Which B-Schools Have The Happiest Grads?

Yale Revives the Bell Curve

It’s hard to find anyone who believes the Earth is flat. These days, you’ll find plenty of executives (and academics) who think evaluations should be curved.

And the Yale School of Management is joining them.

That’s the plan according to Anajani Jain, the school’s Senior Associate Dean. Going forward, the top grades will be awarded to the top 10% of students. What about the bottom 10%? If one group wins, another group must lose. It’s just natural selection, B-school style.

Actually, it’s more corporate style. Yes, the 80s are back, just not with the big hair and bright colors. This time, Yale is dusting off stacked rankings. Like a 80’s horror movie villain, you can shoot, impale, fry, drown, or put a stake through its black heart, but stack rankings will magically return for a sequel.

Popularized by GE Chairman Jack Welch, stack rankings place employee performance into several buckets along a bell curve. Here, managers assume the top ten percent of employees are stars and the bottom ten percent must be replaced. In between, companies peg a certain percentage as B-level and C-level players. As a result, employees are subjectively slotted based on a one-size-fits-all model.

You can probably guess the drawbacks. Aside from encouraging rivalry over cooperation and de-motivating valuable “glue” employees, stack rankings make for an easy way to dump the Cassandras and free thinkers under the guise of “performance.”

So how does this change Yale’s current grading system? Currently, students earn one of four grades: Distinction, Proficient, Pass, or Fail (with a mark between Distinction and Proficient being added in the fall). After analyzing grading patterns, Jain noticed the natural distribution, with 10% of students at the top and another 10% at the bottom. In other words, Yale’s approach is grounded in data science. What’s not to like?

A lot, actually.

First, Yale is an elite business school. It only accepts students with the highest business and academic credentials. If only 10% of students are pulling A’s, then it reflects poorly on the school, according to Herman Aguinis, who teaches at the University of Indiana’s Kelley School of Business. He tells Bloomberg Businessweek, “When you use a normal curve and you limit the number of A’s that you give to your students, that’s probably telling you that your admission system wasn’t that good.” If that isn’t the case, there should be a higher number of stars earning A’s.

Aguinis also co-authored a study with the University of Iowa’s Ernest O’Boyle Jr. that debunks the bell curve. From evaluating professionals ranging from athletes to researchers, they noticed that “hyperperformers” actually carried the majority of the performance burden, reflecting the proverbial “80/20 rule” where 80% of the work is achieved by 20% of the team. As a result, there are actually a greater percentage of high performers – and bottom performers – than the bell curve accounts for. If these findings were applied to business schools, you’d find much smaller class sizes.

In addition, Yale’s approach could undermine learning. For example, Josh Bersin, founder of Bersin by Deloitte, observes that the curve is just a sorting device where grades are almost established before the class begins. To Bersin, a curve discourages students from competing against their talented peers. “So they say, ‘Eh, forget it. I’m never going to be in the top percent, so what the heck. I’ll just do what I’m doing. And they just become sort of complacent.”

While the approach guards against grade inflation, it makes a class error: Assuming every class is equal in terms of talent.

Despite committing to a stack ranking, Jain has signaled an openness to tweaking the system. “We don’t think of any particular grading system to be perfect,” Jain notes. “We remain open-minded about how there might be a need to alter the parameters going forward.” Considering that the use of stack rankings plunged from 49% to 14% between 2009 and 2011 (according to the Institute for Corporate Productivity), those changes could be coming sooner rather than later.

Source: Bloomberg Businessweek