The purpose of this paper is to illustrate the utility of considering
the validation of personality disorders from the perspective of constr
uct validation. It is suggested that the importance of any particular
component of validation depends in large part on the theoretical const
ruct of personality disorder that is being proposed. Illustrations are
provided with respect to content validity, convergent and discriminan
t validity, and predictive validity, and with respect to the substanti
ve issues of selecting diagnostic criteria and establishing a threshol
d for diagnosis.