More carriages have come off the rails in the Roger Pielke Jr train wreck. Pielke finally does a hypothesis test. Trouble is, it's an unpaired t-test, which would only make sense if GISS and HADCRU were independent of each other, i.e. temperature measurements of different planets. Which, uh, they're not.
James Annan explains it here.
And another Pielke carriage comes off the rails here.
Via the Washington Times, we get more Pielke genius.
Roger Pielke Jr: "Climate models are of no practical use beyond providing some intellectual authority in the promotional battle over global-warming policy."
Right, Roger. The real science gets done by writing bombast on a blog in the hopes that some journalist might call you up for a quote.
So, apparently, taking James Annan's advice, Pielke actually talked to someone with a statistics background (an undergraduate), but unfortunately (for himself) did not give that person all the pertinent information.
I'd bet that person is now thanking their lucky stars that Pielke only gave their first name ("Megan"). It would be a real shame if someone else's career was also sent careening off the tracks.