Tuesday, May 09, 2006

Exit Poll Debunking Tied to Right-Wing Fundie?

(Kathy Dopp, this one's for you.)

Febble - well-known exit poll debunker

The folks at Daily Kos know her...

EXIT POLLS "all over estimated the Democratic vote"

"Exit polls again: the mystery of the wandering WPE"

The good folks at DU certainly know her...

"A few corrections on the above Febble-isms"

"Junk Logic? Febble, you are getting very nervous, aren't you?"

And now it's time to find out what you don't know.

Febble's exit poll debunking tied directly to right-wing fundie influence?

There was a time when Febble looked like a Democrat...she spoke out about the depression of Democratic voters in Ohio...she posted, essentially, that it was possible that Bush didn't really win the election...but then it looks like something changed, as she began to debunk exit polls that clearly showed a Kerry victory in 2004.

Now I know you guys have heard of the Mystery Pollster, and I know that Kathy Dopp has heard of Mark Lindeman, aka HudsonValleyMark at Daily Kos - but it's Rick Brady who I'd like to talk about today, the same Rick Brady about whom Febble had this to say on December 23rd, 2004:

"Mystery Pollster cites Rick Brady who contacted Warren Mitofsky, of Mitofsky International, and obtained values of between 50% and 80% for the design effect."

By the way, don't forget that name HudsonValleyMark - we'll come back to him one of these days.

Rick Brady revealed - perhaps anonymity does have it's advantages?

So who is this Rick Brady guy, who the Mystery Pollster gave so much credence to, the same Rick Brady that Febble posted this about?

"my collaborators (Rick Brady has also been working with us) and I are..."

Well, thanks to Rick's Kos profile, we've got a most-excellent place to begin our little adventure, and don't forget to check out his fellow posters - http://stonescryout.org.

1. Looks like he's a planning consultant - I'm sure that prepares one well for solving the mysteries of exit polls - but all humor aside, perhaps it does help out when planning a debunking operation.

2. "In 2001, Rick attended the Horizon School of Evangelism" and "he enjoys leading worship for his home fellowship."

Fellowship - the word just stands out like a sore thumb. Let's look further!

3. On the left side of the page are all sorts of links to interesting right-wing sites and blogs, like NRO, and Weekly Standard, and hey...there's even Captain's Quarters there!

4. But the really cool thing is that Rick mentions an older version of the site, not-so-long ago...

5. At the older site on the right-hand side, we find links to the Washington Times, Drudge Report, Mystery Pollster, Red State, and then we scroll down to this:

Midnight Heartburn

Andrew Sullivan
Joshua Micah Marshall
Talk Left

Wow, I wonder what Markos would think of Rick Brady if he knew that his site was considered unpleasant to one's stomach!

But to be fair, I still haven't proven my case, so let's look even deeper...

6. 1/31/05 - "Our troops have been marching with the Iraqi people towards this day since March 2003...Bush's victory, declared today, is the world's victory...we are living in historic times..."

7. 1/29/05 - "It's a glorious morning. The news from Iraq is better than I think anyone, even the Bush administration, expected...Both Russert and Senator Kerry still don't get it...Iraqis get it... Their neighbors are getting it...When will the left get it?"

8. 12/28/04 - "The Dems are obviously desperate. How petty would the Dems be if they really go through with procedural sabotage? On the campaign trail, John Kerry...isn't it fitting that he and the Dems in the Senate are threatening to become a nuisance. Shoo fly, don't bother me!"

9. 12/22/04 - a long-winded post in support of Bush's plans for Social Security.

10. Also on 12/22/04 - "I am also an activist for the Republican Party because I believe the values of the Party are closest to my family's values."

I'd like to repeat that last one..."I am also an activist for the Republican Party because I believe the values of the Party are closest to my family's values."

Like tax cuts for the rich, massive deficit spending to create the illusion of a decent economy, wars based on a stack of lies and false evidence, cake for the victims of Katrina, illegal wiretaps, yada yada yada?

I wonder if all those Kos folks who supported Rick, Febble, and HudsonValleyMark so aggresively at Daily Kos, like DHinMI, DemFromCT, Timroff, and Lefty Mama, would be so willing to support them after learning that Rick is a Republican activist?

So is Rick Brady a fundie? A member of "The Fellowship?"

Or is he just a right-wing Republican activist who just happened to be warmly greeted at Daily Kos, and just happened to partner with other individuals in search of the perfect formula for debunking the absurd notion that Kerry might have actually won the 2004 election?

Perhaps a kind reader can move this research forward.

Thanks for reading!



Blogger Febble said...

Sure, I can move this "research" forward. I can actually tell you quite a lot.

My name is Elizabeth Liddle. I am a British citizen and I live in Nottingham, UK. I've worked with Mark Lindeman and Rick Brady, and we presented an invited joint paper on the subject of exit poll analysis at the 2005 annual meeting of the American Statistical Association in Minneapolis. The paper is published in the proceedings of that meeting.

I've never met Rick, but I've come to know him well through our email collaboration. I don't share his political views, but that does not stop me having a profound admiration for his incisive and honest intellect. I myself am a member of the British Labour Party, but I've always had an interest in international politics, because, essentially, I believe that we are all citizens of the same planet.

If you want to know my religious beliefs, I'll tell you those too. I'm a catholic, although not at all happy with the way my church seems to be going (or not going). If you search my Daily Kos diaries, you can find out more about those.

I'm also a scientist. That means I'm more interested in getting the right answer than the answer that happens to suit my own desires. I was appalled at the blatant injustices of the 2004 US presidential election, quite apart from being deeply distressed that Bush was re-elected. I still mourn President Gore. Being a trained data analyst I wanted to get to the bottom of what had happened, and, like many people, I was suspicious of the result - the insecurity of computer voting systems, the fact that Gore won the popular vote in 2000, and the big discrepancy in the exit polls made me suspicious.

You probably know the rest of the story. I pored over the Edison-Mitofsky January evaluation, found what I thought was a major problem with the analysis, and posted a paper (the ideas of which I developed in collaboration with Rick Brady, Mark Blumenthal and DemfromCT) proposing a different analytical approach. DemfromCT frontpaged a DKos diary about the paper, Mark Blumenthal did a post on MysteryPollster, and I diaried it myself.

As a result of that paper, I was contracted by Mitofsky to re-analyse the exit-poll data using my suggested measure of exit poll discrepancy. My most important finding was that there was absolutely no tendency for Bush to do any better, relative to his 2000 performance, in precincts with large "redshift" than in precincts with smaller degrees of redshift or blueshift. This finding strongly suggests that vote-shifting fraud was unlikely to have been responsible for more than a small proportion of the the exit poll discrepancy, and implies that likely estimates of the magnitude of vote theft are not in the millions. In other words, the finding suggests that it is extremely unlikely that vote theft occurred on a scale to have stolen the popular vote for Bush.

Why any intelligent person should think that the results of a correlational analysis would depend on whether or not I collaborate with someone with different political and religious views, I have no idea. Last time I checked, correlation coefficients computed by evangelical Christians came out the same as correlation coefficients computed by atheists. And I don't use a rosary when I use SPSS.

Anything else you want to know?

You only have to ask. Research really isn't all that difficult, when you are researching people who don't have anything to hide.

10:11 AM  

Post a Comment

<< Home