Qualitative Research Methods for Data Science?

Why on Earth would a data scientist need to know about qualitative research? There are plenty of reasons. Here are a few.

By Kevin Gray, President of Cannon Gray.

Qualitative research header

Though I’ve had training in qualitative methods, I’m a quant specialist and have been for more than 30 years. However, I’m a user of qualitative research and have been throughout my career. Unless our area of data science has no relationship at all with human beings, it's is very relevant to quantitative researchers. The closer what we do is to marketing research - user experience (UX) and customer relationship management (CRM) being two examples - the more useful it becomes.

Qualitative research provides background and context that makes quantitative analysis such as predictive analytics more useful to decision makers. Consumer surveys do as well and frequently these two kinds of research are combined, with a qualitative phase preceding the survey and sometimes following it in a third phase.

How is qualitative research useful? There are many ways and here are a few illustrations.

First, we should remember that two people may buy the same brand for the same reasons (Miller Lite tastes great). However, they may also buy the same brand for different reasons (Miller Lite tastes great versus Miller Lite is less filling), or buy different brands for the same reasons (Miller Lite tastes great versus Bud Light tastes great) or buy different brands for different reasons (Miller Lite is less filling versus Bud Light tastes great). Consumer behavior can also vary for the same person by occasion or over time, too, as preferences and life circumstances change because of marriage, childbirth, job promotions, re-locations, health concerns, and so on.

Having some knowledge of why people do what they do in certain circumstances can help us both in building and interpreting our final model. For example, in the case of marketing, it helps us understand how to communicate to consumers, not just whichconsumers to target. There are many other ways qualitative can help inform quantitative research, but a fuller elaboration would require much more space than I have here.

Any kind of research has its pros and cons and qualitative is no exception. One of my main criticisms is that the level of expertise among practitioners is unpredictable. A few folks doing qualitative are amateurs, to be blunt. In some cases, lack of training or inexperience is the culprit. Others just don't seem to have it as researchers but are good at client handling, so they get the work.

Unexamined assumptions is another pet peeve of mine, and qualitative researchers seem more prone to this than quantitative researchers. A third criticism often leveled by "quants" is that analysts sometimes draw quite strong conclusions based on their own intuition rather than evidence. Treating speculation as fact is a cognitive error we all make, but I encounter this more among qualitative researchers.

And, of course, we all can be guilty of condemning sweeping generalizations with... sweeping generalizations!

My own view is orthodox: generally speaking, qualitative should only be used as an exploratory tool and conclusions and implications should be more tempered than in quantitative research. However, even with this limitation it can be invaluable in helping design and interpret quantitative studies and uncovering insights no quantitative method is able to.

Though I am not an expert in qualitative research, many of my contacts and business partners are. One is David McCaughan, who is my co-host on our audio podcast series MR Realities. "MR" is short for marketing research and Dave and I discuss a wide range of topics related to marketing research with special guests. We've done more than two dozen thus far and here are the links to the podcasts specifically about qualitative research:

  • "Semiotics: The Problem Child of Qualitative Research" (Sue Bell, Susan Bell Research)
  • "Social Media: Promises, Challenges and the Future" (Professor Raoul Kübler, Ozyegin University)
  • "How to Choose the Right Online Qual Method?" (Jennifer Dale, InsideHeads)
  • "Where Behavioral Economics Fits in the Customer Insight Landscape" (Bri Williams, People Patterns)
  • "Ethnography: Making Meaning from the Mundane" (Gordon Milne, Ipsos)

No registration is necessary - just follow the links.

Those of you who know me know I like books. Here are a few popular ones on qualitative methods I've found especially helpful:

  • Qualitative Research from Start to Finish (Yin)
  • Applied Qualitative Research Design (Roller and Lavrakas)
  • Qual-Online: The Essential Guide (Dale and Abbott)
  • Content Analysis: An Introduction to Its Methodology (Krippendorff)
  • Semiotics: The Basics (Chandler)
  • Doing Anthropology in Consumer Research (Sunderland and Denny)
  • Practical Ethnography (Ladner)

Your favorite bookseller will probably let you have a peek inside, so you can see for yourself if any may be of interest to you. There are also online materials, professional associations and seminars if you'd like to dig even deeper.

The real world is not as tidy as a well-written computer program and I would urge you to keep an open mind about qual. I couldn't do what I do without it.

I hope this has been interesting and helpful!

Bio: Kevin Gray is president of Cannon Gray, a marketing science and analytics consultancy.

Original. Reposted with permission.

Related:

  • Regression Analysis: A Primer
  • Time Series Analysis: A Primer
  • Text Analytics: A Primer



Related articles


0 Comments