Probing the Flipped Classroom: Results of a Controlled Study of Teaching and Learning Outcomes in Undergraduate Engineering and Mathematics
A flipped classroom reverses the paradigm of traditional lecture courses by delivering lectures outside of class ﾖ by means such as videos or screencasts ﾖ and using class meeting time for instructor-mediated active learning. This format has the potential to transform STEM education by increasing student time spent on what research has demonstrated to be the most effective teaching techniques (i.e. active learning) without sacrificing material coverage or educational scaffolding. Many educators are flipping their classrooms, but there is limited data on learning gains currently available. We have rigorously examined the impact of three instructors inverting two STEM courses, in engineering (thermodynamics) and mathematics (differential equations), by measuring student learning gains and attitudes towards the course material and comparing to a 'traditional' active learning control class.
Our expected measureable outcomes were:
1. Higher learning gains;
2. Increased ability to apply material in new situations (transfer);
3. Increased interest in and positive attitudes towards STEM fields (affective gains); and
4. Increased awareness by students of how they learn and strategies that support their learning (metacognitive gains).
Our hypothesis was that increased student learning would arise primarily because of the additional time that students will have with instructors actively working on meaningful tasks in class. The key activities were the measurement of student outcomes in both 'traditional' active learning classrooms and flipped classrooms; details are given in the 'Approach' section below.
The study design was composed of three components: (1) direct assessment measures specific to each of course/discipline in addition to indirect assessment measures; (2) comparison of control and experimental sections offered simultaneously (to reduce student demographic variability) using the same instructor (to limit instructor bias); and (3) direct assessment of learning gains and application both within the course and in downstream courses to determine if learning gains persist.
For nearly all measures across the three-year study, the flipped classroom model showed mostly equivalent results to the traditional active classroom model in terms of student performance. While these findings do not support original expectations of the inverted model, there are possible explanations for these results. It is possible that detecting differences in student performance may be difficult, since students at [school name redacted for review] generally have high academic achievement regardless of the classroom design and the culture of collaboration is already strong. Another possible explanation is that the intervention (i.e., the inverted classroom) was not distinct enough from the traditional active learning course sections and therefore was not a sufficiently strong intervention in comparison to the control. For the first two years of the study, the course formats were similar between the inverted and traditional classes: students encountered similar assignments, homework, and tests regardless of the class format. The main difference for students between the two classroom models was when they had access to the professors for questions, during the lecture (traditional) or while doing homework (inverted). The third year of the study relaxed this requirement but did not show significantly different results.
Our results suggest that the rearrangement of classroom activities and homework may not have a measurable effect on student performance. Some literature suggests that an expansion of the curriculum, rather than a mere rearrangement, is necessary to properly implement an inverted course model (Bishop & Vergeler, 2013). Our data also seem to support the idea that students are impacted the most when an active-learning style of instruction is used, regardless of when they are introduced to new content (in the classroom or at home through video lecture. For the final (fourth) study year, Engineering 82 and Math 45 courses will include an integrated ﾓbest practicesﾔ or hybrid model of instruction for all participating course sections, hence, eliminating distinct study groups. These results will be compared to results from the first three years of the study to see if student performance has improved.
The broader impacts are embedded in the possibility (based on our research results) that students may be equally served by 'traditional' active learning instruction and flipped classrooms. This would mean that instructors could make what they felt to be the best choice in their particular situation without concerns of diminishing student learning.
This project impact has been reasonably wide spread, as the educational community seemed to be hungry for controlled studies on flipped classrooms. Results have been disseminated through traditional peer-reviewed articles, the popular press, and EDUCAUSE case study, multiple conferences, invited presentations at a number of educational institutions and a workshop will be taking place in January 2016 at Harvey Mudd College.
Overall, we did not encounter many unexpected challenges other than one of the original co-PIs (a chemist) withdrawing from the work after taking an administrative position at HMC. We continued the work in engineering and mathematics and have recruited a biologist to run a reduced study in our required biology course.
Yong, Darryl, Rachel Levy, and Nancy K. Lape. 'Why No Difference? A Controlled Flipped Classroom Study for an Introductory Differential Equations Course.' PRIMUS Vol. 25 (919-933).
Lape, N. K., R. Levy, D. Yong, R. Eddy, N. Hankel. ﾓProbing the Flipped Classroom: A Controlled Study of Teaching and Learning Outcomes in Undergraduate Engineering and Mathematics.ﾔ Proceedings of the Annual Meeting of the American Society for Engineering Education, June 2015.
Lape, N. K., R. Levy, D. Yong, K, Haushalter, R. Eddy, N. Hankel. ﾓProbing the Inverted Classroom: A Controlled Study of Teaching and Learning Outcomes in Undergraduate Engineering and Mathematics.ﾔ Proceedings of the Annual Meeting of the American Society for Engineering Education, June 2014.
Popular Press available upon request.