Is exploratory data analysis bad?

Last weekend, I ventured into unchartered territory (for me) and attended the Berliner Methodentreffen, a research conference mostly frequented by social scientists. I participated in a workshop on mixed methods, where the presenter discussed different models of mixing methods with each other (“Methodenpluralität” in German).

She omitted one model that I thought is often used: First to perform exploratory data analysis to detect some interesting phenomenon and then to do some explanatory qualitative research to formulate hypotheses as to why this phenomenon is, was, or keeps happening.

During my question, the temperature in the room dropped noticeably and I was informed that such exploratory data analysis is unscientific and frowned upon. Confused I inquired some more why this is so, but did not get a clear answer.

From a follow-on discussion I got the impression that the main worry was that people would try to sell spurious correlations as research insight. You may remember the funny but bogus correlation between the growth of Internet Explorer and number of pirates around the world or the decline of IE and the decline of number of murders in the US. But why would someone let get bad behavior in the way of useful research?

I have no idea. I still think that exploratory data analysis can help uncover interesting phenomena and guide the development of interesting research questions. Poor or unethical use of such methods should not get in the way of putting them to proper use.



Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.


Posted on

Share the joy

Share on LinkedIn

Share by email

Share on X (Twitter)

Share on WhatsApp

Research projects

Making free and open data easy, safe, and reliable to use
Bringing business intelligence to engineering management
Making open source in products easy, safe, and fun to use