iamom: (Default)
Dustin LindenSmith ([personal profile] iamom) wrote2004-10-22 02:26 pm

on the veracity of analysis used in global warming research

This article discusses a methodological breakdown in the manner in which weather data has been analyzed with respect to global warming research. It's a good read, and references an MIT article that discusses the error, as it were, in detail.

I also posted this thread to a research discussion forum to see what the mathematicians in the group think about this. Apparently the type of analysis used (principal component analysis, or PCA) is a fairly straightforward bit of statistics; I'm skeptical that such a fundamental error could occur without having been noticed or double-checked before now.