The quantitative vs. qualitative divide, take four
EDIT: This is a re-posted entry from my previous blog.
One thing I’ve observed throughout my time in research and academia is that at least in my work and social circles, the shape of the quantitative vs. qualitative research divide is changing. I think this is particularly apparent across generations of researchers, where there seem to be three different waves of social attachments to the categories:
“The method I use (qualitative or quantitative) is a better way of understanding the world.”
“We need both, and we need people that do both (mixed methods everywhere).”
“People should focus on one or the other, but we acknowledge that they just are different and are applied to different questions.”
This last approach sounds okay in some ways, but I still think it has some fundamental problems, which I think manifest themselves in (a) the challenges students have when learning methods and (b) the realistic ways in which quantitative and qualitative researchers treat each other. At best, this approach allows people to tolerate each other while still not really caring about what others have to say, and at worst, it is a hand-wavey way to avoid fixing the original issues with the first wave described above.
Here are some issues that I think this causes:
People not thoroughly reading research using methods different from theirs because they believe it’s not applicable.
The assignment of “objectivity and generalizability” to one method and “subjectivity and actual understanding” to another.
The division of people into “quant” people and “qual” people in a way that means “We think in fundamentally different ways about the world” and not simply “I have more expertise in one method than the other.”
In particular, I think we haven’t been careful enough about how we explain that “different methods are used for different questions,” where “different questions” somehow came to mean “different fields of inquiry” and not “different parts of the process of science and inquiry.”
At least for me, when I teach my students about the difference between quantitative and qualitative research (primarily in quantitative research courses, which are the majority of my courses), I don’t focus on the numbers distinction, but rather on the difference between pre-definition and post-definition. In other words, quantitative research defines concepts in a specific, consistent way across observations before data collection, which allows for certain types of operations and conclusions to be drawn. Meanwhile, qualitative research collects less pre-defined information, and the definitions are identified and constructed after data collection, which allows for other operations and conclusions. The difference in the “types” of questions that can be answered, therefore, is not about the actual thing being studied, but more about where we are in the cycle of inductive and deductive research, which is applicable to anything under study. (I was once told that I had a very distinct eye twitch when someone said “quantitative research isn’t about experiences” at an event.) Of course, there are some exceptions that muddy this definition (e.g., unsupervised learning methods), but I think it works a bit better than the numbers versus people distinction.
For me, there are several practical implications of this type of definition for my own teaching in quantitative methods.
Not defining quantitative as “numbers” and qualitative as “experiences and feelings,” and actually using examples of experience and emotions-centered quantitative research (see: a whole bunch of psychological research).
Not explaining any research as objective, and outlining how quantitative and qualitative research are both subjective, with the difference being where the subjectivity happens. (And, when I happen to know them, bringing up some perspectives from the philosophy of science generally.)
Spending real time talking about measurement and the ways in which human choices are a part of measurement (a concept often glossed over in introductory statistics courses – I took that course three different times, and it didn’t come up once).
Not using the words “qualitative” and “quantitative” as synonyms for “categorical” and “continuous” data, respectively. (Categorical data is still quantitative, by virtue of its consistent definition.)
While being clear that I am not particularly well versed in qualitative methods relative to quantitative methods as a product of my educational experiences, consistently noting places where qualitative research plays a role in my own research field.
Outlining both qualitative and quantitative approaches to studying very similar questions.
In my lecture(s) on causality, covering the qualitative perspective on causal evidence as well (Maxwell 2004 has some good writing here, I think).
Do I think quantitatively-oriented and qualitatively-oriented researchers tend to identify and think about different angles on research by virtue of their approach? Yes. But I also think we’re currently in this weird space where respecting diverse methodological approaches translates into just being polite in public while not really engaging in diverse forms of knowledge-making. And I think that beyond just keeping each other accountable, part of where this can be addressed is in the classroom.