Research Methods

Worried Grandma

Professor Palmer,
I regret to inform you that I won’t be able to take today’s exam as my family has suffered a great loss.

Prof Palmer,
This is the worst possible timing, but sadly my grandmother has died and I will be forced to miss our exam today.

Hey, I missed today’s exam because my grandma died, what should I do?

Every semester right around test time I get emails like the ones above. For some reason grandmothers just start dropping1. It’s uncanny. When I start talking about this fact with my students they all smile and laugh quietly. I tell my students in as earnest a voice as possible that this phenomena is a prime candidate for social research. Fortunately for us this research has already been done by Dr. Mike Adams (1990)2.

Adam’s work is clear: Grandmothers3 hear of their grandchild’s exam, become tremendously worried, and die from the stress. My students laugh at this finding, but I go on in earnest. After a short discussion to define a hypothesis, dependent, and independent variable, I ask the students to work with their neighbors to identify all three in Adam’s dead grandma research. After a few minutes students correctly identify all three and I put up this slide:

Dead Grandma Syndrome Hypothesis

It never fails that at some point a student will say, “Professor Palmer this research doesn’t factor in student lying. Most of the students are making it up.” I muster my acting skills I learned in high school drama and through a perplexed look on my face. “No, that can’t be. Students wouldn’t do that. Would they?” If my acting holds gullible students will tell me that desperate students will say or do just about anything to not miss/fail a test. Then we move into a discussion of spurious correlations.

Students must understand how to ask a research question before students can understand that sociology is a science. Activities like the dead grandma can really help your students grasp these fundamental concepts quickly.


1. While this post and my presentation of the dead grandma research is delivered in a tongue and check way I tell my students after the discussion that I have a great deal of reverence for anyone who suffers the loss of a family member. In fact my dear grandmother’s funeral coincided with a test I had as an undergraduate, but I told the professor days in advance and took the test before the exam date.

2. Adams, Mike. 1990. “The Dead Grandmother/Exam Syndrome and the Potential Downfall Of American Society.” The Connecticut Review. Available at:

3. The research reports that Grandfathers do not expire nearly as often as their heterosexual partners. An interesting fact that I ask my students to explore using a symbolic interactionist lens. Why use grandmas and not grandpas when forming an excuse? There must be some perceived rhetorical value in the gender of the grandparent.

“Uh Professor, I think you are way off on this one. I know what your sociology research tells you, but people round here aren’t like that.”

Teaching students to see beyond the individual and at the sociological level is really hard. Especially in the United States we overvalue anecdotal personal experiences & undervalue empirical social facts. Worse yet, when student’s lived experiences differ from what sociological research finds, they believe that their lived experience invalidates the sociological research. So how can we get our students to see at the sociological level? Easy. Just put it in terms they understand.

I tell my class to imagine that I have just handed back their graded tests for them to review. I tell them that the class average was a 72%. This, I tell them, is an empirical social fact. The trend or in this case the average for the entire class was 72%.

Then I ask them, “would it make sense if one of you told me ‘the average can’t be a 72% because I got a 96% on my test’?” They laugh at the ridiculousness of this question. “Well when I present to you empirical social facts and you say to me ‘well I know this one guy who doesn’t do what your research says’ or ‘well that’s not true in my experience, so your social fact must be wrong’ you are basically arguing that because you got a 96% the class average can’t be a 72%” Many heads nodding in unison. They get it.

While this blog has been mostly focused on social inequality topics, I have taught research methods in one form or another for years now. Teaching students how to design a survey can be tricky because the process is deceptively easy. Students think, “Hey, I have taken tons of surveys before. How hard can it be?” They then proceed to break every rule of good design you talked to them about in class.

A simple, quick, yet effective activity to teach good survey design is to have your students take a survey that is horribly designed. I tell my students that I want no talking and then pass out a survey about internet usage (download it here). Every question on the survey is either double barreled, leading, biased, or has response options that make no sense or overlap.

After a few minutes I tell them to stop and ask what they think of the survey. They uniformly say it’s awful. Students really like this activity- typically they laugh out loud when reading the questions. I then have them pair up and identify everything that is wrong with the questions. As a class we go through each question picking it apart. We then formulate new questions that don’t violate any of the basic survey design rules.

The activity is also beneficial because students get to take home an example of what not to do that they can compare their work to while creating their own survey. Pedagogically I really like this activity because it has the students playing an active role in their education. Also, the “bad survey” is formatted well so you can tell your students that their survey should look like the example you gave them, but with much better questions.

Download the Survey (pdf Version)
Download the Survey (Word Version)