Experimental Design Chalk Talks, a Formative Assessment Employed in a Graduate Molecular Biology Course, Promotes Active Learning and Growth in the Competencies of Experimental Design and Science Communication

Abstract

There is a dearth of literature describing innovative ways to bring active learning activities into graduate life science classrooms. Therefore, we recently described experimental design chalk talks (EDCTs) as an active learning tool, outlined their use as a formative assessment, and shared resources (including a grading rubric) for others to implement this activity in their classrooms (manuscript in press: Heustis et al., Nature Biotechnology, expected December 2019). There is also a need for literature investigating the learning gains that such interventions yield, particularly in graduate life sciences education. This presentation outlines our evaluation of student performance on EDCTs and reveals the learning gains associated with using this activity in our course.For over a decade, we have utilized EDCTs in “Principles of Molecular Biology”, a graduate‐level first‐year molecular biology course offered annually at our private R1 medical school. We use the central dogma of molecular biology as the intellectual scaffold for presenting foundational principles of biochemistry as well as the mechanisms of action for relevant molecules and machines. We emphasize an understanding of how the evolution and application of classical and cutting‐edge techniques has driven discovery and the depth to which we understand fundamental molecular mechanisms. In developing and delivering EDCTs as assessments, students are challenged to respond to an open‐ended research question posed by a faculty expert. We have recently analyzed graded rubrics from 4 years of our course, capturing >600 EDCTs presented by our students (accounting for >300 pairs of rubrics for each of the students who have delivered talks twice each semester).Using quantitative approaches, we highlight the categories on our rubric on which students are most and least proficient, both initially and later in the semester (which we have discovered varies by question, but notably often includes controls); parameters on our rubric which we have discovered generally tend to be linked in student performance; changes in performance at both the student and whole class levels (between the first and second presentation in the semester) using statistical tests to look at changes in both mean scores and variance; and, an assessment of the reliability of the scores we received from the graders in our course, which we assessed by having two “expert graders” re‐grade a subset of videotaped presentations.Our results affirm not only that active learning activities can be incorporated into a graduate life science classroom but also that we can measure learning gains in this setting.Support or Funding InformationThe authors are grateful for funding provided from the Harvard Initiative for Learning and Teaching, in the form of a SPARK grant awarded to RJH and MJV.