We estimate peer effects in college achievement using a dataset in which individuals are exogenously assigned to peer groups of about 30 students with whom they are required to spend the majority of their time interacting. This feature enables us to estimate peer effects that are more comparable to changing the entire cohort of peers. Using this broad peer group, we measure academic peer effects of much larger magnitude than found in previous studies. The effects persist at a diminished rate into follow-on years and we find evidence of nonlinearities in the magnitude of the effects across student academic ability. 1

Women are substantially less likely than men to graduate college with a STEM degree. This paper investigates whether class composition can help explain why women are disproportion-ately more likely to fall out of the STEM “pipeline”. Identification comes from a standardized enrollment process at a large public university that randomly assigns freshmen to different mandatory introductory chemistry lectures. Using administrative data, I find that women who are enrolled in a class with higher ability peers are less likely to graduate with a STEM degree, while men’s persistence in STEM is unaffected by class composition. I also show that the decline for women is most pronounced for those in the bottom third of the ability distribution. I rule out the possibility that this is driven solely by grades because both men and women re-ceive higher grades in classes with higher ability peers. Overall, these results suggest that class composition as an important factor in determining STEM persistence for women and provide a novel explanation for part of the STEM gender gap in post-secondary education.

Support this valuable resource today! This Article is brought to you for free and open access by the ILR Collection at DigitalCommons@ILR. It has been accepted for inclusion in Working

Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. www.econstor.eu The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public. Terms of use: Documents in D I S C U S S I O N P A P E R S E R I E S IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author. IZA Discussion Paper No. 8528 October 2014 ABSTRACT Peer Effects on Childhood and Adolescent Obesity in China Using data from the China Health and Nutrition Survey (CHNS), this study analyzes peer effects on obesity in a sample of 3-to 18-year-old children and adolescents in China. Even after a rich set of covariates and unobserved individual heterogeneity are controlled for, it is evident that such peer effects do indeed exist. These effects are stronger in rural areas, among individuals at the upper end of the BMI distribution, and especially among females. All else being equal, female adolescents whose peers have a higher BMI are less likely to consider themselves overweight, suggesting that peer effects may be working through changed societal bodyweight norms. JEL Classification: I10, I15, J13, C14

forthcoming

We estimate peer effects in college achievement using a data set in which individuals are exogenously assigned to peer groups of about 30 students with whom they are required to spend the majority of their time interacting. This feature enables us to estimate peer effects that are more comparable to changing the entire cohort of peers. Using this broad peer group, we measure academic peer effects of much larger magnitude than found in previous studies. The effects The views expressed in this article are ours and do not necessarily reflect the

Abstract: Using exogenous secondary school assignments to remove self-selection bias to schools and peers within schools, I credibly estimate both (1) the effect of attending schools with higher-achieving peers, and (2) the direct effect of short-run peer quality improvements within schools, on the same population. While students at schools with higher-achieving peers have better academic achievement, within-school short-run increases in peer achievement improve outcomes only at high-achievement schools. Short-run (direct) peer quality accounts for only one tenth of school value-added on average, but at least one-third among the most selective schools. There are large and important differences by gender. In many nations, there is fierce competition for scarce slots at selective schools (Hsieh and Urquiola 2006, Hastings and Weinstein 2008). This is, in part, because students at more selective schools typically have better outcomes ─ giving the impression of sizable benefits to attending selective, and often prestigious, schools. However, because motivated and highachieving students tend to select to these schools, these differences may reflect selection rather than selective schools providing greater value-added. Addressing this selection problem, Jackson (2010) uses a quasi-experimental design and finds that attending a more selective school in Trinidad and Tobago has positive effects on exam performance and high-school graduation.

Using longitudinal administrative data from a large elite research university, this paper separately analyzes the determinants of persistence for life and physical science majors. My results confirm much of the previous research on major persistence in the sciences, but I document that many findings are solely driven by persistence patterns in the physical sciences. For example, I show that the previously documented gender gap in science major persistence is due entirely to a large gap in the physical sciences. Despite large differences in persistence patterns between physical and life science, persistence in both fields is similarly influenced by grades. I provide suggestive evidence that students in both fields are “pulled away ” by their high grades in non-science courses and “pushed out ” by their low grades in their major field. In the physical sciences, analyses using within course and cohort variation show that peer quality in a student’s introductory courses has a lasting impact on the probability of persisting. I would like to thank Ron Ehrenberg, George Jakubson, Joshua Price, Mirinda Martin, Doug Webber, Ken Whelan, and Carrie Ost for very helpful comments. I thank the Sloan Foundation for financial support. Also, I am indebted

comments on early drafts of this paper. The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

(ii) development of