Abstract
Admission processes to access higher education should include the necessary tools to comprehensively measure the competencies and skills related to academic success.
Previous studies have highlighted that relying solely on cognitive tests presents various disadvantages and alternative solutions have been explored, including the application of other types of instruments aimed at capturing not only cognitive competencies. The objective of this study is to analyze the concurrent validity of a classic criterion used in admission tests, such as high school GPA (HSGPA) and cognitive measures, as well as a more innovative instrument called curriculum-sampling tests (CST) that are hypothesized to reflect a combination of cognitive ability, motivation, study time, and tacit knowledge. CSTs have been shown to predict first-year GPA as effectively as HSGPA, but little is known about their relationship to other admissions criteria.
Nine tests were administered to 1,184 applicants during the 2024-25 admission processes for specific degrees at a private university in Spain. Specifically: a) four cognitive tests, which were computed as a single score in cognitive competence, b) four non-cognitive tests, and c) a CST specifically designed.
Of the 1,184 applicants, 369 students enrolled at the university. Correlation and regression analyses were conducted to identify which measures predict HSGPA and CST. The results revealed that: a) predictors of HSGPA were total cognitive score (β=.65, p < .001), organizational skills (β=.13, p = .022), and self-efficacy (β=.13, p = .024); and the model accounted for 47% of the variance; b) predictors of CTS in Psychology were total cognitive score (β=.37, p < .001), and self-efficacy (β=.17, p = .012); and the model accounted for 17% of the variance. The only predictor of CTS in Pharmacy was the total cognitive score (β=.25, p = .009); and the model accounted for 5% of the variance. Predictors of CTS in Medicine were the total cognitive score (β=.29, p < .001) and critical thinking disposition (β=.12, p = .002); and the model accounted for 9,7% of the variance. No significant predictors of CTS emerged in Nursing, but the total cognitive score showed a trend (β=.12, p = .077).
The results of this study show that HSGPA and CST both exhibit strong relationships with cognitive and non-cognitive aspects relevant for success in university. However, for non-cognitive aspects, the relationships are weaker and less consistent. The results suggest HSGPA is useful in admission processes, as it captures cognitive and non-cognitive components of candidates. Similarly, CST appears to be a promising tool to evaluate a combination of cognitive and non-cognitive competencies, although it shows differences across degrees that require further analysis.
This study provides evidence of the importance of including measures in admission tests that capture non-cognitive components of participants, with CST showing great potential. Emphasizing broader admission criteria allows institutions to gather diverse validity evidence to support the intended use of the obtained scores, as well as to gain insights into prospective students that enable universities to implement concrete actions to foster specific competencies or skills.
| Oral presentation | Evaluating the concurrent validity of traditional and alternative measures used in university admission processes in Spain |
|---|---|
| Keywords | University admission, validity evidence, competencies |
