Does this only apply in generic congressional districts?

Posted at 1:56 PM on May 18, 2006 by Bob Collins (6 Comments)

Rasmussen has a poll that just feels like it's full of holes. But I pass it along anyway.

Following the President's Monday night speech, Democrats enjoy a 15-point advantage on the Generic Congressional Ballot. A Rasmussen Reports national opinion survey of 1,000 Likely Voters found that 48% would vote for the Democrat in their district if the election were held today. Just 33% would vote for the Republican.

Why do I think it's full of holes? Well,here's today's congressional vote forecast map from Congressional Quarterly. Red is leaning, strongly leaning or safe Republican. Does that look like the Republicans are in trouble to you?

Comments (6)

welcome to the wonderful world of gerrymandered districts.

Posted by bill | May 18, 2006 2:44 PM

Well, there are a couple of things to consider here. The first is that, in general, when people are polled about their opinion of Congress, their answer tells you their opinion of Congress as a whole, not their opinion of their rep. So although those polled favor a Democratic Congress by a 48-33 margin, that doesn't necessarily translate to how they plan to vote in November.

Which suggests that your skepticism is warranted.

On the other hand, the outcome in November depends in large measure on whether the election is nationalized. In other words, will people be motivated to vote principally on issues specific to their state and district, or will the election be a referendum on Bush and the Republican-controlled Congress? If the answer is the former, then the CQ map you provide may largely correct. If the latter, then the Rassmussen polling may be more meaningful, and we may be seeing a much different map:
http://mydd.com/story/2006/5/15/202917/008

Posted by David Bailey | May 18, 2006 2:49 PM

Could be a case of "throw the bums out." People tend to like their own bums. Kinda like pork. Politicians hate pork -- other lawmakers' pork, of course.

Posted by Bob Collins | May 18, 2006 2:54 PM

It doesn't take a whole lot of competitive seats to produce a dramatic change in party control. Remember that in 1994, more than 90% of House incumbents who ran won reelection.

Posted by David Kirchner | May 18, 2006 9:58 PM

I read somewhere that all the dollars spent polling were totally wasted on congressional seats in 2004, because only one question better predicted who would win than incumbency.

The question?

"Which candidate do you find more attractive?"

Kind of makes you scratch your head, doesn't it?

Posted by Justin | May 20, 2006 10:02 AM

Well, and this map is a spatial map, not a population map. The big swathes of red are large empty spaces, while those blue edges are densely populated coasts. And that deep blue along the Mason-Dixon line is very telling, too. No, that map actually DOES look like trouble for the Republicans.

Posted by Albatross | May 22, 2006 12:43 PM

Services

• Hourly Newscast
• HD Radio
• iPhone App / Mobile
• E-mail newsletters
• RSS feeds
• Podcasts
• Submit your photos
• Videos
• Weather
• Most e-mailed stories