Evaluation of blended learning strategies from exam outcomes

Context: The advantages of online course delivery include easy access to course content utilising multimedia tools to enhance the student educational experience, more rigorous control over course content, and data logging of student online interactions with course content. While higher educational institutions embrace the blended learning model, there are relatively few studies evaluating the effect of blended learning strategies on exam outcomes. We have developed a more objective evaluation procedure, based on statistical modelling the effect of student WAM (weighted average mark) and online lesson activity on exam outcomes. Purpose: We wish to evaluate whether blended teaching strategies translate into better exam outcomes. Approach: This retrospective study compares the effect of blended learning on exam outcomes with a historical control group. Exam scores for traditional face-to-face lectures (2012-2014) were compared to blended course delivery (2015-16). ANOVA was used to assess whether the change to blended learning affects exam outcomes. The effectiveness of online learning was quantified by correlating student online activity with exam outcomes. Results: Student WAM (weighted average mark) was the most significant predictor of final exam score. While there were significant differences between the exam results for each year, we were unable to detect an improvement in exam outcomes after introduction of blended learning. However online study time for each student was an independent predictor of final exam outcome. Conclusions: While we did not detect an improvement in exam outcomes following introduction of online delivery of course content, WAM and online study time were predictors of final exam outcomes. The correlation between online study time and exam marks suggests that online activities were effective.