Student Self-Selection and Test Repetition

Abstract

Student self-selection in deciding to repeat a test was examined by contrasting the test performance of students taking the Scholastic Aptitude Test (SAT) as juniors and again as seniors with the test performance of students taking the SAT only once as juniors. Estimates of expected test performance on a common initial administration in the junior year were derived from separate equating sections and background variables. Residuals of observed minus expected test scores revealed statistically significant differences between students who took a single administration of the SAT as juniors and students who took the same initial administration but also repeated the test as seniors; the initial observed scores of students later repeating the test were consistently lower than their expected scores for both the verbal and mathematical sections. These results indicate that self-selection occurs when students decide to repeat a test and that score changes among these students reflect negative errors of measurement on the initial test administration as well as other factors.