Monday, December 24, 2012

Pollsters’ postmortem: When enthusiasm didn’t matter


THE AFTER-action reports for the 2012 election have been trickling in since the election was over; there’s not much left to debate except the future leverage of President Obama’s winning margin, and the ways the 47 percentile number will be Mitt Romney’s eternal contribution to the national political folklore.

But the pollsters have taken stock of the race; some have come to conclusions that call into question the validity of enthusiasm polling. One analysis in particular, from a conservative-affiliated polling outfit, tells us two things:

(1) For a significant number of voters, enthusiasm — any unalloyed, spasmodic passion for a candidate — didn’t necessarily figure into their turning out at the polls; and (2) enthusiasm as a statistical indicator is too squishy and imprecise to be meaningfully used for presidential polling in the first place.

◊ ◊ ◊

In a Sunday story on campaign polling accuracy, Mark Blumenthal of The Huffington Post notes how pollsters have begun analyzing their own results for the election and found that “questions about the accuracy of self-reported enthusiasm as a predictor of turnout have intensified since Election Day.”

Blumenthal references an infographic produced by Resurgent Republic a GOP-connected polling org. He reports that the infographic “shows that self-expressed enthusiasm by various demographic subgroups just before the election had virtually no relationship to turnout (as measured the change in size of the subgroup in exit polls from 2008 to 2012).”

According to an analysis page at the Resurgent Republic Web site, “Key subgroups of President Obama's winning coalition including Hispanics, young voters, and unmarried women outperformed their 2008 turnout levels, even though these cohorts exuded less enthusiasm to get to the polls than Governor Romney’s core supporters.”

◊ ◊ ◊

The analysis thoughtfully breaks the issue down cohort by cohort:


For voters 18-29: “Despite having low enthusiasm levels throughout the election, voters 18-29 had a 6 percent increase in turnout compared to 2008, up one-point overall to comprise 19 percent of the electorate. President Obama won the youth vote by more than 20 points (60 to 37 percent), but he did so by a smaller margin than 2008 (66 to 32 percent). Latino voters 18-29 propelled the increase among the youth vote, expanding to 4 percent of the overall electorate, while whites and African Americans equaled their 2008 levels.”

For unmarried women voters: “With two-thirds support in November, Democrats hold a commanding lead among this growing segment of the electorate.”

For African-American voters: “Turnout among the black community equaled the sizable percentage increase seen in 2008. From 2004 to 2008, the African-American vote saw an 18 percent change, going from 11 percent in 2004 to 13 percent in 2008. Unlike President Obama’s other coalition groups, African American voters were the most enthusiastic to turnout of all voting groups.”

◊ ◊ ◊

And for Latino voters: “The fastest growing minority group had an 11 percent increase in turnout compared to 2008, up one-point overall to 10 percent of the electorate. Republican support reached a high point in 2004, but has dropped by nearly 20 points over the past two presidential cycles.”

And then, the stunning stat that the GOP better pay attention to: “Meanwhile, the political influence of this voting bloc will increase exponentially as 50,000 Americans of Hispanic descent turn the age of 18 every month for the next two decades.”

Turnout results for other groups — whites in general, seniors and white evangelicals — were more solid for Romney, but their overall participation was diluted relative to a younger, browner, smartphone-dependent, estrogen-powered electorate.

From the Resurgent Republic analysis: “[T]he overarching trend of a less white electorate will continue as Hispanics, Asian Americans, and other minority groups increase their political participation and the white vote continues to age (white seniors increased to 14 percent in 2012).”

◊ ◊ ◊

THE MISTAKES made by the Romney campaign, and its internal pollster mouthpieces, were too numerous to go through here, but chief among them was the assumption that those cohorts of voters had nothing to turn out for — that the, uh, enthusiasm they had for Barack Obama in 2008 was not transferable to the 2012 campaign.

Team Romney & pollsters also overlooked or ignored a corollary of our voting history. It’s a standing assumption that Republicans turn out in greater numbers for midterm elections; it’s also generally held that Democrats tend to turn out strongly for general elections. What made them think this pivotal election would be any different?


The fact that they were wrong about their internal polling, and the campaign strategies derived from that polling, is obvious; the question is, why?

◊ ◊ ◊

Ashanti, commenting on the Blumenthal story at The Huffington Post, offered one possibility in a snarky assessment of the GOP point of view, tongue firmly embedded in cheek:

“According to Republican propaganda and Republican-leaning polling organizations (specifically Rasmussen):

“African American voters were not inclined to vote in the 2012 election. Not only do they not care about voting, but they only wanted to vote in the 2008 to be part of the historic election of the first black President. Besides, African Americans were very disappointed in President Obama, especially in light of the high unemployment in the black community; and since black people only care about themselves and other black people, and since he wasn't being the President of Black People, then they would either stay home or at the pool hall drinking beer, or vote for the white guy! Oh! And without ACORN, the black vote was clearly insignificant anyway, because everyone knows that the only reason Barack Obama won in 2008 was because of ACORN.”

◊ ◊ ◊

REGARDLESS OF how people answered pollsters’ questions about enthusiasm, GOP polling estimates of a lackluster turnout among women, the young and minority voters made more of the idea of voter enthusiasm than it deserved.

The pollsters' perception was off from the start; they made the mistake of confusing enthusiasm with situational enlightened self-interest. For those voters this time, enthusiasm per se never really entered into the equation. They knew what was at stake in this election. The distinctions between Obama and Romney, their policies and personae, were stark and meaningful, and more than enough to spur them into action on Election Day. They knew what had to be done whether they were “enthusiastic” about Obama’s every policy and executive order or not.

For these Obama voters, however disenchanted some of them may have been with the president, voting in the 2012 election was not so much about enthusiasm as it was a civic obligation. Like going to the doctor for a checkup or a flu shot, or getting your teeth cleaned: maybe they didn’t like doing it, but given what was hanging in the balance, the idea of not doing it was even worse.

◊ ◊ ◊

BUT what finally undoes enthusiasm as a reliable barometer of the national mood is its very elastic nature. People have different emotional baselines, political temperaments, tolerances for media consumption, and attention spans. Being enthusiastic isn’t so much a state of mind as it is a state of the moment, and as such it’s highly subject to change. The use of enthusiasm for polling is only scarcely enlightened with the use of survey modifiers like “very” and “moderately.”

Back in August 2010, monitoring the midterm elections then 11 weeks out, I noted that there was talk about an enthusiasm gap for Obama, serious discussion of a form of statistical analysis as gauzy back then as it is today:

“There may be no squishier measurement in the metrics of American politics than the enthusiasm gap, that loosey-goosey way of making a quantifiable out of something that’s more ephemeral than an opinion or a belief.”

Not much has changed since then. Except that certain political strategists for certain humbled political campaigns are now invited, courtesy of hindsight, to disbelieve enthusiasm data altogether. An emerging coalition of Americans knows what counts. The enthusiasm numbers that matter, the ones to believe in, are there in the popular vote. And they’re there for reasons that have less to do with enthusiasm than the pollsters think.

Image credits: Obama-Romney two-shot: via maboot.com. Resurgent Republic info graphic image, logo: © 2012 Resurgent Republic.

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...