The impact of short videos on student performance in an online-flipped college engineering course

The COVID-19 pandemic dramatically accelerated the shift to online learning and teaching (RO-L&T) in higher education. While video is a crucial component of online education, the quantitative effects of video length on student academic performance, particularly in online-flipped college engineering courses, remain largely unexplored. This research aims to address this gap by investigating whether short videos enhance student engagement and learning outcomes in such a setting. The forced campus closures created by the pandemic provided a unique environment for this study. The flipped classroom model, where pre-recorded videos are watched before class, facilitating in-class interactive activities, offers potential advantages. This study focuses on an engineering drawing course and hypothesizes that short videos will lead to increased student engagement (measured by viewing time) and improved final exam scores compared to long videos. The results have significant implications for the design of future online-flipped engineering courses, improving learning effectiveness and student experience.

Existing literature supports the potential benefits of flipped classrooms, emphasizing student-centered activities and increased interaction. The role of educational videos in this model is paramount, and video length is a key factor influencing effectiveness. Cognitive load theory suggests that shorter videos can reduce extraneous cognitive load, improving learning. Previous research on MOOCs and K-12 settings has indicated a preference for shorter videos and increased engagement, but quantitative improvements on exam scores in college-level credit-bearing courses, especially in an entirely online environment, have been scarce. This study bridges this gap by focusing on a fully online, credit-bearing engineering course during the pandemic.

This study used a quasi-experimental design with two groups: a long-video group (n=35) and a short-video group (n=30). Students were randomly assigned to the groups, and their average Gaokao scores (a national college entrance exam in China) were comparable, minimizing the influence of prior knowledge. The course was an Introduction to Engineering Drawing, delivered entirely online using a virtual learning platform. Weekly lecture videos (~55 min) were created, with the short-video group receiving versions trimmed to an average of 8 minutes per video. Student video viewing behavior was automatically tracked. The 2-hour interactive classroom sessions involved group discussions, tutoring, and Q&A. A five-point Likert scale questionnaire (Cronbach's alpha = 0.87) assessed students' perceptions of RO-L&T, video length, and learning experience. A final exam assessed learning outcomes. The Mann-Whitney U-test was used to compare exam scores and viewing time due to non-normal distribution, while a one-way ANOVA test compared Likert scale data.

The short-video group exhibited significantly higher engagement, with a 24.7% improvement in median viewing time compared to the long-video group (0.932 vs. 0.747 normalized engagement time). The short-video group also achieved significantly higher final exam scores, with a median score 7.4% and average score 9.0% higher than the long-video group (p<0.05). The short-video group demonstrated smaller variance in both engagement time and exam scores. The Likert scale results showed no significant differences between groups regarding internet skills or initial attitudes towards online learning or engineering drawing. However, the long-video group reported a higher frequency of rewinding videos and perceived the videos as longer than the short-video group. Both groups reported similar ability to focus and retain information, despite differences in exam scores.

The findings strongly support the hypothesis that shorter videos enhance student engagement and learning outcomes in an online-flipped engineering course. The significant improvements in both engagement and exam scores for the short-video group align with cognitive load theory, suggesting that shorter segments facilitate better information processing. The increased engagement in this credit-bearing course compared to previous MOOC studies likely reflects the greater motivation of students pursuing a degree. The consistency of performance within the short-video group highlights the positive impact of the video length on learning consistency.

This study demonstrates the significant benefits of using shorter videos in online-flipped college engineering courses. The improved student engagement and higher exam scores highlight the importance of video length optimization for effective online learning. Future research should explore additional factors influencing video effectiveness, such as video quality, student device preferences, viewing location, and learning activities. Replicating this study across different engineering courses and incorporating qualitative methods would enhance generalizability and provide richer insights.

The study's limitations include the inability to definitively confirm that students watched the videos attentively, as current technology could not monitor this factor. The sample size, while substantial, might limit the generalizability to other contexts. Future studies can address these limitations by using more advanced monitoring technologies and expanding the sample size to increase generalizability. The current study focused on a specific engineering subject, and exploring its application across different disciplines is needed.