Enhancing retention through learning analytics
1 like352 views
The document discusses how the University of South Wales employs learning analytics to improve student retention and success by analyzing both quantitative and qualitative data. It highlights the identification of key factors affecting student achievement, adaptation of data management practices, and collaborative initiatives to enhance educational outcomes. The focus is on using both internal activity monitoring and external survey data for predictive modeling and ongoing analysis in order to refine educational strategies.
Enhancing retention through learning analytics
- 1. © University of South Wales Enhancing retention through learning analytics Dr Jo Smedley University of South Wales September 2013
- 2. © University of South Wales “University learners sometimes encounter challenges with their learning which can lead them to quit. To enhance retention and success of all students, information technology has enabled the analytical review of considerable quantitative and qualitative learning data. This has informed the identification of several key factors with differential applications, for example, between subjects, between student age groups, which has led to the enhanced targeting of continuing initiatives to maximise overall achievement.” Abstract
- 3. © University of South Wales 3 Is your organisation maximising its information potential? Refining data rich, information poor (DRIP) systems to enhance client experiences Enhancing client experiences Data management Adjusting categories (JACS codes) Adjusting reporting times Cross-University initiative ++++++ External survey data UCAS Admissions Student Experience National Student Survey International Student barometer Internal survey data Module feedback Retention Success Activity Monitoring Virtual Learning Environments Estates Induction ++++++ Dr Jo Smedley Email: jo.smedley@southwales.ac.uk Collaborative opportunities Practitioner case studies Ideas for development Feedback on existing work
- 4. © University of South Wales Retention Induction Activities Internal Survey Data (module feedback, student representation, student experience surveys) Activity Monitoring (Blackboard Interactions, GlamLife Interactions, Missed QMP Assignments, Googlemail Interactions, Logons from student area, Tier 4 Signons, Estates info, Library info) Data Management (Target Setting, Data Sharing)
- 5. © University of South Wales Success & Satisfaction External Survey Reporting (NSS, PRES, International Student Barometer, DLHE, HESA) Data Management (JACS coding)
- 6. © University of South Wales The Undergraduate Learner Journey UCAS Admissions Module feedback x n Student representation End of year surveys x n National Student Survey
- 7. © University of South Wales The Postgraduate Learner Journey Admissions Module feedback x n Student representation End of year surveys x n PRES
- 8. © University of South Wales The International Learner Journey Admissions Module feedback x n Student representation End of year surveys x n International Student Barometer
- 9. © University of South Wales Big Data • Internal data • Activity monitoring • External data Activity monitoring Blackboard Interactions GlamLife interactions Number of missed QMP Assignments Googlemail Interactions Logons from student area Tier 4 sign-ons Estates info (entry etc) Student Representation Library interactions
- 10. © University of South Wales 10 Internal data Module surveys x n Student experience surveys x n Big Data • Internal data • Activity monitoring • External data
- 11. © University of South Wales 11 External data NSS PRES HESADLHE International Student barometerBig Data • Internal data • Activity monitoring • External data
- 12. © University of South Wales 12 Predictive modelling: retention )MonitoringActivity( data)Internal( Retention g f where:- • f and g are a multiplying factors to be determined through data analysis • internal data comprises reported formal and informal data from internal surveys, e.g. module feedback, student experience surveys • activity monitoring comprises data gathered from student interactions, e.g. VLE, Googlemail, Library, Estates
- 13. © University of South Wales 13 Predictive modelling: success/satisfaction )dataExternal( tisfactionSuccess/Sa h where:- • h is a multiplying factor to be determined through data analysis • external data comprises reported data in external league tables, e.g. NSS, PRES, International Barometer, HESA
- 14. © University of South Wales Continuing Work • Analyse categories of existing data to determine model factors • Collaboration – “What works” initiative • Impact • Further dissemination 14Email: jo.smedley@southwales.ac.uk