Research 2000: What We're Going To Reveal Could Be The End of the Daily Kos

Pushback.

Now, what I think could have been the end of the Daily Kos was the assertion that R2K gave Kos good numbers but then Kos changed them, thus creating the suspiciously non-random patterns.

What they really say, though, is that the scandal about Kos is about owing people money, which isn't a good thing, but it's usually not the kind of thing that destroys someone's credibility.

The liberal blogger fired the pollster earlier this month and said he intends to sue in coming days. His State of the Nation polls tracked the favorable and unfavorable ratings for national elected leaders, including Senate Minority Leader Mitch McConnell, R-Ky.

Del Ali, president of Research 2000 in Olney, Md., said he could not respond to the specific allegations Tuesday and referred questions to his attorney, who did not return a call seeking comment.

“I can tell you, we’re fine. What we’re going to reveal, that will be the end of the Daily Kos,” Ali said. “I can say, it has to do with people owing money.”

Still, here's hoping for some popcorn-worthy stuff.

Via Gabe's Twitter feed.

Caution on Falsified Data Claims? Someone in the biz notes that R2K was considered one of the least accurate polls, and asks, "If they were just faking data, couldn't they have averaged Gallup and Rasmussen and faked something closer to accurate?"

Also questions whether some of the offered proofs of falsification are particularly strong.

One thing that I don't get is this paired-odds and paired-evens business. If you have an even number, and split it, you have to get both even numbers or both odd numbers. That's what an even number is.

The polling person also suggests that with sample sizes of 2500, subgroups of 600 or 800 should still yield decent results with a not-too-bad margin of error.

posted by Ace at 09:37 PM

| Access Comments