Motivation: Most STEM courses use summative assessment almost exclusively, and this practice fails to foster student learning and engagement. To improve learning outcomes, we need to develop, test, and implement formative assessment practices.

Intervention: This study uses a proficiency-based assessment tool for statistics education in which students can resubmit assignments to demonstrate successful achievement of learning outcomes. The tool gives scores of either Mastery, Proficient, Developing, or Not Submitted.

Measured Outcomes: Students from the experimental (proficiency-based assessment) and control (summative assessment) group took the Comprehensive Assessment of Outcomes in Statistics (CAOS), a common final exam, and the Survey of Attitudes Toward Statistics (SATS) pre- and post-test. Students in the experimental group were scored based on their level of interaction with the assessment system (proportion resubmitted) and whether students received mastery or proficient scores on the first or latter attempts.

Results: There were no general performance differences between the two groups on CAOS and final exam. CAOS and exam scores were higher among students in the proficiency-based group who chose to resubmit assessments, but this result is likely confounded with student diligence and aptitude (proportion of resubmitted assessments co-varied with Math SAT). Students who achieved proficiency after multiple submissions performed just as well on the CAOS and exam as students who achieved proficiency on the first submission. The experimental group scored much higher on all six subscales of the SATS (particularly for competence and interest in statistics).

Why this is important: Meta-cognitive strategies, such as self-regulation, are crucial to student success, but not all students successfully use these strategies. This proficiency-based assessment tool allows students to test their understanding without penalty. The finding that students who achieved proficiency after submitting multiple times performed just as well as those who achieved proficiency on the first submission is particularly valuable because it demonstrates that students don’t have to be right the first time to achieve learning goals. Most educators would agree that students make errors, but most courses don’t provide opportunities to correct those errors. This article suggests that giving students multiple chances on graded assessments doesn’t hurt learning and improves student attitudes toward the domain.

Posner, M. A. (2011). The impact of proficiency-based assessment and reassessment of learning outcomes system on student achievement and attitudes. Statistics Education Research Journal, 10(1), 3-14.

For more information about the article summary series or more article summary posts, visit the article summary series introduction.