Home > News > When is statistics manipulation evil?
163 views 5 min 0 Comment

When is statistics manipulation evil?

- April 7, 2015

Thomas Lumley points out this misleading headline from the (London) Daily Express:

Here’s what it says in the newspaper:

The biggest vote on this country’s ties to ­Brussels for 40 years saw 80 per cent say they no longer want to be in Europe, the ­Daily Express can reveal.
It marks a huge leap forward in this news­paper’s crusade to get Britain out of the EU.

Actually, the above headline is beyond misleading; it is an outright lie in any meaningful sense of the word. Lumley explains:

This comes from a survey in three Conservative electorates in the southern UK (out of 650 electorates), where 100,000 questionnaires were distributed. About 12 percent said Britain should leave the EU, about 3 percent were opposed, and the other 85 percent didn’t respond.

The 85 percent nonresponse doesn’t bother me so much — that’s par for the course with surveys — but there are obvious problems with (a) sampling from a population that is unrepresentative of the larger population of interest, and (b) not even trying to adjust for differences between sample and population.
Why do I characterize this as “evil,” indeed far more “evil” than the sort of everyday lying we see in politics, from Bill Clinton or Newt Gingrich to whoever happened to be caught in a lie most recently?
To me, the difference is that Clinton, Gingrich, etc., lie because they feel they’re trapped, and they can’t admit the truth. I feel the same way about various liars and plagiarizers in journalism and science and academia. I deplore their behavior and I’m distressed that they won’t own up to it once they’re caught, but their denials are really just a natural continuation of what they did earlier.
With the Daily Express it seems different. They’re creating a lie out of whole cloth.
The next question is: why am I so much more bothered by this lying news item, so much more so than by a comparably misleading effort by a public relations firm (“Americans are buying more potato chips!” or whatever)? Here it seems to be about roles and expectations. Public relations firms are expected to exaggerate, even to lie. It’s pretty much their job to push the envelope of truth. But newspapers are different. People come to newspapers for . . . news!
Maybe this just represents my ignorance of the English press. Perhaps the Daily Express is like the National Enquirer, with people reading it for pure entertainment value and enjoying the outrageous misrepresentations and the P. T. Barnum-like attempts to sucker you into buying a copy of the paper. Hmm . . . I just checked Wikipedia and it seems that the Express isn’t a real newspaper, here it says “The chairman of the Press Standards Board of Finance, which manages PCC funds, described Express Newspapers as a ‘rogue publisher’ . . . in August 2009, the Advertising Standards Authority criticised the company for running advertorials as features alongside adverts for the same products. . . . In January 2010, the Daily Express was censured by the Advertising Standards Authority over a front-page promotion for ‘free’ fireworks . . .”
So I guess it really is just a scam sheet. In that case, I’d put the Express in the same category as those Nigerian spammers. The difference is, perhaps, that from the outside it looks like a legitimate (if lowbrow) newspaper, so it’s polluting the public discourse in a way that more obvious spammers aren’t. Sort of like if some company started selling fake-o computers under some convincing-sounding name such as InterTech, or someone sold fake diplomas from a real-sounding college such as Chatterton University. They’re playing off the credibility associated with legitimate news sources, thus degrading the public discourse.
To me this is even worse than when ignorant billionaires spam people with graphs of fake statistics. Presumably these guys are sincere but clueless, which does not seem as bad as that fake English newspaper deliberately spreading lies.