Predicting performance of pharmacy calculations assessments via an algebra-based pre-test and other variables using a logistical regression model: A pilot study

Objective: Pharmacy calculations are crucial for practice and licensure examinations. This study aimed to be the first to assess the external validity of the correlation between an algebra-based pre-test, previously published, and calculation assessment performances. A secondary objective was to identify variables associated with poor assessment performance. Methods: An 18-item survey and an 18-question algebra-based pre-test were distributed to 144 first professional-year (P1) candidates on the first day of their fall 2024’s semester. This was used to characterize student-specific variables (e.g. work experience). The survey included dichotomous (yes/no) and rating (5-point Likert scale) items; the algebra-based pre-test included open-response items. Data were analyzed using linear regression and logistic regression to determine variables associated with passing (≥70%) or not passing (<70%) assessments. Pearson correlation coefficients were computed between the algebra-based pre-test and assessment performance. Results: Ninety-one candidates (63%) completed the survey, and 139 (97%) completed the algebra-based pre-test. Variables associated with passing assessments (p<.05) included no prior work experience (adjusted odds ratio=6.6, confidence interval=1.3-34.3) and algebra-based pre-test performance >15/18 points (adjusted odds ratio=5.9, confidence interval=1.6-21.8). Algebra-based pre-test performance was moderately correlated with assessment scores (Pearson correlation coefficients: 0.47 for assessments one and two, 0.4 for assessment three). Conclusions: The algebra-based pre-test had a comparable correlation to assessment performance to previously published results and could be a tool to identify those at risk for poor pharmacy calculation assessment performance across the academy. Future studies warrant utilization of this algebra-based pre-test to target interventions to those at high risk of poor performance on calculations content.