Multimedia learning: principles of learning and instructional improvement in Massive, Open, Online Courses (MOOCS)

With good reason, many initial Massive, Open, Online Course (MOOC) studies conducted in the first three years of widespread MOOC hype and adoption have focused on retention rates and completion issues. No longer a new modality, many of the retention questions have now been answered as researchers provided skeptics with myriad examples of success stories and better perspectives on how to examine retention and student success in the massive space [Koller, D., Ng, A., Do, C., & Chen, Z. (2013); Kizilcec, R., Piech, C., & Schneider, E. (2013); Reich, J. (2014); Zheng, S., Rosson, M., Shih, P., & Carroll, J. (2015)]. To fulfill the promise and potential for MOOCs, the enormity of the scale must complement, rather than limit high quality learning outcomes. There has been extensive research (Richard Mayer, et al.) on enhanced learning using multimedia (words and pictures) presentations in clinical settings -- can we see the same success in a MOOC field setting? Consistent with the cognitive theory of multimedia learning (CTML) and Richard E. Mayer’s research with colleagues (Mayer & Bove, 1996; Harp & Mayer, 1998; Moreno & Mayer, 2000; Mayer & Jackson, 2005; Mayer, 2009), I found that learners in the Fall 2015 MOOC offering of “Microeconomic Principles” were able to build more meaningful connections between words and pictures than with words alone as reflected in their performance on practice quizzes across three different course modules. This finding has a number of implications for instructional design. First, we see that designing assessment feedback to only include verification feedback (acknowledgement of only a correct or incorrect answer) does not produce any positive impact on performance and should not be considered a useful treatment for students other than to simply verify their progress. Second, utilizing any type of instant elaboration feedback has an immediate impact on student performance. A text narrative providing the student with additional information about the misunderstood subject matter produces better student performance results, up to 3.4 times better, than a student who did not receive any elaboration feedback (text or multimedia). Third, designing quiz feedback to instantly (dynamically) deploy a multimedia video that covers the topic has the greatest impact on learning performance. Students who had the opportunity to learn the concept visually through the use of pictures, video and audio performed 5.3 times better than a student who did not receive multimedia feedback. This was true of all learners independent of age, gender, level of education and English-language ability. It was also true across four different types of questions reflecting the first four levels of Bloom’s taxonomy. There are a number of follow-up studies that will need to be conducted (discussed later in the dissertation), but these are important findings in a quasi-new delivery format that is still finding its bearings. The results are particularly significant in the MOOC space where scale is observed as an advantage despite its nuanced challenges. In a course with tens of thousands of learners, it is not possible for the instructor, or even teaching assistants and community forum managers to provide real-time content feedback. Spending more time on the already detailed design process for MOOCs would only be worthwhile if we had empirical evidence of actual impact on learner performance. As I conclude by discussing where massive, open, online courses may be headed next, multimedia quiz feedback can now be seen as one of a host of emerging design strategies in the massive space that promotes learning while embracing the scale of the course environment