Citation

Dannefer D, Schutt RK. Am. J. Sociol. 1982; 87(5): 1113-1132.

Copyright

(Copyright © 1982, University of Chicago Press)

DOI

10.1086/227557

PMID

unavailable

Abstract

Studies of race bias in the juvenile justice system have yielded contradictory and inconclusive findings. The diversity of findings, though due in part to inadequacies in the methods used in previous studies, is also attributable in part to the differential possibilities for bias in different settings. This paper develops and tests hypotheses that specify two conditions which affect the likelihood of bias: the characteristics and procedural constraints of processing agencies and the characteristics of their social environments. Log-linear analysis is used, to allow simultaneous control for the influence of prior record, type of allegation, family type, sex, race, and country in analyzing data from police and court records in a populous eastern state. Consistent with the hypotheses, the findings indicate that racial bias is more apparent in police dispositions than in judicial decisions. In the more urban of the two social settings studied, minorities constitute a relatively high proportion of the population; police bias is especially pronounced there. In the same setting, however, this bias may be compensated for, to some extent, by the courts.