Bell Curve The Law Talking Guy Raised by Republicans U.S. West
Well, he's kind of had it in for me ever since I accidentally ran over his dog. Actually, replace "accidentally" with "repeatedly," and replace "dog" with "son."

Monday, June 21, 2004

Thoughts About Polls

Hi Everyone,

Everyone is talking about the electoral horse race (see polling and other polling links on this blog). The national poll horse race is pretty tight. Some polls have Kerry ahead, some have Bush ahead or have them tied. But we should all know by now that the election is not a national popular vote contest. It is a series state by state elections. State by state results are available but you usually have to pay for them (or check out the LA Times interactive electoral vote map/calculator which gives the most recent poll results when you move your mouse over each state). But you can still guess at some things from these nationwide polls. Here are some things I've been noticing and speculating wildly about.

Bush seems to do better in polls that use "adults" or "registered voters" as opposed to "likely voters." Why is this interesting? Republicans usually do better among "likely voters." Republicans do better among those demographic groups that vote the most often (wealthy old white men). Bush support may have a critical weakness: suburbs. Suburbanites are likely voters (and how!) and they tend to be socially progressive, fiscally conservative. States like Ohio are full of them. States like Alabama have very few of them.

Second, Zogby reports a majority (53%) of respondents say it is "time for someone new" in the White House. This number has been over 50% for months. Zogby used this question to accurately predict the Gore-Bush 2000 popular vote (he's the only pollster who successfully predicted that Gore would win the popular vote).

Finally, polls reported in polling (see link) show that voters prefer a Congress controlled by the Democrats by about 50% to 40% (and some undecideds) depending on the poll. Now, this is nationwide poll and nationwide polls are even less useful for Congressional races than they are for the Presidential race. However, these results have been stable for months and may indicate a big turnout for the Democrats' base. The big caveat on this one is that Republicans have been able to redraw the Congressional districts in a lot of states to maximize their chances of winning the most seats. Also, most voters dislike Congress but like their local representative (regardless of party).


The Law Talking Guy said...

On a recent 1200 mile road trip through the American west, I observed not a SINGLE political bumper sticker anywhere. The trip was largely through Bush country. This is an indication that (1) the Republican base is not yet active, or (2) significant parts of the public are not paying attention to the election yet. If the latter, current polls may not line up well with November's results. However, polls also indicate that 8 of 10 Americans are paying attention, a high number by historical standards, and (historically) the June polls are indicative of the November result.

Raised By Republicans said...

I think the 8 out of 10 number is right. I think people are paying close attention to the election. I also think that about 80% of the population has made up their minds already and will not change their minds under any circumstances. However, I don't think that the 80% who are paying attention and the 80% who have made up their minds are the same group.

There are a number of states with very high rates of "undecided" voters. It is common to assume that "undecided" is really code for "uniformed." I disagree. I suspect that many of the "undecided" voters have already decided that they don't like Bush but want to make sure that Kerry doesn't appoint Hilary Clinton (or somone else percieved as being from the left wing of the Democratic party) as his running mate before they fully support him.

Its not that people aren't paying attention. Its that there isn't much new happening. Until Kerry picks his running mate, people will wait and see.

RE: Absence of Bush-Cheney Stickers in Bush Country. I think this indicates low enthusiasm among large segments of the Republican base, especially secular conservatives and libertarians.