Easy A's on the Internet

A surprising Cornell experiment in posting grades; plus a look at recent research into ethical behavior, service charges, and volunteer habits

by Francesca Di Meglio

In a striking example of unintended consequences, a move by Cornell University to give context to student grades by publicly posting median grades for courses has resulted in exactly the opposite student behavior than anticipated.

Cornell's College of Arts & Sciences originally set up a Web site in 1997 where median grades were posted, with the intention of also printing median class grades alongside the grade the student actually received in the course on his or her permanent transcript. Administrators thought students would use the information on the Web site to seek out classes with lower median grades—because, they reasoned, an A in a class that has a median grade of B-minus would be more meaningful than say, an A in a course where the median was A-plus.

Course Shopping Leads to Grade Inflation

However, when Cornell researchers studied about 800,000 course grades issued at Cornell from 1990 to 2004, they found that most students visited the site to shop for classes where the median grade was higher. Plus, professors who tended to give out higher grades were more popular. Students with lower SAT scores were the most likely to seek out courses with higher median grades.

This "shopping" in turn led to grade inflation, Vrinda Kadiyali, associate professor of marketing and economics at Cornell's Johnson Graduate School of Management, one of the authors, explained in an interview. The study, which is undergoing peer review, has not yet been published.

So far, however, the university has posted the median course grades only on the Internet and has not yet put those grades on transcripts. According to an article in the Cornell Daily Sun, the school will start posting the grades on transcripts in the spring. School officials were not immediately available for comment.

The research team hopes the school follows through on its plans. "That will allow Cornell to hold itself to a higher standard because it lets potential employers know where students stand relevant to other students," says Kadiyali.

The presence of the median grade data is well-known to students but less well-known to faculty. The researchers themselves were prompted to do the study when one of them learned of the Web site from a student questioning grades in her course.

Kadiyali says the formula the researchers used to come to these conclusions could easily be applied to Internet teacher rating sites, such as ratemyprofessors.com. It's something educators should consider, she adds, to find out how these posts affect the decision-making of students and, thus, professors and their courses.

Research Eyes Ethical Behavior

Ethics has long been a subject of debate at business school. But recently even more business school researchers are doing their part to determine why people behave the way they do.

In a study published in the November issue of the Journal of Applied Psychology, Scott Reynolds, assistant professor of business ethics at the University of Washington, and doctoral student Tara Ceranic found that moral identity and ethical judgements don't always result in ethical behavior.

In two separate studies, the research teams surveyed about 500 students and managers about their ethical behaviors. One of the studies asked students about past cheating. Those who considered themselves to be moral—using words such as "caring," "just," and "honest" to describe themselves—and considered cheating morally wrong were least likely to cheat. But those who considered themselves to be moral and saw cheating as having ethical justifications were the worst offenders.