Correlates of Externalizing Behavior Symptoms Among Youth Within Two Impoverished, Urban Communities

Abstract

This study examines whether risk factors associated with child externalizing behavior symptoms differ between two similar low-income, urban communities, using baseline parent data of 154 African American youth (ages 9–15) participating in the Collaborative HIV-Prevention and Adolescent Mental Health Project (CHAMP) family program. Separate multiple regression analyses of each city sample indicated that greater child externalizing symptoms were associated with increasing parenting hassles for New York families (n = 46), but greater parent mental health symptoms for participants in Chicago (n = 108). Understanding such distinctions between communities is an important first step toward tailoring services to unique community needs.

KEYWORDS:

• child mental health
• externalizing behavior symptoms
• risk factors

Funding from the National Institute of Mental Health (R01 MH 63662) and the W.T. Grant Foundation is gratefully acknowledged. In addition, this work would not have been possible without the contributions of the CHAMP Collaborative Board members and participants.

Notes

Note. *p < .10; **p < .05; ***p < .01.

^a Sample size is less than 46 due to missing data.

^b Sample size is less than 108 due to missing data.

Note. *p < .10; **p < .05; ***p < .01.

^a Numbers do not add to 46 due to missing data.

^b Numbers do not add to 108 due to missing data.

*p < .10; **p < .05; ***p < .01.

Note. *p < .10; **p < .05; ***p < .01.

^a Pooled B, F, and R ² values were manually computed by averaging statistics from the individual regression analyses of each of the 10 imputed datasets.

^b The “micombine” command in Stata version 10.1 only produces z-score values for pooled multiple regression estimates.

Note. *p < .10; **p < .05; ***p < .01.

^a Pooled B, F, and R ² values were manually computed by averaging statistics from the individual regression analyses of each of the 10 imputed datasets.

^b The “micombine” command in Stata version 10.1 only produces z-score values for pooled multiple regression estimates.