ct senate race polls via pollster

Pollster does a nice job of compiling and averaging polling data about the upcoming election. It will be interesting to see the final numbers, and how badly the polls were off. But mainly I wanted to embed this poll chart thingie from pollster.com (taking over the left side of the first page of this blog entry ... for a while, anyway) because I think it’s cool. A widget for a politics junkie. Like me.

My problem with polls, though, is I know too much about how they can be influenced by the way questions are phrased, but the tone of the poll taker, by the context, and by other things. I took a couple of excellent courses about polling while in grad school at Columbia. One of the things I remember is that in order to get a more accurate response, you need to ask the same question at least twice, but phrased differently each time. And based on what I’ve seen of the questions asked by several of the poll-takers (and I was polled twice via phone during the past two months), this is rarely, if ever, done. Too expensive.

Based on the primary, I’m guessing the difference will be a helluva lot smaller 11% between Liebermouth and Ned Lamont. I suspect it’s 50-50 right now. For one thing, Schlesinger will probably get at least 2% extra than he’s polling by virtue of being first on the ballot. For another, I’ve heard too many stories about how people, especially indies, bullshit the poll-takers just to keep things interesting or because they’re annoyed that anyone is trying to invade the privacy of the voting booth.

I have the feeling that it’s going to be Ned by 2%. I hope I’m wildly off the mark and Ned wins by a landslide. But a win is a win—I just hope it’s definitive enough so there’s no legal crap pulled by a pissed-off Joe.

Posted by .(JavaScript must be enabled to view this email address) on 11/05/06 at 08:54 AM
Commenting is not available in this weblog entry.

<< Back to main