junk social science

Aug 02, 2005 14:56


Read more... )

Leave a comment

Comments 4

labrujah August 2 2005, 15:28:11 UTC
maybe that was a different run of the experiment? the NYer has pretty good fact-checkers.

Reply

screed August 2 2005, 16:56:28 UTC
the throughput got better (near 30%) in one subsequent run, but all the others were low, some so low that he didn't even bother publishing them. part of the problem was that Milgrim was more delighted in finding short chains than he was in doing a rigorous study of social networks.

it would have been accurate to say "Most of the letters, when they reached their destination at all, followed a chain of 4 to 8 people." it still says something about how people interacted in those days: in smaller, more homegeneous networks, or more likely not at all.

i think Gladwell is intentionally misreading the facts to fit his hypothesis; i've found a couple instances of this in his articles, and a few more in Blink.

Reply

squeakymonkey August 2 2005, 18:54:45 UTC
Most social scientists are such bullshitters, and usually no one calls them on it.

Guess what my degree is in?

Reply


substitute August 2 2005, 19:08:37 UTC
That guy has the whole world conned with his sparkling style.

Reply


Leave a comment

Up