Do students cheat?

How to cite

Fullard, Joshua; Boedts, Tristan. 2026. "Do students cheat?." AEA RCT Registry. AEARCTR-0018279. https://doi.org/10.1257/rct.18279-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Joshua Fullard* , Tristan Boedts

* Primary investigator

Registry ID
AEARCTR-0018279
DOI
https://doi.org/10.1257/rct.18279-1.0
First registered
Last updated
Status
Completed
Trial dates
2023-09-01 – 2025-12-31
Countries
United Kingdom of Great Britain and Northern Ireland ()
JEL codes
C93, I23
Registry record
http://www.socialscienceregistry.org/trials/18279

Abstract

We test for peer-to-peer cheating effects on academic performance. Exploiting systematic variation in quiz question reuse across randomly assigned classes, we find that each additional 10 students with prior question exposure increases scores by 0.06 to 0.08 points (0.03 to 0.04 SD). Students tested last - when all peers know the questions - score 1.2 points (0.6 standard deviations) higher than those tested first. These findings show that advance information transmission through social networks meaningfully distorts performance measurement - relevant to any setting where evaluations aim to assess individual capabilities.

Intervention

The data we use comes from three cohorts of postgraduate students at a UK business school. In a compulsory module, students complete three in-class quizzes designed to assess understanding of course material. Each 15-minute quiz consisted of multiple-choice and short-answer questions.

Students are randomly assigned to classes before the start of term (by the university). Within teach class students take part in a series of questions. In this paper we exploit systematic variation in question reuse across cohorts and quizzes.

Experimental design

Exploit variation in question reuse to evaluate the impact of peer-effect on academic performance.

Design details

Randomization method
Randomisation of students to classroom is performed by the university.
Randomization unit
Student to classes.
Clusters
3 cohorts of postgraduate students
Observations
861
Arms
861students across 3 student cohorts. Students take part in three quizzes over the term.
Primary outcomes
If students in later classes perform better on the quiz when the questions are reused.

Cited by

No citation data available for this registration yet.

Keywords