A fe w more polls have been released, giving a clearer, if relatively consistent picture of the strengths of the coalitions competing in the Italian election. The main sources of movement are not movements in public opinion, but rather movements in coalition formation. It seems more likely now that the left will be split neatly between the Partito Democratico and its associates on one hand, and the Sinistra Arcobaleno on the other. On the right, Berlusconi has been trying to provoke Casini into declaring his membership in the centre-right coalition.
Because the coalition arrangements are fluid, Italian pollsters have been trying to gauge voters’ likely reactions to each of them. I’m worried about the reliability of the results thereby obtained. A poll for Repubblica asked voters to say which party they’d vote for under six different scenarios; another, by Coesis, asked voters to choose between five scenarios, only one of which is remotely probable. What’s worse, it’s not clear from the questions what they meant by centre-left and centre-right. A quick Google Scholar search on “hypothetical polling” doesn’t yield much that isn’t related to push-polling, but I’m sure there must be some qualms about reliability out there. i
Post a Comment