.png)
A major study from EVA (Danmarks Evalueringsinstitut), conducted across first-year students at Danish universities, identifies four factors that predict dropout risk, academic outcomes, and student wellbeing. Researchers call them the FEST factors. Together, they offer senior leaders something rare: a research-backed framework for understanding retention that is specific enough to act on.
The pattern across all four is consistent. Each factor connects to retention, academic performance, and wellbeing simultaneously. They are not separate concerns requiring separate strategies. They are four dimensions of the same underlying dynamic.
Look at the wellbeing numbers again.
Academic engagement: 16 percentage points lower risk of poor wellbeing. Self-efficacy: 17 points. Belonging: 16 points.
Student wellbeing correlates as strongly with how students experience their academic work as it does with whether they feel socially connected. This directly challenges how most institutions have structured their wellbeing efforts.
The default assumption in higher education is that wellbeing is primarily a social and pastoral concern. Institutions invest in belonging initiatives, social programming, and peer support structures. These things matter. But the research suggests that academic engagement and self-efficacy are just as powerful drivers of wellbeing, and they respond to entirely different interventions: teaching design, feedback quality, clarity of expectations, and the signals staff send about whether they believe students can succeed.
Most institutions are starting with T. The data says start with F.
EVA's findings align with patterns visible in student feedback data across institutions using StudentPulse.
Engagement is not a stable level that institutions can assume and move on from. It moves. In data from Danish institutions over the past year, engagement scores show significant variation across the semester, dropping sharply at key pressure points before recovering. An institution measuring once a year will average across those movements and miss them entirely. The signal that matters is the dip, not the average.
Self-efficacy is where students express the most concern in their own words. Comments like "I don't feel confident I can do what's expected of me" and "I'm worried I won't be good enough" appear consistently in free-text responses, often months before any formal retention risk is identified. These are early indicators. They are available. But only if someone is looking at the right moment.
The gap between when these signals appear and when most institutions see them is where retention is lost.
The FEST research is valuable precisely because the factors it identifies are measurable and influenceable. But knowing they matter is only useful if institutions can see where they stand at the right moment.
Most cannot. A survey at the end of semester tells you how students felt about a period that has already passed. An annual report tells you what happened last year. By the time either arrives, the student at risk has often already made their decision.
Retention is not lost in the data. It is lost in the gap between when the signal appears and when the institution sees it.
StudentPulse measures the FEST factors continuously, through short check-ins timed to key moments in the student journey. AI processes the responses and surfaces priorities for the teams that need to act.
In practice, this changes three things for senior leaders.
Programme teams see engagement patterns at cohort level during the teaching period, not after it. A dip in week four is visible in week four.
Student services receive early self-efficacy signals before low confidence becomes disengagement. Outreach can happen at the moment it is most likely to help.
Leadership gets a live view that connects student experience to retention risk across programmes and intakes. Not a static report. A view you can steer by.
The FEST framework tells institutions what to measure. StudentPulse makes it possible to act on it while there is still time.
The full picture, including aggregated data from StudentPulse institutions and what the signals look like in practice, is available in our FEST Retention Report.
