October 8, 2006 - October 14, 2006
We have devoted much attention recently to the flood of new national surveys showing small declines in the Bush job approval rating and modest Democratic gains on the generic House ballot question since mid-September. Until today, I had not looked closely at levels of party identification reported on those surveys. It turns out those have also trended Democratic recently, a finding that may explain some of the apparent "house effect" differences among statewide pollsters over the last few days.
The debate over weighting surveys by party identification has been a focus of this blog since its inception. My posts on the subject from 2004 and beyond are worth reviewing but the gist is this: Pollsters typically ask respondents some variant of a question asking whether they consider themselves "a Democrat, a Republican or an independent?" The so called "Party ID" question has been asked, examined and studied for more than 50 years, and an ongoing debate exists about whether to weight (or statistically adjust) survey results by party.
The crux of the debate is whether party identification is more like a fixed demographic characteristic (such as gender or race) or more like an attitude that can change with the prevailing political winds. For most adults, party identification does appear to be highly stable, changing rarely if ever. The problem is that some small portion of voters (perhaps 10% or 15%) appear willing to jump back and forth -- usually between one of the parties and the independent category -- depending on the wording of the question, its position in the survey, how hard the interviewer pushes for an answer, or, in some cases, what has been happening in the news.
Those who argue for weighting by party say that the real trends tend to be slow and gradual and that party weights can adjust dynamically over time to accommodate these slow moving trends (see also the party weighting page maintained by Prof. Alan Reifman). Those who argue against party weighting (a class that includes most of the national media pollsters) worry that such an approach will suppress real but short-term changes that sometimes occur in reaction to news events (such as the period just after the 9/11 attacks or the period just after the 2004 Republican convention).
A look at the party identification data from the recent surveys suggests we may be in the midst of another such short term change. The table that follows shows party identification results for six national surveys conducted before and after the resignation of Congressman Mark Foley. Five of six show some Democratic gain in party identification:
This change may also explain the wide divergence in results reported by the two automated pollsters in two nearly simultaneous surveys conducted this week in Missouri and Ohio. In both states, SurveyUSA showed the Democratic candidates with significantly greater leads (+14 in Ohio and +8 in Missouri) than Rasmussen (+6 in Ohio and -1 in Missouri). While both pollsters use the automated "interactive voice response" (IVR) methodology, Rasmussen weights by party and SurveyUSA does not. Moreover, the most recent SurveyUSA samples have grown more Democratic since August.
Does this shift in party identification represent a real shift in attitudes among the population of adults or registered voters or does it reflect some short enthusiasm among Democrats to be interviewed? Is the change a momentary spike or will it persist until Election Day? These are the questions that professional pollsters are mulling over right now, and the answers are not obvious. We will just have to wait and see (no pun intended).
Of three new polls, two place approval of President Bush at or above 40% while one comes in at 34%. Last week's Cook Political Report/RT Strategies poll, taken 10/5-8/06 finds approval at 41%, disapproval at 54%. The new Fox poll from 10/10-11/06 has approval at 40%, disapproval at 56%. And a Harris Interactive poll done 10/6-9/06 puts approval at 34% and disapproval at 64%. (For a plot with just the trend line estimate and the data points, click here.)
With the addition of these new polls, my trend estimate has continued to decline, now standing at 37.2%. Interestingly this trend estimate is virtually identical whether I use my usual "conservative" blue line estimator or the more sensitive (but easily fooled) estimator. While two of these polls are still in the 40s, the Fox and Harris estimates represent declines from their previous result. Cook is a small rise from their previous reading. The current spread of results is consistent with the usual variability around the trend line of approximately +/-4.8%.
An easier to see comparison of the individual polls with the overall trend is shown in the figure below. There are some clear house effects between the polls, but each tracks the trend reasonably well.
Note: This entry is cross-posted at Political Arithmetik.
Our Slate Election Scorecard update tonight focuses on two new polls in New Jersey that confirm recent gains by Bob Menendez and move the race to lean Democrat status. The overall scorecard tally now indicates 49 seats leaning or held by Democrats, 49 seats held or leaning Republican. Is this change indicative of a larger Democratic surge?
Two new polls out this evening from Survey USA in Ohio and Missouri both show the Democratic candidates in each state leading by much wider margins than on other recent polls. These results and the sometimes improbably wide Democratic margins on the generic House ballot in some recent national surveys leave some wondering whether, as reader Gary Kilbride put it in a comment a few hours ago, "the current poll numbers skew misleadingly toward Democrats due to the Foley scandal." He wonders if the same might be happening to the Majority Watch congressional district results released today.
I will have more to say about all of this tomorrow, but for tonight one quick note about those new Majority Watch congressional surveys. Although they released results from 32 districts today, only nine involved follow-up surveys in districts polled previously using comparable ballot tests. The table below shows the August and October results for those nine districts.
All of these Districts are currently represented by Republicans and all were rated as toss-ups by the Cook Political Report when the polls went into the field (they moved CO-07 to lean Democratic status just yesterday). While Tom Riehle's analysis made much of the apparent Republican improvement in Washington-08, Virginia-02 and Indiana-02, the overall pattern looks more random. Those Republican advances were largely offset by Democratic gains in North Carolina-11 and New Mexico-01. Overall, the average Democratic margin declined by just a single percentage point.
The bigger story may be that the average Republican percentage across these nine districts has not budged from 46% since August or that none of the Republicans in the nine districts holds a statistically significant lead. More on the meaning of these House polls tomorrow.
This "Guest Pollster Corner" contribution comes from Thomas Riehle, a
Partner of RT Strategies
Editor's note: In a 2:30 p.m. press conference, Riehle announced that when
the sum up results of the 63 surveys they have conducted since August and
consider races where a candidate hold a lead beyond the margin of error,
Democrats currently lead or safely hold 217 seats and while Republicans lead
in or hold 198 seats. Democrats will need to win 218 seats to gain majority
control. Full data are now available at www.majoritywatch.com, including a
summary of all top-line results to date. The Pollster.com House Race page
is updated to include all of the new Majority Watch data.
Majority Watch, a project of RT Strategies and Constituent Dynamics, sponsored by Waggener Edstrom Worldwide, is the most comprehensive project ever undertaken to identify and conduct polls in most of the highly contested House races across the country. In August and September, Majority Watch polled in 30 House districts. On October 1, we polled Mark Foley's Florida 16th district with two simultaneous polls, one in which respondents were informed that a vote for Foley would count for the Republican candidate to be named later, and one in which respondents did not get that information.
Today, Majority Watch begins to release results from Round II. In 32 races polled in the current cycle:
- Republican incumbents who seemed to be in trouble in late August have held on or even improved their positions. In Washington's 8th C.D., Republican Rep. Dave Reichert has moved from 3 points behind to 3 points ahead of Democrat Darcy Burner, 48%-45%. In Virginia's 2nd C.D., Republican Rep. Thelma Drake has moved from 8 points behind to a marginal 2-point lead over Democrat Phil Kellam, 48%-46%. In Indiana's 2nd C.D., Republican Rep. Chris Chocola has moved from 12 points behind to only 4 points behind Democrat Joe Donnelly, 46% for Chocola to 50% for Donnelly. In Colorado's 7th C.D., Majority Watch polling shows Republican Rick O'Donnell is tied with Democrat Ed Perlmutter in the race to fill the open seat, essentially unchanged since August. All of these were among the first races Democrats targeted, and that early warning may have given Republicans the head's up they needed to remain competitive and avoid getting swept away.
- Most Republican leaders have survived the worst of Foley's Folly, but in localized areas where there was a local media hook for the story (Florida, New York, possibly Arizona), damage may have been severe for many Republicans, at least at this time -- there's still time to recover.
- On the positive side for Republicans, neither Speaker Denny Hastert (ahead by 10 points, 52%-42%) nor House Page Board chairman Rep. John Shimkus (ahead by 17 points, 53%-36%) seem to have suffered. The highest profile Republican House incumbent closest to Washington, D.C., Rep. Frank Wolf in Virginia's 10th C.D., remains ahead of Democrat Judy Feder, 47%-42%.
- On the other hand, in Ohio, where the Republicans were already beset by the culture of corruption charge, Republican Conference Chairperson Deborah Pryce is behind by double digits, in Ohio's 18th District Republican Joy Padgett trails Democrat Zack Space by 9 points, and even in Ohio's 2nd C.D., Rep. Jean Schmidt is marginally behind Democrat Democrat Victoria Wulsin by 3 points, 45%-48%.
- In New York, NRCC Chairperson Tom Reynolds has stumbled badly, trailing Democrat Jack Davis by 16 points, 56% for Davis to 40% for Reynolds. In the open seat in New York's 24th C.D., Democrat Michael Arcuri has opened a significant lead, 53%-42% over Republican Raymond Meier. Even Republican Rep. Peter King, never shy about pointing out when the leadership is wrong and vocal in his anger at how House leaders have handled the Foley case, seems to have suffered -- his is only marginally ahead, 48%-46% over Democrat Dave Mejias.
- In a surprise, Arizona Republican Rep. Rick Renzi is marginally behind Democrat Ellen Simon, 50%-46%.
- The Philadelphia suburbs remain troublesome for Republicans, with Republican Rep.s Jim Gerlach and Curt Weldon trailing their Democratic challengers.
Majority Watch takes advantage of new technologies, married to the oldest standards of sampling and vote modeling, to extend the practice of public opinion polling down to the level of House races. Calls are made by IVR recordings ("robo-calling"). The sample is drawn from voter lists of active voters, with Majority Watch controlling in-home selection in those households where more than one voter resides. The calls are kept extremely short in order to keep response rates as high as those for many publicly-released telephone interviewer polls (about 20% response rate using the standard AAPOR definition). And consumers are increasingly comfortable pushing buttons to respond to recorded voices -- can any reader say he or she is unfamiliar with the notion of "press 1" for one thing or "press 9" for another? These "robo-calls" perform not much differently than traditional telephone interviewer calls for very short, "horse-race" polls.
Majority Watch is currently polling in ten more House districts for release next week (GA-08, IL-08, IL-10, NH-01, NH-02, NY-19, NY-20, NY-25, NY-29, and OH-01), at which time we will have solid polls, with about 1,000 voters, in each of 55 House races. Depending on developing political circumstances, we may further expand the list and conduct more polls after next week.
Our Slate Election Scorecard update yesterday focused on a new poll in Connecticut conducted by for the Hartford Courant by the University of Connecticut Hartford showing Joe Lieberman leading by an eight point margin (48% to 40%) over Democratic nominee Ned Lamont. Since that update went online, we learned of another new Connecticut poll from SurveyUSA showing Lieberman ahead by a larger margin (53% to 40%). The update makes reference to the results as tabulated by party, which are remarkably consistent across the various polls and tell the story of that race as it stands today.
I was able to gather results by party from four recent polls by Quinnipiac, Rasmussen, SurveyUSA and the Courant/U.Conn. poll (the latter appeared in The Hotline [$], but only for the tabulation without "leaners").
The results by party are remarkably consistent. Lieberman has become the overwhelming favorite of Republicans, receiving an average of 70% of their vote. Even among Republicans, their nominee Alan Schlesinger barely registers at an average of 11%. Remarkably, Democrat Ned Lamont runs ahead of Schlesinger among Republicans on all four surveys (a consistency that suggests statistical significance despite the small sample sizes involved).
With Lieberman winning so many Republican votes, Lamont's inability to get more than 60% of Democrats is what keeps this race from becoming more competitive. The challenge for Lamont in increasing his Democratic support is that many of Lieberman's Democratic supporters voted for him in the August primary. SurveyUSA reports that 80% of Lieberman's primary voters are sticking with him.
There is something about this race that almost defies gravity. With a tight national race for control of the U.S. Senate, two-thirds of Republicans are willing to support a candidate who describes himself as a Democrat who will "caucus with the Democrats," even if Democrats and Republicans split the Senate. And a third of Democrats are apparently taking Lieberman at his word despite his reliance on Republican money to fund his campaign.
Today's flood of new national surveys provides enough raw material for a week's worth of blog posts. The new surveys are from ABC News/Washington Post, CBS/NY Times, CNN/ORC and USAToday/Gallup, plus one more from Newsweek released over the weekend. I want to highlight a few key results, particularly what the new surveys tell us about shifts in the so called "generic congressional" ballot.
As Charles Franklin notes in the previous post, these surveys do indicate an improvement in the Democratic margins. I want to take a closer look at an issue that inevitably confronts us when considering the generic ballot question, whether to watch results among all registered voters or just the sub-samples of "likely voters" as defined by each pollster. The surveys from CNN, Gallup and Newsweek provide both, so the following table provides both.
While some of the surveys (Gallup and CNN) show more change than others, all but CBS/NYTimes poll showed at least some improvement in the Democratic margin since September. When we average the results of the registered voter samples, the Democratic margin increases from 11 to 14 percentage points.
When we shift to likely voters, things get a little murky. Only two of the newest polls -- the ones from Gallup and CNN -- reported results for likely voters in both September and October. And both show bigger swings toward the Democrats, but among both their likely and registered voter subgroups. The change for Gallup among likely voters is simply enormous (from a dead heat to a 23 point Democratic advantage). CNN also shows an eight point gain in the Democratic margin among likely voters (from 13 to 21 points).
So which population should we follow? Frank Newport made the case for the Gallup likely voter model in a post here last week, and a lively debate ensued that will no doubt continue over these new results. Consider the following table that shows how likely and registered voter results have compared since Labor Day on the pollst that reported both.
There is no apparent consistency in the differences between the registered and likely voter samples. On average they seem to make the margin about a point less Republican, but even that disappears when we remove the mid-September USAToday/Gallup poll from the analysis. Consistent with past criticism, the likely voter model appears to be producing more volatile results, particularly for Gallup. But for all the sound and fury of the debate, likely voters and registed voters are looking more or less the same.
The post-Foley Folly polls find an upturn in the Democratic margin in the generic Congressional ballot. Prior to the Foley developments, Democrats held a 10.6 point lead in the polls. (This is the Dem percent minus the Rep percent.) That lead has now jumped to 12.8 points, the highest my trend estimate has reached in the 244 generic ballot polls taken this election cycle. This is all the more important because prior to the Foley Fiasco the trend had moved a bit down, then flattened (though still at or about 10.6, a very strong margin even then.) Whatever possible gains Republicans were beginning to make have now been wiped out.
One important concern is that CNN and Gallup produce extraordinarily high values on the generic ballot in this poll. (Remember, these are now done independently-- CNN uses Opinion Research Corporation for it's polling, while Gallup now polls for USAToday only since the dissolution of the old CNN/USAToday/Gallup partnership.) These values of over 20 points are simply implausible given the rest of the data over the past two years, and the other data from the past two weeks. Could it be that my trend estimate is being unduly influenced by these two absurd results?
No. If I exclude the latest CNN/ORC and Gallup/USAToday polls, the trend estimate is a Democratic advantage of +12.4 points, rather than the 12.8 points if these two are included. That small change in trend would still be the highest Democratic margin of the past two years, and when plotted the line without these polls is visually all but indistinguishable from the blue trend line in the figure.
And so we are back to the key question: how much will this huge (compared to elections since 1994) Democratic lead translate into seats? Will we see movement in polls for individual House races move as well? Stay tuned.
Note: This entry is cross-posted at Political Arithmetik. Typo corrected.
Four new polls find approval of President Bush has declined substantially since the end of September, following revelations of "overly friendly" email and IM messages from Rep. Mark Foley (R-FL) to pre-adult House pages. All these polls were completed before North Korea tested what appears to have been a nuclear weapon on October 9.
The ABC/Washington Post poll, taken 10/5-8/06 finds approval at 39%, disapproval at 60%. CNN/ORC's poll from 10/6-8/06 also has approval at 39% with disapproval at 56%. The Gallup/USAToday poll (10/6-8/06) found approval at 37%, disapproval 59%. The CBS/New York Times poll (10/5-8/06) comes in low among this group, with approval at 34% and disapproval at 60%.
With these new polls, the approval trend estimate has fallen sharply to 38.1%, a bit more than a 2 point drop since mid-September.
Each of the polls shows a decline from the same poll's previous reading. These are too crowded to plot all four in one image, so here they are by poll.
The CBS/NYT poll is quite low compared to the others, raising the possibility that it is another outlier, as was the Newsweek poll over the weekend. The graph below checks this, and finds that CBS/NYT is NOT an outlier, despite it's low reading of approval. With the new data, Newsweek remains an outlier, though less extreme than before the trend was revised by the four new polls.
The conclusion is obvious. Approval has now given up most of the gains made in the August-September rally.
Note: This entry is cross-posted at Political Arithmetik.
Our Slate Election Scorecard update for this evening focuses on two races for Governor: Michigan, where two new polls showing Governor Jennifer Granholm leading restore that state to "lean" Democrat status, and Oregon where Democratic Governor Ted Kulongoski's recently narrowing lead has shifted the race into the toss-up category. Read it all.
Dr. Cobb is an assistant professor of political science at North Carolina State University who specializes in survey research and polling methods.
In a recent poll that Dr. William Boettcher and I conducted with the Institute for Southern Studies, we asked some unique questions about Iraq and we received some unexpected answers. The poll was administered by Knowledge Networks, Inc., included 1,342 respondents, and was conducted from Sept. 19-26. The margin of sampling error for the national sample is plus or minus 2.7 percent.
Bill and I are interested in the perceived goals and probability of success in Iraq, but when we looked around we could not find many surveys asking people what they thought was the primary US goal. Questions about goals were mostly limited to asking whether Iraq was part of the war on terrorism, and a few asked about whether Iraq was already experiencing a civil war.
So, in one question on our survey, we asked respondents what they thought was the primary US goal in Iraq. We asked respondents to choose one goal among four possible alternative mission objectives, but we also permitted respondents to answer "something else" and to subsequently type in their response (one advantage of using Knowledge Networks is that this facilitates the ability to record open ended responses). The question is as follows:
The following are commonly cited as US goals in Iraq. Which one do you think is the primary reason the US is in Iraq today?
- Iraq is the central front in the war on terrorism
- We are promoting Democracy in Iraq and in the Middle East
- We are preventing Iraq from sliding into a civil war
- We want to ensure access to oil
- Something else_____________
What caught our attention was the answer distribution that resulted in a plurality of respondents identifying "access to oil" as the main reason we are currently in Iraq. After we re-coded a majority of the "something else" responses into one new "objective" or back into one of the original four goals, the breakdown was as follows (before we coded the something else responses, 15% had said something else and oil was just 30%):
Oil = 34%
Terrorism = 26%
Democracy = 25%
Preventing Civil War = 7%
General cynicism about/towards Bush = 7%
To be perfectly honest, we had a hard time determining a non-administration preferred mission frame, and belatedly added "oil" as our fourth alternative. We certainly did not expect oil to be the plurality response. Not only is the goal of access to oil rarely mentioned in the mainstream media coverage about Iraq, but also when oil is mentioned as a goal, it is often framed as a crackpot conspiracy.
So my question is this: why are people picking oil as the goal? One explanation is based on the concept of pluralistic ignorance: people privately feel this way but fail to recognize that others also feel the same way, so they are reluctant to speak out and we fail to recognize shared beliefs. Another similar explanation is that the answer of ensuring access to oil is indeed literally what people think we are doing in Iraq but that elites, the media and academics have failed to notice this indicator of deep cynicism about Iraq,s o we don't discuss it. A third explanation is that many people who picked oil do not literally believe this is the primary goal, but the answer options we provided left them with little choice but to choose oil since (a) they reject the administration goals of terror and democracy and (b) preventing civil war is still a noble cause. Some people have argued that we provided respondents with the opportunity to say, "something else", but we know that respondents will choose among the alternatives given even though they would behave different in given an open-ended response option (keep in mind, a surprisingly large percentage of the sample (15%) indeed originally chose "something else".
To help with an analysis of this question, here are two cross-tabulations of the perceived goal with (1) the importance respondents' place on the US achieving that goal and (2) party identification (note: we don't have importance ratings for respondents who originally said, "something else".):
We have our interpretation of these data, but don't want to bias the discussion, so we offer our thanks in advance for any comments on our provocative data...
Not quite two weeks ago, I wrote about a survey of Indiana's 8th Congressional District conducted by for the Evansville Courier & Press by Indiana State University's Sociology and Research Lab. Of the registered voters that responded to to that survey, 63.5% were women, and along with Bill Cullo at Crosstabs.org I wondered whether that unusually high proportion of women might have skewed the results. I wrote that the editors at the Courier & Press "owe their readers some indication of how this very unusual result may have affected the results." About a week ago, they did just that.
A follow-up article explained that Thomas L. Steiger, the sociology professor who conducted the poll, "didn't see the need to adjust for gender in this case" but "for the sake of discussion . . . did a calculation that gave more weight to men." The result:
There was only a fractional change in the spread between Hostettler and Ellsworth.
"There is no evidence of sex bias in this sample related to candidate preference," Steiger said.
"I do disagree with altering poll results. What is important is to test what is being asserted ... and there is no statistical evidence from this sample that females" bias the survey's results in Ellsworth's favor.
In fact, women and men were nearly equal in how they responded to the question about which candidate they preferred: 48.2 percent of men favored Ellsworth, and nearly the same percentage of women - 47.1 percent - favored him, too.
The author went back to Bill Cullo,who agreed with Steiger:
Cullo, however, now agrees that gender had a very minimal effect on this poll.
"I didn't have a problem with Hostettler trailing," Cullo said, "but to not adjust for gender calls things into question. It would have been wise to go ahead and weight the data for gender, though. But it's not nearly the disparity I thought it would have to be. I thought it would be a much more stark contrast."
In fact, when Cullo learned of additional polling data from the Courier & Press, he said Hostettler could be in bigger trouble.
See the full article for details, and lets give the Courier & Press credit for trying to shed more light on this subject.
And as long as we are on this subject, I want to clarify something that confused at least one valued reader about my initial post. I wrote:
Most media pollsters begin with a sample of all adults, and weight the adult sample
that to match the highly reliable demographic estimates from the U.S. Census. They then select a pool of registered or likely voters from the larger adult sample, allowing the demographics of the sub-sample to vary.
I did not mean to imply that the process of selecting registered for likely voters involves a second round of random sampling. Pollsters simply select the subgroup of interest (self identified registered voters, or voters that they classify as "likely") from the larger sample. The process is analogous to selecting any other subgroup (women, 18-30 year olds, union members, etc.).
And I should again make clear that weighting or adjusting a sample of all adults by demographics like gender and age is not controversial among media and political pollsters, because, as I wrote in the first post, we can base those adjustments on highly reliable U.S. Census estimates of the adult population. The practice of seperately weighting the subgroup of registered or likely voters -- the issue in the Courier & Press survey -- is more controversial, because the demographics of those populations vary slightly from election to election, and estimates are less reliable.
My point was that campaign pollsters are typically willing to make educated guesses about the demographics of voters. Media pollsters, including ISU's Steiger, are far less willing.
A new Newsweek poll [story, results] finds approval of President Bush at 33%, a record low for the Newsweek organization, though not for other polls. The poll, taken 10/5-6/06 also found disapproval at 59%. As the graph above makes clear, however, the Newsweek poll is far below other recent polls and the current trend estimate. Before concluding that approval has "really" fallen to 33%, a good deal more data will have to confirm this extremely large drop.
That is not to say that approval has not turned down. Indeed, after the last two weeks of unremitting bad news for the White House, it is hard to imagine how if could have failed to do so. The question is whether the Newsweek poll is a reasonable estimate of current opinion or if it is too far away from other results to be plausible.
With the addition of the Newsweek poll, my current estimate of approval (the dark blue line) stands at 39.4, a full point below the recent high of 40.4 on 9/20. (This value is constantly being revised as new polling allows better estimates of the peak. On 9/19 the estimate was 41.1, while on 9/21 it stood at 41.2. Three polls completed 9/25-9/27 agreed on 42%, driving the estimator that day to 42.2. However, in light of all the polling since, the revised estimate puts the recent high at only 40.4 as of 9/20.)
While a one point drop may seem small, this is a drop in the trend estimate, NOT the very noisy raw polls which move by much more from poll to poll but which also contain much more random variation. One point in 16 days is almost 2 points per month, which is quite a substantial rate of decline. For comparison, during the first five months of 2005, the President's approval declined by one point each 13.5 days, as estimated just before approval reversed and started back up on May 15. I hasten to add that we lack enough data to be at all confident as to what the current rate of decline actually is. One point per 16 days is merely the best guess given current data. It might be less or more and will certainly change as new data come in.
So granted approval is now clearly going down, is the Newsweek poll telling us that the drop is of historic rapidity and that new polls will also show similarly dramatic drops, or is Newsweek's poll a statistical outlier-- getting the direction right but seriously overstating the magnitude of decline?
Based on all polling since 2002, the Newsweek result is clearly a statistical outlier. It is well below the 90% confidence interval in the figure below and is among the larger negative outliers of the past four years. The figure plots the residuals, the deviation from the trend estimate, for each poll taken since January 1, 2002. Orange points are polls that fall outside the 90% confidence interval, indicated by the low and high horizontal lines. The mean of all residuals is zero, the middle horizontal line. Newsweek polls are in red for easy identification.
Newsweek has a good polling track record, producing only one previous outlier in 70 polls. Here "outlier" is any poll outside the 90% confidence interval. By definition, we expect 10% of cases to fall outside that range. In fact, Newsweek has now produced two out of 71 polls since 2002 that are outside that 90% CI, a 2.8% rate. This latest poll, however, is clearly well beyond the lower limit of the interval.
Two recent polls have also fallen outside, but above, the 90% CI. Both of these were completed prior to the revelations about Representative Foley's IM messages to House pages, though after the now estimated peak of approval on 9/20.
Even if we focus on the four polls completed (at least partially) since the Foley Folly began, we see that all four are well below the trend estimate, but that the Newsweek poll is substantially below even these polls.
The conclusion is that the Newsweek poll is implausibly low, given the other data we have seen so far. A possible (but statistically less likely) alternative is that approval is falling very rapidly so that the unusually low Newsweek result reflects a much sharper rate of decline than the 1% per 16 days that I currently estimate. Such a rate of decline would be much higher than any yet seen in the Bush administration. I find this quite unlikely, but the data will answer this one way or another within a couple of weeks. For now, I'd bet the approval trend will fall another half point by the end of next week. If it falls much more than that, the Newsweek poll will look like a harbinger rather than a statistical fluke.