Evidence Brief · Research
Does testing help students learn, or only measure them?
The EdJournal · 4 min read · June 2026
Strong evidence: it does both, but its underused power is as a learning strategy. Actively recalling information builds retention far more than re-reading does.
The testing effect is among the most replicated findings in cognitive psychology. A systematic review by Agarwal and colleagues (2021) found retrieval practice consistently benefits learning across real schools and classrooms. Each act of recall strengthens the memory trace, making future recall more likely.
The key shift is conceptual: here, “testing” is not assessment. It is the learning itself. Low-stakes quizzing, brain-dumps and recall prompts are study events, not grading events — and they work even though they feel less fluent than reviewing notes. Spacing is the natural partner: the same study time, distributed across sessions, retains far better than the same time massed into one.
What the data shows
Weeks–months
longer retention for students using retrieval practice versus re-reading — yet testing is still mostly treated as a measurement tool, not a learning one. (Roediger & Pyc, 2012; Dunlosky et al., 2013; Agarwal et al., 2021)
What it means for schools
- Replace some re-reading and review with frequent, low-stakes recall.
- Separate practice testing from graded assessment so retrieval feels safe.
- A 2025 Frontiers in Psychology study confirmed the benefit with 5th graders in ordinary classrooms — this is not lab-only.
What we don’t yet know: format effects (free recall vs. multiple choice), feedback timing and ideal frequency are still being refined — the core effect is settled; the optimisation is ongoing.
The EdJournal’s view
What teachers call a quiz, cognitive scientists call a learning event. The cheapest, best-evidenced lever in the classroom is also the one most likely to feel like wasted time.
Education. Intelligence.
