October 3, 2012

Politico Pair: ‘If (Cooked) Polls (With 91% Non-completion Rates) Hold Steady,’ ‘Obama Will Have Another Four Years’

In an item which talks about a “secret retreat” planned by eight senators which is so “secret” that it’s getting a two-page story, the Politico’s John Bresnahan and Jake Sherman write that “If polls stay steady, (House Speaker John) Boehner will be at the helm of a House filled with Republicans disappointed that Obama will have another four years in the White House.”

Uh, last time I checked, pollsters’ results can hold steady or go in whatever cooked or uncooked directions they wish, and they still won’t determine the outcome of the election. Ballots by voters and the presumably accurate inclusion and counting of such ballots will. Besides, as will be shown, there are even more valid reasons to question poll results now than in the past. Several paragraphs from the rest of B&S’s BS, which is apparently designed to get the country ready to accept “revenue” (i.e., tax) increases, follow the jump (bolds are mine throughout this post).

Gang of 8 plots secret retreat

… the senators in the Gang of Six — plus two newbies — will have a secret retreat next week hosted by Sen. Mark Warner (D-Va.) to start cobbling together a plan to avoid the looming fiscal cliff.

For this round of talks, the bipartisan group has added Sen. Lamar Alexander (R-Tenn.), a former member of Minority Leader Mitch McConnell’s (R-Ky.) leadership team, and Sen. Michael Bennet (D-Colo.), who has grown close to Obama as he tries to capture Colorado this election season.

Senate insiders confirmed the upcoming session, although the group will not disclose its location.

“This group continues to meet and work toward a bipartisan solution,” Warner spokesman Kevin Hall said.

Other Gang of Six members include Sens. Dick Durbin (D-Ill.), Kent Conrad (D-N.D.), Saxby Chambliss (R-Ga.), Mike Crapo (R-Idaho) and Tom Coburn (R-Okla.).

The group came together in the summer of 2011 to try to hammer out a deal to solve the debt ceiling crisis but failed to reach an agreement in the end.

… With Reid, Obama and Senate Democrats strongly opposed to extending the Bush tax rates for anyone making more than $200,000 annually and families making over $250,000 annually, the lines are set for a huge political clash. Insiders on both sides of the aisle are bracing for a rocky lame-duck session no matter who wins on Election Day.

If polls stay steady, Boehner will be at the helm of a House filled with Republicans disappointed that Obama will have another four years in the White House.

No, guys. If the voters vote as the cooked polls you favor indicate they might, Boehner and Republicans will be facing Obama for four more years. Unfortunately for the polls uber alles crowd, there are stronger reasons than ever to believe that such will not be the case this year.

On Sunday, PJ Media’s estimable, inimitable Zombie (“We Are the 91 Percent”; also the origin of the graphic at the top right) noticed an inconvenient item Pew Research published in May. Pew is the organization which was considered, along with Rasmussen, one of the two pollsters who “got the (final) result dead right” in the 2008 presidential election. Unlike Rasmussen, however, cooked Pew polls showed Obama with absurd 15- and 14- point leads over John McCain in that year’s final two pre-election weeks. This year, Pew’s preelection polls look every bit as cooked as they were four years ago.

In May, Pew told readers that over 91% of those pollsters attempt to contact do not complete the related surveys, and then tried to say “that’s okay anyway”:

PewTable9percentPollCooperation2012

It has become increasingly difficult to contact potential respondents and to persuade them to participate. The percentage of households in a sample that are successfully interviewed – the response rate – has fallen dramatically. At Pew Research, the response rate of a typical telephone survey was 36% in 1997 and is just 9% today.

The general decline in response rates is evident across nearly all types of surveys, in the United States and abroad. At the same time, greater effort and expense are required to achieve even the diminished response rates of today. These challenges have led many to question whether surveys are still providing accurate and unbiased information. Although response rates have decreased in landline surveys, the inclusion of cell phones – necessitated by the rapid rise of households with cell phones but no landline – has further contributed to the overall decline in response rates for telephone surveys.

A new study by the Pew Research Center for the People & the Press finds that, despite declining response rates, telephone surveys that include landlines and cell phones and are weighted to match the demographic composition of the population continue to provide accurate data on most political, social and economic measures.

Zombie’s summary, in a slightly revised sound bite: “38% of people are unreachable — and out of every 7 people actually contacted by pollsters, only 1 will answer the polling questions, while the remaining 6 refuse to answer.”

Pew’s brave conclusion that it’s all okay anyway is based more on hope than fact. Its reference point covering the 2010 elections comes from the National Council on Public Polls, which tells us that in the 2010 elections, candidate-related polls during the final three weeks before Election Day varied from the final result by 2.4%, while final-week results varied by 2.1%.

That’s fine, but look at what has most likely happened to the poll completion rate since then. Drawing a straight line from 2006 to 2012 (reasonable, since the table shows a 6-point decline every three years), it appears safe to say that the current completion rate is about 30% lower than it was just two years ago (9% vs. an interpolated 13% in 2010; the 4-point reduction divided by 13 rounds to 31%) and a whopping 47% below 2008 (an 8-point reduction from an interpolated 17% in 2008).

So we’re really supposed to trust poll results already typically skewed toward oversampling of Democrats when pollsters are able to complete surveys at barely half the rate seen in 2008?

Zombie also logically asserted that because Republicans are less than half as likely to trust the media than Democrats, they are probably twice as likely to not respond to polls, which more people, especially Republicans and conservatives, now see as media tools of news manipulation instead of as sincere attempts to gauge public opinion. (One could go further and say that if Republicans and conservatives dominate the ranks of non-completers, media distrust is even higher than the polls indicate.)

Unless someone can make the case that Pew’s experience is atypical, the Politico pair, in their breezy assumption that the polls reflect what people are really thinking, could still conceivably be right. But if so, it’s clear that it will only be by accident. What’s annoying is an apparent perception on the part of many journalists and those on the left (but I repeat myself) that if this year’s election results do vary significantly from the polls, it will be the fault of the electorate or the counting process, and not the supposedly sacrosanct polls. Horse manure.

Cross-posted at NewsBusters.org.

Share

3 Comments

  1. I hope your analysis is correct. Here in Ohio the Columbus Dispatch “Ohio Poll” has a reasonably good track record on sampling the pulse of the Ohio preference for POTUS. The CD poll last week showed a 9% Obama lead (51 – 42%) — but the sample contained +7% dem respondents (i.e., 703 dem; 563 repub., and 362 other). This poll relies on questionnaires sent out “randomly” to names on the Ohio voter registry. I think dems might out number repubs by less then 3% on state-wide Ohio voter rolls. Since most dems support Obama and most repubs are for Romney the only two ways this could happen is that a smaller percentage of repubs returned the questionnaire or the CD’s process of selecting registered voters is not a “random” as they claim? Alternatively it could mean despite what we hear repubs are really NOT more motivated in 2012 then dems?

    Comment by Fargo44 — October 3, 2012 @ 3:40 pm

  2. I took a more detailed look at the Cbus Dispatch poll in a Watchdog post which is under review.

    Comment by Tom — October 3, 2012 @ 4:03 pm

  3. In the end the only poll that counts is at the voting both. As far as opinion polls are concerned they are useless unless it excludes both parties to show only the responses of independents/moderates. They will decide the election in combination with voter turn out of the base.

    IF one were able to parse the opinion poll data for only independents, what would the percentages be? This IMO would be the only realistic indicator of how the candidate was connecting with those who will make the final decision.

    Comment by dscott — October 4, 2012 @ 2:42 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.