Saturday, September 30, 2006

Fascinating article in the New Yorker about how some Cystic Fibrosis centers are much more effective and why that is, not just because I have a personal stake in this issues.

The author doesn't come out and say this, but it seems to me the answer comes down to the talent of the people on that team. The best CF center has an extremely talented doctor who is the driving force behind the center. The Cincinnati center highlighted by the article could improve itself incrementally by doing things similar to how the best centers do things, but they will never be as good, because they don't have that doctor.

We're willing to acknowledge the importance and impact of talent in some contexts. Any serious baseball fan knows that Albert Pujols is several times more valuable to the Cardinals than a utility player like Aaron Miles. Nobody thinks that if the Pittsburgh Pirates could significantly improve themselves by following the same daily routine as the New York Yankees. Teamwork and chemistry are important, but you have to have the horses.

But in the businesses where most of use work, we pretend this isn't the case. I'm a software engineer. Employers do "pay for performance" which usually means that the best engineers get a 5% raise while the average ones get 3% raises, resulting in the best engineers making maybe 10% more than average engineers. This in spite of reasearch showing that the best developers are 10 times more productive than average engineers. Yet, if an organization is looking to improve its software quality, it won't try to hire better engineers, it will refine its processes.

Other industries are worse. The compensation system for public school teachers works on the assumption that the only measure of a teacher's effectiveness is years of service. But if the best baseball players are much more valuable than average baseball players, and the best software engineers are 10 times more effective than average engineers, is there any reason to believe that the best teachers are many times more valuable than average teachers? And shouldn't their compensation refelect that? And if not, won't great teachers become average teachers?

We don't want to believe this, though, because it means that we have to do the work of identifying the best. Something in me wold like to believe that all CF centers are about equal. The idea that changing centers could extend Meagan's life by many years is a bit scary. It would mean that I would need to find out what the best center is, and maybe move our family there. Can't we just assume they're all abot equal and worry about something else? Isn't that why chains are so popular? Walk into a McDonald's and you know what you're going to get. If you're OK with that, then you don't need to research each town you visit to find a restaurant.

Except McDonald's food is mediocre, not excellent. For a quick meal on the road, mediocre might be good enough. But is mediocre good enough for educating our children, caring for a degenerative fatal disease, or producing a new software products. We need excellence, but we're using tools and techniques designed to achieve a baseline of mediocrity rather than excellence.

There's also a rationing problem. By definition, there are few people on the right side of the curve. How can they meet the large demand for their services? For commercial software engineers, the market can determine it -- whoever's willing to pay the most gets the elite engineer's services. But is this how we want to determine who gets access to the best CF centers? the best high shool algebra teacher?

So I guess I have more questions than answers....
Post a Comment