A conversation with Sandeep Jayaprakash…
A few weeks ago I reached out to Sandeep Jayaprakash in order to connect and learn more about the work that he has been doing in the field of learning analytics. Today, we met via Google Hangout and he shared his project with me.
The learning analytics project that we discussed was the early alert system that was developed at Marist College and this work was published in the inaugural issues of the Journal of Learning Analytics which is published by SoLAR out of Athabasca University in Canada. Early Alert of Academically At-Risk Students.
The goal of this open source learning analytics project was to build upon the original framework proposed by SoLAR in 2011.
Sandeep explained that there are several agencies that are working to develop standards for data compliance SoLAR is one of these agencies, as well as IMS Global, and ADL (through the TIN Can API or Experience API). Once there are data standards across systems, much like the banking industry or medical records then the data mining will be much more standardized rather than customized.
The project takes a look at a wide range of data dimensions (trends and patterns) regarding learning. The software is quite complex and can handle many complex tasks starting with extraction, transformation and analysis during which point is handles missing values. Then using matching learning the data mining looks for correlations after being trained using historical data and becomes “smarter” the more data it churns. The final output categorizes the learners into levels of high, medium and low risk of failure.
After building the model to be ideal for Marist College, they tested the model using a couple of community colleges and a HBCU and found the model to be somewhat (65% – 80%) portable with modifications being necessary to improve the accuracy of the early alert system based upon the particular student population.
While there is still much work to be done this team is leading the charge on some very innovative thinking and it was my pleasure to talk with Sandeep. I should note that Sandeep is a member of a larger project team and he continually reminded me that the his team is very talented. This project was funded by an Educause and Bill and Melinda Gates Foundation Grant.
Jayaprakash, S. M., Moody, E. W., Lauría, E. J. M., Regan, J. R., & Baron, J. D. (2014). Early Alert of Academically At-Risk Students: An Open Source Analytics Initiative. Journal of Learning Analytics, 1(1), 6–47.