June 30, 2010
Daily Kos: Stat Detectives Prove R2K Poll Results Bogus
It's a detective story for statistics geeks: A group of math brainiacs dissect a major polling organization's suspicious results to reveal statistical anomalies that could not have happened by chance. Read the article "Research 2000, Problems in Plain Sight" at the Daily Kos.
Here is part of the introduction:
A bit over two weeks ago, a group of statistic wizards... approached me with a disturbing premise -- they had been poring over the crosstabs of the weekly Research 2000 polling we had been running, and were concerned that the numbers weren't legit... The investigators' report is below, but its conclusion speaks volumes: "We do not know exactly how the weekly R2K results were created, but we are confident they could not accurately describe random polls."
Even if you don't have much math or statistical background, you will find the process these statistical detectives used fascinating. A mystery with statistical clues! I love this stuff. (Thanks to Katie Paine for the hot tip. She loves this stuff too.) -- Bill Paarlberg, Editor
See more coverage of this situation (Hint: Some lawyers are going to be very busy, and a pollster needs a new job.) in TPM Muckraker:
- Kos Alleges Top Pollster Provided Bogus Data, Will Sue
- Kos Lawyer On R2K: 'He Handed Daily Kos Fiction And Claimed It Was Fact'