A four-factor structure of posttraumatic stress disorder (PTSD) has been proposed for DSM-V based on empirical evidence that it is superior to the three-factor DSM-IV structure. However, most studies reveal multiple structures fit the data well in adolescent samples, and high factor correlations have been reported. Within two national samples of adolescents, we tested eight PTSD factor structures, which have never been compared in a single study. Confirmatory factor analyses (CFA) of PTSD symptoms were conducted in two national samples of adolescents: the National Survey of Adolescents (NSA; N = 4023) and the NSA-Replication (NSA-R; N = 3614). CFA revealed that all models provided very good fit to both samples (RMSEAs = .021-.039), though the one-factor model can be rejected, and correlations between factors were high (rs = .80-1.0). Potential interpretations of these findings include: (1) the indicators (i.e., symptoms) need refinement; or (2) relevant symptoms have yet to be identified.
Copyright © 2010 Elsevier Ltd. All rights reserved.