This week, we’re going think about internal and external validity in study design. We’ll also think a bit about how statistical evidence can inform theory and vice versa! (Internal validity: does the study make sense on its own terms, i.e. is it a true statement about the sample it studies? External validity: can we generalize the results of the study beyond the sample it looks at?)

Discuss the following questions as a group. Professor Rao will visit each group to help clarify any questions you may have. We’re going to try something a bit different this time: rather than having a longer small-group discussion period and a short full-class discussion period, we’re going to do a random seminar for the big group time. To give you more time to read the article and formulate your own questions/thoughts, there are fewer questions than usual in the exercise here.

Random seminar

Each of you will be selected randomly, without replacement, to share your thoughts/questions from your small-group discussions for (up to) 2 minutes. You can use your time to share your thoughts and questions, respond to others, or both.

When your time is up, the alert will flash, and Professor Rao will say it’s time to move on. If we finish with time to spare, we can have a less-formal chat. You can skip your turn if you’d like by saying “skip”, and you can also finish early if you’d like.

How do minimum wage increases affect employment?

We’re going to do something a bit differently this time. Rather than having all members of the group read the same articles, you’ll each read this article about bunching estimators for minimum wage effects by Arin Dube and coauthors. Then you’ll each choose to read sections of a different article from the following set:

Make sure each article gets read. You should each read the bunching estimators piece and one of the other articles. The goal here is to learn from each other!

1. Internal validity and design
  1. How do Card and Krueger design their study to learn about how minimum wage increases affect wages and unemployment? What category does this study fall into? What about the bunching design Dube describes?
  2. What are the key confounding variables to worry about in Card and Krueger’s study, based on what you understand of its design? Would the same issues apply to a bunching design like Dube describes?
2. External validity and theory
  1. Suppose you were asked to advise on two potential minimum wage increases: a 10% increase and a 75% increase (both relative to the current minimum wage). How would you use the existing evidence to determine what the effects of each might be?
  2. The empirical results from the minimum wage studies seem to contradict basic economic theory about competitive labor markets. What gives? Is the theory just wrong, and if so why? Or are the empirics inaccurate, and if so why?
3. Your questions/thoughts
  1. What did you think about how minimum wage increases affect unemployment before these readings? What do you think now? What questions are you left with?