Slashdot Mirror
Daily Kos Pollster Made Up Numbers
jamie found a story up on Daily Kos revealing that the polling firm they had contracted with for 18 months, Research 2000 or R2K, apparently made up or at least manually tweaked its polling results. The blog published a preliminary report by a team of statistics gurus (Mark Grebner, Michael Weissman, and Jonathan Weissman), and it is an exemplar of clarity and concision. The team reports, "We do not know exactly how the weekly R2K results were created, but we are confident they could not accurately describe random polls." Daily Kos will be filing a lawsuit against its former pollster. "For the past year and a half, Daily Kos has been featuring weekly poll results from the Research 2000 (R2K) organization. These polls were often praised for their 'transparency,' since they included detailed cross-tabs on sub-populations and a clear description of the random dialing technique. However, on June 6, 2010, FiveThirtyEight.com rated R2K as among the least accurate pollsters in predicting election results. Daily Kos then terminated the relationship. One of us (MG) wondered if odd patterns he had noticed in R2K's reports might be connected with R2K's mediocre track record, prompting our investigation of whether the reports could represent proper random polling. ... This posting is a careful initial report of our findings, not intended to be a full formal analysis but rather to alert people not to rely on R2K's results."
In Neal Stephenson's Cryptonomicon, there's a scene early in the book where the Allies are assembling the personnel for Station X (aka Bletchely Park). Statistician, turned Nazi codebreaker Lawrence Waterhouse, points out that his Nazi counterpart Rudy von Hacklheber, would notice something was amiss with the Allied personnel changes based the statistics of people being transfered to Bletchely Park, and then quickly deduce that the Allies are attempting to break the Enigma code. To camouflage the transfers, Waterhouse suggests creating ficticious personnel and have some of them transfered to Bletchely Park as well. However the military can't just make any random fake person, the fictious people must be statisitically drawn from a distribution that when added to distribution of real Bletchely Park personnel, the combined distribution is statistically insignificant (i.e. fail to reject the null hypothesis) than any other large military base.
If Research 2000 did what is suggested, they failed to taint the polls with the right kind of fake data, just like what the novel warned about.