Even simple theories need to be tested.
Yesterday, I heard a piece on NPR where a simple premise was tested: pairing academically weak students with strong students has an overall positive effect due to peer pressure. This is formally known as peer effects.
Duh! I thought. After a written test at Parris Island, my drill instructors would make a list of all the recruits in my platoon based on how well they scored. Then, they'd pair the top performer with the bottom performer, number two with the second from the bottom, and so on. It seemed brilliantly simple. This social engineering theory was recently tested at the Air Force Academy over a two year period. But the results weren't as anyone expected.
Creating a squadron by pulling out the middle-of-the-pack cadets and having the strongest cadets live in the same squadron with the weakest had a negative effect. It splintered the squadron into two different social groups where the good students hung out with the other good students and the weak students hung out with fellow weak students. This resulted in a decrease in performance of the weak students and the experiment was ended to prevent further harm.
It turned out the average performers were the glue between these two groups. I also found it surprising that the average performers did better in a squadron of all average performers and worse in the mixed group. From the NPR interview...
... this presents a dilemma, because it suggests that in regular squadrons where everyone is mixed together, the performance of the middle students is actually adversely affected. You pull them out, they start to do better. But when you pull them out, the weakest students start to do worse.
Let's hear it for mediocrity. ;^D