How temperature data was analyzed

The AP contacted University of South Carolina statistics professor John Grego, a longtime reliable statistics source. In addition, the American Statistical Association sent an e-mail request from the AP seeking statisticians willing to examine certain sets of numbers and look for trends without being told what those numbers represented.

Three professors of statistics agreed: David Peterson, retired from Duke University; Mack Shelley, director of public policy and administration at Iowa State University; and Edward Melnick from New York University.

Each was given two spreadsheets, neither of which had any indication they were temperature data.

One spreadsheet was year-by-year global temperature changes from 1880 to 2009, adjusted through most of this year from the National Oceanic and Atmospheric Administration's ground measurements. The other was year-to-year temperature changes from 1979-2009 gathered by scientists at the University of Alabama in Huntsville from atmospheric measurements by satellite.

None of the experts detected a downward, or cooling, trend in the numbers. All saw a distinct upward trend.

This type of blind test is a valid way of seeking statistical help by trying to keep the statisticians' personal beliefs out of any analysis, said Alan Karr, director of the National Institute of Statistical Sciences. But there is a downside from keeping statisticians in the dark because it ties their hands a bit and may make it difficult to determine trends from variation, he said.

After their analysis, all the statisticians were told that the numbers represented temperature changes. All stuck by their assessments.

___

Oct 26, 2009 - 1:43 p.m. EDT

Copyright 2009, The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.