Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
As I suspected all along, here is another exit poll (TNS-SOFRES) which contradicts IPSOS about the 25-34 age group.

the exit poll numbers (age group, Royal's %, Sarko's %):

  • 18-24 60 40
  • 25-34 54 46
  • 35-49 51 49
  • 50-64 41 59
  • 65+ 36 64

tns-sofres exit poll, see page 2

There is now even less doubt that Sarko is the president of the retirees who were most sensitive to his security/anti-immigration propaganda. The IPSOS exit poll numbers ascribing the 24-35 age group to Sarko at 57% didn't make much sense, and the difference with this poll is so large (an 11 point difference) that one can only wonder what happened ...

by Fete des fous on Mon May 28th, 2007 at 02:35:49 PM EST
[ Parent ]
These exit polls are all over the place, so it's hard to know what really happened. Two things are cristal clear in all:

  • the young voted strongly for Royal
  • the old voted even more strongly for Sarkozy

In between, it's a lot hard to say.

In the long run, we're all dead. John Maynard Keynes
by Jerome a Paris (etg@eurotrib.com) on Mon May 28th, 2007 at 04:38:46 PM EST
[ Parent ]
Which exit polls are you talking about? I looked for other polls a couple of weeks ago and couldn't find any apart from the IPSOS one. A 10 point difference among the 25-34 y.o. between 2 polls is huge and cannot be due to random error. Either the methodologies are very wrong or worse.

If I am right that it means the elections were decided primarily by the security/immigration issues, it doesn't say much for the effectiveness of the declinist propaganda.

I note the unwillingness of the media to even mention the age divide in the vote. It seems the left should discuss this point as widely as possible before the legislatives.

by Fete des fous on Mon May 28th, 2007 at 05:20:48 PM EST
[ Parent ]
The TNS/Sofres poll linked to just above (thanks for the link, btw) may not be an exit poll; according to the fiche technique its a telephone poll conducted on election day.

One could make that case that on a nice spring day, younger voters are less likely to be at home answering the phone and presumably, an actual exit poll has a much lower margin for error since it gets a better (and larger) sample of actual voters, but simple common sense suggests an error in the IPSOS age cross-tabs.

by desmoulins (gsb6@lycos.com) on Tue May 29th, 2007 at 02:32:28 AM EST
[ Parent ]
Both the TNS-SOFRES and the IPSOS-DELL were conducted over the phone on the evening of the election. They also share the quotas method, the only difference in the method appears to be sample size (1200 versus 3600 people interviewed).
by Fete des fous on Tue May 29th, 2007 at 07:42:30 AM EST
[ Parent ]


Occasional Series