| Description |
We developed an introductory chemical engineering design laboratory course coupling traditional, hands-on, and virtual learning techniques. The course is centered on a set of interactive, online simulations mimicking laboratory experimentation, which the students complete for homework prior to completing a corresponding hands-on experiment as a team in the lab. These simulations track all student input, ranging from mouse movements to field alterations. We analyzed these data, in order to determine how student behaviors correlate with success in this course. In this analysis, we determined that the time of day and how early students submit their homework are strong predictors of their success on the assignment and in the course; we also used this analysis to uncover some common student misconceptions about the theory behind the simulations. Additionally, we developed an online system to serve as a hub for a variety of course material, including the interactive simulations, automatically graded homework problems, student peer evaluations, safety training quizzes, pre- and posttests, and more. We also devised a grade normalization technique designed to transform scores to a desired mean and standard deviation while preserving scores of 0% and 100%. We use this transformation to help compare student scores between sections of our course taught at different times or by different instructors. We performed regression techniques to help identify which student characteristics are predictive of students' final scores in the course, using this transformation and a wide variety of student performance and iv demographic data. In this analysis, we found that the strongest predictors of student success were receiving high student peer evaluations and submitting homework assignments earlier relative to their due date. |
