Monday, April 23, 2012

A Question about the Polls and the Vote

Several commenters have suggested that the discrepancy between the FN vote and the polls shows that people are still reluctant to own up to voting for the extreme right (a version of "the Bradley effect," as it were). But no one has explained why the polls overstated Mélenchon's vote. Were left-wing voters perhaps ashamed to say that they were voting for the bland Hollande rather than the fiery Mélenchon? Is it prerferable on the left to be thought a revolutionary rather than a pragmatist?

A different sort of explanation would look at the polls themselves. French pollsters rely on a method of quotas, which requires considerable understanding of the structure of the electorate before sampling begins. Since the parties in France are unstable, especially the minor parties, that knowledge, based on analysis of previous elections, may be incorrect, leading to biased sampling. I don't know enough about the methods used to say why there might have been systematic errors in favor of Mélenchon and against Le Pen. Perhaps the urban concentration of JLM voters as opposed to the periurban concentration of MLP voters has something to do with it, as demographic patterns have been shifting too. In any case, I am skeptical of the "Bradley effect" stories and would suggest that the flaws are in the polls instead. And this raises doubts about the round 2 polling. The polls were actually quite accurate for the top-tier candidates in round 1, but round 2 depends on getting the transfer of votes right, and the poor results for the second- and third-tier candidates indicates that the polls may not have a good handle on this. So I'm not ruling out a significant surprise on May 6.


Anonymous said...

YOur wish is my command:


My friends/acquaintances who wanted to vote Mélenchon ended up voting for Hollande at the last minute because so many people around them spoke of Marine Le Pen. I don't know how pervasive this last-minute change was in the French population at large but it's surprising how many people switched between last week and today.

Anonymous said...

Alain Juppé, un des piliers du gouvernement et de la campagne de Nicolas Sarkozy s’est vendu : « Si Nicolas Sarkozy perdait, nous serions un certain nombre à tout faire pour que l’UMP garde sa cohésion ».

On n’évoque jamais la défaite lorsqu’on croit à la victoire.

Marine Le Pen, FR2: "J'ai entendu Nicolas Sarkozy dire à mes électeurs : "je vous ai entendu". Ca me rappelle le "Je vous ai compris", et ça, ça s'était mal terminé." "celui qui aura perdu ce sera de sa faute, uniquement de sa faute, c'est trop facile de dire que c'est la faute des autres."

FrédéricLN said...

Well, sampling is based on demographics, including of course the town size ("tranche d'agglo"), so the Paris vs countryside factor should not have generated any bias.

The post-survey weighting is, the pollsters say, based on the declaration of previous votes.

But the largest difficulty altogether is that (in France as in the USA or elsewhere) less and less people agree to answer to surveys. And the "acceptance" factor may be sharply correlated with future vote, also within the previous electorate of some candidate.

In short: people who reject "le système au pouvoir" may also reject opinion polls, as a part of this system.

Anonymous said...

More news:
Sarkozy promises a May 1st rally about "real work".As of now, the concept sounds kind of confusing even to UMP representatives.
Royal asked NKM about her definition of "real work" and got no answer except that "real workers" are those not in unions. A problem I see with this is that in industrial areas, if we're to believe the various news segments, unionized factory workers voted... Le Pen and Sarkozy's trying to get their vote.
Libération weighs in.

bernard said...

I am no great fan of Sarkozy to say the least, but he was right when he said that looking at the polls on any given day did not give you a result. You had to also look at the dynamics revealed by polling over time.

As a result I am not sure that polls over-estimated the vote for Melenchon. I had noted in the back of my mind that he had seemed to top around 14% and, then, in the last 2 weeks prior to the first round, seemed to be edging down. The first round may have been a continuation of this retreat witnessed by numerous polls. It may, of course, possibly be explained by the accentuation of the "vote utile" campaign in the last 2 weeks prior to voting.

bernard said...

not to mention the standard error of estimation with a 5% degree of confidence decision rule of course...