Do “gifted and talented” sections improve educational outcomes?

Maybe not, according to Sa A. Bui, Steven G. Craig, and Scott A. Imberman (ungated versions here):

In this paper we determine how the receipt of gifted and talented (GT) services affects student outcomes. We identify the causal relationship by exploiting a discontinuity in eligibility requirements and find that for students on the margin there is no discernable impact on achievement even though peers improve substantially. We then use randomized lotteries to examine the impact of attending a GT magnet program relative to GT programs in other schools and find that, despite being exposed to higher quality teachers and peers that are one standard deviation higher achieving, only science achievement improves. We argue that these results are consistent with an invidious comparison model of peer effects offsetting other benefits. Evidence of large reductions in course grades and rank relative to peers in both regression discontinuity and lottery models are consistent with this explanation.

Do the smarter kids choose a peer group to maximize…– what variable?  Maybe it’s a mistake to hang around with people who are too much smarter than you are and maybe a lot of kids know that.  How well do they choose their peer groups?  Can the authorities improve the outcomes of non-troubled kids by manipulating their peer groups?  Here is another new paper on peer effects.

In yet another new paper, early kindergarten seems to hurt not help children’s subsequent educational outcomes.