One of the standout lessons we should all have learned from the 2017 General Elections is that most of the polling companies got the result spectacularly wrong.
Aside from Survation who called in almost spot on, the rest of the big polling companies ranged from poor to absolutely woeful. The final result was 42% - 40% for the Tories, but in the final week most of the big polling companies totally embarrassed themselves by predicting landslide Tory victories. Yougov predicted a +7 point win for the Tories, Ipsos Mori and Panelbase went +8, ComRes went +10, ICM went +12, and BMG went +13.
We now know that they called the election so wrong because they weighted their results on the assumption that the younger generations wouldn't be arsed to get out and vote, an assumption that was destroyed by Labour's spectacularly successful online campaigning, and their decision to actually offer a manifesto of hope for the younger generations.
Fast forward five months to November 2017 and right-wing Twitter suddenly erupted into a massive collective orgasm over a Kantar poll showing a 4 point poll lead for the Tories.
From "centrist dad" Blairite types through to the toxic new Tory Blue-kip Tory demographic, it seemed that every right-wing Twitter account I followed was suddenly pushing this outlier poll result.
I thought the result was more than a bit odd, so I decided to have a look at Kantar's polling methodology to find out how they generated this result. The first thing to note is that their unweighted results actually showed a Labour lead, but the real kicker is the way they simply decided to ignore what happened at the 2017 general election completely, and predict a pitiful 19% turnout for the 18-24 age group, and a feeble 37% turnout for the 25-34 group.
So the only way they managed to generate such a positive poll result for the Tories was by assuming that the 2017 General Election didn't actually happen, and that the massive surge of support for Labour amongst the under-50s is suddenly going to dissipate back to nothing.
Kantar have publicly shamed themselves by using such an obviously outdated and inaccurate methodology in order to generate an outlier poll result, but one of the most interesting things is the way that the flood of right-wingers who shared this absurd rigged poll didn't even have a "that's odd" moment, and didn't bother to even remotely investigate how such an outlier result was created before they reposted it on Twitter with their snarky anti-Corbyn / anti-Labour jibes.
They just loved the rigged poll result because it confirmed their personal biases, and shared it without bothering to look into the details at all, which means they've all embarrassed themselves just as much as the Kantar pollsters who cobbled this pitiful nonsense together in the first place.
The whole episode just goes to show the incredible levels of reality-aversion some people are suffering, from the Kantar pollsters who used such an embarrassingly out of date methodology, to the unthinking right-wing Twitter mob who unquestioningly Tweeted it out because it confirmed their personal political biases.
Another Angry Voice is a "Pay As You Feel" website. You can have access to all of my work for free, or you can choose to make a small donation to help me keep writing. The choice is entirely yours.
No comments:
Post a Comment