Markos Moulitsas of the Daily Kos says he'll sue Research 2000, the polling firm that has conducted a weekly State of the Nation poll for the site for a year and a half.
Research 2000 -- whose client list includes the Washington Post, the New York Times and the Wall Street Journal -- tells Talking Points Memo it has done nothing wrong. "I will tell you unequivocally that we conducted EVERY poll properly for the Daily Kos," company president Del Ali says.
Moulitas, aka Kos, disavowed every poll done by Research 2000 after getting a report from three "statistics wizards," who wrote: "We do not know exactly how the weekly R2K results were created, but we are confident they could not accurately describe random polls."
Among other evidence, the stats wizards present the following results as almost statistically impossible. Take a look and see if you can spot the same weirdness that caught their attention. I'll put the answer after the jump.

ANSWER, for those who ready to read it.
In every instance, the result is either even for both genders or odd for both genders. The investigators write that the pattern is repeated, a lot:
For their favorable ratings (Fav), the even-odd property matched 776 times. For unfavorable (Unf) there were 777 matches.
Common sense says that that result is highly unlikely, but it helps to do a more precise calculation. Since the odds of getting a match each time are essentially 50%, the odds of getting 776/778 matches are just like those of getting 776 heads on 778 tosses of a fair coin. Results that extreme happen less than one time in 10228. That's one followed by 228 zeros. (The number of atoms within our cosmic horizon is something like 1 followed by 80 zeros.) For the Unf, the odds are less than one in 10231. (Having some Undecideds makes Fav and Unf nearly independent, so these are two separate wildly unlikely events.)





I can't think what sort of raw data manipulation would result in this 'feature'. But a report writer rounding defect... that I can visualize.
I want Nate Silver's take on this one.
Nate has an article up on this one already:
http://www.fivethirtyeight.com/2010/06/breaking-daily-kos-to-sue-research-2000.html?utm_source=twitterfeed&utm_medium=twitter
I bet if you give this info to Jeff (Forrest Gump) Sessions he can figure out away to blame it on Elana Kagan in the hearings.
Figures don't lie, but liars figure.
I also find it interesting that all male results across horizontally equals 100, which is sensible. However; Boehner's male results add up to 110. Can anyone explain this oddity?
*98% of all statistics are wrong* Always found this to be a funny quote.
Yes, it was a typo on my part. Doesn't affect the even/odd stats.
Oops, I'm sure the story meant 10e228, 10^228, or 10[superscript]228, all meaning 10 to the 228th power.
Also I'm pretty sure the numbers for Boehner among men do not add up to 100%.
26+51+33=110. Is this right? I checked it with Excel !
"the numbers for Boehner among men do not add up to 100%."
Which is why Pfizer invented Viagra.
Please tip your waitress, I'll be here all week.
i noticed that, too
Ba-Dum-Bum. [Rimshot for KevinTP]
"There are three kinds of lies: lies, damned lies and statistics."
---Mark Twain
Which is a quote Twain attributed to Benjamin Disraeli (who is unlikely to have attributed it in turn to William Gladstone, I imagine).
Research 2000 : 60% of the time, it works every time. (http://www.sex-panther.com/)
Sounds like someone has the rarely seen three sided coin. And they are using it!
If you're going to post this, you have to format the numbers properly. It's 10^228, if you can't do superscript. Not 10228.
Wow
Like duh!
And Boehner's numbers add up to 110%.
Discounting the source of the statement (Boehner), the intent is the correct step. i.e. raising the retirement age to at least 70, albeit gradually.
clearly boehner has a stiff drink each time he opens his mouth and ppl needs one right after that hahaha
I see the women are strongly opposed to Senator McConnell, Senator Boehner, the Republican Party and the Republicans in congress
BP Spills Coffee you tube1
follow the bp spills coffee link on you Tube rather funny.....sally from the UK
Is Frank Lutz in charge of the stats dept now? I thought he was polling walruses in the Gulf.