Content area
Abstract
The objective of this study was threefold: (1) to identify variables that correlate with dropout in a residential vocational education institution, (2) to predict which applicants would drop out, (3) to shed light on the influence of qualitative, contextual factors in research with disadvantaged Hispanic, Black, and White youths (ages 18--24). This study fills gaps in psychological assessment and vocational education literature by investigating dropout in residential vocational education institutions and calling into question the validity of the Wonderlic Personnel Test, Rosenberg Self-Esteem Scale, and the Holden Psychological Screening Inventory for use with disadvantaged youths.
There were 147 participants in the study (50% Hispanic, 25% Black, 15.8% White, 6.7% Asian, and 2.5% Native American). I used Point Biserial correlations to assess the relationships between dropout and variables associated with dropout in like contexts. I used logistic regression in an attempt to discriminate between students who were likely to drop out and those who were not. The four predictors in the study were general mental ability, self-esteem, psychopathology, and antisocial traits.
The results of this study suggest that predictors of dropout in secondary schools may not generalize to residential vocational education programs. None of the predictors yielded significant correlations with dropout or high enough reliability coefficients to warrant their use in decision-making in vocational education institutions. Additionally, researchers cannot assume that disadvantaged adolescents can follow instructions, read or grasp the nature of assessments, or provide accurate information about themselves. Disadvantaged adolescents may need detailed instructions and guidance when completing psychological questionnaires. Fifty-four participants produced invalid WPT profiles, which calls into the assessment's use with disadvantaged adolescents with potentially low reading ability or English fluency. The results may also preclude data collection in a group setting with disadvantaged adolescents. Disadvantaged adolescents may not engage in research unless they receive incentives they value and an explanation for the study that is relevant to their lives. The results highlight an overall need for increased focus on contextual factors in research and in the provision of institutional services.