<p>Items from two measures of antisocial cognition (moral neutralization and cognitive impulsivity) served as indicators in a series of confirmatory factor analyses (CFAs) and random intercepts cross-lagged panel model (RI-CLPM) analyses. Two samples of middle school children served as participants in this study: 824 (392 boys, 432 girls<b>)</b> youth (mean age = 11.21&#xa0;years) from the Pocono Bullying Study (PBS) and 2,550 (1,213 boys, 1,337 girls) youth (mean age = 12.13&#xa0;years) from the Gang Resistance Education and Training (GREAT) study. One- and two-factor CFAs revealed that fit was significantly better in the two-factor model where moral neutralization and cognitive impulsivity were treated as separate constructs. There were no significant differences in fit between a one-factor model and two-factor model in which items were randomly assigned to factors. An RI-CLPM analysis was performed to test for reciprocity over three separate time periods. Eight out of eight cross-lagged correlations involving moral neutralization and cognitive impulsivity achieved significance across two samples of early adolescent respondents. These results provide evidence of a prospective bidirectional relationship between within-person latent variable measures of moral neutralization and cognitive impulsivity that have potentially important theoretical, practical, and policy implications.</p>