At a talk given at UNC last night, UVA climatologist Patrick Michaels mentioned that this news was going to be all over the paper today and as usual, he exposed it for the misuse of statistics that it is. First he pointed out that it was for a 34-month period beginning in the middle of 2002. Clearly, in terms of climatological history such a short period is meaningless for establishing a trend, as was noted in the quote that Paul cited. But there is more to the story. While it might be meaningless for establishing a trend it is quite useful for reinforcing a mind set. The question might be asked, why didn’t the author go back further than 34 months? Pat answered this by putting up the extended data for the thickness of the Antarctic ice sheet and low and behold for a period of ten years prior to the middle of 2002 there had been a sharp increase in its thickness. It is well known in statistics that a person can show any trend he desires if he picks his beginning and end points judiciously–if not honestly.
BTW–over those ten years how many stories did we see in the Washington Post discussing the dramatic increase in the thickness of the Antarctic ice sheet? go do a Google search and let me know.
