In Massive Open Online Courses (MOOC), it is impossible for instructors to teach thousands of students on a personal basis. While students are encouraged to discuss with each other using discussion forum and other tools, participations have not been active and frequent for most. Why is that so?
Three of the most popular MOOC platforms: Udacity, Coursera, and edX.
To answer this question, we conducted user research with interviews and surveys to understand students’ expectations and motivations underlying their peer learning behaviors and usages. We uncovered 3 main frictions caused by the learning context: 1) learning goals mismatch between students and peer activities, 2) excessive time and effort demanded by peer activities, 3) unmanaged students’ scale and diversity. We translated the findings into design opportunities to make peer learning more effective and engaging.
MOOCs have revolutionized the landscape of online learning, providing high-quality education to a massive scale of people around the world. Unlike physical schools, it is impossible to provide one-on-one teaching to every student taking MOOCs. Therefore, an obvious solution is to encourage students to learn from and with each other in MOOCs.
Despite encouragement and incentives from the instructors, as well as the presence of features to do so (e.g., discussion forum, Google Hangout, Facebook Groups, etc), multiple pieces of research have found that not only most students are not using them actively and frequently (through quantitative usage data), but there is no improvement in learning outcomes and engagement. It remains unclear how students perceive the act of learning with others in the MOOC context, and what could motivate them to do so.
In this project, we tackled the following problem: How to encourage and facilitate students from around the globe to learn from/with each other in the MOOC context?
Users & Audiences
This project directly benefits MOOC platforms, instructors, and of course, the students. Our lessons and implications allow them to understand why existing pedagogies and features are not very effective, and what can be done to make peer learning more engaging and effective. As more institutions are adopting this strategy, our findings could change the educational landscape by making peer learning more adaptable in massive online classes.
There were four members in this project. I was the research lead, together with three research advisors. As the leading researcher, I first reviewed research and reports related to the use of peer learning features in MOOCs, as well as established pedagogical framework to understand this phenomenon.
A fantastic resource to learn more about peer learning is Peer Learning in Higher Education by David Boud, Ruth Cohen, and Jane Sampson.
Through multiple meetings, the advisors assisted me in narrowing down the research gaps. After determining the specific questions to answer, I designed and conducted pilot interviews with 5 experienced MOOC students to unravel interesting general themes using the inductive thematic analysis approach. During the interviews, we focused not just on their behaviors, but also their underlying motivations, expectations, and assumptions made.
The themes strengthened our study focus, which we used to revise our interview questions and method for the actual interviews with 10 more students. Interview recordings and written notes were transcribed, analyzed, and triangulated with existing literature to generate a bigger and clearer picture. We concluded that there are three main contextual frictions impeding the effectiveness and engagement of peer learning in MOOCs. To validate our interview findings with a larger and more diverse student population (because MOOCs are huge!), I conducted another large-scale survey with 174 students from different countries and courses.
Main Findings & Suggestions
We discovered three points of frictions between how peer learning is currently being designed and how MOOC students expect peer learning to work.
- Existing peer learning features and activities are incorporated in a general manner. However, students expect peer learning to address a specific learning goal for them.
- Existing designs of peer learning features are inherited from previous forms of online courses. However, students find them time-consuming and effortful to use in MOOC.
- Existing peer learning activities include students regardless of their diversity and progress. However, students carry certain expectations of their peers in a group.
To reduce the above friction, we propose the following design implications after reviewing established literature on peer learning design.
- Differentiate and support both active and passive form of peer learning engagement. Both serve different purposes in MOOCs.
- Design peer learning activities and features that target and integrate with specific instructional activities (e.g., videos, quizzes, assignments, etc.) on both pedagogical and user interface level.
- Manage the scale of peer learning features by curating students’ diversity profile
- Reconsider the roles and instructors’ expectations of discussion forum
- Enable better grounding and expressivity in the forum
In the final stages of the project, where we have conducted all interviews but haven’t analysed the data, we faced severe time constraints as the deadline loomed near. With up to 20 hours of audio recording to transcribe in 2 weeks, we decided that in order to tackle this challenge without sacrificing the quality of the work, we had to be laser-focused on the main questions we sought to answer. I first spent a day looking back at the summaries of the research literature to get a clearer picture of the gaps and questions. This enables me to adopt a targeted approach while coding the interview and survey data. Looking back, it was a taxing period that involved transcribing and coding the data, generating and organizing the themes, and writing the findings themselves.
By going through this process, I learned the importance of focus and clarity in disentangling constraints, and I have been applying this mindset in my work since then.
Retrospective & Lessons Learned
The main lesson I learned in this project is the importance of being self-critical and open-minded with my own work. I made this mistake in the first iteration of this project, where we did not anticipate the difficulty of establishing generalisability on our interview findings in a massive online course context with just 15 interviewees. In retrospect, we could have anticipated this by being self-critical on the limitations of our methodology. In the end, we complemented our interviews with a large-scale survey. At the same time, I learned that we should have not been afraid to walk the extra mile (to do the survey) if it could lead us to a better path. The quality of the work should be the north-star, not the conveniences of the methods used.