why did the pollsters underestimate the historic breakthrough of the RN?

An unexpected…and historic breakthrough. The National Rally is in the process of obtaining “by far the largest group in the history of (his) political family”as declared by the president of the party, Marine Le Pen, on the evening of the second round of the legislative elections, Sunday June 19.

>> Results, analyses, reactions… Follow the election evening of the second round of the legislative elections in our live

The far-right party won 90 seats in the future hemicycle of the National Assembly, according to an Ipsos-Sopra Steria estimate for France Télévisions, Radio France, France Médias Monde and parliamentary channels. A score much higher than that predicted by the polling institutes, which credited the RN with 20 to 50 seats. Franceinfo interviewed Mathieu Gallard, director of studies for the Ipsos institute, to try to understand why the pollsters were wrong.

franceinfo: how to explain the difference between the projections made by the polling institutes and the first results very favorable to the National Rally?

Matthew Gallard: It is true that we announced up to about 50 seats for the National Rally, and it will be significantly higher. This is explained in particular by the fact that the vote reports for the candidates opposed to the RN candidates were very bad. The far-right candidates did not arouse fear among voters who opposed them. When an RN candidate had a duel with the Nupes, the voters of Together! did not go to vote for the candidate of the left. And when an RN candidate faced a candidate from Ensemble!, the voters of Nupes did not move either. There is no longer a Republican front, except on the margins for the LR candidates who were facing the RN candidates.

Why was this split in the Republican front difficult to anticipate?

The fact that members of the government and LREM candidates attacked Jean-Luc Mélenchon and Nupes in a very violent way has developed very antagonistic relations between these two electorates. This is a trend that we had seen, but whose extent we had underestimated. For a left-wing voter, Together! is as hated as the RN. And a voter of Together! does not feel anything in common with a voter of Nupes. This is proof of the tripartition of political life, which we had already noted during the presidential election, with three large blocks that are completely sealed: RN, Together! and Nupes.

Abstention is a little lower than in 2017. However, we know that high abstention is traditionally harmful to the RN score. Is this another explanation for the party’s higher than expected score?

No, I don’t think so, because abstention is still higher than in the first round. And the RN had obtained a good score last Sunday despite strong abstention, which we had well anticipated.

Does this forecasting failure call into question your methodology?

We cannot yet say definitively why we were wrong. There may be other reasons than the poor vote transfer from which the non-RN candidates benefited, we will have to work on it and it will take a little time. But I don’t think it will fundamentally challenge our methodology. This evening’s results show what we already know: the projections for the number of seats are fragile because a variation of a few points in the vote reports of certain electorates has major consequences on the results, since a lot of seats are up for grabs. a point or two. And at the same time, if we give ranges in the number of seats that are too wide, that doesn’t interest many people…


source site

Latest