Is cheating on tests inevitable?
Teaching English with Oxford
Providing education leaders and decision-makers with solutions and insights into English learning and assessment.
Take a moment to think about these questions: What percentage of high school students do you think admit to having cheated? Does that number go up or down when they get to college? Do women cheat more than men? Do psychological characteristics predict cheating? What about national culture?
According to one survey in the USA, 95% percent of high school students admitted to some kind of cheating. This figure drops to 60% for college students but whether this reflects an increase in honesty (i.e. cheating less) or a decrease (i.e. lying about cheating less) is open to question. And while I’m confident that all the readers of this blog fall into the 5%, the fact is cheating is endemic and a grave threat to test integrity. Understandably, researchers have responded by seeking the root causes of cheating in order to implement measures to stop it.
Of course, anti-cheating measures are nothing new. Imagine this scenario: a test taker on exam day arrives at the secure test centre and presents their biometric ID. They are divested of any contraband items such as hand-written notes and are led to their randomly allocated desk, separated from fellow test takers to prevent copying, all under the watchful eyes of the invigilators. While this may sound familiar, the scenario I am describing dates back over one-and-a-half millennia to the Imperial exam for the Chinese civil service, when many of the security protocols we take for granted today were developed.
Does culture influence cheating?
Jump back to the present day and picture a different scenario. Hundreds of parents are protesting angrily outside their children’s high school, chanting “We want fairness!,” in what would ordinarily be a laudable sentiment, given their children were completing their university entrance exam, but the chant continued, “There is no fairness if you do not let us cheat." The high school in question, in China’s Hubei province, had been attaining unusually impressive pass rates resulting in a government crackdown on cheating at the school. Measures included checking biometric IDs, divesting test takers of contraband such as mobile phones and secret transmitters, and the use of invigilators drafted from other schools. The parents rioted. But given it was rife across the country, wouldn’t the only fair thing be to, er, let the children cheat?
Feeling justified in cheating because other people cheat is called the ‘perceived peer cheating effect’ and it is one of the stronger motivators for cheating. There is evidence that this phenomenon is linked to dimensions of national culture, with greater prevalence of the perceived peer cheating effect found in high ‘Power Distance’ cultures (as opposed to more pluralistic ones), ‘Collectivist’ cultures who view themselves as members of larger groups (as opposed to ‘Individualistic’ ones where ties between people are looser) and strong ‘Uncertainty avoidance’ cultures which have a need for clarity and structure (as opposed to weak Uncertainty avoidance cultures which are comfortable with ambiguity and chaos).
However, cultural explanations can be essentialist, so in practical terms culture has been addressed at a more local level to address the issue of cheating. For example, the International Center for Academic Integrity has proposed building a culture of academic integrity ‘in the classroom, at the institutional level, and beyond, based on the six principles of honesty, trust, fairness, respect, responsibility, and courage. As today’s students are tomorrow’s leaders, nothing could be more desirable. Related anti-cheating initiatives include honour codes in which students commit to academic integrity, but research results are mixed on the effectiveness of these, and at best promoting academic integrity is only a partial solution.
Given the USA cultural profile (pluralistic, individualistic and low uncertainty avoidance) is the opposite of a ‘perceived peer cheating effect’ culture, and yet boasts a 95% high school cheating record, it’s no wonder that some researchers have focused instead on more individual characteristics in search of the root cause of cheating.
Does your personality type make you more likely to cheat?
While the motives for cheating are wide ranging, poor self-discipline is a common personality trait associated with cheating, where the student lacks the discipline to start or complete their studies. Other personality traits linked with cheating are risk tolerance, poor verbal ability and psychopathy. Additional variables include language background (those studying in a country whose language they don’t speak fluently), finance (cost-benefit analysis of study fees vs. risk of getting caught), and lack of exam preparation. An increasingly popular option for these students is contract cheating.
When cheating gets professional
Contract cheating refers to third party providers (henceforth ‘contractors’) offering cheating services covering everything from ‘essay mills’ to provision of ‘ghosts’ to take an exam on the test taker’s behalf to lawyers submitting dubious results appeals. It’s big business. This professionalization of the cheating industry poses a major challenge to test providers who in response are bringing an arsenal of countermeasures, including everything from test design such as computer-adaptive tests like the Oxford Test of English Advanced, which present different questions to each test taker to the use of AI to identify human-generated vs computer-generated essays. However, despite the threat of contractors to test integrity, it may be the contractors themselves who bring about their own demise.
Blackmail, fraud and extortion
A test taker who has employed a contractor may breathe a sigh of relief when they receive their Pass result. However, the relief may be short lived, because to engage a contractor is to engage a criminal, and some may not be satisfied with just the original payment for their service. Out of the blue, the test taker may receive a communication from the contractor threatening to expose the test taker as a dishonest cheater if additional payment, or something worse, isn’t provided. Or perhaps the test taker suddenly finds that the credit card they used to pay the contractor has been maxed out. Or perhaps their identity has now been stolen. The possibilities are endless.
Alerting test takers to these dangers can significantly reduce the amount of contractor engagement, though one study reported this only reduced engagement by 50%. The number of risk tolerant students shouldn’t be underestimated.
领英推荐
A magic solution to cheating?
Unfortunately, there’s no silver bullet. The reasons for cheating are myriad, as are the ways of cheating, so test providers need myriad ways to combat it. This includes test providers coming together to share knowledge and experience. That’s why I’ll be talking at this year’s Conference on Test Security (COTS). I hope to see you there.
So, who cheats more – men or women?
If you got to the end of this blog, you’ll be wondering about the question I posed at the start about whether men or women cheat more. The balance of (self-reported) evidence suggests there is no difference, though if one sex did cheat more, wouldn't they also be unlikely to admit it ..?
References
Conference on Test Security 2024 https://conferenceontestsecurity.org/
The Fundamental Values of Academic integrity (3rd edition)
Oxford Test of English Advanced https://elt.oup.com/feature/global/oxford-test-of-english/advanced?itm_source=oup-elt-site&itm_medium=hero-banner&itm_campaign=ww-ote-advanced&cc=gb&selLanguage=en
Riot after Chinese teachers try to stop pupils cheating https://www.telegraph.co.uk/news/worldnews/asia/china/10132391/Riot-after-Chinese-teachers-try-to-stop-pupils-cheating.html
Oxford English Assessment Research https://elt.oup.com/feature/global/oxford-english-assessment/research?cc=gb&selLanguage=en
?
Colin Finnerty is Head of Assessment Research at Oxford University Press. Colin holds an MSc in TESOL and was instrumental in the design and validation of the Oxford Test of English and the Oxford Placement Test for Young Learners. Previously Colin has worked for Cambridge Assessment on IELTS and for the British Council, running professional development courses for teachers. His research interests include corpus linguistics, Natural Language Processing and machine learning. Colin has presented at international conferences including Language Testing Forum (LTF), Association of Language Testers in Europe (ALTE) and International Association of Teachers of English as a Foreign Language (IATEFL).
?