Disruptive classmates

You can do great work on education in New Zealand's Integrated Data Infrastructure. It lets you link kids to their parents and families, so full family background from parents' education to criminal records can be used as explanatory variables.

But you can't do this. From Scott Carrell, Mark Hoekstra and Elira Kuka in the November AER: The Long-Run Effects of Disruptive Peers.

A large and growing literature has documented the importance of peer effects in education. However, there is relatively little evidence on the long-run educational and labor market consequences of childhood peers. We examine this question by linking administrative data on elementary school students to subsequent test scores, college attendance and completion, and earnings. To distinguish the effect of peers from confounding factors, we exploit the population variation in the proportion of children from families linked to domestic violence, who have been shown to disrupt contemporaneous behavior and learning. Results show that exposure to a disruptive peer in classes of 25 during elementary school reduces earnings at age 24 to 28 by 3 percent. We estimate that differential exposure to children linked to domestic violence explains 5 percent of the rich-poor earnings gap in our data, and that each year of exposure to a disruptive peer reduces the present discounted value of classmates' future earnings by $80,000.

The IDI can tell you which school a kid attends, but cannot tell you which classroom that child is in. You can neither link children to their teachers nor to their classmates because that information isn't held in central government's administrative data. So the best you could do is get the school's proportion of kids with relevant CYF notifications or family policing records.

It's a bit of a shame - I wonder whether much of what parents seek in higher decile schools isn't educational quality, but avoiding disruptive peers.