Pollster.com

Articles and Analysis

 

Generic Ballot Update

Topics: 2006 , The 2006 Race

Three of the last six national polls have found sharp downturns in the Democratic lead on the congressional generic ballot. After rising steadily since the week before the Foley scandal, the Democratic advantage has now begun to turn down. USAToday/Gallup, ABC/Washington Post and Pew Research Center all find substantial drops. Newsweek, Time and last week's CBS/New York Times polls do not find that decline, but rather show stability at around a 15-point Democratic lead.

SixPanelGenericBallot1105sml.jpg

While these shifts this MAY signal a sharp change of opinion going into the weekend, the magnitude of the drop is quite uncertain with only three polls. We routinely see lots of variation across polls, especially when looking at the generic ballot margin. Nonetheless, the shifts have been enough to convince my "local regression" estimator (the blue line in the figure) to turn down for the first time in a while. Since the blue trend line considers ALL the polls, it is not overly sensitive to single polls, though the combined weight of Gallup, ABC/WP and Pew is enough to move it down about 4 points, from +15 to +11 for the Dems. It is likely that the individual polls are overstating the extent of the downturn. The trend estimator captures the "signal" among all the "noise" from poll to poll. It would take more polls to "know" how much this downturn really represents. But the "poll" taken on Tuesday will answer the question for sure.

However, the current estimate of the Democratic lead based on the trend of all recent surveys remains at roughly +11. While down from the peak of early October, check my post and comparison graphic from earlier this week.  The final Democratic advantage has not been over 10 points (or even close) in the last 12 years. 

 

Comments

very informative site...
My name is Mark Johnson, and I've been visiting your site for past few months
I?m a recent UC Berkeley Political science grad and I along with some fellow Princeton alums have been working hard to launch our own internet startup called Rizzleweb.com.
Rizzleweb is basically an online political community where people can log on and write performance reviews\comments for congressmen, senators, the president, and various other local and state officials across the country. I was hoping that if it would not be too much trouble you could place a link of our site on your blog. If this is not possible (which we completely understand), we still hope you will check out our site, and post some reviews.
Your contribution will encourage us to put more effort in improving our website.
Mark Johnson
www.Rizzleweb.com
markjohnson2020@hotmail.com

____________________

debcoop:

I posted this at MyDD re the geographic distribution in the Pew poll. Maybe you can tell me if my take on the seemingly too large a percentage of southern voters polled has any validilty.

My post

These are Nov. likely voters percentages.

Total 100 40 53 7=100 43 47 10=100 3 (1,795)

Region R D Other change Number
Northeast 20 40 49 11 6 (337)
Midwest 25 40 51 9 -1 (467)
South 34 48 42 10 7 (651)
West 21 44 47 9 2 (340)

Is it typical for the South to have 34% of the total numbers of voters contacted? And you will see that in that region the breakdown is 48%R and 42%D. I don't know if that is standard and what the regional breakdown is in other polls.

But the other regions are only in the low to mid 20's. The population numbers just can't be that the South has more absolute number of voters than the dense Northeast or Midwest.

So while I think this poll does indicate a stiffening of Repub support in this region, the south isn't a region in which we have too many races in contention. Their turnout increase isn't going to impact the 50-60 contested races in the disputed regions of the Northeast, Midwest and the Mountain West.

If you look at those disputed regions you see that the Dem lead in the Northeast is 49D-40R (decrease from 60-34), Midwest 51-40 (same as Oct!) and the WEst 47-44 (though 52-42 in Oct)

Yes a decline but still big enough leads in the districts and state where we need to win.

____________________

Michel:

I was astonished in 2002, disgusted in 2004 but I will not get over it if you don't vote those thugs out of power this time. Come on America, don't fall of my map please.

____________________

Steve:

The generic Congressional poll seems to me to be a total waste of effort and one of those urban myths that need dispelling. Just look at the WashPo generic polls in October 1994--every one had Dem +5 at least. How many Seats did Reps gain? 60?

Since 1981 that poll has never had a Rep lead of 5 or more--yet guess what--sometimes the Reps win Congress!

Why is the generic poll even taken if it seems to have zero predictive value?

____________________

Mark Lindeman:

Steve, no such luck: Charles Franklin and others have clearly demonstrated that it has greater than zero predictive value. That said, it's obviously very noisy. (Pollsters ask the generic poll question because they can -- and they don't want to ignore the House race entirely.)

It would be interesting to see whether differences in question wording or context might account for the apparent difference in trends. For instance, asking the generic question later in a survey might yield more predictively accurate results than asking it early, because more respondents might be thinking about the campaign and recalling the party identification of the actual candidates in their district.

____________________

charlotte:

good point, steve.. i think we need a gentle reminder that all this data and $.75 will still get you a cup of coffee (except it's $4.00 at starbuck's)... the only poll that matters is the one in the ballot box on election day!

____________________

Patrick:

Michel: Does the phrase "fromage mangeant singe de reddition" resonate?

____________________

Ben Ross:

A question slightly off-topic:

What is the effect of automated, interactive push polling on polls, especially IVR polls? It seems to me that an IVR poll (and perhaps a conventional phone poll) is going to oversample people who listened all the way through the push poll -- no matter how much you correct for demographics.

Looking at poll results for Maryland, Tennessee, and Montana, I don't see any systematic difference between IVR and phone polls (SUSA makes Steele look stronger in Maryland, but that has been true since before push polling started). But I still wonder about this -- especially since this comparison wouldn't detect a bias that affected both phone and IVR polls.

____________________

Aaron:

Pelosi cautioned that the number of Democratic House victories could be higher or lower and said her greatest concern is over the integrity of the count -- from the reliability of electronic voting machines to her worries that Republicans will try to manipulate the outcome.

"That is the only variable in this," Pelosi said. "Will we have an honest count?''

In other words, if the Republicans win, it is because they cheated.

____________________

Matonte:

Who knows what this means. Maybe these are statisical outliers or maybe this is a sign that people haven't had enough yet, though if they haven't had enough yet now, when will they?

____________________

Aaron:

Guys, calm down. This is just natural tightening of the race, as Republican voters come home to their party. The older polls showing Democrats +15 to +20 just weren't realistic. Even in the Republican romp of 1994, the GOP won by less than 10 points. We're still well on our way to the biggest blowout since the Watergate midterms.

____________________

Aaron:

A new CNN poll shows the Democrats with a 58-38 lead:

The percentage of likely voters who plan to vote for Democrats in Tuesday's congressional elections increased in the past week, and those voters supporting Democrats also seem less likely to change their minds before casting ballots, according to a CNN poll conducted over the weekend.

Democrats hold a 58 percent to 38 percent advantage over Republicans among likely voters in the survey released Monday morning, compared to the 53 percent to 42 percent advantage reflected in the poll a week ago. Pollsters asked people who wires.send-nat-b wires.send-nat-b identified themselves as being likely to vote which party's candidate they would you vote for in their congressional district if the election were being held today.

____________________

bob:

in a phone poll in which the area code and prefixes are chosen and the last four digits are random, i suspect there may be a more than insignificant effect where the range of Air America or some independent talk radio signal overlaps with the default ubiquitous right wing talk radio.

____________________

Matthew:

Patrick: Your French is terrible. It should be "singe de capitulation qui mange du fromage." If you're going to try to insult our French allies, at least do it correctly, ya goddamn nimrod.

____________________

Matthew:

Patrick: Your French is terrible. I think you're trying to say "singe de capitulation qui mange du fromage." If you're going to try to insult our French allies, at least do it correctly. Machine translation (sort of like robot calls) doesn't lead to very accurate results.

____________________

Dan:

Let me also remind, many of these polls are based on "likely voter" models which become less reliable on a sliding scale the lower the turnout. We really have no idea who is going to turn out to vote. That is the value of aggregating these results.

____________________

Karen:

It seems the only thing Americans are more sick of than campaign commercials is the MSM telling them that Dems will win. The blatant partisanship of the networks and CNN has made many people mad as hell, and they are savoring an opportunity to watch Wolf Blitzer cry that his dems aren't going to win, that Brian Williams finally realizes the Army Times is USAToday and remains unsold in many PX, and that Nancy Pelosi is scarier than Helen Thomas.

____________________

brutalist:

What about the CNN poll?

____________________

Jmarcuscampbell:

I posted this at MyDD re the geographic distribution in the Pew poll. Maybe you can tell me if my take on the seemingly too large a percentage of southern voters polled has any validity.
My post
These are Nov. likely voters percentages.
Total 100 40 53 7=100 43 47 10=100 3 (1,795)
Region R D Other change Number
Northeast 20 40 49 11 6 (337)
Midwest 25 40 51 9 -1 (467)
South 34 48 42 10 7 (651)
West 21 44 47 9 2 (340)

Dear Deb Coop,
I will attempt to give you a concept of what these numbers represent. First and foremost a flawed distribution model to begin with. As you noted the representative sample is not representative of the population being sampled. So why would a company produce such a flawed sample. Perhaps the poll was commissioned. Perhaps the sample came from a larger sample, and this particular subset of the sample produced the desired results. Perhaps, Pew is trying to get some of that political money gravy train, in the last days of the mid-term election cycle.

My back ground is in Operations Research, and have an extensive background in statistics. My favorite book from college, was called "How to lie with statistics" It may be out of print now, but it was a classic. It is always important to know ho commissions of particular set of numbers. If the numbers come from a newspaper, or TV station late in an election cycle, and they show a race getting closer, be very suspect. The TV station and newspapers make their money from selling ads to politicians. They also benefit from ratings, of new shocking poll numbers, film at 11pm.
I applaud you for pointing out the flawed sample subset of this poll. This poll is being used in corporate media, without attribution of the sample subset, only the confidence level or margin of error, referenced. It is very misleading, lacks truthiness as Colbert would say. Great work Deb. I will send this up to Josh at Talkingpoints memo to see if they think this deserves wider distribution.

____________________

Tired:

>It seems the only thing Americans are more sick of than campaign commercials is the MSM telling them that Dems will win. The blatant partisanship of the networks and CNN has made many people mad as hell, and they are savoring an opportunity to watch Wolf Blitzer cry that his dems aren't going to win, that Brian Williams finally realizes the Army Times is USAToday and remains unsold in many PX, and that Nancy Pelosi is scarier than Helen Thomas.>It seems the only thing Americans are more sick of than campaign commercials is the MSM telling them that Dems will win. The blatant partisanship of the networks and CNN has made many people mad as hell, and they are savoring an opportunity to watch Wolf Blitzer cry that his dems aren't going to win, that Brian Williams finally realizes the Army Times is USAToday and remains unsold in many PX, and that Nancy Pelosi is scarier than Helen Thomas.

Gee, are we just a tad out of touch with our fellow Americans these days?

Yeah, the huge mess in this country was created solely by the liberal media. Gimme a F-ing break, lady. I voted a straight Democratic ticket for the first time EVER in this election, and I didn't vote for Clinton, Clinton again, or Gore. Buh-bye, GOP!

____________________

Get out the handiwipes. Someone is going to have a lot of egg on their face Tuesday night. CNN and Pew both can't be right on the generic ballot question.

____________________

Matonte:

Though, it would be interesting to see what would happen if the democrats can't even win the house. Would this mean that the dems have become fully irrevelant at the national level?

____________________

Matonte:

What is interesting how all the polls were seeing about a 45-55 split just a week ago and now they vary, on the last day of the campaign, from 48-52 to 40-60.

____________________

Also, FOX News has just released its Election Eve poll with Democrats leading the Generic Congressional Ballot by 13.

http://www.foxnews.com

____________________

Matonte:

hmmmm....so there is a bipartisan inconsistency in the polls...hmmm...

____________________

Frankie:

This is one the more bizarre distribution and variation of polls in recent memory. Good thing this damn thing is going to be over tommorow and we'll see who was right and wrong.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR