Friday, September 10, 2010

Don’t Get Sprayed by the Poll Cat

Amazing what the inattentive can miss.

Once again, we are close enough to an election that every newscast is likely to offer us a new poll on one race or another. The thing about many of these polls is, they always seem to come in handy to prop up the media’s accepted narrative of an election cycle. Of course, the media’s accepted narrative this year is that the Democrats are going to be buried in an avalanche of Tea Party enthusiasts who are all upset about Democratic accomplishments in health care, financial regulation, and preventing the recession from becoming a Depression.

What has been lacking in all this coverage is a little Political Science 101. Yes, the Tea Partiers make for amusing news coverage. For TV news, they are better B-roll material than the media have enjoyed since Chicago 1968. But the first question should be: what, exactly, do they bring to the GOP table? Look (if only briefly) at the crowd at the typical Tea Party event. The first thing you notice is, they are overwhelmingly white, and the handful of blacks the TV cameras find often look like they’re out past their curfew from the nursing home. If you do listen to interviews, or even the speeches of the “leaders,” you realize pretty quickly that the educational standard of these mobs is ... a little lacking. But they do show up for rallies and political events, so we know they have an above-average level of political participation. Finally, they obviously express extreme, not to mention confused (“Keep the government away from my Medicaid!”) conservative ideals.

Just to put it simply, if we put this data before a competent political scientist who had just emerged from a three year coma, and never heard of the “Tea Party,” what would he or she predict about these Tea Partiers’ 2008 voting behavior? If you guessed first, that this group probably voted ≥ 98% for McCain, you get a free pint of Publius’s BBQ Sauce, when I get around to a production run. Perhaps equally importantly, given their current high level of activism, our political scientist would tell us that these people were already in the active voter pool in 2008. Modalities of activism may change - 2008’s phone bankers are 2010’s Tea Party demonstrators - but overall participation levels are more stable. Somehow, though, it is always the unspoken assumption not only of media commentators, but of the “experts” they trot out for their stories, that these protesters are all newly emergent voters who will bolster the GOP’s weak 2008 and 2006 showings. All too often, this “analysis” has gone unchallenged by Democratic spokespersons.

What has stricken me so far in this election cycle is how dominant has been the presumption that Tea Partiers represent a response to widely held views, and not that widely held views have been formed by the misleading narrative of Tea Partiers as a “populist groundswell of opposition to the Democrats.” Voters with shallow ideological roots (an impolitic way to describe “swing voters”) are easily swayed by the dominant themes they see in news coverage, and the Tea Party has been Story One in the media since mid-2009. One thing we do know about these voters is that their attention to politics sharpens as an election approaches, and they begin behaving more consistently with their interests and prior predispositions, than with the wind that was blowing them around in the spring and summer.

Which brings us to what the inattentive may have missed.

In all the impending-Republican-deluge hype, one little story has been overlooked. This week, the Gallup Poll released a poll indicating a sudden shift in the generic Congressional ballot. Whereas the Republicans had moved out to a 51%-41% lead a week earlier, the poll this week shows a dead heat at 46%. Part of this is that the 10% GOP lead may look a little like one of those outliers that from time to time intrude on the best polls. It was something of a jump in the GOP total when we look at the poll’s results so far this year:

But these results bode well for the Democrats, and are consistent with an historic structural trend in polling and voting patterns. In order to get an idea of how this has historically worked, we need look no further back than the Gallup daily tracking poll from the 2008 Presidential election:

In this poll, we see something that many have forgotten: as late as early September 2008, McCain actually had a lead over Obama! Most of us have fixed memories of a near-landslide 52.9%-45.7% win, as represented in this map, which adjusts the state size for electoral votes:

Seeing those 365 electoral votes on the map in 2008 has left a lot of people forgetting how close the whole thing was a few weeks earlier. But it’s an old story in American politics. People respond to the dominant theme in the weeks and months before Labor Day, because most people have relatively low levels of political attention until an election is imminent. This year, the low-information voter is asked by the pollster how he plans to vote, and his or her vague thought (in the 3 seconds or less before answering) is something like, “All I been hearin’ is ever’body gittin’ all over Obama,” so he or she indicates a Republican preference.

Then, something funny happens on the way to the local precinct. As ads fill the airwaves, and yard signs dot the drive to work, the voter becomes more aware of the impending election. On the margins, voters begin paying more attention to the detailed news coverage of the election, not all of which is horse-race drivel. Suddenly, the reality of Ron Paul or Sharron Angle wanting to dismantle Social Security makes it to the radar screen of the lower-information voter, and suddenly those previously expressed GOP preferences start dropping off. The end result starts looking more like November 2008.

There are two important points to take away from all this. One is that it is important for us to push back, as Democrats, against the accepted wisdom of a GOP tsunami. A majority of voters - I hate to quote Nixon, but the silent majority - agree with our positions on the issues, when they are focused on them. The second thing, and this may be more important, is to keep heart. There are doorbells to be rung, and envelopes to be stuffed. 2008 wouldn’t have happened if we’d given up when McCain was “ahead” in September.

Today’s post is dedicated to the memory of JR (1945-2010) - union man, and a proud Democrat.


  1. To me, it looks like you're taking national numbers and applying them to the red state of Alabama. I remember sitting at a restaurant in Alabama on Election Day 2008 and overhearing three different tables complain that "he's not my president". I can't put my hands on it right now, but I read an article some years ago that said after an undecided person voted three times for the same political party that person then begins to descibe him/herself as a member of that party.

  2. You’re making the fundamental Beltway Error of classifying a state on the basis of its Presidential vote. The aggregate Democratic vote for state legislative seats was about 54% in 2006. Democrats hold the Lieutenant Governorship, Ag Commissioner, all three seats on the PSC, and the overwhelming majority of Sheriffs, Probate Judges, and Circuit Clerks. In 2008, Democrats won 3 of the 7 U.S. House seats (and would have won a fourth if the shining hyphenated-American intellects at the DCCC had funded Segall properly). That’s hardly a profound shade of red.

    As to a voter forming party identification from his or her own recent voting history, that is a widely reported phenomenon, and is one of several demonstrated variables in the process. But I am not writing about voters who have that sort of consistency. I have written earlier about the fact that there's a ~5% aggregate Democratic boost across the ticket in gubernatorial years (though I was writing there about one position). Those voters, by definition, aren’t going to vote for the same party three times in a row.

  3. Great Poli Sci 101 tutortial. It illustrates how the media uses polls and misinformation to keep the dumb bigots in line.
    I concur;
    What has stricken me so far in this election cycle is how dominant has been the presumption that Tea Partiers represent a response to widely held views, and not that widely held views have been formed by the misleading narrative of Tea Partiers as a “populist groundswell of opposition to the Democrats.”

    Because they are misinformed and uniformed they can be counted on to vote against their best interest time and time again. I'd be willing to bet the people over heard saying President Obama wasn't their President were on Medicaide/care or without health insurance.
    Unfortunately, and for what ever reason Democrats (collectively) can't/won't/don't push back against the misinformation, and those who dare push back are marginalized or suppressed.

    Until and unless the bigotry is exposed instead of excused we are going to continue to be the reddest of the red states. Where voters can be counted on to vote against their best interest.

    Thanks for pushing back in memory of JR~union man and proud democrat.

  4. This comment has been removed by the author.