The news is from the US and allows people in Britain to give themselves a quick pat on the back because from what’s come out so far it very much looks as if the British rules on polling transparency would stop any similar situation arising (at least as far as polls commissioned from British Polling Council members go):
I’ve obtained a copy of the lawsuit that Daily Kos just filed against Research 2000, and this going to get nastier than you thought.
The suit contains striking new details about Research 2000’s alleged reluctance to release its raw data and its alleged money problems, directly alleging that the firm committed deliberate fraud by selling DailyKos years of data that was “phony” and “falsified.”
You can read the full story here.
4 Comments
I wonder if the Daily Kos is going to apologise to its readers and critics for years of publishing stuff based on duff data? :o)
It’s good that Markos Moulitsas has been very open and hasn’t attempted to cover up the problems with DailyKos’s pollster.
Their coverage http://www.dailykos.com/tag/Research%202000
Yeah, that’ll go down really well when your client was, not that long ago, “nothing more than a fringe blogger”. Besides which, Nate Silver stopped being a “fringe blogger” what, 2 days after his first post? Three?
Sure, their UK coverage wasn’t brilliant, but it was a lot more incisive than most of the UK media.
It seems that in the US anybody can advertise himself as a pollster, and since there is a dearth of statistical expertise, and pollsters are not required to reveal their internal data or their weighting methodologies (indeed, they regard them as proprietary secrets), they can get away with a great deal. All they are asked for by their clients is to provide plausible topline numbers; and all those who hire them can go on is (a) reputation (which really means that they delivered the numbers expected) and (b) accuracy. In the case of the Research 2000 work for Daily Kos, they were asked to provide tabular breakdowns of subgroups as well. It was analysis of these breakdowns that revealed that there had been a fishy manipulation of the numbers, or possibly even wholesale invention of data. While Daily Kos can be faulted for not doing a deeper investigation of the pollster it hired, in fact there was no expectation that there was anything to check. Pollsters were considered to be accurate or inaccurate, possibly distorting the results by dodgy questions or inappropriate weighting, but the concept of outright fraud was not considered. Presumably that will now change.