Parsing the Polls: How Low Can Congress Go?
The fallout from Rep. Mark Foley's resignation and the House Republican leadership's subsequent struggles to explain what they knew and when they knew it remains difficult to fully gauge at this point. But it's a safe bet that Congress's job-approval numbers will drop as Election Day draws closer.
And the lower that congressional approval ratings sink, the better chance Democrats have of retaking the House majority in the fall. Although Republicans insist that voters have adopted a "pox on both your houses" mentality in this election, it is the GOP that controls the House, Senate and White House -- meaning that the blame is likely to fall disproportionately at the party's feet.
Let's parse the polls!
Over the past month, eight national polls have been conducted that asked voters for their opinions about Congress. The average approval rating for the institution during that time was 28.75 percent, with the average disapproval score sitting at 68.75 percent. The lowest approval rating in any poll -- 20 percent -- came in the NBC/Wall Street Journal survey in the field Sept. 8-11. The highest -- 40 percent -- was in an ABC poll done Sept. 5-7.
In five of the eight surveys, congressional disapproval was 60 percent or higher -- topping out at 68 percent in a Sept. 9-11 AP-Ipsos survey. The lowest disapproval score came from a Fox/Opinion Dynamics poll, showing 53 percent of the sample unhappy with Congress's performance.
These most recent numbers are a continuation of the struggles Congress has endured for much of the past year. In the 93 polls listed on the indispensable Polling Report site, not a single one showed Congress with a net positive approval rating. In fact, the bets approval showing for Congress over the past year is 43 percent -- reached three times (twice in Dec. '05, once in Jan. '06). Since June 1, Congress's approval rating has been measured above 31 percent just twice -- the aforementioned ABC poll and an early August ABC/Washington Post survey where 36 percent of the sample approved of Congress's performance.
How do these miserable ratings stack up with historical patterns, and how predictive are congressional job approval/disapproval numbers when it comes to election results?
Thankfully, we can make some historical comparisons because a few major polling organizations regularly release data from cycles past.
In 2002 -- the first midterm election of George W. Bush's presidency -- a mid-October NBC/Wall Street Journal poll showed 44 percent of voters approving of the job Congress was doing while 40 percent disapproved. A Gallup survey in the field in early October 2002 had 50 percent approving and 40 percent disapproving. A CBS News/New York Times poll of likely voters in early November had a less optimistic view -- 42 percent approval and 45 percent disapproval. Republicans picked up a net of two Senate seats and eight House seats that election.
Go back to 1998, the second midterm election of the Clinton presidency: The final NBC/WSJ survey -- conducted in late October -- showed 48 percent approving and 39 percent disapproving of Congress. Gallup had Congress's approval at 44 percent and its disapproval at 47 at that time, while CBS/NYT had it 41 percent approve/48 percent disapprove. Democrats gained five House seats and broke even in the Senate.
Four years ago voters were generally happy with the performance of Congress, as the good will following the Sept. 11, 2001 terrorist attacks had yet to wear off. President Bush and congressional Republicans smartly turned the elections into a referendum on national security -- a move that allowed them to buck historical trends and pick up seats in a midterm election.
In 1998 Republicans saw a chance to grow their majorities in the House and Senate by turning the election into a referendum on President Clinton and the impeachment proceedings against him. But voters never seemed to buy what House and Senate Republicans were selling; while congressional approval numbers dipped in the run-up to the election, Clinton's job approval numbers remained strong -- helping fuel the Democratic gains.
The problem with comparing 2006 to 2002 or 1998 is that the numbers for Congress are considerably more lopsided this cycle than they have been in the recent past. There's a major difference between a 44 approve/47 disapprove for Congress and a 28 approve/65 disapprove. We simply don't have a good way of quantifying what such a huge disparity between those approving and disapproving of Congress means in terms of voting results this fall.
What we do know is that -- atmospherically -- this election is setting up to be a very difficult one for Republicans. Traditional measures of the electorate's temperature (presidential approve/disapprove, congressional approve/disapprove, generic ballot etc.) show that Americans are extremely unhappy with the party in power and ready for a change. Republicans insist that these macro-measures matter little in individual campaigns that are influenced more by local concerns than national issues.
We'll know whether they were right in 34 days. Looking for a way to gauge how strong the anti-Republican sentiment nationwide is? Start by monitoring House races like the ones in Connecticut's 2nd District, Florida's 22nd District, New Mexico's 1st District and Ohio's 1st District.
Each race features a competent Republican incumbent against a credible Democratic challenger in a district that should be competitive for both sides. If all (or the majority) of those incumbent lawmakers lose, you can safely claim that Democrats succeeded in nationalizing this election. If the incumbents win, Republicans will likely hold the House thanks to their candidates' focus on local issues.
October 4, 2006; 6:00 AM ET
Categories: House , Parsing the Polls
Save & Share: Previous: Allen Attempts to Start Over
Next: 2008: Halperin and Harris Describe the Winning Formula
The comments to this entry are closed.