BackgroundThe Adult-Adolescent Parenting Inventory – 2 (AAPI–2) is a widely used assessment of parenting attitudes associated with child maltreatment for research, clinical services, and program evaluation. The overrepresentation of African Americans in child welfare systems paired with a substantial body of research describing the contextual, cultural, and ethnic differences in African American parenting underscore the need to assess the validity of assessment tools in specific populations. ObjectiveThis study specifically addresses the construct and predictive validity of the AAPI –2 form A (pre-test) within an urban, low-income African American sample.Participants and Setting: The AAPI –2 was collected as an evaluation tool for a free parenting class administered by a non-profit child abuse prevention agency in a large mid-Atlantic city and de-identified for this study. Participants identified as Black (n = 218) and resided within city limits. The majority were female (72 %) and unemployed (52 %), where 22 % were married or partnered. MethodsConstruct validity of the measures was assessed via Confirmatory Factor Analysis (CFA) of the developers 5– factor model and, due to the complexity of the model (40 items and 5 subscales), fit indices of each individual subscale were also assessed. To assess predictive validity for child maltreatment, participant scores were matched to CPS records. ResultsThe developers 5– factor model was less than optimal fit for the data. When assessed individually, the E– Power and Independence subscale revealed a poor fit, where D– Role Reversal was a good fit, with remaining 3 subscales a fair fit. There were no associations between AAPI–2 scores and CPS records. ConclusionsThe uneven fit of the 5 subscales of the AAPI–2 with this sample suggests varying success in measuring the underlying constructs. Further, the lack of predictive validity in this sample recommends a nuanced interpretation when used as a risk assessment tool. The specific applications of the subscale constructs are discussed.