So in my stats class, one of the quizzes I give presents a number of research scenarios. The student’s responsibility is to determine the appropriate statistical procedure they’d need to use to correctly solve the problem. The key to correctly answer the question is typically contained in the problem itself. More often than not, I use silly or disgusting examples. I was just reviewing quizzes my TAs graded for me and found the funniest answer.
I gave the following scenario:
On a willingness to eat disgusting things test, the possible scores range from 1 to 10 (the higher the score, the more tolerant a participant is of eating disgusting things), international data show that people typically score 6. Psychologists recently tested a group of 149 male college students who are members of Greek letter social organizations. This group scored an average of 9.3 and their standard deviation was .62.
Now the procedure they’d use to solve this is a single sample t-test because we have to estimate the standard deviation in the population based on our sample. They got the test right, but the explanation was awesome. To explain why they decided to do a t-test, they said:
…Greeks tend to be a smaller population who eat nasty things when joining the fraternity.
I laughed so hard tea came out of my nose.