The posting below looks at designing research studies that make use of both qualitative and quantitative data. It is from Chapter 8, Mixed-Methods Research, in the book, Introduction to Educational Research, by Craig A. Mertler, Arizona State University, and Mertler Educational Consulting . Published by Corwin, A SAGE Company, 2455 Teller Road, Thousand Oaks, California 91320, (800) 233-9936, Fax: (800) 417- 2466, www.sagepublishing.com Copyright © 2016 by Sage Publications, Inc. All rights reserved. Reprinted with permission.
UP NEXT: Opposite Day: Counterintuitive Advice from Chairs to New Faculty for Professional Growth
---------- 2,550 words ----------
This chapter deals with mixed-methods research, a group of approaches to conducting educational research studies that combines both quantitative and qualitative data. While that description may seem somewhat basic and straightforward, there are many important aspects to consider. In this chapter, we will look at the characteristics of mixed-methods research, along with various designs and other important decisions to be made during the process of conducting mixed-methods research studies.
Characteristics of Mixed-Methods Research
The major characteristic of mixed-methods research is that it combines quantitative and qualitative approaches by including both quantitative and qualitative data in a single research study (Gay, Mills, & Airasian, 2009). Creswell and Plano Clark (2011) define mixed-methods research as those studies that include at least one quantitative strand and one qualitative strand. A strand is a component of a study that encompasses the basic process of conducting quantitative or qualitative research: posing a research question, collecting and analyzing data, and interpreting the results. However, these are merely “surface-level” descriptions of the characteristics of mixed-methods research. We will look more closely at specific characteristics in a moment.
Before examining various characteristics of this approach to conducting research, it is important to understand when and how mixed-methods research began. Creswell and Plano Clark (2011) date the beginnings of mixed-methods research back to the mid- to late 1980s. Methodology experts and writers from all around the world seemed to have been simultaneously working on similar ideas regarding the combination of quantitative and qualitative methods. Up to this point in time, many qualitative researchers and quantitative researchers did not see the legitimacy in the other approach to doing research. However, members of both research camps began to realize, on a deeper level, the value of the alternate approach. For example, quantitative researchers began to see that qualitative data could play an important role in quantitative research; similarly, qualitative researchers began to see that reporting only qualitative views of the world – and of a few individuals – would not permit generalization of the findings to many other individuals and audiences (Creswell & Plano Clark, 2011). Over the past decade or more, interest in the use of mixed-methods research as a means for studying educational topics and phenomenon has grown substantially.
Between the late 1980s and today, definitions and descriptions of mixed-methods research have shifted and morphed, and they continue to do so. While having a singular definition is desirable for many researchers, Creswell and Plano Clark (2011) have instead offered a definition of core characteristics of mixed-methods research. They suggest that their core characteristics provide a broader definition of mixed-methods research, since they combine methods, philosophies, and a research design orientation. These characteristics also highlight the key components that should be considered when designing and conducting a mixed-methods study. These six core characteristics focus on activities of the mixed-methods researcher and include the following actions:
- Collecting and analyzing persuasively and rigorously both qualitative and quantitative data, based on research questions [emphasis added]
- Mixing – or integrating or linking – the two forms of data either concurrently by combining or merging them, sequentially by having one build on the other, or embedding one within the other
- Giving priority to one or to both forms of data, again based on the research questions and the emphasis of the research [emphasis added]
- Using these procedures in a single research study or in multiple phases of a program of research
- Framing these procedures within philosophical worldviews and theoretical lenses
- Combining the procedures into specific research designs that direct the plan for conducting the study (p. 5)
These core characteristics provide an extremely comprehensive perspective on the critical aspects of engaging in mixed-methods research. You will see them integrated into our discussions as we proceed through this chapter.
Not unlike any other approach to conducting research, when preparing a research study that will use a mixed methodology, the researcher must provide a justification for the use of this approach. Researchers would need to do this even if they were engaged in a study that was purely qualitative or purely quantitative. There are specific situations that would more likely warrant a research approach that capitalizes on the combination of quantitative and qualitative data. Creswell and Plano Clark (2011) have described six scenarios or examples of research problems that are best suited for mixed-methods research:
- A need exists because one data source may be insufficient. As you know, qualitative data provide understanding through greater depth, whereas quantitative data provide broader, more general understanding. Each approach has its advantages and limitations. Qualitative data may provide a deep examination of a phenomenon of interest but only with respect to a handful of participants. On the other hand, quantitative data can provide information across a much broader sampling of participants, but the depth of that information is certainly limited. Depending on the goals of a research study – as well as its guiding research questions – one type of data alone may not tell the complete picture or adequately answer the research questions. Additionally, the results from the analysis of qualitative data and those from the collection of quantitative data may be contradictory, which could not have been discovered if only one type or the other was collected and analyzed. Using both types of data in a single research study provides depth as well as breadth.
- A need exists to explain initial results. Sometimes researchers find themselves in situations where the results of the study do not provide complete understanding of the research problem; further explanation is needed. This additional explanation can be provided through the collection and analysis of a second set of data that helps explain the results of the initial set of data. For example, quantitative data can be used to provide numerical expressions of the relationships among variables or differences between groups, but detailed understanding of what those relationships mean or from where the differences came (i.e., the meanings behind the results of the statistical tests) can be provided only by qualitative data collection and analysis, as a follow-up to the initial collection of quantitative data.
- A need exists to generalize exploratory findings. As you know from your studies of qualitative research methods, in some research investigations entered into by researchers, the research questions are not known, the variables cannot yet be identified, and the goals of the research cannot be specified at the outset of the study. In these scenarios, an initial phase focused on the collection of qualitative data is necessary simply to explore the setting or participants involved. Once there is enhanced general knowledge of the research situation, the qualitative phase can be followed up with a quantitative study to generalize and test what was learned from the initial exploration.
- A need exists to enhance a study with a second method. In some research situations, a second method can be added to provide enhanced understanding of some phase of research that has been conducted. For example, a researcher could add a qualitative component to enhance an experimental, correlational, or causal-comparative study. Similarly, quantitative data could be added to enhance the findings of an ethnographic, narrative, or grounded theory research study. In these situations, however, the second method is embedded or nested within the primary method. The design of this approach should not be confused with the one described above, where the second method is used as a follow-up to the initial method of data collection.
- A need exists to best employ a theoretical stance. There may be a particular research situation where a theoretical perspective dictates the need to collect both quantitative and qualitative data. All data could either be collected simultaneously or sequentially, with one form of data building on the other. The application of a particular theoretical viewpoint may determine this specific need.
- A need exists to understand a research objective through multiple research phases. Many research studies require multiple research phases – which may or may not be viewed as individual, separate studies – whereby researchers may need to connect several seemingly independent studies to achieve the overall research goal. This is a common approach used in comprehensive and/or multiyear evaluation or other types of longitudinal studies. As with the previous need, data may be collected simultaneously or sequentially. If the phases of data collection are simultaneous, or occur relatively close in time, we refer to the study as a multiphase mixed-methods research study; if the phases of data collection are distinctly separated by substantial periods of time, we might refer to the study as a multiproject mixed-methods study.
These scenarios illustrate situations in which mixed-methods research would be an appropriate design for investigating the particular problems. Although this list is not necessarily exhaustive, these cases and explanations can certainly serve as justifications for the researchers’ need to use a particular mixed-methods research design. In many cases, researchers may combine some of these six explanations to provide the most accurate justification for the use of mixed-methods designs. As you will also see later in the chapter, these six research problems lay the groundwork for and parallel the various designs of mixed-methods research we will examine shortly.
The Mixed-Methods Research Process
As you might expect, the process for conducting mixed-methods research closely parallels the general process for conducting educational research, presented in Chapter 2. That being said, there are some unique aspects to consider in the process of conducting mixed-methods research. The entire process is outlined and described below (Creswell, 2005; Fraenkel, Wallen, & Hyun, 2012); however, if you compare these steps with those described in Chapter 2, you will notice the additional, unique aspects in the process of conducting mixed-methods research:
1. Identification of the research problem to be studied. As we have seen numerous times in this book, the clear identification and specification of a research topic is the first step in any study. Consider that you may want to include both quantitative and qualitative data; however, do not become overly concerned about balancing the two forms of data at this point in the process.
2. Determination of whether a mixed-methods study is feasible. If you believe that your study will benefit from the use of quantitative and qualitative data, you must consider what this entails. First and foremost, you must have well-developed skills in gathering both quantitative and qualitative data. Gathering both types of data will also be more time-consuming; so you must factor in the desired timetable for your study. Additionally, you must have the appropriate skills to analyze both types of data. Finally, it is important to consider the make-up of potential audiences for your research – those audiences should be able to understand and have an appreciation for the complexity of mixed-methods designs. If any of the above conditions are not adequately satisfied, a mixed-methods study is likely not feasible.
3. Development of a clear and sound rationale for doing a mixed-methods study. Provided the study is feasible, you should consider and be prepared to answer questions of why you are collecting both quantitative and qualitative data, why both types of data are necessary, and how the study will be enhanced as a result of doing so. Again, if you cannot be clear and explicit in providing a rationale, a mixed-methods study may not be appropriate.
4. Identification of the appropriate mixed-methods design to guide your data collection. We will discuss mixed-methods designs shortly, but for now you will need to determine the following aspects of your data collection strategy:
- The priority you will give to quantitative and qualitative data
- The sequence of your data collection, if you do not plan to collect both types of data simultaneously
- The specific forms of quantitative and qualitative data you will collect
The determinations you make regarding these three items will typically align with a particular mixed-methods research design, which you should identify for inclusion in your final research report.
5. Development of research questions for both quantitative and qualitative methods. In a mixed-methods study, researchers typically delineate research questions that pertain specifically to the analysis of quantitative data and ones that pertain specifically to the analysis of qualitative data. It is possible also to add research questions that can be answered by the combination of the interpretations of both kinds of analysis. Depending on the nature of your study, some of the research questions may need to emerge during the course of the study. For example, if you are collecting quantitative data to be followed with collection of qualitative data, those qualitative research questions will likely depend on the outcomes of the quantitative data analysis. (It is important to note that Steps 4 and 5 may occur in reverse order or concurrently.)
6. Review of related literature and development of a written review. Reviewing related literature provides the same benefits we have discussed in previous chapters – guiding aspects of your study and contextualizing your study. You should develop a thorough written review of the pertinent body of literature to be included in your final research report.
7. Collection of data. Qualitative and quantitative procedures for the collection of data, which will be described in full in Chapters 11 and 12, are appropriate for mixed-methods research as well. In fact, there are no differences in data collection procedures – quantitative data in a mixed-methods study are collected just as quantitative data in any study would be, and the same holds true for qualitative data. The only caveat, however, is that care must be taken to ensure that data are collected so they parallel the mixed-methods research design you specified earlier.
8. Analysis of data. Similarly, data analysis proceeds just as presented in Chapters 11 and 13. The exception is that you must determine – based on the mixed-methods design you are using – whether you will analyze quantitative data separately from quantitative data or integrate the two types of data analysis. Again, you must ensure that you are following the particular process outlined by the specific mixed-methods research design you chose.
9. Development of conclusions and recommendations. You must draw conclusions, inferences, and recommendations directly from the interpretation of results of the data analysis. Once again, however, you must ensure that you are interpreting the data appropriately; in other words, you must determine if the interpretations of your analytical results will be drawn separately and sequentially, or if they will be done in an integrated, concurrent manner.
10. Preparation of a final research report. The final step in conducting a mixed-methods research study is to prepare the final research report. There is some variation in developing a report of mixed-methods research, compared with a report of just quantitative or just qualitative research. Specifically, the report should parallel your data analysis and interpretation of results. For example, if your study involved separate data collection, analysis, and interpretation for your quantitative data and qualitative data, your report should contain two separate sections for the collection, analysis, and interpretation – one for each type of data. In contrast, if your analysis and interpretation were integrated into one process across both types of data, you should include only one section reporting the combination of quantitative and qualitative data. Thus, the data analysis section is an attempt to converge both types of data into a single set of results and interpretations, relating directly back to the research problem and guiding questions.
Creswell, J.W. (2005). Educational research: Planning, conducting, and evaluating quantitative and qualitative research (2nd ed.). Upper Saddle River, NJ: Merrill/Prentice Hall.
Creswell, J.W., & Plano Clark, V.L. (2011). Designing and conducting mixed methods research (2nd ed.). Los Angeles, CA: Sage.
Fraenkel, J.R., Wallen, N.E., & Hyun, H. (2012). How to design and evaluate research in education (8th ed.). Boston, MA: McGraw-Hill.
Gay, L.R., Mills, G.E., & Airasian, P. (2009). Educational research: Competencies for analysis and applications (9th ed.). Upper Saddle River, NJ: Merrill.