I'm writing a CSS book

JavaScript archives Compatibility Thidrekssaga QuirksBlog Donations CSS book Politics Mobile About COH

contact

Polls, polls, polls

New poll found, interesting stats surfaced, and the role of the SGP as a canary. Welcome to Political Quirks, poll edition.

Fifth pollster: Buzzpeil

Yesterday I found that there is a fifth pollster: Buzzpeil. It’s a “social media” pollster that doesn’t actually ask people what they’re going to vote, but instead looks at their tweets and Facebook posts — or something.

The poll is mainly remarkable for the complete absence of 50Plus — apparently the elderly don’t tweet enough. Also, a few weeks back the left block got 76 seats, something it’s never got in any other poll.

I don’t really believe in Buzzpeil’s methodology since there’s no way of correcting for those voters that don’t use social media, but I guess we should give them a chance and see what their error is after the elections. So I added it to the polls page.

The plan seems to be to publish a poll per day, but I’ll restrict myself to copying one per week. Oddly, the last published poll is from 20th of August, nearly a week ago. Let’s hope Buzzpeil will continue polling.

What do party voters do?

Translation of legend:

Zelfstandig ondernemer

Working for themselves

Werkzaam in loondienst

Working for a boss

Werkzaam bij overheid

Civil servants

Arbeidsongeschikt

Jobless for reasons of health

Werkloos/bijstand

Jobless because they can’t find a job

Gepensioneerd/VUT

Pensioners

Studerend

Students

Huisvrouw/huisman

Housewives or ... househusbands, I presume

De Volkskrant and TNS-NIPO did interesting research about whether voters for the six big parties work or not. Unfortunately the research is not online; this picture is the best I got.

The take-aways are the following:

• For CDA and PvdA the share of pensioners is largest, while the share of working people is smallest. (Smallest meaning 37-42%, so it’s not actuall small.)
• The percentage of jobless is largest for the PVV and SP. Apparently the jobless (whether for reasons of health or because they can’t find a job) are most inclined to cast a protest votes. These groups are smallest with VVD and CDA.
• D66 has the largest group of working people. That is somewhat unexpected: plenty other parties claim to be for the hard-working Dutch.
• The PvdA has the largest groups of civil servants, followed by VVD. So much for the leftist church that dominates our civil service.

Take a look yourself; I hope the translation of the legend helps. (These terms are not easy to translate, I can tell you, so it sounds a bit stilted.)

The SGP as a canary

If a new poll is released, the first thing you should do is look at the SGP — the ultra-orthodox protestant party that’s against women in politics, TV, and doing anything on Sundays. It will win two seats in the elections, but recently quite a number of polls from all pollsters gave the SGP three seats.

Why? Don’t the pollsters know that the SGP always wins two seats? Yes they do, but the problem here is turnout.

Back in 1998 the SGP won three seats because turnout was so low in that election — only about 70%, if I recall correctly. Since SGP people always vote no matter what, they can help their party to a third seat if the rest of the electorate is lukewarm.

As far as I can see that’s what’s been happening in the polls recently. The number of SGP voters remained stable, but the number of voters for other parties declined. Turnout is low, in other words.

It’s clear that this turnout problem exists mostly on the right. The left is doing well with a healthy score of about 70-74 seats. The right is doing lousy with a score in the low sixties.

That’s understandable: the Rutte government consisted only of right-wing parties, so right now disappointed right-wing voters have no place to go. (Well, to D66, but that’s not for everyone.) An abstention is logical.

Now the really interesting part is that the SGP is going back to two seats in three out of five polls. That’s not unexpected in itself, but I expected the parties on the right to do somewhat better now: apparently turnout is rising, which, with most non-voters residing on the right, should mean that the right gets more seats.

To be honest I don’t know what’s going on. It could be that the pollsters tweak their SGP scores a bit because the elections are looming near and they all know the SGP will win two seats. It could also be that significant numbers of right-wing voters shift to the left, but I don’t believe that. It could be that all pollsters underestimate the right, which is possible, but not really convincing.

Let’s wait for the next set of polls and then look at our canary again.

<— New parties and their success | Party profile — CDA —>

Comments (closed)