As with last time round, I’m weighting the lowess trend line seen here by the number of respondents in the sample. Unfortunately, one of the polling companies, TermometroPolitico, is using an internet panel; because of the differences in make-up of internet panels compared to phone panels (and neither is perfect), certain cases tend to be weighted more heavily, which means the standard error might still be high even if the number of respondents is large.

At the moment, termometropolitico seems to be weighting on past vote recall and demographics; most phone interviewers just weight on demographics. My only concern about this is that respondents will falsely remember that they voted for the ‘winners’  in 2008, which might depress the centre-right’s vote share. Indeed, Termometropolitico’s estimates for the PdL tend to be on the low side.

That was fun.

This year, I’m aggregating the polling data for the European Parliament election — the Italian part, anyway.

This is less fun.

As everyone knows, European Parliament elections are second-order elections. Voters use them not to vote on the issues — they don’t know about the issues — but instead to give the government a progress report.

And whilst progress reports are interesting, your view of government performance might not be rocked that much by campaign events. And that’s when there is a campaign — so far, the parties have promised to spend less “in light of the current economic climate”.

With that out of the way, I present you with some flat lines:

Unfortunately Le Parisien does not provide the text of the questions asked; Le Monde is somewhat more helpful, but one would have to see the questions to ensure that the wording wasn’t skewed.

If there was no movement over the last two weeks – a huge assumption – then we can see that, contrary to last time round, pollsters did not underestimate support for the main party of the centre-right, the PdL. Rather, they overestimated PDL support, but had problems  in estimating the success of the Lega.

Intriguingly, the Lega was under-estimated by about as much as the Sinistra Arcobaleno was over-rated. The Lega Nord now claims to be the working class party (partito operaio) of the North. Certainly, it would be interesting to find out how many former Communist voters, losing traditional voting cues with the disappearance of an overtly Communist party, switched their vote to the Lega at the last moment, but this is based on pure speculation.

Of course, this only works if actors respect the law. Berlusconi doesn’t. Instead, he’s been leaking publicising polling info about a continued seven to nine point advantage for the PdL.

• Polling accuracy: what happened last time.
• Where the battleground regions are
• A post  on methodology

Further notes on methodology are available in the last polling round-up, available here.

In a search to understand what might happen between now and election day, I’ve taken a look at the polling from the last election. Then, the pollsters got it wrong – either that, or people changed their minds in between the polling embargo and the day of the election.

The answer, I think, is a little bit of both. I’ve replicated the wonderful analysis of Simon Jackman for the 2006 polls, using Italian data from Callegaro and Gasperoni as a starting point, and adding data from the ITANES tracking survey from the 40 days before the election.

Key points?

1. People did change their mind in the last fourteen days, and they shifted to the centre-right;
2. All but five of the 73 polls overpredicted the left’s support;
3. The infamous Penn, Schoen, & Berland poll was not accurate at the time it was taken;
4. The three most accurate pollsters were Piepoli, TNS Abacus, and Ispo; unfortunately TNS Abacus hasn’t provided a poll this time round, and Piepoli has only supplied one
5. The three least accurate pollsters were Tql, Simera, and Brunikmedia. Only Brunikmedia is still providing polls.

Response to Callegaro and Gasperoni, pdf.

How exciting: the trend is exactly where it was before the polls were added, with a gap of 7.3%. And, indeed, before the previous batch of polls was added.

If you believe that there will be a uniform national swing – and if you believe that the Destra’s role in Lazio will be inconsequential – then bet on a narrow majority for the PdL in the Senate, with 164 seats compared to the cut-off point, 158. Hopefully over the next couple of days I’ll be looking at the polls in retrospective, and seeing how much confidence we can attach to each pollster’s estimate.

I’m still projecting a majority for the PdL, but I’m less confident of this judgement for reasons noted in the last post – that, essentially, the game will be won or lost in Lazio, and we just don’t have information to judge that.

Additionally, seat estimates for the Sinistra Arcobaleno and the UDC are below what the parties themselves are predicting.  Not too surprising, perhaps,  but the UDC claims that their votes will be determining in the Senate, so they clearly expect more than 2 seats. A strong showing in Sicily, certainly, but where else?