Mapping student engagement in a massive open online course (MOOC)
ISBN 978-85-88783-11-9
Authors
1Quinn, S.; 2Robinson, A.
1THE PENNSYLVANIA STATE UNIVERSITY Email: sdq107@psu.edu
2THE PENNSYLVANIA STATE UNIVERSITY Email: acr181@psu.edu
Abstract
In the past several years, the massive open online course (MOOC) has appeared as an alternative form of distance instruction offered by institutions of higher education. Many universities have embraced MOOCs as a means to educate a broader range of learners, as well as a novel mechanism for promoting a school's visibility, brand, and value in a time when university expenditures face increased scrutiny and risk of cuts. Recent research has debunked the idea that MOOC students are otherwise uneducated individuals in the developing world who have no chance of receiving any other kind of post-secondary education. Indeed most of the people who take MOOCs have already earned a college degree. But though we know a fair amount about about who takes MOOCs, we know much less about the geography of their participation: where people sign up for MOOCs, where people actually complete MOOCs, and how learners in different parts of the world approach the MOOC with specific strategies and intentions. The focus of our research is to characterize and explore spatial patterns of learner engagement with MOOCs. Here we make progress towards this goal by mapping student locations, attributes, and learning behaviors from a MOOC called Maps and the Geospatial Revolution. This MOOC, offered by Penn State University for the first time in 2013, has attracted over 73,000 enrollments to date. The course is a 5-week introductory class on Cartography and GIScience. As one of the first geography-focused MOOCs, it attracted over 48,000 students in its first offering, with over 3,000 completing the course with a passing grade (a seemingly low, but actually common rate of completion for a MOOC). These students represented over 200 countries. Data was collected throughout the course in the form of discussion forum posts, quiz scores, grades from assessments, and basic demographic information on enrolled students. Our research associates these course participation statistics with the students' locations (geocoded using their IP addresses), revealing a geography of learner engagement in MOOCs that was previously unexplored. For example, among students who complete the first quiz, we observe spatial patterns of dropout rates that show us where students tended to skip the remaining quizzes. We see a cluster of students in eastern Europe who are much less likely to drop out of the course after the first quiz than their counterparts on the west coast of North America. We have also identified other informative geographies of learner behavior such as forum posting frequency and number of times each quiz was attempted. Our approach aggregates the geocoded student points into hexagonal bins of roughly equal area that can then be filtered and compared. This helps preserve anonymity, while ensuring that the figures are normalized by similar areas. For example, if we want to look at the rate of student engagement in the course forums, we can tally the number of students in a bin that posted at least once during the course, and divide it by the number of students in the bin who enrolled in the course. To avoid high percentages resulting from low-enrollment areas, we can filter the bins so that we only display bins where some minimum number of participants enrolled. Our hypothesis is that geographical differences will reveal distinct approaches toward learning in the 'Maps' MOOC. The long term goal of our project is to develop an interactive framework in which the geography of MOOC learning engagement can be comprehensively evaluated across space and time. We seek findings that can benefit designers of MOOCs and educators of all kinds who teach on a global scale and need to adapt educational materials to a worldwide audience. We expect our findings will shed light specifically on learning engagement in a massive course on cartography, and can help learning designers better understand MOOC student behaviors that appear in the global classroom.
Keywords
education; MOOC; cultures