Date of Award


Degree Type


Degree Name

Master of Arts (MA)




Predictors of academic achievement among urban low-income African American adolescents have primarily been investigated by examining "main effects," or limited interactions with conventional statistical techniques. This paper adds to the literature by examining the factors that influence academic outcomes among this population within an ecological systems framework. This allowed for a comprehensive understanding of how numerous protective and risk factors, across ecological settings, interact to influence academic outcomes.

Optimal Data Analysis (ODA) was employed to create prediction models for mathematic and reading achievement. ODA allowed for the examination of a vast number of variables in one statistical model without increasing statistical error. In total, 111 variables across seven constructs (e.g., individual level characteristics, psychopathology, family structure, family functioning, social support, life stressors, and community/neighborhood) competed within the ODA model for the strongest predictor of academic outcomes (both success and failure). In addition, ODA allowed for the creation of the "pathways" towards academic outcomes by creating classification tree models. Data was collected among a final sample of 167 low-income fifth through eighth grade urban African American adolescents (53% females and 47% males). Parent report, self-report, and in vivo accounts of the adolescents' daily experience (e.g., daily distress) were collected during a one week time frame.

Contrary to expectation, family structure and family functioning variables were weaker predictors of academic outcomes than community variables. The strongest predictor of both mathematic and reading achievement was school factors (e.g., school socioeconomic status). In addition, common factors in the literature predictive of academic outcomes were not found to be associated with academic achievement in our sample (gender, family income). Several individual level characteristics' (e.g., academic self-efficacy, social problems, and openness to other racial groups) main effects were predictive of academic achievement.

When accounting for ecological contexts, family functioning variables were found to interact with each other and with each individual's school to predict mathematic outcomes. For example, within the context of a low socioeconomic status (SES) school, increased time spent with family may be protective of academic outcomes, if the student's family is not engaging in high-risk behaviors (e.g., high levels of parental alcohol consumption). With regard to reading achievement, individual level characteristics (e.g., delinquency, social problems, openness to other racial groups) were found to interact among each other in the context of specific school environments. The most consistent predictor of academic achievement was the SES of the school a student attended. Within the school context, characteristics at a Microsystem level were found to interact and either protect, or place students at greater risk for academic achievement.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.