Are Within-Individual Causes of Delinquency the Same as Between-Individual Causes?

David P. Farrington, University of Cambridge
Rolf Loeber, University of Pittsburgh
Yanming Yin, University of Pittsburg Medical Center
Stewart J. Anderson, University of Pittsburgh

This paper investigates causation using a new method in criminology. Since the influential research of Glueck and Glueck (1950), almost all studies of the causes of delinquency have compared risk factors of delinquents and non-delinquents or have correlated risk factors with levels of delinquency. In both cases, between-individual differences in risk factors are compared with between-individual differences in delinquency.

There is, however, another way to investigate the causes of delinquency, by comparing within-individual changes in risk factors over time with within-individual changes in delinquency over time. This method has rarely been used, because it requires repeated measures of both risk factors and delinquency in a longitudinal study. Nevertheless, it is arguably a better way of studying causes. In order to advance knowledge about these two different methods of studying the causes of delinquency, this paper compares -- for the first time in criminology as far as we are aware -- between-individual and within-individual correlations between risk factors and delinquency. Because of their novelty, these analyses are mainly illustrative. This paper also compares results obtained with simultaneous correlations (where the risk factor and delinquency are measured at the same time) with forward-lagged correlations (where the risk factor is measured before the delinquency).

(Return to Program Resources)

Updated 05/20/2006