Abstract

In an attempt to understand what makes blended learning (BL) more effective than Classroom Instruction (CI), this research looked more closely at the Flipped Classroom (FC) model of BL. The FC takes a relatively consistent approach to course design by flipping what is traditionally done in the classroom (i.e., lecture) with what is traditionally done as homework (i.e., application).
Numerous studies have been conducted comparing FC with the CI on student achievement in higher education without conclusive results. To synthesize the literature, this dissertation implemented a systematic review and meta-analysis to measure the average effect size and the direction of the impact and to determine the conditions under which students learn more effectively. To ensure a transparent process the potential for bias in each step of a meta-analysis was acknowledged and addressed.
Through a systematic review of the literature from 2000 to 2017, 114 studies were included and 125 effect sizes were calculated. Using meta-analysis these effect sizes created a weighted mean effect-size of +0.30, which was statistically significant at p < 0.05 and educationally significant.
Study features were analyzed to determine if there were any attributes that made a difference but none were found to be significant. The use of quizzes, however, showed an interesting pattern and near significant difference (p = .058) when the effect sizes were grouped by STEM, non-STEM and Health-related disciplines. No publication bias was found, no outliers were found from the sensitivity analysis, and there was no significant difference between the effects from quasi-experimental and experimental designs.
While the FC significantly outperformed CI it was not to a greater extent than general BL outperformed CI. Future research is encouraged between levels of treatments, instead of between FC and CI, in order to provide more nuanced results about how to improve instructional design in future courses.