Protective factors for inner city adolescents at risk: Family factors and social support

Abstract

The national estimated school dropout rate is 25%. A wide variety of variables have been associated with the occurrence of school dropout. This study compared three groups of adolescents on a continuum of risk for school dropout: adolescents in the mainstream school program (n = 91); adolescents in an at risk program (n = 78); and adolescents in an alternative evening program (n = 25). Participants were low-income, inner-city, African-American youth from the same high school. The purpose of this study was to develop a model that could discriminate between the groups identified. The study focused on family variables and social support variables. Measures included the Family Adaptability and Cohesion Evaluation Scale (FACES II), a subscale of the Perceived Social Support Scale (PSSS) which measured peer support, a direct adaptation of the PSSS which measured support from adults outside the immediate family, and a questionnaire that gathered demographic information and information previously associated with school dropout. A discriminant function analysis was performed. Participants in the at risk program could be discriminated from participants in the mainstream and alternative programs in that they were more likely to repeat a grade, be older, be a parent or parent to be, have a higher grade point average and come from less cohesive families. Students in the alternative program were more likely than students in the mainstream to be males, be at a lower current grade, have lower intelligence scores, and lack adult and peer support. Interestingly, they were more likely to be satisfied with family cohesion and adaptability, and come from more adaptive families. Results suggest that adolescents at the highest level of risk experience more educational difficulties, including interactions with peers and adults, and adolescents at a lower continuum of risk are more likely to experience difficulties in the home environment. Classification rates for the three groups were as follows: 40% for the alternative program, with a prior probability of only 12%; 68.7% for the risk program, with a prior probability of 40%; and 82.7%, with a prior probability of 48%. The model greatly increased correct classification of membership into each group over that of chance.