Positive contribution of focus groups to mixed methods research: An example for discussion

Positive contribution of focus groups to mixed methods research: An example for discussion

Review of Focusing on what counts: Using exploratory focus groups to enhance the development of an electronic survey in a mixed-methods research design by Galliott & Graham.

By Ryan Rominger, Ph.D., LCPC-PIT

Galliott, N., & Graham, L. J. (2016). Focusing on what counts: Using exploratory focus groups to enhance the development of an electronic survey in a mixed-methods research design. Australian Educational Researcher, 43, 567-585. Doi: 10.1007/s13384-016-0216-5

Introduction

Galliott and Graham (2016) used a sequential exploratory mixed method design to investigate career planning within adolescent populations in Australia. This post is an invitation to readers to explore the structure and use of the research method utilized by Galliott and Graham, and deepen the reader’s understanding of both mixed methods designs and the specific use of focus groups within such designs. You may access the article, once logged into the University of Phoenix system, by clicking on this link.

Examining this Mixed-Method Research Study

To start, let us clarify what ‘sequential exploratory mixed method design’ means. Creswell and Plano Clark (2011) note that a mixed methods design is one in which multiple types of research methods are combined within one study. For example, a study may combine a phase of quantitative data collection and analysis, and a separate phase of qualitative data collection and analysis. Alternatively, a study may contain a single phase wherein both quantitative and qualitative data are collected and analyzed (either separately or together). The important point here is that the methods are ‘mixed’ or exist together in some form within a single research endeavor.

The next construct to be aware of is the term ‘sequential.’ A sequential design is one in which the phases of data collection, and often analysis, occur separately and one after the other. The corollary is a design where both occur at the same time, which is called a concurrent design. In the case of Galliott and Graham’s study, the authors first use a set of focus groups (qualitative data collection) to gain a better understanding of the variables and thus create a better survey (quantitative data collection) to administer to participants during the second phase. As the data collection occurred separately, and one after the other, it was a sequential design.

The next term of import is ‘exploratory.’ Within Creswell and Plano Clark’s (2011) framework, there are two ways of using the combination of quantitative and qualitative data collection and analysis. One way is to use one set of data to explain another set of data. This, the ‘explanatory’ type, may use qualitative data (such as interviews with participants) to illuminate details and qualia regarding a set of initial quantitative data (numerical data reported or collected by groups of individuals). When group characteristics are not clear, or more information is needed regarding the subjective experience (after quantitative data has been collected) then an explanatory model will be assumed. However, if qualitative data is collected first and is used to explore in more depth sets of variables or constructs prior to collecting quantitative data, then an ‘exploratory’ method has been used. The collection of qualitative data and analysis is used to explore the topic, before launching into the second phase of the study. In the case of Galliott and Graham, the authors used the qualitative focus groups to explore the variables, and then inform creation of a survey to distribute to a larger set of participants.

And now we have come full circle, back to the focus groups. Knowing the meaning of the mixed method design, a reader may understand the utility of the specific practice of using a focus group. Focus groups are a group of participants who share some similarity or experience with the topic at hand, and who may be able to provide information to researchers about the phenomenon. The participants are brought together and interviewed as a group, rather than as individuals, so that participants may talk among themselves, trigger awareness within each other, discuss similarities and differences, and feel more comfortable with the presence of other participants/peers. In the present study, Galliott and Graham (2016) brought together groups of adolescents who had engaged in career preparation within high schools. The groups were able to discuss, together, the questions posed by the researchers, thus providing in-depth information which was fuller and more detailed than singular interviews could provide. Indeed, the authors wrote:

The aim of the focus group phase was to pilot questions drawn from the literature, and to better understand what was and was not developmentally appropriate. As the research questions were not sensitive and unlikely to cause embarrassment, focus groups provided an ideal method to collect data that could then be analysed [sic] and the results used to tailor an instrument for use in a large-scale cross-sectional survey. (Galliott & Graham, 2016, p. 570)

Delving more deeply into the article itself, there are numerous examples of how use of the focus group provided valuable information in the research endeavor. First, the researchers noted that information from the focus group helped identify subgroups (students with no career plan versus students who had a plan but no details of how to execute the plan; Galliott & Graham, 2016, p. 573). When working with participants, it is vital to identify potential subgroups, as these groups may respond differently to survey questions. Also, identifying subgroups within the focus group data allowed researchers to build and ask survey questions to help identify group members, and thus conduct analysis within and between subgroups.

Second, focus group responses revealed a complexity to which the researchers had not been privy. This complexity, once identified, was integrated into the survey through “multi-stranded questions” (p. 574) designed to flesh-out the complex components of the variable. Third, the authors write:

Our participants also indicated that some students may temper their future aspirations according to their perceived academic ability and self-efficacy, a trait that has been noted in recent Australian research (Hawkins 2014). This finding prompted us to include two items designed to tap into student perceptions of their academic abilities relative to others in their year group, and their own self-efficacy and problem-solving abilities. (p. 575)

Thus, focus group data revealed alignment with more recent research, and thus the researchers included two additional questions in the survey to target this information. These two questions allowed researchers to identify subjective and objective components of career choice capabilities, which then allowed for statistical comparison within and across groups particularly as this component compares to other variables (prior achievement, self-efficacy, and career certainty; Galliott & Graham, 2016, p. 575).

Fourth, information from the focus group informed researcher understanding of the difference between schools. This difference led to creation of survey questions based on skip-logic (i.e., if yes, then go to question 12, if no, go to question 14). Additionally, the researchers identified students with high motivation to engage in career preparation activities and students with lower motivation, in addition to differences in presence of types of preparation activities depending on the location of the school and economic standing of the school. Further, feedback in the focus groups raised awareness that some of the programs, even in lower class schools, were highly competitive and not accessible by a majority of students (Galliott & Graham, 2016).

Thus, it is clear that the focus group played a significant role in both researcher understanding and formation of the larger survey which was distributed during the second phase of the study. The authors name four specific ways that the focus group helped their study:

  1. Firstly, interaction with representatives of the target group enabled us to explore the career-related ideas of high school students in different socioeconomic regions and various school types (including government and non-government schools).
  2. Secondly, focus groups assisted in familiarizing [sic] the researchers with the language used by the students in the study.
  3. Thirdly, focus groups helped us to partly fill the gaps in the research literature with regard to the variety of educational and life experiences across a range of students, highlighting the need for us to sensitively frame demographic questions.
  4. Finally, focus groups enabled us to pilot some of the planned survey questions and to adjust individual questions based on students’ ease of interpretation. (Galliott & Graham, 2016, p. 581)

One critique of the article, however, is that the authors could have been clearer when first introducing the structure of the method. Creswell and Plano Clark (2011) suggest that a method be clearly articulated when first introducing the method. Thus, the authors could have improved the article in two ways. First, the authors could have described, as I did above, at the beginning of the method section that the study was a sequential exploratory mixed method design. The authors did mention this in the Abstract, but the information should have been carried into the main discussion in the Methods section of their article. Second, a reader may have benefited from a graphical representation of the method itself. This is particularly effective for readers who are more visual learners. It is also a common practice when clarifying mixed methods designs. In this case, a simplified version of the graphical representation would have been qual->QUANT, as the qualitative focus group came in the first research phase, and the quantitative survey came in the second phase. Additionally, the QUANT, representing the quantitative research phase, is all capitals because this was the primary focus of the study. The qual (qualitative) phase was meant to support the quant phase. Unfortunately, the weighting of the quant and qual must be implied based on the overarching study; it was not explicitly stated by the authors.

A larger graphical representation may include the separate focus groups (as their own boxes) in the qual phase, and the survey (as its own box) in the quant phase, with specific indicators of where the analysis of each data set occurred. In this case, the data analysis occurred right after collection, and thus it would be represented as (qual [qual analysis] à QUANT [quant analysis]). This graphical representation further acknowledges how the analysis combined to inform the overall study. In some mixed methods research, the researchers must be explicit about how the strains of research (quant and qual) are analyzed in respect to each other, as analysis can both bias and inform following analysis. In some mixed methods, the data are analyzed together, at the end of the study with pre-designed methodological practices (based on the ontology and epistemology of the researchers).

For further explanation of mixed methods designs, please watch my video on mixed methods designs;  or, for a more in-depth discussion of weighting of qualitative and quantitative, mapping the design, and additional typologies, see this video.

What Next?

My hope is that through analysis of this article, you have become better informed about conducting mixed methods research. Specifically, my hope is that you, the reader, are better able to assess the utility of using a focus group for mixed method research, and thus determine if a focus group would facilitate a stronger research design for your own studies. Please use the discussion herein this blog area to engage in further discussion of the article and methods by commenting below. I look forward to our discussion. 

References and More

Creswell, J. W., & Plano Clark, V. L. (2011). Designing and conducting mixed methods research (2nd ed.). Los Angeles, CA: Sage.

Galliott, N., & Graham, L. J. (2016). Focusing on what counts: Using exploratory focus groups to enhance the development of an electronic survey in a mixed-methods research design. Australian Educational Researcher, 43, 567-585. Doi: 10.1007/s13384-016-0216-5

If you are interested in mixed-methods, I invite you to join our Research Methodology Group.