Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

This article presents estimates of the effect of private school competition on public school performance. Using data on school districts in Georgia, the authors estimate models relating tenth- and third-grade test scores for either reading or mathematics to the level of private school competition. Test scores are not measurably or significantly higher in areas with greater private school competition, a result robust through multiple estimations using three measures of private school competition and a variety of control variables. The authors address the possible endogeneity between test scores and private school competition using instrumental variables estimators, with percentage of the population that is Catholic, county population in 1980, lagged competition, and various other measures as alternative instruments.

In 2001, the Early Childhood Study began to examine the development of four year-olds in early childhood programs, preschools and child care centers in Georgia, which are referred to in this report collectively as preschools. Purposes for the study included:

* *following the development of young children attending publicly funded and private preschools;

* *ascertaining the quality of their experiences; and

* *estimating the effects of Georgia's Pre-K Program.

This article presents estimates of the effect of private school competition on public school performance. Using data on school districts in Georgia, the authors estimate models relating tenth- and third-grade test scores for either reading or mathematics to the level of private school competition. Test scores are not measurably or significantly higher in areas with greater private school competition, a result robust through multiple estimations using three measures of private school competition and a variety of control variables. The authors address the possible endogeneity between test scores and private school competition using instrumental variables estimators, with percentage of the population that is Catholic, county population in 1980, lagged competition, and various other measures as alternative instruments.