Research and method

A while back I was working in an academic department where my job was to support faculty research, answer methodological questions, design projects, write grants, and so on. One faculty member asked me a question regarding mixed methods research. My response was to clarify the purpose of research and method more generally. Below I include a revised version of my response.

Research is an activity that one undertakes in order to answer a question one has regarding a phenomena of interest. In order to answer that question one needs to effectively understand how to study that phenomena and put processes in place so as to collect information regarding it. The outcome one hopes for is to understand and perhaps explain the phenomena. The process of information gathering and sorting through that information to ensure that the information you have gathered is accurate and valid is a research method.

So what is mixed methods? The question the colleague who contacted me posed was: Is it using different methods to answer one question, or using different methods to answer different questions within one study?

Some people argue that there is no such thing as “true” or “pure” mixed methods. One’s view on this will likely be determined by how dogmatic/orthodox one’s personal approach is to research broadly conceived, and research methods more specifically. Without getting too epistemological (the branch of philosophy that studies how to know, and which is often thought about in relation to ontology, the study of what can be known) the answer is “yes” either of these would be mixed methods research.

In general mixed methods refers to the use of multiple approaches to collecting and interpreting information in order to answer questions on a topic of interest (e.g., toward a particular research project). Creswell (2015) who is a recognized mixed methods expert, defines mixed methods as “An approach to research in the social, behavioral, and health sciences in which the investigator gathers both quantitative (closed-ended) and qualitative (open-ended) data, integrates the two, and then draws interpretations based on the combined strengths of both data to understand research problems.” (p.2). Note that Creswell distinguishes between quantitative and qualitative methods, this is because each provides very different types of information about a phenomenon, and each has its own limitations and implications.

Research methods at the most basic level are about ways of knowing something. If you have a particular phenomenon of interest there are many ways to know it and understand it, the only limitation is your own creativity, training, and ability/motivation to expend effort. It is impossible to know something completely and the choice you make in your approach to knowing will truncate (cut off parts) of what you will come to know. We recognize this when we talk about limitations to a particular research method as we write up our results. In order to deal with truncation sometimes we decide to take multiple approaches to knowing the phenomena of interest. This may mean we have to pose multiple questions in relation to that phenomenon (indeed any single research project is often an iterative process of asking new questions to better understand the phenomenon of interest), but by adopting multiple approaches we are employing multiple methods in our effort to know it, that is, using a mix of methods or taking a mixed methods approach.

For example, statistics are based on averages and variance from the average. As such, they are good for getting a broad picture of a phenomenon, speaking about things in generalities, and getting to know trends in variables related to a phenomenon of interest. However, they truncate the everyday lived experiences of people, they miss nuance, context, emotion, and many other aspects of almost any topic. This is why we will sometimes complement statistical studies (either at the same time, or in subsequent studies) with more in-depth qualitative approaches.

When it comes to mixed methods there are different approaches. Commonly, some people will use interviews to develop questionnaires and then do a survey of a larger sample. Others do the opposite by first doing a survey and then doing interviews to add some narrative/nuanced details to the broader picture they have painted with the statistics.

Hopefully I haven't muddied the waters for people with this post.

References

Creswell, J. W. (2015). A concise introduction to mixed methods research. Los Angeles: SAGE.