Preliminary validation of the dropout risk inventory for middle and high school students

Purpose: The aim of the present study is to describe the Dropout Risk Inventory (NU) and provide preliminary validation of its psychometric properties using a normative sample. Method: Participants were recruited from a large urban southwestern U.S. district with diverse demographics. Across the 343 students included in the analytic sample, the participants were primarily male (61.22%, n = 210) and Hispanic (44.02%, n = 151). We conducted factor analytic, reliability (including test retest), and predictivity validity tests were conducted. Results: Results of factor analyses found the support for the sub-scale structure of the DRI composed of 58 items and nine subscales (student-teacher relationships, behavioral disengagement, psychological engagement, academic and cognitive engagement, self-efficacy, economic stress, positive and negative peer influences, and goal setting and problem solving). Most sub-scales possessed good internal consistency reliability (alpha > 0.80) and test-retest reliability from grade 9 to subsequent grades. Predictive validity tests indicated consistent relationships (e.g., r's between 0.10 and 0.40, p < 0.001) between subscales and behavioral and academic subject performance grades. The economic stress and peer index subscales were found to be largely unrelated to these outcomes. Conclusions: Findings provide initial evidence for the Dropout Risk Inventory for use with middle and high school students. Findings suggest that the DRI is reliable and possesses useful predictive validity with respect to behavioral and academic risk.