The present study focused on the consequences of cut-off scores in personal
ity disorder diagnoses for their association with criminal behavior. Using
ICD-10 personality disorder criteria eliminating offence-related symptoms,
we studied the distributions of categorically diagnosed personality disorde
rs and of dimensional personality disorder scores in a group of offenders a
nd a noncriminal control group. Whereas the dimensional scores of the offen
der group differed significantly from those of the control group for all pe
rsonality disorders under study, the frequency of categorical diagnoses dif
fered significantly for two personality disorders only. Moreover, predictio
n of group membership (offenders vs. nonoffenders) from personality disorde
r scores was substantially more precise than prediction from categorical di
agnoses. It is concluded that a dimensional approach to personality disorde
r diagnosis Is not only superior theoretically but also yields more precise
information about the specific associations with criminal behavior.