Year of Publication

2005

Document Type

Thesis

College

Arts and Sciences

Department

Psychology

First Advisor

Donald R. Lynam

Abstract

This study examined validity of the Life History Calendar by comparing retrospective and prospective reports of adolescent substance use. Agreement was calculated using kappa and phi coefficients for dichotomous variables, and Bivariate correlations for average substance use. Effects of potential personality, psychopathology, and demographic moderators on agreement were assessed through hierarchical regression analyses and curvilinear relations determined. Results reflected moderate agreement between retrospective and prospective reports of substance use, moderated by personality and psychopathology variables, particularly Agreeableness, symptoms of antisocial personality disorder, and symptoms of substance abuse. Agreement between retrospective and prospective reports was adequate for reports of alcohol and marijuana use for at least six years after initial reports of use. Agreement for cigarette reports was adequate a year after initial reporting.

Share

COinS