This study sought to replicate the results of our earlier study, which were published in this Journal (Willoughby et. al 2011), that used mother-reported items from the Achenbach System of Empirically Based Assessment to develop a screening measure of callous unemotional (CU) behaviors for use with preschool-aged children. We further sought to extend those results by exploring the predictive validity of the CU measure with aggression trajectories in early-/mid- childhood. The current study involved secondary data analysis of the NICHD Study of Early Childhood and Youth Development (NICHD-SECCYD) dataset. Factor analyses included N = 1,176 children who participated in the age 3 year assessment of the NICHD-SECCYD. Predictive models included N = 1,081 children for whom four of the six possible teacher ratings of aggressive behavior were available from annual assessments spanning 1st–6th grades. Consistent with prior work, a three-factor confirmatory factor model, which differentiated CU from oppositional defiant (ODD) and attention deficit/hyperactive-impulsive (ADHD) behaviors, provided the best fit to the data. Among children with disorganized attachment status, the combination of high levels of mother-rated ODD behaviors and CU behaviors, was predictive of stable elevated levels of teacher-rated aggression from 1st–6th grade (predicted probability = 0.38, compared with a base rate of 0.07). These results demonstrate that CU behaviors can be reliably measured by parent report in young children and are dissociable from more commonly assessed dimensions of disruptive behavior. Three-year-old children who exhibit elevated levels of ODD and CU behaviors, and who have disorganized attachments, are at increased risk for exhibiting elevated levels of aggression across middle childhood. Results are discussed from the perspective of early assessment and intervention.