Psychometric Properties of the Modified Parenting Attitudes Measure Among African American Adolescents in Public

Background/Purpose: Adolescence is a challenging stage of development, particularly among African American youth living in low-income urban areas. Evidence suggests that knowledgeable and supportive parents increase the likelihood of successfully navigating this tempestuous period, but within the context of public housing developments marked by crime, illicit drug activities, violence and social disorder, these parental resources are compromised. Given the unique social and environmental conditions faced by families living in public housing, there is need to accurately assess the relationship between parenting practices and adolescent outcomes in a culturally sensitive way. Valid and reliable measurement reflecting how maternal and paternal parenting practices are similar, different, complementary, or additive in the study of youth outcomes are critical for building knowledge on supportive parenting practices for African American youth living in public housing. Therefore, we examine the psychometric properties of a modified Parental Attitudes Measure (PAM) looking specifically at the scales’ performance with maternal and paternal caregivers.

Methods: Data on the PAM was collected from the perspective of 660 African American youth ages 11 to 20 (M = 15.4) living in urban public housing across three northeastern states in the United States. Youth were asked to report on their attitudes about both maternal and paternal caregivers. Reliability was determined using internal consistency. An exploratory factor analysis (EFA) was used to examine the dimensions and loading of the construct with this population. A confirmatory factor analysis (CFA) examined the validity of the measure using four goodness-of-fit indices. Pearson correlations were used to test convergent construct validity with depression and self-efficacy.

Results: The EFA indicated that a two-factor structure was recommended based on eigenvalues greater than 1.0. All items loaded above .3 with excellent KMO and Barlett’s Test of Sphericity values for maternal [KMO=.914; Bartlett’s Test 3219.967(66) p = 0.00] and paternal [KMO=.945; Bartlett’s Test = 5649.614(66) p = 0.00] attitudes. Each factor demonstrated good reliability (Cronbach α of .86, .84, for maternal encouragement and supervision, respectively and .93, .92 for paternal encouragement and supervision, respectively). The goodness of fit indices for the CFA showed that with the exception of chi-square, the two-factor model had acceptable fit for mothers (χ²(53) = 234.791, p < .00; RMSEA = 0.075, CI [0.065, 0.085]; CFI = 0.97; SRMR = 0.04); and fathers (χ²(53) = 198.519, p < .00; RMSEA = 0.086, CI [0.058, 0.078); CFI = 0.99, SRMR = 0.02). The measure had acceptable convergent construct validity with maternal caregivers, but only partially with paternal caregivers suggesting differences in youth's attitudes on paternal parenting practices and its association with depression and self-efficacy.

Implications: Findings demonstrate that the PAM is a valid and reliable tool for examining maternal and paternal parenting attitudes among African American youth living in urban public housing. Findings suggest differences in youth's attitudes on parenting practices between maternal and paternal caregivers, and differences by parent in its association with self-efficacy and depressive symptoms. Future studies should include youth and their fathers in knowledge creation about paternal caregiving practices that support families in public housing communities.