Monday, August 23, 2010

Epochcentric bias

From Charles Murray's Human Accomplishment. A very intriguing and rigorous discussion of what constitutes accomplishment, how do we measure it, and how do we explain it.

Such lists exhibit the same bias that plagues more sophisticated histories and chronologies, found whenever historiometricians have looked for it: The recent past gets more attention than it will prove in the long run to have deserved. Dean Simonton, one of the first scholars to document this effect, named it epochcentric bias. The magnitude of the bias for general histories is large. Two scholars who measured it for a news almanac published in 1978 found the decay in attention given to preceding eras to be exponential. Applying their findings, the implication is that (for example), the half-century from 1800-1850 gets only 35 percent as much coverage as 1900-1950, for reasons having nothing to do with the potential amount of material that might have been included, but simply because those events were a century further back in time.

No comments:

Post a Comment