Datasets

The Center for Academic Innovation (CAI) is responsible for creating and stewarding learning analytics datasets that provide unique research opportunities. We invite researchers across U-M to partner with us in conducting rigorous, responsible research into the efficacy and impact of learning technologies and higher education. All data access requests will be reviewed to ensure they:

As part of U-M’s partnerships with various online learning platforms, we manage large datasets depicting learner enrollment and engagement. We leverage these datasets to power platforms like Michigan Online, support operational needs like tracking learner progress through micro-credentials and make possible research into all phases of web-based learning. 

 

Educational Technology Datasets


Each of our educational technology platforms produces a dataset that we snapshot and can make available for research purposes. The datasets can be paired with data from both Canvas and the
Student Data Warehouse. We encourage you to read more about our Educational Technology portfolio to assess what platforms best align with your research interests, and reach out to us to discuss new research studies.

Online Learning Datasets

Coursera Phoenix
The higher education community experimented and iterated MOOCs over their first four years of life. In 2016, changes in design practices were matched by Coursera transitioning from their Spark technology to a new Phoenix platform, complete with mobile-friendly features and a new data structure. The Coursera Phoenix dataset includes all U-M MOOCs launched since 2016, their structure and design, and learner engagement and completion. 

Coursera Spark
Coursera and U-M first launched a partnership to provide Massive Open Online Courses (MOOCs) globally in 2012. Learning experiences launched between 2012 and 2016 are recorded in the Coursera Spark dataset. The widely publicized launch of MOOCs yielded millions of enrollments in these first courses and the data continues to provide a remarkable lens on the advent of the global MOOC experiment.  

edX

U-M joined edX as a partner school in 2015. Our edX open learning dataset includes records of all learning experiences launched on the platform, learner engagement, and completion. 

 

Global Surveys

At the beginning and end of each of our open learning experiences, we invite learners to answer surveys regarding their interests, motivation, and experience with the content. The surveys provide a rich set of data that can be paired with the platform engagement data to create a more holistic lens for research. 

 

Online Learning Data Warehouse
The Online Learning Data Warehouse is a single source for engaging with cross-platform MOOC enrollment and completion. We have collaborated with the ITS Data Innovations, Systems, and Cloud (DISC) team to extract data from the other standalone datasets and transform it into a common structure. The Online Learning Data Warehouse currently houses courses, course enrollment, and course completion data on the U-M open learning experiences on Coursera and edX platforms, while in-course engagement data can be accessed through the platform-specific datasets.