THE UNIQUE FEATURES OF FOCUS GROUPS

A focus group is a data-collection method that brings together a small group of individuals (i.e., six to eight people) to discuss a series of open-ended questions. A moderator asks the questions and facilitates the ensuing discussion. ^{Footnote 1} The goal is to understand how people talk about topics of interest. Data are generated through the conversations that unfold.

Focus groups are a unique method of data collection for three interrelated reasons. First, they are inherently social in nature. Data are generated through the conversations and interactions that take place among participants. As a part of this process, individuals can react to one another’s comments. They may refine their ideas or build off of what others say. The social dynamic of focus groups provides a synergistic quality. It also enables the researcher to capture nuance and tensions surrounding potentially contested or highly contextualized subjects.

Second, focus groups generate emic data—that is, data that emerge with little interference or guidance from the researcher or the research setting (Krippendorf Reference Krippendorf1980). Although a set of questions guides focus-group conversations, participants determine how those conversations unfold. The open nature allows for spontaneity and a free discussion on specific topics (Onwuegbuzie et al. Reference Onwuegbuzie, Dickinson, Leech and Zoran2009, 2).

Third, focus groups produce data at three levels of analysis (Cyr Reference Cyr2016). Researchers can focus on discrete, individual responses to questions, collecting data at the individual level of analysis. This typically involves assessing participants’ initial responses to questions. They can assess the extent of consensus within the group, collecting data at the group level of analysis. Here, researchers focus on how conversations for each question end. What does the group conclude about a certain topic? Is there agreement? How much agreement? Finally, researchers may want to explore specific exchanges that take place between individuals, which might spark new ideas or hypotheses that the researcher can explore later. This represents the interactive level of analysis.

Taken together, these three characteristics—their social nature, the emic data produced, and the three levels of analysis that can be accessed—endow focus groups with certain strengths that make them useful in mixed-methods research designs. The focus group’s social form allows a researcher to study group processes and dynamics (Stewart, Shamdasani, and Rook Reference Stewart, Shamdasani, Rook, Bickman and Rog2009). However, it also provides an environment in which multiple individuals can grapple with complex or contested subject matter. With emic data, the voice of the subjects—rather than that of the researcher—is privileged. These two features allow the researcher to capture greater nuance, which can be vital for measurement validity (Adcock and Collier Reference Adcock and Collier2001). Finally, the distinct types of data that focus groups generate allow researchers considerable flexibility regarding how they use the data produced.

Certainly, the features of focus groups that enable researchers to understand social- and group-level dynamics also present certain challenges. Specifically, two weaknesses merit consideration. ^{Footnote 2} First, focus groups typically come together using a type of nonprobability sampling. It therefore is difficult (although not theoretically impossible; see Fern Reference Fern2001) for them to be representative of a population. Moreover, focus groups produce “situated accounts” in which participants are asked to consider questions in an artificial setting (Sim Reference Sim1998, 349). It therefore is possible that participant responses do not reflect what individuals might actually say in a “real” social setting. ^{Footnote 3}

Second, focus groups may generate certain pressures or forces that can affect how participants respond. Individuals may feel censored as a result of particular group dynamics. They may feel pressure to conform. This element of “groupthink” can suppress participant opinions, which biases the data collected (Carey and Smith Reference Carey and Smith1994, 124). Additionally, the dynamics and composition of each group are inevitably unique. An individual might not make the same statement in a focus group composed of different individuals or affected by different dynamics.

It is possible, of course, to mitigate these problems. A moderator can (and should) work to create an open, permissive atmosphere. Participants should feel that they can share their point of view (Morgan and Krueger Reference Morgan, Krueger and Morgan1993, 8–9). A researcher also may ask participants to record their answers before voicing them. This allows the researcher to measure whether participants’ spoken answers diverge from what they originally wrote.

Crucially, focus-group proponents note that the social pressures that permeate a focus-group setting are, in fact, broadly reflective of actual processes of opinion formation and expression (Schutt Reference Schutt2011, 309). Compared with other methods, focus groups more closely reflect daily conversations (Gamson Reference Gamson1992). Because of this, for certain questions—especially those that tap into group dynamics and processes—groupthink can inform rather than obscure the data (Hollander Reference Hollander2004, 611), giving the method more external validity than, for example, one-on-one interviews in asocial settings (Albrecht, Johnson, and Walther Reference Albrecht, Gerianne Johnson, Walther and Morgan1993).

Ultimately, as with any data-collection method, researchers must carefully justify their use of focus groups and address their weaknesses. This may mean more carefully delimiting the scope of one’s findings. It also may mean crafting a research design that highlights the strengths of focus groups while deemphasizing their weaknesses. In this sense, focus groups are particularly well suited for mixed-methods research.

FOCUS GROUPS AND MIXED-METHODS RESEARCH

Focus groups often are used alongside other methods. For example, they can be combined with other qualitative methods, most typically with one-on-one interviews (Morgan Reference Morgan1996, 134). Focus groups allow researchers to access diverse perspectives on multiple aspects of a topic. An interview that thoroughly examines one’s views on a topic is unlikely to achieve that type of broad coverage.

Focus groups also allow a researcher to observe and analyze interactions on a particular question—a feature that is lacking in an individual interview (Duggleby Reference Duggleby2005, 832). They can address the intersubjectivity of certain concepts and “draw out [the] shared life experiences” of participants (Fern Reference Fern2001). McDonnell (Reference McDonnell2010), for example, used interviews and focus groups to speak with Ghanaians about a series of AIDS billboards displayed in Accra. He used in-depth interviews to assess individual opinions about specific billboards. Focus groups allowed participants to enter into a more comparative debate about the effectiveness and impact of each (McDonnell Reference McDonnell2010, 1811). Here, the depth achieved through interviews was complemented by the breadth and complexity of focus groups.

Mixing qualitative methods is only one approach to integrating focus groups into a mixed-methods research design. The remainder of this article focuses on combining focus groups with large-N studies, especially those that address the microfoundations of politics, including surveys, experiments, and survey–experiments. ^{Footnote 4} A recent study found that in the top political science and sociology journals, focus groups were twice as likely to be combined with these as with qualitative methods. ^{Footnote 5} Given their prevalence—and the general upward trend in integrating qualitative and quantitative methods (Seawright Reference Seawright2016)—it makes sense to delineate how focus groups work with large-N studies.

I suggest that focus groups are useful in at least two ways when used with large-N work: they (1) construct valid instruments; and (2) explain the correlations that statistical work uncovers. In both cases, focus groups are useful because they offset weaknesses of large-N studies.

IN CONCLUSION, OR WHY THE RESEARCH QUESTION (STILL) MATTERS

Focus groups are a useful complement to other data-collection methods. This article suggests that focus groups have characteristics that make them particularly useful for certain mixed-methods research designs. When used in conjunction with interviews, the social setting of a focus group allows a researcher to achieve greater breadth and range on a set of issues. In combination with large-N work, the emic data they generate provide depth and nuance. Their social form helps participants to work through more complex or intersubjective concepts and questions. Consequently, they can be useful for achieving greater measurement validity and for understanding why individuals think as they do. In any mixed-methods research design, the several types of data generated by focus groups give researchers flexibility and potentially open up multiple avenues for exploration.

To be sure, the utility of focus groups when mixing methods is conditioned by the research we want to undertake. This article suggests that focus groups are particularly useful, for example, when scholars study complex, intersubjective phenomena or when they want to understand individual or group attitudes and behaviors. As always, the research question matters. On this point, little has changed, despite the growing tendency to engage in multimethod research.