Student performance on the Test of Scientific Literacy Skills (TOSLS) does not change with assignment of a low-stakes grade

Abstract

Objective

Response-validated multiple-choice assessments are used in college courses to assess student learning gains. The ability of a test to accurately reflect student learning gains is highly dependent on the students’ effort. Within our institution, lackluster student effort is common on response-validated multiple-choice concept assessments that are not included as a portion of the semester grade but are used to inform curricular changes. Thus, we set out to determine whether increasing testing stakes by assigning a grade on student performance had an effect on student score and self-reported effort. The Test of Scientific Literacy Skills (TOSLS) is a response-validated multiple-choice assessment used to measure scientific literacy in undergraduates. We administered the TOSLS to students enrolled in a general education Biology course, both during the first 2 weeks (pretest) and the last 2 weeks (posttest) of the course.

Results

Self-reported effort and TOSLS performance were significantly correlated in the ungraded cohort. This relationship did not exist in the graded sections. Our data indicate that assigning a low-stakes grade has no significant effect on mean student performance or self-reported effort on the TOSLS within our general education course.