2   When to Use Focus Groups
In Chapter 1, we learned about three characteristics of focus groups that, when taken together, give the method certain unique advantages. Focus groups are social in form. They create data through emic processes. Data can be generated at three different levels of analysis: the individual, the group, and interactive levels.
In practice, these traits help make focus groups uniquely suited for understanding group processes and dynamics. They can also enable a researcher to understand how individuals think or act in social settings. Either way, it is important to underscore that the social setting of the focus group cannot be set aside. Even when researchers analyze data at the individual level, they must keep in mind that individual responses are likely shaped by the group environment. Individuals will be influenced by the responses of others; they will also be influenced by the simple fact of speaking in front of a group.
The three characteristics of focus groups also have the effect of empowering their participants. The researcher puts forth a set of questions for participants to consider. Ultimately, however, the participants decide what they want to address and if they wish to take the conversation in a different direction. Having acknowledged specific focus group characteristics and certain advantages that these bring, we can discern when focus groups are useful for the researcher. Specifically, we can identify the kinds of questions that focus groups can help answer. We can also, I should note, detect when focus groups may not be appropriate.
This chapter is dedicated to these tasks. It first considers the kinds of questions that focus groups can effectively address. It then offers multiple alternatives for how to address them: either as a stand-alone method, in conjunction with other qualitative methods; or in conjunction with quantitative methods. Finally, the chapter highlights certain challenges that can arise when using focus groups. These challenges are important to keep in mind, because they allow us to identify situations where focus groups would not be appropriate. By the end of the chapter, the researcher should have the tools to decide if and when to use focus groups in their research design.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
Research Topics That Are Focus Group Friendly
19
Research Topics That Are Focus Group Friendly
It is not particularly instructive for you, nor feasible for me, to produce an exhaustive list of questions that focus groups can effectively address. It is possible, however, to provide an expansive list - one that can provide the researcher with a good foundation for assessing whether her specific topic could be addressed using focus groups. It may not surprise the reader to know that this list is based on what we have already learned thus far about the data collection method. And, in fact, much of what we will discuss more formally below was first mentioned in Chapter 1.
First and foremost, the list of questions that follows takes into account the two focus group advantages that we have highlighted: focus groups are suited for understanding group dynamics and processes, and they can be empowering for their participants. It also takes into account the three focus group characteristics that help define those advantages. What follows, therefore, is a list of six different types of questions that focus groups can help to address. Again, while this is not an exhaustive list, it should serve as a blueprint for the kinds of questions that are amenable to focus group-based work.
Table 2.1 provides this list. It explains that focus groups are useful for: measuring socially constructed phenomena; understanding how groups think, come to decisions, and/or process information; understanding how individuals think and act within social settings; facilitating conversations on difficult or sensitive topics and/or with vulnerable groups; contextualizing concepts; and exploring a new topic of interest. Let us examine each topic in turn.
Measuring Socially Constructed Phenomena
First things first: What is a socially constructed phenomenon? We saw in Chapter 1 that partisan identities emerge through social processes. That is,
Table 2.1 Focus Groups Can Be a Useful Data Collection Method If...
A phenomenon under investigation is socially constructed; The researcher wishes to understand group processes;
The researcher wants to know how individuals think and act in social settings; The research topic is sensitive or taboo in and/or is associated with vulnerable groups; The researcher needs to ensure that concepts are properly contextualized both within
and across research settings; Little is known about the topic of interest.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
20      I When to Use Focus Groups
we learn what it means to be, for example, a Democrat by observing how other Democrats speak and present themselves to the world. When individuals feel affinities with the beliefs and/or actions of Democrats with whom they interact or observe, they may choose to identify as Democrat. Partisan identity, like all group identities, is forged through social processes. Its meaning is, like many phenomena or concepts, socially constructed (see, e.g., Abdelal et al. 2009).
Socially constructed phenomena are those things that we learn about or understand thanks to our interactions with others. They are concepts created by humans and, often, serve to impose some sort of order on our social world (Yanow 2015, vii). How do we, as individuals, think about race, gender, or culture? What about power? Corruption? These are all examples of socially constructed phenomena. We know this for (at least) two reasons. First, their meaning is highly contextualized. What is perceived as an act of corruption in the United States differs from one in China or Saudi Arabia (see, e.g., Andersson and Heywood 2009, 749-750; Chadda 2004; Brown and Cloke 2004). Second, their meaning can change over time (Yanow 2015, vii). What it means to be a Republican can evolve as different candidates emerge to represent the party in national or subnational positions. President Donald Trump, a Republican, is different from, say, Ronald Reagan or John McCain in terms of his rhetoric and governing style. If more and more "Donald Trumps" emerge to represent the Republican party, our understanding of what a Republican is, in general, might change.
Because socially constructed phenomena are highly contextual and variable in their meaning, it makes sense to study them with a data collection method that can capture context and nuance. Focus groups are eminently suited for this task. Participants are allowed to speak freely about the topic in question. The conversation may reveal disagreement about a concept's meaning, contradictions over that meaning, and whether, ultimately, consensus can be forged. Socially constructed phenomena are complex by nature; focus groups allow the researcher to capture that complexity. They also tend to require high-effort cognitive thought (Chaiken 1980; Krosnick 1991; Tourangeau 1984). Socially constructed phenomena are not easy concepts to grasp. Because of this, individuals tend to look for easy shortcuts (called satisficing) when dealing with difficult ideas on their own (Krosnick 1991; Krosnick et al. 1996). The social form of focus groups allows multiple individuals to share the burden of working through complicated ideas. Consequently, they can be more suitable than one-on-one interviews or surveys for addressing these kinds of concepts.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
Research Topics That Are Focus Group Friendly
21
Understanding Group Processes
It perhaps comes as no surprise that focus groups - a group-based data collection method - are highly effective at studying group processes, including how groups think, process information, come to decisions, prioritize, and construct meaning. Focus groups are useful for studying groups who share a common identity or shared norms and goals (Merton 1987, 555). We can think here of Paluck and Green (2009), who undertook focus groups with genocide survivors, the Twa people (a Pygmy minority), prisoners in Rwanda, and members of eight different communities. In each set of focus groups, the participants shared a common identity: they had survived the genocide, belonged to an ethnic group or community, or were prisoners. The researchers wanted to see if each group responded differently to a radio program that challenged norms of deference and legitimized dissent. The groups themselves, in other words, were the subjects of study. Focus groups were one vehicle through which the researchers could access each group.
Focus groups can also consist of more heterogeneous groups. In my work on political parties in Peru, Bolivia, and Venezuela, I measure whether citizens attribute certain traits or ideas to electorally weak parties (Cyr 2017b). I hypothesize that parties that retain a broadly recognized identity or "brand" (Lupu 2013) are more likely to survive a period of poor electoral performance than those that do not. I undertook four to five focus groups in each country and asked participants to discuss each electorally weak party. The only requirement for participation was that the individuals be of voting age for the election when the parties were voted out of national government. These were extremely heterogeneous groups. My goal was not to get a sample of any particular sector of society. Instead, I wanted to show that everyday citizens could (or could not) speak meaningfully about the parties in question. Where this occurred across multiple groups, then I had evidence of a persistent party brand.
Overall, when researchers wish to understand group dynamics and processes, a focus group is an appropriate data collection method to use.
Understanding How Individuals Act in Social Settings
Individuals tend to act differently in a group than they do on their own. We can thank longstanding social science research for this finding (see, e.g., Allport 1920; 1924). Sometimes, a researcher may want to exploit the effect of the group dynamic on how individuals act, think, or speak. In this situation, the researcher privileges the individual as the unit of interest in the focus group, but she is actively engaging with its social form.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
22      I When to Use Focus Groups
Let us return to the example of Paluck and Green (2009) to understand how this can work in practice. The researchers wanted to see if community members in post-genocide Rwanda were more willing to express dissent in private versus public settings. To test for this, they compared the levels of dissent expressed by individuals in one-on-one interviews versus in focus groups. They found that certain subsets of individuals were more likely to express dissent in private. Here, the researchers exploited the social setting of the focus group to measure individual behavior. Theirs is a prime example of using focus groups to understand how individuals act in social settings.
Facilitating Conversations on Sensitive Topics or with Vulnerable Groups
Focus groups represent a space where individuals with similar backgrounds or experiences can share and communicate with each other. While this can allow a researcher to broach any number of topics particular to a specific group or collectivity, focus groups become particularly useful when discussing a sensitive or difficult topic or when working with vulnerable groups.1 By sensitive topic we mean any topic that might be "intimate, discreditable or incriminating" in nature (Renzetti and Lee 1993, ix, as taken from Liamputtong 2011, 108). By vulnerable groups, I refer to those that are potentially subject to discrimination, including (but not limited to) children, the elderly, ethnic minorities, immigrants, the homeless, and the LGBTQ community (see Liamputtong 2011, 109).
Why are focus groups especially useful when undertaking research on sensitive topics or with vulnerable groups? Again, the social setting of the focus group is key. Individuals may be reluctant to discuss sensitive topics, such as a traumatic event that they may have experienced, in a one-on-one interview with a researcher. When an individual knows that she is in a focus group with others like her, however, some of the pressures or discomfort associated with the topic can be shared and, therefore, diluted. In this sense, focus groups are a "forum for mutual support" (Kroll et al. 2007, 697). Liamputtong (2011) helpfully makes this point:
Focus groups create layers of communication and, therefore, provide respondents with a safe environment where they can articulate their experiences, opinions and beliefs in the company of people who share similar experiences and hold similar beliefs (p. 110).
When it comes to sensitive topics or vulnerable groups, the influence of the focus group's social setting can actually be positive for participants, allowing
1 Liamputtong (2011) addresses these topics in great detail and provides multiple examples.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
Research Topics That Are Focus Group Friendly
23
them to speak on matters that they might not otherwise be comfortable addressing (Kitzinger 1995, 111).
Contextualizing Topics
As we will see below, focus groups have regularly been used as a pre-test for large-N, and especially survey-based, research. With focus groups a researcher can: refine question phrasing; revise close-ended questions to include the full range of responses; and ensure that all dimensions of a particular topic are covered in the survey. Focus groups are useful for these tasks because of the emic process through which data are generated. A researcher can measure how individuals in a surveyed area talk about the topic of interest and therefore feel confident that the survey protocol reflects the linguistic idiosyncrasies of the research site. Focus groups, in other words, are useful for contextualization, and they help a researcher achieve greater measurement validity in their survey protocol (Cyr 2017b; see also Morgan 1997).
For example, a group of sociologists wanted to understand how racial attitudes affected the neighborhood preferences of white and African American families in the United States (Krysan et al. 2009). They conducted an experiment in which hundreds of individuals in Chicago and Detroit were randomly assigned to watch one of several videos that showed one neighborhood but varied the racial and class composition of its residents. To ensure that the videos conveyed proper racial and class cues, the researchers first pretested the videos in focus groups. Once the focus groups confirmed the intended cues, the researchers were more confident that their videos accurately portrayed race and class in the two populations (Chicago and Detroit) of interest (Krysan et al. 2009, 537-538).
This example demonstrates the utility of focus groups for ensuring that survey and/or experimental protocols are properly contextualized for their intended audience. If the focus groups participants had not interpreted the videos properly, then the data gleaned would have instructed researchers on how to improve the video's content so that it validly conveyed race and class. Additionally, the focus groups in Detroit might have interpreted the videos differently from those in Chicago. In this case, the focus groups would have allowed the researchers to ensure cross-sectional comparability, and one or both videos would have been revised accordingly. Focus groups can help researchers contextualize both within and across research sites.
Although this example has shown that focus groups can be useful as a pretest for larger-N studies, it is important to keep in mind that the context they provide can be useful in other ways. Focus groups can be used to contextualize or ground unexpected findings from a larger survey or experiment. In other
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
24      I When to Use Focus Groups
words, they can be a useful follow-up orptwf-test (Merton and Kendall 1946). Focus groups can also provide useful context prior to undertaking qualitative fieldwork. They can, for example, introduce a researcher to the idiosyncrasies and dynamics of a new field site, providing useful tips on how to undertake subsequent participant observation and/or in-depth, individual interviews (Morgan 1997, 23-24).
Exploring a New Topic of Interest
Finally, focus groups can be useful for researchers embarking on a new research agenda, especially when little is known about it (Kidd and Parshall 2000). This is because the researcher can rely upon focus group participants to do the heavy lifting with respect to revealing what might be important or salient about a particular research question. Focus groups are empowering for participants precisely because they have control of what is said. In this case, researchers can use that empowerment in their favor: In letting participants take the reins, researchers can pinpoint or isolate what might be noteworthy about a new topic of interest.
We see this happening in a project on youth drug use. Although the authors were not new to this research agenda, they were new to focus groups. They decided to organize a focus group of adolescent LSD users to see, among other things, if any new information would be produced (Agar and MacDonald 1995). Although skeptical that focus groups could generate quality data (they intuited that "a few hours with a few groups guarantees only that the 'quality' in 'qualitative' will go the way of fast food" [!]), the researchers discovered something new in the focus group (p. 78) - namely, that many teenagers used the over-the-counter drug, Robitussin, as a substitute for LSD. This new information emerged thanks to group interactions - one characteristic of focus groups that makes them "shine" (Agar and MacDonald 1995, 80). The authors concluded that group interactions allowed for insights that might not (and, in their case, had not) emerge via other qualitative methods, such as interviews and ethnography. It is for this reason that researchers use focus groups for idea or hypothesis generation (Fern 1982).
One final note about the focus group that Agar and MacDonald (1995) organized. It consisted entirely of LSD-using youths who were seeking treatment for their addiction. This was a homogenous group of kids with a common, shared experience. The researchers explicitly sought to tap into that group dynamic - something they were hard-pressed to do with other qualitative methods. Additionally, the researchers were collecting data on a vulnerable and hard-to-reach group: young, illegal drug users. In one
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
How to Use Focus Groups I 25
research design, therefore, multiple topics for which focus groups are appropriate overlapped. This is not unusual, and the set of six topics addressed here should not be treated as mutually exclusive. Instead, they mutually reinforce each other and, where they overlap, make the use of focus groups even more relevant.
How to Use Focus Groups
We have identified key topics that focus groups can be useful for addressing. We know, in other words, when to use focus groups. We must also consider how to use focus groups within a research design. On this point, there are two initial alternatives. A researcher may use focus groups as a stand-alone method. She may also use them in conjunction with other data collection methods. Should a researcher wish to design a mixed (i.e., multiple) methods research project, she will then have to decide how to use focus groups with other qualitative or quantitative methods.
To be sure, it is not feasible to explain how to use focus groups in every potential mixed-methods combination. Instead, the following pages help the researcher understand the advantages that focus groups can provide vis-a-vis other qualitative methods and in concert with quantitative methods. It also spends some time considering two different approaches to mixing methods in general: triangulation and integration. By the end of the section, a researcher should know how her particular project might effectively include focus groups.
Before we examine in detail the multiple approaches to using focus groups in a research design, it is useful first to consider how work has used focus groups. What we see is that a large majority of publications use focus groups in combination with other types of methods and most typically with in-depth interviews or as pre-tests to surveys. The use of focus groups as survey pre-tests has a long trajectory in the social sciences (Copsey 2008; Morgan 1996; Stewart and Shamdasani 1990). This was true in the 1990s (Morgan 1996, 133). It is also true in the early twenty-first century (Cyr 2016).
Indeed, in their original formulation of focus groups as a data collection method, Merton and Kendall (1946) saw them as supplementing, not supplanting, quantitative methods. They highlighted the role focus groups played in revealing the mechanisms behind the correlations that statistical methods uncovered (p. 543). Still, focus groups can also be used - and, certainly, they have been used - as a stand-alone method in the social sciences. In all, the most important thing to keep in mind is that focus groups, like all data collection
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002
26      I When to Use Focus Groups
methods, should be used in ways that are consistent with the objectives and purpose of the research project. This can mean using them as a stand-alone method, as we will consider immediately below. It can also mean, however, using them in conjunction with other methods.
Focus Groups on Their Own
Although less common in practice, focus groups can be used as a stand-alone data collection method. Perhaps the most important thing to keep in mind when undertaking a study that relies exclusively on focus groups is that the results gleaned must comprise a "sufficient body of knowledge" for the research question at hand (Morgan 1997, 21). When does one attain this kind of sufficiency from focus groups? A person whose research wishes to understand the particular dynamics of a specific group - and in particular, how that group sees itself and the world - can use focus groups (Morgan 1997, 20-21). They also serve as a stand-alone method when the experiences and perspectives sought out by the researcher are only accessible via group interaction (Liamputtong 2011, 88).
For example, Kitzinger (1994) used multiple focus groups to examine the effect of media messages about AIDS and, specifically, how each group reacted to the messages and why they reacted as they did. Focus groups were the preferred data collection method, because the researcher wanted to explore the "social context of public understanding" (Kitzinger 1994, 104). In this example, the focus group was uniquely suited for addressing the research question at hand, since the method is inherently social.
Finally, focus groups are often used in the early stages of a broader research agenda. They are useful, in other words, for undertaking an exploration of a topic about which you (the researcher) and your field know very little. For example, Prabhakar (2012) uses focus groups as a first step toward understanding public attitudes about taxation in the United Kingdom. He finds that those attitudes changed as a function of whether tax policies were viewed separately or as a group. A next phase of his project, he concludes, would use surveys to see if these findings are broadly representative (Prabhakar 2012, 87).
When used for exploratory purposes, two additional points on focus groups as a stand-alone method bear mentioning. First, as evidenced by Prabhakar (2012), this kind of exploratory work is publishable! Second, it typically represents the first of what will probably be multiple phases of a research project. In that sense, while focus groups are, strictly speaking, a self-contained method when it comes to exploratory work, their use is often oriented toward a larger project incorporating multiple methods.
Downloaded from https://www.cambridge.org/core. Pontificia Universidad Catolica de Chile, on 03 Aug 2020 at 23:12:53, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316987124.002