responses from the questionnaires and the focus groups than between the focus groups and the role-play.
Of greater interest would be the comparability of data from interviews and focus groups; however, Fern's (1982) research emerged as the only relevant publication from the literature search. Although Fern's comparison of data from focus groups and interviews would be of great concern to market researchers, it is of little value to those with an interest in pediatric psychology who are interested in a comparison of psychological insights afforded by focus groups and interviews. Morgan (1993) also noted the absence of such studies.
The issue of credibility (similar to aspects of reliability) is also addressed by very few of the studies reviewed. One exception is a study by Kidd, Townley, Cole, McKnight, and Piercy (1997), which explored the role of children in potentially dangerous farm chores. In their study, credibility was established by comparing information from focus groups about children's involvement in farm work with data on pediatric injuries on farms in the area. Another form of credibility checking employed by the authors involved presenting each focus group with themes that had emerged from previous groups for clarification. The reliability of the coding system was checked by comparing coding between raters on 10% of the data set. This provided a kappa coefficient of the agreement between raters. The thoroughness with which Kidd et al. (1997) approach the issue of credibility provides an excellent model for researchers who wish to apply the same level of rigor to their data.
Current Status and Future Development
From the foregoing discussion, it should be clear that focus groups can be used in a variety of ways in research with children. However, much research remains to be conducted on the group process itself. At present, “most of our knowledge about focus groups comes from personal experience rather than systematic investigation” (Morgan & Krueger, 1993, p. 3). This is true for adult and child participants alike. Currently, many guidelines exist regarding the conduct, utility, and outcomes of focus groups, yet there is limited scientific evidence to support these claims. For example, we know nothing of the differing outcomes of focus groups that last a number of sessions (e.g., Wayland & Rawlins, 1997) in comparison to the majority of groups, which take place on one occasion only. Nor do we have research on the value of warm-up activities designed to stimulate group participation (e.g., Beck et al., 1987; Hoppe et al., 1994).
Methodological studies involving large numbers of participants need to be designed to investigate the impact of systematic manipulation of focus group composition (e.g., group size, age of participants, gender) and the use of stimulus materials. Such an approach would allow us to identify the value of large versus small groups, mixed gender versus homogeneous groups, “ friendship” groups versus “stranger” groups and other group composition factors. The outcome of these groups could be evaluated according to the quality and quantity of relevant responses that emerged during the group interaction, ratings of the participants' comfort during the different scenarios, and variables relevant to understanding the underlying group dynamic (e.g., silences, spontaneity, number of interruptions). Only when such empirical evidence is available can we truly capitalize on the potential productivity of focus groups.
Another feature of the focus group method that warrants further investigation is the extent to which children enjoy their participation in such groups. Children's evaluations of the research method employed may have implications for their willingness to share information and to participate in future research. There is a need to investigate children's views of and preferences for different research methods in order to facilitate their participation and to ensure their experience is a positive one.
In addition to the need for more research on focus group methodology, there is also a real need for standardization of the reporting of focus group findings. A number of studies reviewed in this article failed to document information on the basic composition of their groups, including size, gender balance, and method of recruitment. Omissions of this nature would not be regarded as acceptable if the authors were reporting on quantitative methods, and there does not seem any justification for its omission when qualitative methods are used.
In the prevailing climate of child-centeredness and child inclusiveness, there is a need to examine research methods that will encourage and assist children to respond fully and in a manner conducive to their needs, wants, and desires. This must also be balanced against two other important considerations, one pertaining to the method that best fits the research question and the second to the needs and abilities of the researcher and the constraints within which the researcher must operate. The aim of this article is to highlight the potential of focus groups as an appropriate research tool in eliciting children's views and acknowledging children as experts. It highlights the potential benefits of focus groups both to the researcher and the participants and the diversity of purposes focus groups can fulfill. Focus groups are not the methodological remedy for all research problems; however, they are unique in terms of their group context, their ability to allow collection of a large amount of data in a short period of time, and their ultimate potential in eliciting children's views.
We acknowledge the receipt of funding from the Health Research Board, Ireland. We thank Adell Hickey for her assistance.