Pedro Magalhães

Margens de Erro

PS and PSD in the Portuguese polls

Posted May 14th, 2012 at 6:16 pm4 Comments

As usual, very few polls, nothing since April 20th (!), everything very dependent on the leverage of particular polls from particular pollsters. And yet, note the recent unusual pattern when we take into account normal polling patterns for PSD and PS: one party's loss has ceased to be the other party's gain.



I'll leave the deep interpretation to you, but it doesn't seem too difficult.

by Pedro Magalhães

French polls, 2nd round (May 4th update)

Posted May 4th, 2012 at 2:30 pm4 Comments

Including polls made public yesterday and today:




















And zooming in on the last month:



















So I would not open the bottle of champagne right now.

by Pedro Magalhães

France 2nd round polls, May 1st update

Posted May 1st, 2012 at 2:24 pm4 Comments




















There's a petit frisson around Sarkozy "narrowing the gap," and yes, there may be something going on along those lines. But a little perspective: looking at the last two polls of each of the eight pollsters, Holland has declined in just four of them, and by a maximum of two points. The lowest score for Hollande in any poll is 53%, i.e, about what Sarkozy had against Royal at about this point in the campaign in 2007, or what Chirac had against Jospin in 1995. Sarkozy ended up with 53.1% of the vote and Chirac with 52.6%. Having said that, considering his record, it's quite extraordinary that Sarkozy manages to keep this even vaguely competitive. And I can't help thinking about the famous  "shy Tories" phenomenon in the UK 1992 election: could it be that polls are underestimating the rightist vote? French pollsters have a good track record, but that's only if you decide to discount the FN vote. A Socialist candidate as favorite is, as we know, an unusual phenomenon. And a look at the technical reports of the French polls shows quota sampling all over the place, and it's difficult not to be a little freaked out about that.

by Pedro Magalhães

France 2nd round polls, April 30th update

Posted April 30th, 2012 at 2:55 pm4 Comments

Eight pollsters now (Ifop, LH2, Harris, Ipsos, BVA, CSA, Harris, TNS Sofres). Source: Sondages en France. Since January 15th. Last polls: Ipsos and LH2 (fieldwork completed on the 28th):


by Pedro Magalhães

France, 2nd round polls, April 27th update

Posted April 28th, 2012 at 3:41 am4 Comments


by Pedro Magalhães

The last (official) Greek polls

Posted April 26th, 2012 at 5:40 pm4 Comments

Here. The extreme values, rounded:

ND 22-25.5%
PASOK 17-19%
Syriza 9-11%
KKE 9-11%
Independent Greeks 8-10%
Democratic Left 5-9%
Golden Dawn 4-5%
LAOS 3-4%
Ecogreens 3-4%
Democratic Alliance 3-4%
Drazi 1-2%

Lots of undecided too (20% or above). Election day is May 6th.

by Pedro Magalhães

French 2nd round polls

Posted April 26th, 2012 at 4:30 pm4 Comments

A quick look at the French 2nd round polls. Data taken from Sondages en France, 2nd round voting intentions for Hollande, six pollsters (Ifop, Ipsos, Opinion-Way, Harris, BVA, CSA), 63 polls total since January 22nd. I show two Lowess lines, one more sensitive than the other. Vertical lines are the 1st and 2nd round election dates. It takes a lot of zooming to get any inkling of change, but there it is: Hollande declining until early April, no clear evidence of change ever since, certainly too early to detect any post-1st round changes.


by Pedro Magalhães

The polls in France

Posted April 23rd, 2012 at 3:18 pm4 Comments

The last French polls did well, by comparative standards at least. In 2007, no pollster managed to estimate the vote for the first four candidates with an average absolute error below 2 points. In some cases, like in the CSA last poll, things were even worse, with Sarkozy's vote underestimated by almost 6 points and Le Pen's vote overestimated by more than 6 points. Things went better this time: Hollande and Sarkozy did slightly better than expected, and the major problems came with Le Pen and Mélenchon doing, respectively, better and worse that what the polls were suggesting, but still with deviations that were not so large as the ones that occurred in 2007.














A pollster-by-pollster look suggests that, in most cases, whatever may have been behind the over- or under-estimation of candidates' results was common to all pollsters: Le Pen was consistently underestimated, Mélenchon was consistently overestimated and, with few exceptions, the underestimation of both Hollande and Sarko were also common (and relatively small).











Things continue to look bleak for Sarkozy. In 2007, for the 2nd round results, Sarkozy got a majority of the votes in almost every single poll published since the beginning of the year:
























Now, Sarkozy faces the exact opposite situation: he has been behind Hollande in every single poll published since at least February, and the polls conducted yesterday place him behind Hollande at least 6 points.

by Pedro Magalhães

Eurosondagem, 11-17 Abril, N=1036, Tel.

Posted April 21st, 2012 at 1:08 pm4 Comments

PSD: 35.3% (-0.7%)
PS: 30.5% (0.9%)
CDS-PP: 10.7% (-1.3%)
CDU: 9.1% (+0.6%)
BE: 6.4% (-0.5%)

Aqui. Para a jornalista Cristina Figueiredo, do Expresso, mudanças abaixo de um ponto percentual numa sondagem significam "Passos e Governo caem, Seguro e PS sobem." Realmente, gastar massa numa sondagem para depois vir dizer que está tudo exactamente na mesma é chato.

by Pedro Magalhães

Portugal, abreviado

Posted April 21st, 2012 at 5:07 am4 Comments

"Imagine que tem um problema na canalização da sua casa. Chama um técnico que lhe diz que terá de pagar 100€ pelo arranjo. Se quiser factura, serão 123€. Pedia ou não factura?

Pedia: 31%
Não pedia: 64%
Ns/Nr: 5%"

Aqui.

by Pedro Magalhães