Delinquency and Gender
Melbourne Institute Working Paper No. 13/01
Date: November 2001
This paper investigates the determinants of juvenile delinquency for males and females using the 1958 Philadelphia Birth Cohort Study. Ordered probit models for juvenile arrest are estimated separately for males and females. An adaptation of the EM-algorithm is used to estimate the model for females in order address a problem of missing values for the variable linking demographic and arrest data. The results indicate that juvenile arrests for both males and females are more likely for nonwhites and for those who leave education early. Males and females behave differently, in that males are more likely to be repeat offenders.