Resumen
This study reports the impact of examining either with digital or paper-based tests in science subjects taught across the secondary level. With our method, we compare the percentile ranking scores of two cohorts earned in computer- and paper-based teacher-made assessments to find signals of a testing mode effect. It was found that overall, at cohort and gender levels, pupils were rank-ordered equivalently in both testing modes. Furthermore, females and top-achieving pupils were the two subgroups where the differences between modes were smaller. The practical implications of these findings are discussed from the lens of a case study and the doubt about whether regular schools could afford to deliver high-stakes computer-based tests.