Published On: Mon, Mar 13th, 2017

Political polls

The parliamentary elections on Wednesday in the Netherlands are not only a test for the political parties that have been angling for favors from the electorate. They are also a test for the polling agencies that have been entertaining the country with one poll after the other. But how reliable are these pseudo-scientific exercises?

In 2012 Maurice de Hond was pretty close with his predictions, in the sense that the parties that were leading in his polls also turned out to be the winners in the real elections. But there were, of course differences because an opinion poll is not full proof.

In 2012, the VVD beat the De Hond poll by five seats (winning 41 instead of the projected 36), while the Socialist Party fell short of expectations (15 instead of 20). The PvdA did slightly better than projected (38 vs. 36), while the PVV won 15 instead of 18 seats. With the other major parties – CDA, D66 and GreenLeft De Hond was reasonably on the mark.

So what does his poll have in store for us this time around? If we have to believe the polls the PvdA is heading for a dramatic loss. It will fall from the 38 seats it holds now in the Second Chamber to a miserable 9 – at best a fifth wheel on the wagon for the next cabinet.

Its coalition partner VVD also stands to lose a big chunk of its current 41 seats and drop to 24, just enough to keep André Bosman in the Parliament.

Of course the most spectacular projection has to do with GreenLeft and its young leader Jesse Klaver. In the De Hond poll, GreenLeft will skyrocket from its current 4 seats to 20 and make it a serious contender for participation in the next government.

The CDA is also doing well in the poll. It is scheduled to improve from its current 13 seats to 22, on a par with the party of the man who made the most noise in the political landscape – Geert Wilders and the PVV.

The PVV now holds 18 seats and will win 22 on Wednesday, the De Hond poll projects

The Socialist Party will hang on to its 15 seats and is thus treading water, while D66 will improve from 12 to 17 seats.

All this, of course, according to the De Hond poll.

Peilloos.nl reported in 2012 about the reliability – or lack thereof – of political polls. The site quoted Joop van Holsteyn, a professor in voter-research, who noted that the so-called ‘floating voters’ (those who have not made up their mind yet) are probably the most involved in any election. “He knows what he wants but he makes his choice at the very last moment because then he has the maximum amount of information about the balance of power between the parties at his disposal.

“The source of all knowledge?” Van Holsteyn wrote. “Polls. The strategic voter takes all polls serious as a yardstick for the power ratio. He can hardly do anything else. The polls are the point of departure on the road to the political destination; to get there sometimes requires a strategic vote. And that vote could result in a different result that earlier polls suggested.”

According to Van Holsteyn’s argument polls are wrong by definition. And he may have a point there.