Posted 6:53 AM by steve

A corollary to GIGO (Garbage In, Garbage Out) is that more output confirms one's bias, the less likely the inputs are double checked (and the reverse holds true, the more the output varies from what one expected to see, the more likely one is to try to figure out what is 'wrong' with the program).

This is how we end up with situations like this, where reports of reductions of Artic ice were overblown due to what is being called 'sensor drift'. The initial reports backed up what the researchers expected (and likely wanted) to see, that ice coverage was in fact shrinking (due to global warming) and thus didn't bother checking to see if their measurements were in fact accurate. It was only when presented with incontrovertible evidence (photographs of allegedly open sea that were covered with ice) did researchers go back and examine their testing.

So remember this the next time someone attempts to claim that science has 'proven' X or Y. There's a chance they're right, and there's a chance they're wrong but just don't know it... yet.