This study addressed non-traditional student completion and attrition in a correctional academy. The study examined the influence of academic/institutional support factors, background characteristics, environmental support, academic outcome (academic performance), and psychological outcome (motivation and satisfaction) on persistence. The sample consisted of 102 students enrolled in a correctional academy in Central Florida between 1994 and 2001. Johnson's (1991) theoretical model of student persistence and attrition was modified for use in this study.
The modified instrument was examined by a panel of criminal justice training center directors and pilot tested on a subset of students in the correctional academy. The relationships between the variables and the amount of variance explained in the endogenous variables and outcome variable were analyzed using the path analysis multivariate technique and the Statistical Analysis System's Covariance Analysis of Linear Structural Equations procedure.
The study found that academic outcome (e.g., grades) had a statistically significant influence on attrition and persistence. Academic outcome had the strongest direct effect of any single predictor on attrition and persistence. The study also revealed that gender was moderately correlated ( r = -.43) with attrition and persistence. The descriptive statistics relating to academic/institutional support (climate of the environment) revealed that the correctional academy students need more interaction with instructors and coordinators to feel comfortable seeking help with both academic and personal issues. The paramilitary structure of the academy might inhibit some students from seeking help and consequently affect their ability to persist.
Although psychological outcome (or satisfaction and motivation) did not have a significant influence on persistence/attrition, it was correlated ( r = .12) with academic outcome. While the sample size was sufficient to meet the power criteria of the study, the study could be replicated on a larger population to increase the generalizability.
My Research