The Evidence on Teaching

A Marked Improvement? A Review of the Evidence on Written Marking (EEF/Oxford)

Elliott, V., Baird, J., Hopfenbeck, T., et al. · 2016

grade Dreviewindependentunreplicated
Sample
Systematic review + survey of 1,382 teachers
Population
Primary/secondary; much underlying evidence from higher ed/EFL.
Design
The audit of the largest feedback time-sink. The follow-on FLASH marking RCT (103 schools) cut workload but lost its GCSE attainment arm to COVID.
Key findings
Teachers spend hours per week marking; the causal evidence that written marking raises attainment is close to EMPTY. What little exists supports: comments without grades, actionable targets, class time for pupils to respond, and abandoning acknowledgment/triple marking. Mark less, mark better.
Genetic confound
N/A (evidence audit).
Replication notes
The negative meta-finding (no robust marking evidence) remains unrefuted; EEF's 2021 feedback guidance retained it.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Written marking on attainmentqualitativeevidence base too weak to estimate; no large K-12 RCTs of marking as practicedunknownn/aunclearnot-applicabledomain-skill
Acknowledgment marking (ticks, 'good effort')qualitativeunlikely to enhance progress; grading every piece may REDUCE marking's impactmixedshort-termunclearend-of-treatmentdomain-skill

Cited by