HEPI / Advance HE breakfast seminar: What is the best way to measure student success?
Overview
On Thursday 12 February, HEPI and Advance HE host their first in-person seminar of the 2026 series at the Houses of Parliament.
There are many barometers by which to evaluate student success – including academic achievements, student satisfaction, levels of wellbeing, graduate outcomes, retention rates, other sector metrics and qualitative judgements – but what are the best ways?
Policymakers are often drawn to one-dimensional outcomes but how should we measure student success in a broader sense? And can developments in AI and more sophisticated use of data improve how we measure success?
Speakers
Professor Chris Millward
Interim Director for Fair Access and Participation, Office for Students
Professor Karen Stanton
Vice-Chancellor, University of the Arts London
Alex Stanley
Vice President, Higher Education, National Union of Students
Rose Stephenson (Chair)
Director of Policy and Strategy, HEPI
Partner
Comments
Brenda Shankey says:
Can I express an interest for invite please.
Reply
Replies
Carole Cox says:
Thank you Brenda. I have now emailed you directly.
Kind regards
Carole
Reply
Jonathan Alltimes says:
What should be the pattern and to what extent the student fits the pattern depends on ones purpose. Just because we think we understand what causes the pattern, does not mean we can control it. The language of ‘outcomes’ commonly used in policy language refers to an implicit model of stochastic processes, whereby different combinations of the same variables can cause different distributions of outcome effects, but is that the best model? One of the many weaknesses in the use of metrics is the application of the average or mean, another is how scale changes the power of the metric, as does the timeframe. The most valuable idea to remember is the qualitative precedes the quantitative, so thinking in terms real practical examples from direct experience should guide our thinking about student success and a historical model should be articulated.
Reply
Add comment