Effects of parenting, father absence, and affiliation with delinquent peers on delinquent behavior among African-American male adolescents

Adolescence,  Spring, 2003  by Mallie J. Paschall,  Christopher L. Ringwalt,  Robert L. Flewelling

• E-mail
• Print
• Link

RESULTS

Descriptive Statistics

Descriptive statistics are provided in Table 1. The mean age of the adolescents in 1996 was 15.6 years. Fifty-four percent of the adolescents were receiving free or reduced-price school lunch. More than half (53%) of the adolescents reported that they were not living with a father or father surrogate. Descriptive statistics indicated that key study variables (e.g., adolescents' delinquent behavior, mothers' monitoring of sons' behavior) had adequate variability, and all multi-item scale measures demonstrated adequate internal consistency (Oronbach's alpha [greater than or equal to] .85).

Bivariate Analyses

Correlations among study variables are presented in Table 2. The 1997 measure for adolescents' delinquent behavior was associated with mothers' monitoring of their sons' behavior (r = - .25, p < .01) and mothers' perceived control over their sons' behavior (r = - .35, p < .01) in the expected direction. Neither mothers' communication with their sons nor mother-son relations was significantly associated with the 1997 measure for delinquent behavior at the .05 level, though the relationship with mother-son relations approached statistical significance (r = - .14, p = .07). Father absence was not associated with the adolescents' delinquent behavior. Adolescents' affiliation with delinquent peers was associated with the 1997 measure for delinquent behavior (r = .48, p <.01) and with several of the parenting measures in the expected directions. Receiving free or reduced-price school lunch was modestly associated with the 1997 measure for delinquent behavior (r .16, p <.05).

Multivariate Analyses

Results of multiple regression analyses for the entire sample are provided in Table 3 (standardized beta coefficients). After adjusting for age, SES, father absence, and the 1996 measure for delinquent behavior, mothers' perceived control over sons' behavior was the only parenting variable predictive of adolescents' delinquent behavior (beta = - .24, p <.01). Together the parenting variables explained an additional five percent of the variance in adolescents' delinquent behavior. Adolescents' affiliation with delinquent peers also was a significant predictor of delinquent behavior, explaining an additional two percent of the variance in delinquent behavior. The addition of this variable to the regression model did not, however, attenuate the effect of mothers' perceived control over sons' behavior.

Results of regression analyses conducted separately for mothers and sons in father-absent and father-present families are presented in Table 4 (unstandardized beta coefficients). The effect of mothers' perceived control over their sons' behavior appeared to be stronger in father-absent families (beta = - .40, p < .01) than in father-present families (beta - .28, ns), both before and after including affiliation with delinquent peers in the regression models. However, additional analyses with an interaction term indicated that these differences were not statistically significant. None of the other maternal parenting variables were predictive of delinquent behavior in either father-absent or father-present families. Affiliation with delinquent peers was predictive of delinquent behavior in father-absent families (beta = .55, p = .05), but not father-present families (beta = .53, ns); however, this difference was not statistically significant. Noteworthy was the differential association between our SES measure an d delinquent behavior in father-absent versus father-present families. Receiving free or reduced-price school lunch was more strongly associated with delinquent behavior in father-absent families (beta = 1.09, p < .05) < .05) than in father-present families (beta = .25, ns); this difference was statistically significant.