Search for: All records

Abstract While formative assessments (FAs) can facilitate learning within undergraduate STEM courses, their impact likely depends on many factors, including how instructors implement them, whether students buy-in to them, and how students utilize them. FAs have many different implementation characteristics, including what kinds of questions are asked, whether questions are asked before or after covering the material in class, how feedback is provided, how students are graded, and other logistical considerations. We conducted 38 semi-structured interviews with students from eight undergraduate biology courses to explore how various implementation characteristics of in-class and out-of-class FAs can influence student perceptions and behaviors. We also interviewed course instructors to provide context for understanding student experiences. Using thematic analysis, we outlined various FA implementation characteristics, characterized the range of FA utilization behaviors reported by students, and identified emergent themes regarding the impact of certain implementation characteristics on student buy-in and utilization. Furthermore, we found that implementation characteristics have combined effects on student engagement and that students will tolerate a degree of “acceptable discomfort” with implementation features that contradict their learning preferences. These results can aid instructor reflection and guide future research on the complex connections between activity implementation and student engagement within STEM disciplines. 

The General Biology–Measuring Achievement and Progression in Science (GenBio-MAPS) assessment measures student understanding of the Vision and Change core concepts at the beginning, middle, and end of undergraduate biology degree programs. Assessment coordinators typically administer this instrument as a low-stakes assignment for which students receive participation credit. While these conditions can elicit high participation rates, it remains unclear how to best measure and account for potential variation in the amount of effort students give to the assessment. To better understand student test-taking motivation, we analyzed GenBio-MAPS data from more than 8000 students at 20 institutions. While the majority of students give acceptable effort, some students exhibited behaviors associated with low motivation, such as low self-reported effort, short test completion time, and high levels of rapid-selection behavior on test questions. Standard least-squares regression models revealed that students’ self-reported effort predicts their observable time-based behaviors and that these motivation indices predict students’ GenBio-MAPS scores. Furthermore, we observed that test-taking behaviors and performance change as students progress through the assessment. We provide recommendations for identifying and filtering out data from students with low test-taking motivation so that the filtered data set better represents student understanding.