A LONGITUDINAL TEST OF SOCIAL-CONTROL THEORY AND DELINQUENCY

Recent longitudinal research suggests that cross-sectional studies have exaggerated the importance of Hirschi's social control theory. This longitudinal research, however, suffers from one or more of the following problems: (a) measures of questionable validity and/or reliability; (b) misspecified causal models, including models that omit important variables and fail to examine the reciprocal and contemporaneous effects between variables; and (c) the failure to consider certain methodological problems peculiar to panel analysis, such as autocorrelation. Most of these problems reduce the likelihood of finding a causal effect from social control to delinquency, and so make the findings of the longitudinal studies suspect. This article uses data from the first two waves of the National Youth Survey to overcome these problems, and provide a more accurate estimate of the effect of social control on delinquency. © 1991, SAGE PUBLICATIONS. All rights reserved.