Richard Neumann: Three Questions about the 2006 Elections
Were the polls accurate?
Mostly no. The key question polls ask before election day is the “generic ballot” about the House of Representatives: If the election were held today, would you vote for the Democrat or for the Republican in your district? Here are the last polls taken before the election:
|Dates||Rep%||Dem%||Unsure %||Dem |
|CBS/N.Y. Times||Oct 27-31||34||52||14||18|
On election day, it appears that about 53% of the electorate voted for Democratic candidates for the House, and about 45% voted for Republican candidates. The remaining 2% voted for third-party candidates.
Of the nine polls, six wildly overestimated the Democratic lead, predicting it to be 13 to 20 percentage points, rather than the 8 percentage points it turned out to be (53% to 45%).
One poll (Pew) underestimated by half, predicting the margin would be 4 percentage points rather than 8.
Only two of the nine polls were essentially accurate: USA Today/Gallup and ABC/Washington Post.
(For two reasons, the 53% and 45% figures are not yet precise. First, the counting is still going on, with absentee ballots, etc. Second, for 34 seats only one party put up a candidate, and the media did not report the votes that one candidate got, although in nearly all states those votes eventually will be officially recorded. To get a preliminary national number now requires estimating the votes in those 34 uncontested districts, based on what unopposed candidates have gotten in the past, and then adding those estimates to the preliminary vote numbers for the other 401 districts reported in the days after the election by CNN and the New York Times. Because no media outlet totals the numbers for those 401 districts, each district’s preliminary votes have to be fed into a spreadsheet. The final precise numbers will be available when each state announces its official vote count, which will be weeks from now in some instances. Next spring, the Clerk of the House will publish a report compiling those final, official numbers.)
2. Did the number of seats each party won in Congress accurately reflect that party’s percentages of the popular vote?
No. The Democrats were robbed in both the Senate and the House.
In the Senate, they have a one-vote majority but got nine million more votes in the elections that produced the 100 members of the new Senate (94 million Democrat to 85 million Republican votes, 2002-2006, one-third of the Senate being elected every two years). How can the Democrats get so many popular votes but have only a one-vote majority? Constitutionally, the Senate is exempt from Baker v. Carr (requiring legislative districts to be equal in population). Each state gets two senators, and a voter in tiny Wyoming (with two Republican senators) thus has over 20 times the power of a voter in New York (with two Democratic senators).
In the House, this week’s 53%-45% popular vote split would historically have produced 250 to 270 seats for the Democrats. They’ll end up only with 231 to 234, depending on how the recounts go in the close districts. That’s because Republican gerrymandering limited Democratic seats in states like Texas, Georgia, and North Carolina.
The Republicans were similarly robbed in 1994 (when they took over Congress) because of Democratic gerrymandering. Since 1994, with the aid of computers, Republicans have turned gerrymandering into such a precise art that in some states neighbors can end up in different districts because they are registered with different political parties.
3. Did the Democrats win a decisive victory?
Yes and no.
Yes: Obviously, the Democrats persuaded the public to give them control of Congress after 12 years of Republican control. That upends what’s been happening in Washington, and the next two years will be very different from recent history. (But you already knew all that.)
No: In the new Senate, the Democrats will have a bare majority of 51 out of 100 senators. From 1959 to 1981 and from 1987 to 1995, when the Democrats controlled the Senate, their average during those periods was 60 seats. That is a good benchmark for a decisive majority, especially because it takes 60 votes in the Senate to end a filibuster. In the new House, the Democrats won about 53% of the popular vote and will have 231 to 234 seats. During the 40 years from the 1954 election until the 1994 election, the Democrats never had fewer than 232 seats. In three of those Congresses, the Democrats had as many as 291, 292, and 295 House seats and got 56% to 58% of the popular House vote. In American politics, winning 58% of the national popular vote is a landslide. No president has ever won more than 61% of the popular vote. Fifty-three percent is not a landslide. It is just a victory.
comments powered by Disqus
Lawrence Brooks Hughes - 11/16/2006
Everyone should favor a verifiable paper trail, everywhere. Everyone should favor a sound voter I.D. system everywhere, too, but the Democrats steadfastly resist the latter to protect their hundreds of thousands (or millions) of votes in net advantage from illegal aliens. When the subject shifts to vote fraud, it would help to start by stipulating it is widespread and always helps the Democrats, not only because of aliens, but also because of all the big city machines and Indian reservations... It is very unlikely, adjusted for fraud, whether Al Gore actually did win the popular vote in the 2000 election.
John Edward Philips - 11/15/2006
that the polls were off because electronic voting machines were providing the Republicans with a small percentage of extra votes? How can we know unless there is a verifiable paper trail in every precinct?
Carl Becker - 11/14/2006
BOTH parties estimating the extent of their power is nothing new.
One of the reason the Republicans lost is because they did "not know how typical Americans live, work, think and are motivated."
Lawrence Brooks Hughes - 11/14/2006
As for the liberal media polls overestimating the Democrats' power, this is nothing new. They do it in every election. I used to think it was an attempt at self-fulfilling prophesy on their part, but this year they actually showed polls closing toward the end in favor of Republicans. That would never be done by the wilfully unethical. Now I think the answer is chronic false assumptions by the type of people who work in the liberal media--cloistered Ivy League idiots for the most part, who do not know how typical Americans live, work, think and are motivated.
Lawrence Brooks Hughes - 11/14/2006
I agree with you in general, but you should have mentioned much of the gerrymandering is by mutual consent of the two major parties. Both always seek to protect incumbents. You might also have mentioned the black House districts, too, where a liberal Supreme Court has helped the GOP. There is no better formula for making a GOP district than to pack all the black voters on top of each other in support of a John Conyers or an Alcee Hastings.
- NYT History Book Reviews: Who Got Noticed this Week?
- Researchers have discovered a previously unknown 149-page manuscript defending homosexuality.
- What Counts as Historical Evidence? The Fracas over John Stauffer’s Black Confederates
- Israeli journalist-turned-biographer, Shabtai Teveth, is remembered for his attack on the New Historians
- Harvard’s Drew Faust says the Civil War marked the start of large-scale industrial war, not WW I