Empirical analysis of peer effects on student achievement has been open to question because of the difficulties of separating peer effects from other confounding influences. While most econometric attention has been directed at issues of simultaneous determination of peer interactions, we argue that issues of omitted and mismeasured variables are likely to be more important. We control for the most important determinants of achievement that will confound peer estimates by removing student and school-by-grade fixed effects in addition to observable family and school characteristics. The analysis also addresses the reciprocal nature of peer interactions and the interpretation of estimates based upon models using past achievement as the measure of peer group quality. The results indicate that peer achievement has a positive effect on achievement growth. Moreover, students throughout the school test score distribution appear to benefit from higher achieving schoolmates. On the other hand, the variance in achievement appears to have no systematic effect.
