Enhancing retention interventions with real-time data – predictive learning analytics [167] Home : ALT Online Winter Conference 2018

Enhancing retention interventions with real-time data – predictive learning analytics [167]

Public Group Active 4 years, 1 month ago

Wed, Dec 12 2018, 10:00am – 11:00am

Description

Ulster University is piloting Blackboard Predict (Predict), a predictive analytics solution which uses historical student data and outcomes to identify and compare characteristics with the current cohort of students.

Ulster is using Predict to help inform interventions designed to increase the numbers of students completing a programme of study within a specific time frame and increase numbers of students who progress in and beyond education.

Predict ingests over 200 data elements about each student from Banner, Ulster’s Student Records System, and combines it with interactions in Blackboard Learn to specify a random forest model which is compared against previous students who have exited with known outcomes.

Predict enhances existing longitudinal, research informed, intervention strategies by providing real-time predictions which are recalculated as teaching progresses throughout the academic year.

Implemented in Spring 2018 our evidence base is developing however we are seeing significant benefits in the conversations we are having about data literacy, data hygiene, data cleansing and data informed decision making. Our expectations for the project are not revolutionary and we accept that a prediction is not a prophecy but an opportunity to target conversations with students who show similar data characteristics to previous students who may have failed a module or programme.

This session will be an opportunity to share what we have learnt to date and recommend resources that we have used to help us with policy, governance and roll out.

Go Back

Supported by