A new study by Raquel Gutiérrez-González and Alvaro Zamarron, neurosurgeons and educators affiliated with La Paz University Hospital and the Autonomous University of Madrid, tackles this question with unusual granularity. Their work, published in BMC Medical Education, links the digital footprint left by each medical student on each video lecture directly to that student’s accuracy on each corresponding final-exam question. Rather than treating engagement as a vague aggregate, the researchers connected the dots at the level of individual videos and individual test items, offering one of the most finely detailed portraits to date of how watching behavior relates to learning outcomes in a flipped medical course.

The setting was an undergraduate Neurosurgery course delivered in a flipped format across two consecutive academic years, 2021/22 and 2022/23. The course relied on 51 video-based lectures covering 11 topics, hosted on the Edpuzzle platform, a tool that allows instructors to track exactly how much of each video each student watched and how much time each student spent with it. In total, the researchers analyzed records from 51 students in the first cohort and 44 in the second, pooling 95 learners. The final examination contained 26 items, each of which could be mapped to the video material that prepared students for it. Every exam answer was dichotomized simply as correct or incorrect, with unanswered questions counted as errors.

The analytical machinery behind the study was correspondingly careful. The authors computed two engagement metrics for every student-video pair: the percentage of the video viewed and the minutes of time spent. They then examined associations with exam accuracy at three distinct levels, the individual item, the individual student, and the topic, using point-biserial and Pearson correlations. To compare the two cohorts, they applied Fisher’s r-to-z test. To model the probability of answering an item correctly, they turned to logistic regression with standardized predictors, and they went a step further by adjusting standard errors for the fact that multiple exam items cluster within each student, using a cluster-robust estimator and generalized estimating equations. A pre-specified sensitivity analysis addressed a subtle problem: students who were excluded from the primary sample might simply have disengaged entirely, so the researchers reclassified them as complete non-engagers with zero percent viewed and zero minutes, expanding the analytic sample to 111 and 114 students across the two cohorts.

The headline finding is consistent in direction across every analysis. Both engagement metrics were positively associated with exam accuracy in both cohorts, and the differences between the two years were not statistically significant. When the two cohorts were pooled, the association proved small at the item level: the correlation between percentage viewed and item accuracy was just 0.08, and between time spent and accuracy a nearly identical 0.07, although both were highly significant at p less than 0.001 given the enormous number of student-item pairs involved. In practical terms, engagement with a specific video explains less than one percent of the variance in whether a specific related exam item is answered correctly.

Zoom out from individual items to whole students, however, and the picture sharpens considerably. At the student level, the correlation between overall percentage viewed and overall exam performance reached 0.32, and the correlation for total time spent was also 0.32, both significant at p equals 0.002. A correlation of that magnitude corresponds to roughly ten percent of shared variance, a moderate effect by the standards of educational research, where countless confounding factors, from prior knowledge to test anxiety to general motivation, muddy every signal. The convergence of the two cohorts on essentially identical effect sizes adds a degree of confidence that this is not a one-off statistical fluke but a reproducible pattern in how students learn from video in this setting.

When both engagement metrics were entered into a single multivariable logistic model on the pooled sample, an instructive subtlety emerged. The two metrics are deeply intertwined, correlating with each other at the item level at 0.77, and this collinearity, reflected in a variance inflation factor of 2.4, meant the model had to arbitrate between two largely redundant signals. Percentage viewed emerged as the stronger and more informative indicator, associated with an 18 percent increase in the odds of answering an item correctly for every standard deviation of additional viewing coverage, a result significant at p equals 0.031. Time spent, by contrast, lost its statistical significance once percentage viewed was accounted for. The lesson for learning-analytics enthusiasts is clear: how completely a student watches matters more than how long, a distinction that raw platform dashboards reporting minutes rarely make.

The authors’ statistical rigor also tempers any temptation toward overclaiming. When standard errors were adjusted for the clustering of exam items within students, the univariable association for percentage viewed held steady, with an odds ratio of 1.21 and a 95 percent confidence interval of 1.06 to 1.38, but the multivariable estimate became non-significant, with a confidence interval spanning 0.95 to 1.45. Meanwhile, the sensitivity analysis that reclassified all excluded students as complete non-engagers left every association unchanged in direction and slightly stronger: the student-level correlation rose to 0.36, and the multivariable odds ratio rose to 1.23 with p equals 0.004. In other words, the most conservative reading of the data still supports a genuine positive association, and the least favorable assumption about missing students does not undermine it.

Equally important is what the study explicitly does not claim. Because the design is correlational and retrospective, greater engagement with the videos is associated with, but cannot be assumed to cause, better performance. It remains entirely possible that the strongest students, already destined for high exam scores, are also the most diligent video watchers, rather than the videos themselves lifting performance. The authors are candid that the item-level effects, at r of 0.08 to 0.10 and less than one percent of variance, should not be interpreted as educationally large. Still, the study’s ethics approval from the Puerta de Hierro University Hospital Ethics Committee, its adherence to the Helsinki Declaration, and its transparent reporting of every statistical caveat place it firmly on the constructive side of the reproducibility ledger.

For educators designing flipped classrooms, the practical implications are tangible. Embedding lectures on platforms like Edpuzzle that log viewing coverage gives instructors a measurable, meaningful signal of student engagement, one that correlates moderately with exam outcomes at the student level and could flag disengaged learners long before test day. The finding that percentage viewed outperforms raw time spent suggests that dashboards should prioritize completion coverage over stopwatch metrics. And the modest item-level effects serve as a healthy corrective to techno-optimism: watching a video closely improves the odds on a related exam question, but no single video transforms an exam. In the end, this Madrid neurosurgery course offers a quiet but valuable data point in the debate over digital learning: students who watch to the end tend to know more, and now educators can see it in the data, one video and one exam question at a time.

Subject of Research: The association between video lecture engagement and exam performance in a flipped undergraduate neurosurgery course

Article Title: Watching to learn: video engagement and item-level exam performance in a flipped classroom

Article References: Gutiérrez-González, R., & Zamarron, A. (2026). Watching to learn: video engagement and item-level exam performance in a flipped classroom. BMC Medical Education. https://doi.org/10.1186/s12909-026-10553-8

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10553-8

Keywords: flipped classroom, video-based lectures, medical education, learning analytics, exam performance, student engagement, Edpuzzle, neurosurgery, logistic regression, correlational study, active learning, Watching

The study exploited a natural feature of the Spanish medical school system. Students are assigned at the start of their degree to one of four academic units, each with its own teaching hospital and faculty, and this assignment remains fixed throughout their training. All four units follow the same curriculum, share identical learning objectives, and sit the same final examination simultaneously. That structure allowed the researchers to run a quasi-experiment: three units continued with conventional PowerPoint-supported lectures, while the fourth adopted the flipped classroom from 2021 onwards. Because the same instructors taught their respective units across all three years, instructor variability was effectively eliminated, and any drift in outcomes could be attributed to the evolving design rather than a change of personnel.

The flipped model was not static. In its first year, students watched video-based lectures and completed online questionnaires at home, then spent class time on problem-based and cooperative learning activities built around clinical cases. In the second year, the team consolidated all eleven course topics into a single interactive presentation on the Genial.ly platform, linking videos hosted on Edpuzzle and quizzes created with Wooclap, alongside learning objectives and take-home messages. In the third year, informed by student feedback and engagement analytics, every video was re-edited to run under six minutes—the duration some researchers consider optimal for keeping attention—while the content itself remained unchanged. A short quiz of two or three questions drawn from the MIR examination, the compulsory national test that gates entry to medical residency in Spain, was added to the end of each in-class session to sharpen the stakes for senior students.

Academic outcomes were measured through the shared final examination, which contained 26 Neurosurgery questions alongside Neurology items. Only the Neurosurgery results were analyzed, avoiding contamination from the parallel subject. Questions were all multiple-choice with objectively determined correct answers, eliminating subjective marking, and scores were adjusted for negative marking—correct answers earned one point, blanks zero, and errors cost 0.33. Across the full sample, students taught with the flipped classroom answered, on average, one more question correctly than their lecture-taught peers, 19.2 versus 18.2, a difference that was statistically significant (p = 0.009). The adjusted final exam score showed a similar gap of 1.2 points, 18.1 versus 16.9 (p = 0.005). Dropout rates, by contrast, were statistically indistinguishable between the two arms—8.0 percent in the flipped group versus 5.7 percent in the traditional group—suggesting the more demanding methodology did not push students away.

Those differences may sound modest, but in the context of medical education research they carry weight. Most published studies of the flipped classroom in medicine enroll well under 400 students, and many rely on comparisons with historical cohorts rather than concurrent controls, making it impossible to separate the effect of the methodology from cohort quality or year-to-year examination variation. A large scoping review of educational innovation in clinical neurology found that only 28 percent of studies even included a comparison group, and most stopped at measuring student perceptions rather than actual knowledge gains. The Madrid team’s design—a concurrent control group in every year, a large sample, objective scoring, and longitudinal follow-up—addresses precisely the gaps that reviewers have repeatedly flagged as undermining confidence in the field.

The engagement data told a more complicated story. Participation was logged across three channels: the proportion of each video watched on Edpuzzle, completion of the Wooclap questionnaires, and attendance at in-class sessions. Participation held steady between the first and second years but dropped noticeably in the third year, even as the course materials were being polished to a level students rated more highly. The researchers offer several possible explanations, none of which can be confirmed with the data at hand. One candidate is the novelty effect: the flipped approach was genuinely new in 2021, and as it became routine over three years it may have lost some of its power to excite. Another is the social proof effect, the tendency of students to engage more when they see their peers doing so. Perhaps most intriguing is what the authors call a paradoxical effect of optimization—by making videos shorter and materials instantly accessible, the team may inadvertently have encouraged a just-in-time approach to study, in which students felt free to consult content at the last moment rather than prepare systematically before class.

Correlation analysis added nuance to the picture of what drives learning. The percentage of video a student watched correlated positively with exam performance, as did the proportion of embedded quiz questions answered correctly, but both relationships were weak, with Spearman coefficients of roughly 0.20. Video engagement explained only about four percent of the variance in examination scores. The message is that watching the videos is part of the machinery, not the engine itself. This aligns with the satisfaction survey results, in which the video lectures were consistently rated the most demanding component of the course, while the clinical cases worked through in class emerged as the clear favorite, followed by the interactive questionnaires. Notably, satisfaction ratings rose steadily across the three years—video ratings improved especially after the six-minute edit—even as participation declined, meaning students appreciated the refinements even when they used them less uniformly.

The authors frame their findings through several theoretical lenses. The flipped classroom does not succeed because it embodies a single educational theory, they argue, but because it coherently integrates several at once: constructivist active learning, social collaboration through cooperative work, cognitive load management via short, rewatchable videos, and situated authenticity through real clinical cases. Clinical disciplines may be particularly fertile ground because they naturally supply the complex, authentic problems that activate these principles simultaneously. Case-solving activities also push students up Bloom’s taxonomy toward applying and analyzing rather than merely remembering, and prior reviews have identified case-based learning as a powerful driver of clinical reasoning and decision-making skills. In the Madrid experience, the problem-based sessions with peer interaction and frequent instructor feedback were described as crucial to the method’s success—arguably more so than the out-of-class videos that give the flipped model its name.

The study is candid about its limits. Students were not randomized to conditions, a structural constraint of the institution rather than a design choice, so definitive causal claims are off the table. The authors note mitigating factors: assignment to academic units is administrative, demographic characteristics were similar across groups, dropout rates did not differ significantly, and all units shared the same curriculum and simultaneous examination. Still, subtle differences in student populations or in the enthusiasm and pedagogical skill of individual instructors cannot be fully excluded. Satisfaction and participation data were not collected in the three lecture-based units, so it remains unclear whether the flipped group’s higher satisfaction reflects a genuine preference or the freshness of a different approach. And the drop in participation in the final year resists full explanation, limiting the conclusions that can be drawn about engagement over the long term.

Even so, the practical implications are clear enough for educators weighing the investment. Building a flipped course demands substantial upfront time and effort, especially from instructors, but the payoffs—better exam performance, high student satisfaction, reusable materials, and professional development—suggest a good return for motivated teachers and supportive institutions. The approach uses fewer resources than high-fidelity simulation while being more demanding than plain lecturing, and its sustainability ultimately depends on an institutional culture that values and rewards teaching innovation. The research team proposes two priorities for the field: external validation through multicenter studies spanning different universities and disciplines, and evaluation of whether flipped learning changes later clinical behavior and even patient outcomes—the higher rungs of the Kirkpatrick evaluation ladder that undergraduate studies can rarely reach. In the meantime, the Madrid experiment offers a template for how educational innovations should be tested: not as a single dramatic intervention, but as a methodology continuously tuned, measured against concurrent controls, and honestly reported with both its gains and its puzzles intact.

Subject of Research: Flipped classroom methodology in undergraduate neurosurgery medical education

Article Title: Strengthening medical education through the flipped classroom: evidence from a three-year quasi-experimental study

Article References: Gutiérrez-González, R., Zamarron, A., & Royuela, A. (2025). Strengthening medical education through the flipped classroom: evidence from a three-year quasi-experimental study. Journal of New Approaches in Educational Research, 14(1), Article 26. https://doi.org/10.1007/s44322-025-00050-z

Image Credits: AI Generated

DOI: 10.1007/s44322-025-00050-z

Keywords: flipped classroom, medical education, neurosurgery, quasi-experimental study, problem-based learning, student engagement, video lectures, gamification, active learning, student satisfaction, exam performance, educational innovation