How Classroom Flipping Affects Coast Guard License Students in Engineering

Abstract

Flipped classroom is an active-learning strategy in which student activities that are traditionally completed at home are exchanged, or “flipped”, with activities traditionally completed in class. In engineering, this approach usually means class time is used for student-led problem solving and interpreting results-a situation that promotes higher levels of cognitive work in the classroom under the tutelage of the instructor and peers. The flipped classroom model was recently instituted at State University of New York (SUNY) Maritime College in a junior-level engineering course with a student population of largely varying abilities and career goals, including students who seek U.S. Coast Guard licensure. The motivation for adopting the flipped classroom model was threefold: (1) The traditional classroom may be a disconnect for license students who often work in teams and spend many hours in active-learning environments like ship simulators and on a training ship. Thus, it is reasonable to suspect that the flipped classroom model may resonate well with license students who seem to thrive in these hands-on environments. (2) Prior research demonstrates the flipped classroom model benefits a wide variety of students with wide-ranging abilities, which aligns well with the course's student body. (3) Prior to flipping, the course was showing a concerning trending decline in student performance and a wide gap between student and instructor expectations, motivating the instructor to consider alternative teaching strategies. The objective of this paper is to estimate the effect classroom flipping has on engineering students, particularly license students, and determine the statistical significance of the observed effect. This objective is met by first retrieving the final student grades in the selected course in a timeframe that includes a pre- and post-flipping period (2016-2019), as well as the final student grades in a similar control course that was not flipped during the same time period. The data also include whether or not each student was a license student, which allows for the construction of three datasets: one of only license students, one of only non-license students (“interns”), and one of all students. Then the well-established difference-in-differences (DiD) technique is employed to each of the datasets in order to measure the effect flipping the selected course has on final course grades. Lastly, random inference is applied to calculate p-values and determine the statistical significance of the observed effect. It is found that license students see an average increase of +0.583 grade point average (on a 4-point scale), at a confidence level of 99%, which is a greater effect than what intern students see, which is a +0.474 increase on average, at a confidence level of 95%. Also included are comments and results of a poll that was completed by students in the flipped classroom meant to gauge their satisfaction in the course and what they believe the effect the flipped classroom had on them in meeting the course learning objectives.