Toward the measurement and prediction of victim proneness

Abstract

Two samples (each of 500 households) were drawn from the data gathered by the National Crime Survey (NCS). Utilizing a nonlinear response model, it was found that the resulting victimizations fit well a theoretical construct of victim proneness. Further it was discovered that some aspects of proneness can be predicted from social characteristics of the households. It was also discovered that for some purposes direct nonlinear regression is more useful. It was concluded that this approach might be fruitful in resolving difficulties arising from assignation of crimes to households or persons; and by providing a simplified parameterization of victimization, it would allow the more graceful testing of alternative hypotheses about the structure of victimization. This has important consequences for the testing of the efficacy of intervention programs. © 1983, Sage Publications. All rights reserved.