Tuesday, April 14, 2009


No, not those guys on TV, though they do kick ass.

What I'm referring to are some myths we live by in the lab. One of cherished one among some people is that competition inside the lab improves productivity. That's complete bullcrap. Sure, there are lots of PIs whose management styles use it, but it's simply wrong. Apparently Candid Engineer's PI feels this way. Which sucks, even if it is all too common.

Unfortunately, what these jerky PIs ignore is the actual data that suggests that competition within groups hurts overall productivity. Teresa Amabile is a Harvard business school prof who has tracked the daily work of people in high tech, chemical, and consumer products industries, and the results run completely counter to many of the preconceived notions we have about creativity. An article in Fast Company discusses the 6 Myths of Creativity. All of them are good, but here's the money quote for this discussion:

5. Competition Beats Collaboration

There's a widespread belief, particularly in the finance and high-tech industries, that internal competition fosters innovation. In our surveys, we found that creativity takes a hit when people in a work group compete instead of collaborate. The most creative teams are those that have the confidence to share and debate ideas. But when people compete for recognition, they stop sharing information. And that's destructive because nobody in an organization has all of the information required to put all the pieces of the puzzle together.

I wonder how well this observation scales beyond individual lab groups to science as Science. How much competition is good, and when does it start to be detrimental? Certainly the last sentence here can be applied to Science.

Next, Bob Sutton is a Stanford B-school professor, who wrote a book called "The No Asshole Rule" (how great is that? Plus he has a kickass blog, which has been on the Googly Reader for some time). His recent post highlights another group's paper
...using quantitative analysis to uncover patterns across large numbers of studies -- in this case, 72 studies of nearly 5000 groups. The overall findings aren't a surprise, that groups that engage in more information sharing enjoy better performance, cohesion, knowledge integration, and satisfaction with decisions made
Now sure, these aren't academic lab groups, but as all the crankass PIs out there seems to insist on saying, academic science is the real world. So for any of you PI types who buy into this line of fallacious thinking, just stop. You're a scientist, go with the data.

And if not, then do us all a favor and wear a goddamn button that says, "I'm the Michael Vick of pitting my trainees against one another." Then we'll all be fairly forewarned.


Candid Engineer said...

I am fortunate in that the incident I blogged about was the only occurrence of this sort of situation in the year + that I've worked in my lab.

I completely agree with you (and the data). For the most part in my lab, I am amazed at the ideas that are generated when I sit down with some other people and we brainstorm. Anyone who has ever actually done this knows that internal competition is not a good situation.

ScientistMother said...

its totally true. I've worked in lab that fosters competition which meant the left hand did not know what the right hand was doing. And i've worked in labs were its all about collaboration, the collaboration labs are way better, publish in better journals and have way cooler ideas.