Why Students Cheat?

If one asks Google this exact question she will get about 22,000,000 results in 0.50 seconds. As our alumna Anastasiia Faikina (Bachelor’s programme 'Economics'2016) discovered, yet there is no theoretical model that would explain the phenomenon. So she decided to take charge

Anastasiia Faikina: - My aim was to explain which factors influence human’s decision to cheat or plagiarize. There are a lot of empirical works on this topic but no good theoretical models so far. In short, modifying the standard principal-agent model, I found out that higher individual's abilities and commitment to social norms make cheating less probable. At the same time, the principal (boss, professor) can decrease the level of cheating not only through different punishment and rewarding schemes, but also through an increase of trusting atmosphere among workers or students (which I name 'peer trust').

I also got a very interesting result in my empirical study of HSE students regarding the peer trust. While inequality in abilities among students influences their trust to each other, income inequality does not have any effect on it at all as it usually has on the level of general trust to people.

So, the first step for theoretical explanation of cheating behaviour has been done but there is still a space for significant sophistication of my model.  

Paper Title: Peer Effects and Dishonest Behavior in the Individual Projects

Thesis Advisor, Kosmas Marinakis, Ph.D.: Nastya has done excellent work tackling a very interesting problem. Why people make a conscious decision to exhibit dishonest behavior? What makes them substitute honest effort with cheating? First, those agents neglect the beneficial effects of honest effort as an investment on human capital focusing only on the short-term benefit of a higher grade. Second, agents tend to substitute costly effort with risk of getting caught according on their risk attitude. In Nastya's paper, effort, cheating, risk shifting and economic choice write a story of "crime and punishment". Great research - worth reading.

Abstract: This paper presents a theoretical model of the agent’s optimal decision between honest behavior and cheating in the individual projects while being located in peer environment. I construct the model using a classical principal–agent relationship given a possibility of cheating that has not been properly analyzed before. I consider two different settings, cheating as a crime of opportunity and a simultaneous decision about efforts and cheating, which give the same intuitive results. The conclusions of my model suggest that higher able agents, who are more productive in honest efforts and more adherent to social norms, have a significantly lower probability of being engaged in cheating behavior. At the same time, the principal can also reduce the level of cheating through an increase of a size of the punishment or manipulation of risks. The most interesting solution of the problem of cheating is a creation of atmosphere of peer trust since agents always refer to the behavior of others. A possible illustration of the model is a problem of cheating by students whose decisions are very consistent with the implications of the model due to my empirical analysis.

Keywords: Cheating, compensation scheme, contract theory, efforts, incentives, peer effects, peer trust, performance evaluation, social learning