Image by Glenn Carstens Peters on Upsplash

Remember that particular methodologies answer particular kinds of questions, and each method has its own limitations.

Quantitative research designs are focused on answering questions about what, how much, how many, and to what extent already identified variables are related. Sophisticated designs can test the strength of relationships between variables or even establish causal relationships. For example, if I had access to a database that documented people’s blood types, vaccination history, and health events, I could test the correlation between blood types, exposure to vaccines with fetal matter, and adverse vaccine reactions. I can crunch those numbers and then interpret the results for statistically significant relationships. So, quantitative research designs are about the measure of things and processes.

Qualitative methodologies, on the other hand, are focused on the meaning of things. The aim is to reach greater understanding of the problem or phenomenon by identifying what is essential for understanding the problem or phenomenon. The motivating questions are about how and why does this work or not work and the “data” is generally narrative text rather than counts and measures. For example, rather than asking how many people are bored with their jobs, I might ask, “What does it mean to feel bored with a job?” Or rather than asking how many or to what extent are leaders in public service transformational leaders, I might ask, “What does it mean to be a transformational leader in public service?” Aspects of the essences identified can later be tested in quantitative research but testing the relationships between themes or aspects or the experience identified in the explication of the narrative would not be the focus of a purely qualitative methodology.

When it comes to distilling the meaning of a phenomenon or even relationships between the themes explicated, you need to be clear from the outset if you are approaching the narrative with a pre-existing lens or theory or allowing the essence to emerge by intentionally suspending your preconceptions about the phenomenon and relationships within it. So, there are two ways of approaching qualitative data, one in which you are honest about the lens with which you are attempting to gain understanding, or second, you will attempt to suspend your preconceptions based on the literature review and personal experience and allow answers to the research question and resolution of the problem to arise from the data itself. Arguably, human beings are perspectival, in other words, as human beings, we always experience phenomena from a perspective or point of view, and the researcher as a human being can therefore only be honest about the preconceptions with which he or she arrived at an answer to the research question.

Since I completed my PhD, apps have been created, like NVivo, to make the processing of the narratives text, easier, but here is the dilemma: While its convenient to have an app like NVivo count the themes and categories you are either looking for via your theoretical lens or that emerge from the narratives, NVivo sometimes misses a truth that just one person uttered but could be subsumed under lesser truths expressed by all the participants. For example, only one participant in my sample for my PhD included the metaphor of Mother as Snake Woman, but the essence the image evoked, namely, the persecutory mother, was a strong theme expressed in all participants’ dream series and descriptions of their relationships with their mothers gathered through amplification of their dream series. Strictly speaking, if you use NVivo or any of its alternatives to process the narratives, you have reduced the narrative to measures and counts and are, technically, using a mixed-method design. So, bear in mind that qualitative methodologies are not about how many people expressed the same idea but about the depth of understanding reached in the explication of what participants said. Moreover, these apps process the data rather than distill the meaning.Mixed method designs can occur on several levels other than the data gathered and the strategies used for processing that data, for example, counting themes that emerge from a series of narratives. A mixed method design can be primary. Social surveys, for instance, can, from the outset, collect both statistics and narrative data. In trying to assess mine employees’ trust of management in the 1980s, one question was, “To what extent do you believe management attends to your physical welfare?” We used a Likert-type scale to measure how many (white and black) miners perceived that statement about management to be true and to what degree they thought it true, or how much. That’s a very simple quantitative design. However, the question, “Why do you say this?” was also asked with an open-ended follow-up question. The question, “Why do you believe that?” evokes narrative data that not only alerts the researcher to what aspects of their physical welfare participants deem important but also for what aspects of their physical welfare they hold management responsible. So, in this example, a mixed-method data collection strategy was used: We collected both quantitative and qualitative data, but the processing of the narrative or qualitative data was quantitative: the number of times a theme emerged was counted, and the themes were listed in the order of frequency or occurrence. So, bear in mind that, for the most part, research is about making strategic decisions in your efforts to answer the research question and solve the research problem. Choosing and developing the methodological design is about finding the best means to approach and gather information about the topic and process it to answer the research question.

Besides the research question, a number of other issues will influence your choice of research design, for example, one consideration is access to reliable data, be that medical records or people who can describe their experience. Access may also depend on the sensitivity of the information the research requires. For example, in a society where same-sexed partnerships are illegal and even persecuted, one is unlikely to have people volunteer to share their experience of their same-sexed relationships, and such people are unlikely to be honest in a general survey. Another consideration is the amount of time, money, and energy available to do the research. Still another is what is possible at your institution. For instance, it would silly to propose doing research about the meaning of boredom and creativity if the department specializes in neuropsychology. But you could, if you have access to the equipment, measure brain wave activity and other physiological responses among bored and creative people.

So a methodological design may well be a mosaic of quantitative and qualitative data collection procedures and processing strategies that offer the best chance of answering the research question, and the kind of methodology for answering the question may well be constrained by considerations related to the amount of time, money, and energy that can be invested in the research. For example, a longitudinal design following the degree to which participants’ stress levels decreased after a workshop about preventing burnout over five years is not achievable if your research report must be submitted within six months of being approved. And regardless of what methodological design you choose to do, you have to show that the design you chose is as good if not better than any other design you could have chosen given your research question and circumstances.