CoChat

Artificial Intelligence for Student Assessment: A Systematic Review

Open access
González-Calatayud, Víctor, Prendes-Espinosa, Paz, Roig-Vila, Rosabel · 2021 · Applied Sciences

The DOI matches its Crossref record; 60 of the 61 references it cites were found in a registry or an index; 19 of the 73 sources its claims cite were checked for support.

Abstract

This paper analyzes the use of AI for student assessment based on a systematic review and concludes that pedagogy is often not reflected, formative evaluation is common, and more educationally focused research is needed.

Overview

CoChat Summary

TL;DR

This systematic review of 22 studies, selected from 454 initial records, finds that AI in student assessment is used mainly for formative assessment and automatic grading, while pedagogical models are often not reflected. It concludes that teacher training and more pedagogically focused research are needed, especially beyond higher education.

Key findings

  • From 454 papers found, 449 remained after duplicates were removed, and 22 were selected; most selected studies focused on higher education, with some examples at secondary level, and the predominant author country was the USA (4 contributions), followed by Greece (3).
  • Of the 22 selected papers, 15 had AI for formative assessment as their starting point, and 10 explicitly stated that the system used grades students automatically.
  • Only one study clearly addressed pedagogical models underpinning AI use; three others mentioned pedagogical rationale, mainly collaboration or argument maps, and recent work focused more on technical aspects than pedagogy.
  • Three studies compared AI use with non-use: Goel and Joyner found students who used AI performed better; Grivokostopoulou et al. found AI and manual grading were very similar and correlated; Samarakou et al. found those who used AI scored 17% higher than those who did not.

Methods

  • Systematic review following the PRISMA Statement, searching Scopus and Web of Science in March 2021 with terms including “Artificial Intelligence,” “Machine Learning,” “Education,” “Assessment,” and “Evaluation”; inclusion criteria were peer-reviewed empirical research published 2010–2020 in English or Spanish.
  • Titles and abstracts were evaluated independently by two reviewers using Rayyan in blind mode, with a third party resolving discrepancies; full texts were then reviewed, and coding/data extraction used Rayyan tags, a spreadsheet, and Mendeley.
  • After screening, 339 studies were excluded after title/abstract review, 68 full texts were discarded for not meeting criteria, and 10 were excluded for lack of access, leaving 22 studies for analysis.

Limitations

  • The paper states that the study found a small number of articles focused on student assessment relative to the large number of AI applications, and it calls for further research, mainly at educational levels other than higher education, plus teacher training.
  • The paper states that the method was applied rigorously to guarantee validity relative to the eligibility criteria.
  • Evident from the text: the search was limited to Scopus and Web of Science, English and Spanish, and 2010–2020; 10 full texts were excluded because they were not accessible; the final set of 22 studies was mostly higher education and STEM/health contexts, limiting generalizability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/app11125467

Cites

60 of 61 references verified0 citation errors · 1 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

16 Valid · 4 Not valid · 24 Not checked · 43 Own results and interpretations

36 more claims · 2 with their cited sources checked

Create a free account to see all 87 claims and to verify the 3 references still open.

See all claims — free account
  • Probably not a substantial claim. 30
  • It may state two separate claims as one. 16
  • Borderline: one of the tests for a substantial claim is unclear. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

4 of 4

Citations

Cites

Show all 61 references

Research works printed but not linked: 1

  1. Ouguengay, 2015, A neuro-fuzzy inference system for the evaluation of reading/writing competencies acquisition in an e-learning environnement, J. Theor. Appl. Inf. Technol., 81, 600

Cited by

See all 503

Metadata

Authors
González-Calatayud, V., Prendes-Espinosa, P., & Roig-Vila, R.
Affiliations
  • Department of Statistics, Mathematics and Computer Sciences, Miguel Hernández University, 03202 Elche, Spain
  • Department of Didactic and School Organization, University of Murcia, 30100 Murcia, Spain
  • University of Alicante (ES)
Venue
Applied Sciences
Publisher
MDPI AG
Published
2021-06-12
Volume
11
Issue
12
Pages
5467
ISSN
2076-3417 (ISSN-L)
OpenAlex
W3171148899
Cited by
503
References
59
Citations by year
2026: 87 · 2025: 167 · 2024: 169 · 2023: 51 · 2022: 24 · 2021: 5
FWCI
9.46
Type
journal article
Language
en
Open access
Locations
  • Applied Sciences (publishedVersion, cc-by)
  • Repositorio Institucional de la Universidad de Alicante (Universidad de Alicante) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Social Sciences › Education › Artificial Intelligence in Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
  • Social Sciences › Education › Educational Technology and Pedagogy
Keywords
artificial intelligence in education, student assessment, automated grading, formative assessment, AI-based tutoring, teacher professional development

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Artificial Intelligence for Student Assessment: A Systematic Review. https://app.cochat.ai/papers/10.3390/app11125467
The paper
González-Calatayud, V., Prendes-Espinosa, P., & Roig-Vila, R. (2021). Artificial Intelligence for Student Assessment: A Systematic Review. Applied Sciences. https://doi.org/10.3390/app11125467

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.