Abstract
The tenure system in the United States places significant importance on teaching effectiveness. To date, students' evaluations of teaching (SETs) have been the reigning mechanism for assessing effective teaching. However, prior work has shown that SETs are often biased against underrepresented groups and minorities. The present study analyzes options for effective teaching assessments, which include evaluating final grades and measuring the differences between students' pre- and post-tests (normalized gain) using standard instruments. The content area and the instrument used in this study originated in the computational thinking field, which has a widespread presence in engineering, where minorities are at a disadvantage. This study obtained a total of 88 student participants from four sections of an introductory engineering course at a Southwestern institution. The study utilized a computational thinking diagnostic (CTD) to inform the course teaching approach (the intervention). Results show that (a) normalized learning gains correlated moderately with SETs, (b) final grades correlated strongly with SETs, (c) final grades correlated strongly with normalized learning gains, (d) the educational intervention based on the CTD significantly affected student learning, and (e) SET comments affect evaluations. The implications include the notion that standardized instrument-driven instruction and evaluations can increase the success of minorities on both sides of the classroom. The purpose of this manuscript is to invite the Heliyon readership to get involved in the development of related instruments and to incorporate these measures of learning into their instruction so biases are avoided or minimized.