The Predictive Analytics Reporting Framework: Mitigating Academic Risk Through Predictive Modeling, Benchmarking, and Intervention Tracking Bill Bloemer, Vickie Cook, & Karen Swan University of Illinois Springfield Beth Davis and Ellen Wagner, Predictive Analytics Reporting (PAR) Framework
While Big Data raise expectations, student data drive big decisions in.edu
National Non-profit Multi-institutional Collaborative Institutional Effectiveness + Student Success
From Hindsight to Foresight
Game-Changer: Common Data Definitions PAR institutions jointly develop and agree to conform to a set of common data definitions for all variables in PAR data set. The PAR Framework common data definitions work as a Rosetta Stone of student success. PAR Framework team openly licensed and published the PAR common data definitions:https://community.datacookbook.com/public/institutio ns/par The PAR Framework project s release of the Data Definitions as a Creative Commons license is a major step forward and a possible competitive advantage over commercial competitors (Lowhendahl, Education Hype Cycle, 2014. Gartner Research)
Descriptive and Predictive Insight PAR Benchmarks Descriptive Analytics Cross Institutional Student/degree/major level insight into: 1. What did the retention look like for students entering in the same cohort 2. How does your institution compare to peer institutions / institutions in other sectors 3. How did performance vary by student attributes PAR Models Predictive Analytics Institutional Specific insight into: 1. What students are being retained over time? 2. Which students are currently at risk for completing and why? 3. Which factors are directly correlated to student success? 4. What is the predicted course completion rate for a particular program?
Value of Benchmarks from a Global Perspective The data will create conversations about challenges facing student success These conversations will lead to strategic initiatives, interventions, and best practices Best practices will lead to improving student success at the institution and at the national level
Predictive models reduce guesswork to find students at risk Percentage of students correctly predicted 100 90 80 70 60 50 40 30 20 10 30% Lift curves 0 0 10 20 30 40 50 60 70 80 90 100 Percentage of students Model: Non-Persisters No Model Model: Persisters
Predicting retention aimed at taking action - finding the most important factors
PAR anonymized ID Risk they will not be retained Actionable information at the student level 1 st, 2 nd and 3 rd most important factors contributing to risk
Student Success Matrix (SSM X ) Review Inventorying & categorizing student success interventions/ supports using a common framework Based on known predictors of risk and success In the context of the academic life cycle Addresses Now What? by linking predictions to action Enables cross institutional benchmarking Supports local and cross institutional cost/ benefit analyses. PAR Framework 2013
modeling retention & progression instructor behaviors Data Driven Institutional Response/Interventions learner characteristics course characteristics fit/learner perceptions of belonging learner behaviors retention/ progression other supports Data Driven Institutional Response/Interventions
predictors time connection entry progress completion learner characteristics learner behaviors fit/feelings of belonging other learner support course/program characteristics instructor behaviors
PAR Framework 2013 Comprehensive view completed SSM x
PAR Framework 2013
Examine interventions by predictor category PAR Framework 2013
Isolate interventions Find gaps PAR Framework 2013
Applying interventions at the point of greatest need/value A fundamental objective for developing common language and frameworks for reviewing student interventions is to apply the most effective interventions at the points of greatest need to remediate student risk. PAR has paved the way for creating common understanding of student risk and common tools for diagnosing risk The road to developing consistent and applied measurement to student impact of intervention will take time and vigilance.
Thank you! http://parframework.org @PARFRamework