What the Pollsters Aren’t Telling Us
- 1 October 2012 by Author 2 Comments
What the Pollsters Aren’t Telling Us
By Richard Larsen
Published – Idaho State Journal, 09/30/12
Most presidential election polls indicate that Barack Obama is leading Mitt Romney from two to eight percentage points. A notable exception is the Rasumssen poll which has Romney up slightly. While the pollsters are unlikely colluding, as the mainstream media do, to support their candidate of choice, there are three significant quantitative reasons the polls paint a skewed picture.
For a little perspective, with just a week to go before the 1980 presidential election, challenger Ronald Reagan trailed incumbent President Jimmy Carter by three to six percentage points in the polls. Then came the one and only presidential debate, and pre-election polls indicated the race was “too close to call.” The final results were significantly different than “too close to call,” as Reagan won by ten points, even with Independent Candidate John Anderson picking up nearly 7%, and Reagan won an electoral landslide, 489 to 49, with only 6 states voting for Carter.
The first issue that challenges the current polling methodology is based in party affiliation. This is always tenuous and in a state of flux, depending on sometimes exogenous factors that incline voters to lean one way or another with regard to political party identification. According to Rasmussen Reports, the latest data indicate a quantitative spike for Republicans (37.6%) and a drop for Democrats (33.3%) representing a 4.3% spread advantage for Republicans, with Independents coming in at 29.2%.
To provide context, at the time of the 2010 midterm elections, party affiliation was Republican 36.0%, Democrat 34.7%, for a 1.3% advantage for Republicans, and Independents at 29.3%. At the time of the 2008 presidential election, Democrats led significantly with 41.4%, versus Republicans at 33.8%, for a 7.6% Democrat advantage, and Independents at 24.7%. Election results for both 2008 and 2010, based on self-declared party affiliation, portend significantly different results for 2012 than what polls are currently indicating.
In spite of the current affiliation advantage for Republicans, the major polls are nearly all oversampling Democrats, as they explain in their methodology. Their rationale is based primarily on 2008 voter demographics gleaned from exit polls. Pollster’s explanation for their methodology is an exercise in statistical and verbal gymnastics as they justify their sampling rates.
Dick Morris, who engineered Bill Clinton’s successful elections as governor and president, said recently, “All of the polling out there uses some variant of the 2008 election turnout as its model for weighting respondents and this overstates the Democratic vote by a huge margin.”
Morris explains why the 2008 data is not an accurate model for the 2012 election. “But 2008 was no ordinary election. Blacks, for example, usually cast only 11% of the vote, but, in 2008, they made up 14% of the vote. Latinos increased their share of the vote by 1.5% and college kids almost doubled their vote share.”
“But polling indicates a widespread lack of enthusiasm among Obama’s core demographic support due to high unemployment, disappointment with his policies and performance, and the lack of novelty in voting for a black candidate now that he has already served as president. If you adjust virtually any of the published polls to reflect the 2004 vote, not the 2008 vote, they show the race either tied or Romney ahead, a view much closer to reality.”
Which brings us to the second anomaly as it relates to the current crop of polls. There’s a significant shift in the enthusiasm gap with this election cycle. In 2008, voters that leaned Democrat were 61% more enthusiastic than usual to vote, while the Republicans were at 35%. In 2010, the enthusiasm gap had evened out at nearly 50% for each side. For this election cycle, Gallup indicates those that lean Republican are 51% more enthusiastic, versus 39% for the Democrat leaners.
Gerald Seib of the Wall Street Journal pointed to this enthusiasm gap this week. “Among those who voice the highest interest in the election — in other words, those who seem most intensely interested in voting — Mr. Romney leads by three percentage points.”
The third quantitative anomaly is in how the demographic disparities just don’t add up to confirm most poll results. The independent vote, 29.2% who self-identify as such, favors Romney by 15 points (James Carville’s Democracy Corps) and 14 points (CNN). Politico indicates Romney is favored by middleclass voters by 14 points and male voters by 6. Rasmussen says Romney leads the white vote by 20%. The 45 and older voters favor Romney by 7%, white women favor Romney by 9% representing a 16% gain from what McCain got in 2008. Romney is making gains with Hispanics at 40%, and leads with Catholics at 51%. A Democrat has never won the presidency without a majority of the Catholic vote. Obama is losing 25% of the Jewish vote he garnered last time, and the youth vote, which voted 78% for Obama last time, gives him only a slight edge, 49% to 41%, this time around. Also, Dick Morris claims that historically, as much as 75% of the undecided vote breaks for the challenger, which is between 7-10%.
Questionable under or oversampling, a widening enthusiasm gap, and demographic breakdowns that don’t add up, paint a different picture than what most polls are indicating and cast a pall of doubt over their validity. Ultimately it all boils down to who shows up to vote, and the best thing for conservatives to do is ignore the pollsters and show up en masse on November 6.
AP award winning columnist Richard Larsen is President of Larsen Financial, a brokerage and financial planning firm in Pocatello, and is a graduate of Idaho State University with a BA in Political Science and History and former member of the Idaho State Journal Editorial Board. He can be reached at firstname.lastname@example.org.