Monday, February 14, 2011

More fallacies in the media about cooperative groups

With the release of Edward Glaeser's book Triumph of the City, and with a new University of Michigan study on group cooperation, the supposed virtues of cooperative groups are once again the talk of the media—as seen, for example, in a recent New York Times Op-Ed by David Brooks on "The Splendor of Cities" and a recent article in last weekend's Wall Street Journal on the "Sunset of the Solo Scientist". 

You can be sure that anyone in the education establishment who reads these articles will find just what he or she is looking for: more reasons to make students spend significant time working in groups. Confirmation bias blinds most people to the underlying fallacies:

1. Cooperation does not equal collaboration.
I’ve made this point many times over: Many (and I’m guessing most) group collaborations, while they include collective brainstorming, ongoing conversation, and, ultimately, some sort of collective wrapping up, have people spending the majority of the time working on their own. Construction sites and film sets aside, how often is the bulk of the work, and of the time spent on it, accomplished by people occupying the same open (cubicle-free) space and constantly interacting and doing things together?

2. Carefully chosen cooperative games aren’t representative of most collaborations.
A corollary to 1: The specially-designed cooperative games of the University of Michigan's recent psychological experiments aren’t representative of most human endeavors. The fact that highly cooperative groups (those with higher social or “group intelligence”) outperform less cooperative ones on certain highly cooperation-dependent games says nothing about collaborations in general. Were the same experiments performed on real-world collaborations in STEM or business, would “group intelligence” still be the most important factor?

4. Even ostensibly “cooperative” groups can be arenas for competition and bullying.
(Another point I’ve made many times over and won’t belabor here).

3. Density and frequent contact do not equal cooperation.
I haven't read Glaeser's book, but some of the reviews of it I've seen suggest that brainstorming and competition, and not necessarily group cooperation, are key ingredients in how cities advance civilizations via increased human interaction.

5. The rarity of solitary geniuses does not equal their demise.
As one WSJ letter notes in response to the WSJ’s recent article on the “Sunset of the Solo Scientist”:

Minds like Einstein’s or Newton’s come infrequently, and the gap in time between theirs and the next true greats may be now. Or maybe not. What about Stephen Hawking? He may yet eclipse Einstein.

Maybe you just don’t know who today’s greats are. Some may not be recognized until well after they’re dead. Ideas are still the realm of individuals.

Great ideas typically spring from a single mind. Teams of researches may trudge through mountains of data but not have the “Aha!” moment… ever.

I think that encouraging individual, critical and creative thought is the only way to create true greats.


Lsquared said...

I agree. My reference points are mostly math. Andrew Wiles proved Fermat's Last Theorem. He couldn't have done it if Ken Ribet hadn't proved that if the Taniyama-Shimura conjecture was proved, then it would also prove Fermat's last theorem, so it wasn't a result that was done without knowing other people's work, but when he was working on it, he worked completely alone, and didn't talk to anyone about what he was doing until he believed he had the solution. Science needs both the sharing of knowledge, and the persistence of the individual working alone. Collaborators aren't rare, but they aren't the only way to go, either.

Deirdre Mundy said...

Also, it seems to me that the most productive TRULY cooperative efforts in science have come from a partnership, or perhaps a trio. And these fruitful collaborations usually grow organically, out of a mutual interest, rather tahn being imposed from above.

So we get the Curies, and Watson,Crick and Franklin (and notice Franklin always gets left out.) and Miller-Urey, etc. etc....

So partnerships, not groups. And freely chosen ones. Totally UNLIKE what happens in a classroom environt when a teacher decides that "Groupwork is necessary for learning!!!"

Also, as an aside, in an elementary classroom, brainstorming is usually an exercise in stupid. Coming up with 20 lame ideas is NOT a good substitute for one or two GOOD ones.....

ChemProf said...

Since grad school, I've published many collaborative papers. Never did we sit and brainstorm. Sometimes we'd talk and argue a point, but often it was more iterative -- one person would write a draft, and the other person would read and comment on it, pointing out questions or problems with interpretation. It never resembled elementary-school "group work." In fact, I've found in general that the least productive meetings are the ones without some kind of first pass on paper or an agenda. Without that focus, even educated adults tend to just talk around the issue and waste time. Anyone who has ever sat through a faculty meeting without an agenda should be familiar with this phenomena.