Study orientation tools are frequently grounded in the notion that achieving person-environment (PE) fit is key to academic success. Nevertheless, the literature reveals two notable gaps: the focus on predictive rather than explanatory role of PE fit within a broader set of variables, and its varying impact on outcomes across study programmes. This study aimed to address these gaps by investigating the relative importance of PE fit within a comprehensive set of pre-enrolment predictors to predict programme-specific persistence. We analysed data from 1305 prospective first-year students across five study programmes, with at least 200 students per programme. Data analysis included propensity score weighting and logistic LASSO regressions with cross-validation. The results indicated prediction accuracy in each programme ranging from 67% to 77% in the training data, which reduced to 50–75% in the test data, reflecting good prediction of persistence but challenges in predicting dropout. Inspection of the retained predictors revealed varying predictors across study programmes, with interest and skill fit variables representing the largest effects. This study underscores the necessity of programme-specific predictions to understand the relationship between PE fit and first-year persistence. The findings lay the groundwork for more personalised feedback in study orientation tools.
Document
OpenAccess
