Borderline and schizotypal personality constructs were compared as predictors of dissociation for 971 nonclinical university undergraduates who completed 11 self-report behavior inventories. Structural equation modeling and multiple regression were used to (a) test the hypothesis that the quantitative interaction of borderline and schizotypal constructs would be the strongest predictor of dissociation and (b) evaluate each construct alone as well as their additive effect for prediction of dissociation. The interaction hypothesis was rejected; the quantitative interaction was only a substantively trifling and statistically nonsignificant predictor of dissociation. In a commonality analysis, the additive borderline and schizotypal effect was equivalent to schizotypy alone and marginally larger than the borderline effect alone for the prediction of dissociation. However, all three effects separately were substantial predictors; thus, dissociation is an untenable discriminator for the borderline and schizotypal constructs. Finally, when schizotypy was compared directly to the borderline construct such that the common variance was distributed dependent upon comparative predictive power (i.e., beta weights), schizotypy was greatly superior to the borderline construct for dissociative predictability. Consequently, reconsideration of the existing paradigm that dissociation is more strongly associated with the borderline construct than with schizotypy is warranted.