Uh, oh, this isn’t good.
Kos, at the Great Orange Satan:
I have just published a report by three statistics wizards showing, quite convincingly, that the weekly Research 2000 State of the Nation poll we ran the past year and a half was likely bunk….
Much to his credit Kos is being very open about the situation stating, “…I hereby renounce any post we’ve written based exclusively on Research 2000 polling…”
From 1997 to 2002 I helped run a mom and pop operation that did relatively small sample legislative district polling in Missouri. We were held together with spit, baling wire, and duct tape. Large amounts of caffeine helped, too. The opinion research game is brutal. I’ve always said that the mechanics are straightforward, it’s the logistics that are a nightmare.
I cannot imagine the nightmares and pressure of doing weekly national polls.
Maybe the lesson in this is that if you’re going to commission a lot of polling (and spend a lot of money) do a lot of digging before, during, and after.
This is not good.