Probabilistic analysis of academic engagement: Exploring student motivation through a massive open online course 

Importance Tools for engaging students in higher education include the use of online learning management systems (e.g., Moodle) intended to support face-to-face classroom learning and Massive Open Online Courses (MOOCs). MOOCs involve online learning environments that allow an unlimited number of participants open access – yet the completion rates are dismal. Educational psychology researchers are now showing interest in the MOOC context as one way to advance understanding of student motivation.Purpose Using self-determination theory (SDT) as a framework, this study advancesunderstanding of context-specific learning experiences by providing a probabilistic description of basic psychological need satisfaction within a MOOC on dinosaurs (DINO 101).Methods Students (N = 1037) from 151 countries voluntarily completed one questionnaire on the SDT basic psychological needs of autonomy, competence, and relatedness. In order to reveal the best structural understanding of SDT within a MOOC learning environment, data analysis involved the development of a Bayesian Network (BN) probabilistic model.Results Participants (50% female, 50% male) ranged from 18 to 74 years of age (59% under 45). The highest frequencies among participants were university graduates (31%) and participants between the ages of 25 and 34 (27.1%). We developed a BN graphical model to achieve the best Bayesian Information Criterion value (BIC = -3400.1), resulting in a probabilistic accuracy of 77.41%. This model revealed connections among competence (e.g., I had the chance to show my capacities during DINO 101) and autonomy (e.g., I have been able to freely decide my own pace of learning during DINO 101): participants with high autonomy had an 80.01% probability of having a moderate level of competence. Relatedness, though distinct from the autonomy and competence relationship, consisted of intra-connections among items.Limitations This research relied on data from 10% of the total number of students who enrolled in DINO 101 at any given time. Data on students who left the course part way through were missing and including this valuable perspective in future research could help inform course design efforts and retain more learners.Conclusions Overall, participants responded positively to the MOOC offering of DINO 101. Given the nature of the MOOC design and previous SDT research, it was no surprise that autonomy and competence revealed strong inter-connections. Most interesting was the finding of relatedness as a distinct need. To increase MOOC appeal and retain more learners, we recommend MOOC designers and educators embed more activities that tap into relatedness (i.e., discussion boards).