Citation

Shavit Y, Rattner A. Am. J. Sociol. 1988; 93(6): 1457-1470.

Copyright

(Copyright © 1988, University of Chicago Press)

DOI

10.1086/228907

PMID

unavailable

Abstract

The age distribution of criminal activity peaks in the mid to late teen years and declines thereafter. Recently, Hirschi and Gottfredson have argued that the shape of the age distribution is invariant across social groups and that this shape is unexplained by any known st of sociological variables. They also assert that longitudinal data and models are not necessary to explain delinquency. The present investigation employs longitudinal data on the early life histories of an Israeli male birth cohort to test these assertions and fails to reject them. The shape of the age distribution does not vary significantly across ethnic, socioeconomic, and religious orthodoxy groups; nor can it be explained in terms of age variations in marital status, employment, and schooling. However, contrary to Hirschi and Gottfredson's assertion concerning longitudinal data, it is demonstrated that in the absence of such data, several parameters of the estimated models would have been biased.