In this paper, we estimate the effects of neighborhood crime and in-school crime on educational outcomes for elementary and secondary schools in the city of Atlanta over the period 1999 to 2002. We specify a model that accounts for the joint determination of both types of crime along with school outcomes. Despite the large empirical literature on both education production functions and crime incidence, there has been little empirical work on crime's effect on school outcomes. One exception is Grogger (1997) who used individual data from the High School and Beyond study to estimate the effect of school violence on measures of individual student performance. After controlling for individual and school characteristics, he found that moderate and severe levels of school violence had substantial negative consequences for school outcomes. Our study both updates and expands on his work, using current data and better measures of neighborhood violence. Working Paper 07-19
More