This chapter focuses on the construction of risk indices and their use in predicting correlates of child maltreatment by exploring the relationship between the indices and the outcome measures of the Adolescent Parent Services Evaluation. This 3-year evaluation, which began in the fall of 1987, compared seven new parent programs in seven States offering education support services for pregnant and parenting adolescents. A total of 488 adolescents participated in the study. Using a nonequivalent control group design with each site serving as a comparison group for the others, researchers determined the impact of services by testing the adolescents within 30 days of intake, at the close of intensive services, and 6 months after the close of services. The risk indices were based upon staff assessment of adolescents at intake. Assessment instruments included the Family, Friends, and Community Questionnaire, Parenting Questionnaire, Questionnaire on Infants and Young Children, Youth Self-Report, and Coopersmith's How I Feel Questionnaire. Items from the staff assessment were grouped into economic stability, potential for abuse or neglect, primary relationships, attitude toward services, and health measures. Findings indicate that the risk indices were not good predictors of the scores of adolescents on self-administered questionnaires measuring correlates of child abuse and neglect. Results of regression analysis, however, suggest that the indices had some construct validity. Specifically, the best predictor of all subcomponent and total scores assessing social support was primary relationships; potential for abuse and neglect was the best predictor of subcomponent and total scores evaluating parenting practices, knowledge, and expectations; and economic stability was the best predictor of the score on self-esteem. 16 references and 8 tables.
Do you have something you think is appropriate for the library? Submit Library Resources.