U.S. flag

An official website of the United States government, Department of Justice.

Scientific Justification and the Race-IQ-Delinquency Model (From Critique and Explanation, P 91-131, 1986, Timothy F Hartnagel and Robert A Silverman, eds. - See NCJ-102349)

NCJ Number
R A Gordon
Date Published
41 pages
This paper establishes the construct validity of a model that relates differences in IQ parameters to differences in delinquency between blacks and whites, and it identifies delinquency variables as mediators of IQ.
Developments since the first test of the model confirm its construct validity. There is independent evidence that the assumption of normally distributed intelligence is warranted and an abundance of research that indicates IQ tests are not biased against blacks. The number of studies reporting correlations between IQ and delinquency have increased. The studies agree about the magnitude of the IQ-crime correlation in samples presumably centered at about IQ 100. Insights have been gained about the variation in the magnitude of IQ-crime correlations between blacks and whites. Major new estimates of IQ parameters support and improve on the estimates used in the model's first trial. They further demonstrate the meaningfulness of race classification in the model. The latest test of the model was applied to the delinquent sample that appeared in Philadelphia's juvenile court between 1949 and 1954 and to delinquents committed to training schools nationwide in 1964. The model's explanatory power was confirmed. The model holds in the face of other variables impacting delinquency because these variables are strongly related to IQ. 3 tables.