So…”causal Fridays” are totally going to be a thing now, because a. I think that program evaluation is cool and important, and b. I love nerdy puns. If you want to blame someone, go find Zach Weinersmith, since he totally started it. The general idea is that, each Friday, I’ll dig up a paper that has an interesting causal result and put it here for you to ponder (and hopefully discuss at work happy hours and hot Friday night dates). This week we’ve got a paper on peer effects and educational investments:
Leonardo Bursztyn, Robert Jensen
When effort is observable to peers, students may act to avoid social penalties by conforming to prevailing norms. To test for such behavior, we conducted an experiment in which 11th grade students were offered complimentary access to an online SAT preparatory course. Signup sheets differed randomly across students (within classrooms) only in the extent to which they emphasized that the decision to enroll would be kept private from classmates. In non-honors classes, the signup rate was 11 percentage points lower when decisions to enroll were public rather than private. Sign up in honors classes was unaffected. To further isolate the role of peer pressure we examine students taking the same number of honors classes. The timing of our visits to each school will find some of these students in one of their honors classes and others in one of their non-honors classes; which they happen to be sitting in when we arrive to conduct our experiment should be (and, empirically, is) uncorrelated with student characteristics. When offered the course in a non-honors class, these students were 25 percentage points less likely to sign up if the decision was public rather than private. But if they were offered the course in one of their honors classes, they were 25 percentage points more likely to sign up when the decision was public. Thus, students are highly responsive to who their peers are and what the prevailing norm is when they make decisions.
You can have the paper emailed to most university email addresses, or you can find a non-gated version of the paper by googling the title. Note that this paper fits the “causal” requirement in a trivial but clean way, since the researchers essentially run a randomized trial (not to be confused with a randomized controlled trial, which is apparently a subtly different animal) such that, in each type of classroom, they have equivalent groups where the only systematic difference is what type of privacy is offered. (It’s then pretty intuitive that any differences in outcome are attributable to the difference in treatment.)
From a content perspective, this is…depressing but not surprising? I specifically like that the researchers tested for the effect in both regular and honors classes, since I’ve thought for a while now that this difference in peer pressure and incentives is underappreciated and needs to be studied more in order to understand differences in outcomes such as educations levels, earnings, etc. (The idea isn’t entirely new- Roland Fryer has a theoretical analysis of “acting white,” for example, but it’s important to see the empirical evidence as well as observe that this effect doesn’t only pertain to minorities.) I think I started thinking about this when somebody asked me if I was bullied in high school- I went to what I will refer to as “nerd school,” so the notion hadn’t really occurred to me before, and I think I responded with “does Brian S. trashing me in his campaign speech for National Honor Society President count?” (Apparently the answer to that question is a resounding no accompanied by an eye roll.) But this got me thinking- given that people respond to incentives, where would I have ended up if I had been given negative reinforcement every time I excelled at something nerdy? It’s hard enough to stay in and do work now when my friends want to go out and give me crap for my choices, and I doubt that my self-control as a teenager was markedly higher than it is now.
As I am writing this, I am watching an episode of the Carrie Diaries (don’t you dare judge me, the outfits are fantastic) where the class vixen is afraid of people finding out that she’s smart because it would jeopardize her chances of becoming prom queen. (Oh, Donna LaDonna, you are so my spirit animal.) I guess that this is a decent strategy- as the paper result suggests, if one can’t change one’s environment, the best option likely is to keep one’s utility-maximizing choices secret. This is not ideal, but what worries me more is that these same environments can also provide incentives to not make the nerdy investments in the first place. If this is the case, then early outcomes lead to different environments (eg. regular versus honors classes) that provide different incentives, which then magnify the initial differences in outcomes.
From a policy standpoint, it’s unclear what the best way to mitigate this discrepancy in incentives is. On one hand, it’s possible that throwing the honors kids into the regular classes could create a critical mass of what I will call “achievement culture” and drown out the negative peer pressure. On the other hand, it’s hard to have students at significantly different levels of achievement in a single classroom, and, if such a critical mass of achievers is not reached, such a policy could just end in the honors students smoking weed in the parking lot rather than going to class. (On that note, somebody should investigate whether the relationship between weed and performance suffers from a reverse causality problem.)