In a 16-week quasi-experiment, A-Level STEM students receiving high-frequency automated marking scored 16.8 percentage points higher on a blind-graded final mock exam than control students with human marking¶
Claim · All claims
Evidence · 1 study · causalr2·q2·i3large
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i3 Controlling for Z-scored pre-test baselines, the intervention group achieved an adjusted post-test mean of 81.2% versus 64.4% for the control group (F(1,139) = 24.32, p<.001, η²p = .149). → Matey Yordanov 2026
Evidence¶
Matey Yordanov 2026¶
Matey Yordanov, Mikhail Bychkov, and Andrei Kuchma. (2026). The Effect of High-Frequency, Automatically-marked Formative Assessments on Student Outcomes in A-Level Sciences. https://arxiv.org/abs/2607.23566 doi:10.48550/arxiv.2607.23566
q2 · i3 · causal · r2
ANCOVA on post-test mock scores in a 16-week quasi-experiment with intact classes (intervention N=72, control N=70), Z-scored pre-test as covariate and cluster-robust standard errors. The article reports a statistically significant main effect with effect size η²p = .149, and adjusted means of 81.2% (SE=1.2) versus 64.4% (SE=1.4).
"The analysis revealed a statistically significant main effect,F(1,139) = 24.32,p<.001,η 2 p =.149."
Discussion¶
Related Claims¶
- Step-by-step automated annotations accelerated correction of procedural misconceptions, with intervention students rectifying fundamental errors weeks before control peers — related
- Automated marking reduced assessment turnaround from 11.2 days to 0.1 days and enabled intervention students to complete 4.1 times the practice volume of controls — related
- Intervention students reported significantly higher exam confidence than controls (M=4.3 vs 3.4 on a 5-point scale, d=0.76) — related
- Student interpreters who practiced with an automated scoring system outperformed the control group on total interpreting scores at post-test — related
- Non-layered AI feedback produced significantly higher post-test performance than layered AI feedback after controlling for pre-test — related
- Peer assessment through tests, marks, and grades produces learning gains and reduced anxiety, with most participants reporting learning benefits — related
- STEM students using real-time AI feedback achieved nearly double the mean gain score of students receiving conventional human-led instruction — related
- Post-test mean learning outcomes were 81.81 in the ChatGPT class versus 70.45 in the conventional class — related
- AI-generated (GPT-4) feedback produced no greater linguistic progress than human tutor feedback among ENL students over six weeks — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
automated-marking-16-point-mock-gain-a-level-stem |
title |
In a 16-week quasi-experiment, A-Level STEM students receiving high-frequency automated marking scored 16.8 percentage points higher on a blind-graded final mock exam than control students with human marking |
description |
In a 16-week quasi-experiment, A-Level STEM students receiving high-frequency automated marking scored 16.8 percentage points higher on a blind-graded final mock exam than control students with human marking |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — matey-yordanov-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.