Volume 67, Issue 7

Economically Eradicating Cheating
by Commentator Staff

At the behest of Associate Professor of Economics, Dr. Elias Grivoyannis, Yeshiva College mathematics major Uri Hertzberg devised a Game Theory model displaying an approach that decreases incidents of cheating.  Game Theory is a relatively new topic that explains the strategies that players in a game use, and analyzes the final payoffs of a given game.

Setting up a game tree where a given student (A) decides whether or not to cheat, and gives the payoffs for each decision, can help elucidate the reason for that student’s decision.  With this understanding, it’s possible to devise a strategy that deters student A from cheating.

The game tree has three decision nodes.  First, the tree branches out into two branches: the student’s decision to “cheat” and the student’s decision “not to cheat.”  The second decision node has two branches from student A’s decision to cheat: student A “doesn’t get caught” and student A “gets caught.”  The payoffs from the three branches of the tree are the determining factors in a student’s choice.

Taking the payoffs from the game tree, we see that student A will decide to cheat if pG – qC > M, where p=the probability that student A wont get caught if he cheats, G=the gains to student A from cheating successfully, q=1–p=the probability that student A gets caught if he cheats, C=the cost to student A of getting caught cheating, and M=the value the student places on his morality not to cheat.

In other words, student A will decide to cheat if his gains from successful cheating multiplied by the probability of being successful minus his cost from getting caught multiplied by the probability of him getting caught is greater than the value he places on the morality of not cheating.

There are many factors that go into variables G and C.  The gains to a student who cheats successfully includes such factors as profits from cheating, his ability to get a better job later on, the guilt of cheating, his ego, and more.

The cost to a student of getting caught cheating includes factors such as paying the consequences of cheating, embarrassment, getting a bad reputation, and more.  Each student has a lot to take into account, and the gains and costs of cheating are different for each particular student.

Knowing that a student will cheat if pG–qC>M, a university can deter students from cheating by either raising M, or lowering pG–qC.  Since a university has very little control over the factors of G, and it is very difficult to raise M (because most students who cheat already know it is morally wrong and do it anyway), the best option is raising q or C.  The optimal strategy involves increasing q as high as possible, because if q=100 percent, then p decreases to 0 percent, and students will never cheat.

Another possible strategy includes raising C by making the punishment for cheating as harsh as possible.  This approach, however, is not as guaranteed as the first, because regardless of how high C is, if a student doesn’t think that he will get caught, then the cost of getting caught is not taken into account.

On the other hand, it is very easy to raise C by making harsh punishments, while raising q is more difficult.  A university can attempt to raise q by having strict test guidelines and by adding proctors to watch the students during tests. 

“If Yeshiva University wants to eliminate cheating, it is in their interest to hire more proctors, set up strict test guidelines, and make the punishments for cheating as harsh as possible” explains Hertzberg.  “While there is some cost to the university in taking these steps, it seems to be a minimal price to pay to lessen the amount of cheating in the university and raise its integrity.”¨

What do you think? Click here to send a letter to the editors.
All content is copyright © Yeshiva University Commentator.