Callous-Unemotional Behaviors in Young Girls: Shared and Unique Effects Relative to Conduct Problems

Abstract

Among girls, little is known about the shared and unique associations that callous-unemotional (CU) behaviors and conduct problems have with aspects of emotional and behavioral dysregulation and with parenting practices. This study examined these associations using a large community-based sample of young girls (N = 990). The findings revealed that hyperactivity–impulsivity and negative emotionality were positively and uniquely associated with conduct problems, but not with CU behaviors, after controlling for co-occurring conduct problems. Conduct problems were also positively associated with both generalized anxiety and panic/somatic anxiety. In contrast, CU behaviors were negatively related to generalized anxiety problems after controlling for co-occurring conduct problems. The results also indicated that conduct problems were more closely associated with harsh punishment and low parental warmth among girls with low versus high CU behaviors.

This work was supported by grants from the National Institute of Mental Health (MH056630, MH66167, MH07179) and the National Institute on Drug Abuse (DA012237) and by funding from the Office of Juvenile Justice and Delinquency Prevention. Dustin Pardini received support from the National Science Foundation (SES-0215551) through the National Consortium on Violence Research. Any opinions, findings, conclusions, or recommendations expressed in this article are ours and do not necessarily reflect the views of the National Science Foundation.

We thank the participants of the Pittsburgh Girls Study for their commitment to this research and our dedicated research team for their continued efforts. We also acknowledge the support of the Pittsburgh Public Schools as well as all other school districts that have made the collection of teacher data possible.

Notes

¹Because low internal consistency can reduce statistical power and increase Type II errors (i.e., decreasing the likelihood of detecting a significant effect), a power analysis was conducted examining the association between the CU construct and a predictor variable with the lowest reliability (low parental warmth α = .69). For a sample size of 990 and a significance level of .05, these results showed that there was still adequate power (99.8%) to detect a small effect (r ² = .10).

²The analyses presented here used untransformed variables. When the analyses were rerun using the square root transformations the results remained largely unchanged. These analyses are available upon request from the primary author.

Note: CU = callous-unemotional; P/T = best-estimate parent-teacher report; CP = conduct problems; P = parent report.

Note: CU = callous-unemotional; P/T = best-estimate parent-teacher report; CP = conduct problems; P = parent report.

Note: Different subscripts (a versus b) indicate that the standardized regression coefficients were significantly different at p < .01. pr = partial correlation controlling for demographic variables (girls' age, race, parental education, and receipt of public assistance); β = standardized regression coefficient controlling for demographic variables and the overlap between callous-unemotional (CU) behaviors and conduct problems (CP); sr ² = semipartial correlation squared controlling for demographic variables and the overlap between CU behaviors and CP.

∗p < .01.

∗∗p < .001.

Note: CU = callous-unemotional.

∗∗p < .001.