Our Blog

Good Measurement Needs Good Leadership

Posted August 20, 2012 12:13 PM by Dylan Miyake

Every morning, I run through Emory University. The evolution of this institution over time has been incredible. From a small Methodist college in a small town in Georgia, Emory has emerged as a major international university in the largest city in the southeast. The undergraduate, law, medical, and business schools attract students from literally around the world, and speakers like the Dalai Lama and President Carter have spoken at the university.

So it was with a heavy heart that I read the article in the Atlanta Journal Constitution this weekend that Emory may have been misreporting student data for over a decade. It appears they have been falsely inflating its test scores by reporting data for admitted students instead of enrolled students, cutting off the bottom 10% of the class, and other data shenanigans. Why would they do something like this? Because the stakes at the level Emory plays are high -- you're either a top 20 school or not, and for the past 10 years, Emory has straddled that line.

According to the investigation, there's no evidence that the president, provost, or any of the deans had any knowledge of the cheating, but at the end of the day, that's not the point. Just like in Atlanta Public Schools, this is fundamentally a failure of leadership. The job of leadership is to put in place a system that pushes people to the best possible job they can do, not find ways to cheat or short-cut their way to results.

The lesson here is not a new one. People are people, and there will always be someone looking for a shortcut. But organizations need to build a culture and environment where shortcuts are not accepted or tolerated. Typically, these organizations are also ones that are extremely open and transparent -- after all, it's really hard to hide in a house with glass walls.

So, my advice to Emory (and other universities around the country): Use this as a wake-up call. Think about the measures you have in place and how they could be gamed. Run the "black hat" risk management scenarios to understand how someone could hack the measure. And fix the short-term problem so it can't reoccur.

But patching the hole is not enough. We also need to fix the conditions where punching a hole in the wall is considered okay and could be overlooked for 10 years. So, let's understand the behaviors that drove gaming the measure. Understand the culture of the admissions office. Engage the entire campus in the conversation on how to fix it. And then fix it.

Related Blog Entries