[ARCHIVED] Predictive Analytics

[fra2007]

Dear All,

Our University is looking into doing some predictive analytics for student performance.  I wanted to get some suggestions and ideas from people who have or in the process of implementing such a thing at their universities.  Where to start? What's worked? What hasn't? Are there any guides?

I understand this is a very vast subject and that there are possibly many ways to accomplish such a thing, but I wanted some feedback from the community on where to start.

Thank you,

Fraz Ali

[bhewitt797]

yes, i too would like to use this

[DanielleRose]

@fra2007 @bhewitt797 Do you already have some sort of analytics in place? i.e. analytics that help you understand the current behaviour so you can have an idea of what to want to predict. Whether it is what sort of grades the student is likely to have in a specific course or the final outcome or whether the student is likely to withdraw, etc.. Check out Linsights which is a learning analytics platform that provides you quite a lot of these answers.

[Chris_Munzo]

"Predictive Analytics" is a broad subject.  Solutions range from year-long integration efforts that extract/analyze data from all of your school systems, to off-the-shelf solutions that can be up and running in matter of days that focus on which students are currently at risk.  We provide a solution -- Dropout Detective -- that falls into the latter category.  We've been a partner of Canvas for over seven years and you can find out more about us by either searching the Community or by visiting our Partner Page:  https://community.canvaslms.com/t5/Partner-Listings/Partner-Listing-AspirEDU-Inc/ta-p/428557