StATS: The danger of using humor on a web page (November 14, 2006). Category: Website details

I got a nice email yesterday from someone who had read my web page with a definition of a retrospective study.

I had made a humorous comment about the difficulties with prospective studies.

A prospective study is slow. The subjects are uncooperative and refuse to die fast enough to make our study successful.

The person who read that page suggested that the comment was insensitive and possibly hurtful for someone who was researching their own child's illness. She pointed out that some level of humor is probably needed for a dry topic, but that this comment was "more likely to offend than to amuse."

It's worth keeping in mind that when you make a humorous comment that is taken the wrong way in a live seminar, you have the opportunity to immediately qualify the statement. But on a web page, unless the person reading it is nice enough to offer some feedback, you will never know and never have the opportunity to correct your error.

I rewrote the comment:

A prospective study is slow. Unless you are studying a rapidly fatal disease, you have to wait years or even decades to accumulate sufficient data to draw any strong conclusions.

and I hope that is better.

The point that wanted to make is an important one. Often the needs of the research study is at odds with what everyone really wants. Anytime you are studying a discrete event, you need to accumulate about 25 to 50 of these events in each group in order to get reasonable power and precision. Most of the "events" that we study in research are bad things, like death, relapse, device failure. You need to plan the research carefully so that you study a large enough group of patients for a long enough period of time so that you do get a sufficient number of these bad events happening. If you study mortality in a group of ten thousand patients, and in one group four patients die and in the other group only two patients die, you have indeed seen a doubling of mortality risk, but it is impossible to declare anything about a data set with only six total deaths.

Researchers will often start their research programs in patients who are most seriously ill for just this reason. A group of patients who are seriously ill are more likely to experience bad events like re-hopsitalization, and provide you with the sort of precision that you need to make meaningful statements about risk.

Creative Commons License This work is licensed under a Creative Commons Attribution 3.0 United States License. It was written by Steve Simon and was last modified on 04/01/2010.

This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at