Climate scientists should take some basic courses in statistics
From the GWPF – London, 6 February: A recent paper in Nature has received worldwide media attention because of its claim to have shown that the recent hiatus in surface temperature rises was the result of natural variability. The lead author, Jochem Marotzke of the Max Planck Institute, also claimed that his work dealt a fatal blow to suggestions that computer simulations have systematically overestimated the global warming caused by rising carbon dioxide concentrations.
However, Nic Lewis, an expert in this area of climate science, today pubished an article demonstrating that there are serious errors in the paper, and that its conclusions cannot be sustained. Lewis said:
“As well as having some basic statistical errors, Marotzke’s study can be shown to utilise circular logic. This means that its conclusions are unsound. Moreover, the stability of estimates for at least one of the two key structural model properties used is so poor that even were he able to rework his paper without the circularity – which appears impracticable – it would very likely be impossible to draw meaningful conclusions. I think the authors have no scientifically-defensible choice but to withdraw the paper.”
Lewis’s findings, which have been published at the influential Climate Audit blog, have been reviewed and confirmed by two statisticians: Professor Gordon Hughes of Edinburgh University and Roman Mureika, formerly of the University of New Brunswick. Professor Hughes said of the Marotzke paper:
“The statistical methods used in the paper are so bad as to merit use in a class on how not to do applied statistics. All this paper demonstrates is that climate scientists should take some basic courses in statistics and Nature should get some competent referees.”