Turnout, political science and popular debate
Michael McDonald provides an entertaining bit of snark in his overview of the 2008 turnout figures.
The 2008 presidential election poses a challenge to voting scholars. The myth of declining turnout rates has been laid to rest, and participation has been increasing for three consecutive presidential elections, contrary to a large number of theories advanced to explain the myth. Unless I have missed something, membership in bowling leagues is not surging; ESPN, the Food Network, and the SciFi Channel are not broadcasting political news; political commercials remain negative; citizen trust in government remains low; income inequality is increasing; and turnout among those apathetic baby boomers and Gen-Xers is rising along with all other age categories.
I dunno that he’s right about the Sci-Fi channel (in my opinion, Battlestar Galactica did a better job of covering the GWOT than large chunks of the mainstream media during 2005-2006)1 but his point still stands. There are a lot of theories out there that explain declining voter turnout in secular terms, pointing to large scale changes in media and society as the culprit. These claims look to be at best of limited explanatory power, and at worst fundamentally problematic, given that voter turnout is rising again while none of these trends has reversed. As McDonald suggests, parties’ voter mobilization efforts and (perhaps) early voting and other measures aimed at making voting easier are far more plausible explanatory factors.
But this begs a question: if voting decline can’t be traced back to declining social capital, falling television news consumption etc, why does a miniature industrial complex of semi-popular books, magazine articles and op-eds say otherwise? Here, I’m not blaming the research underlying some of these claims (which seems, as best as I can tell, competent on its own terms), so much as the willingness of the commentariat to take up and distort and simplify these claims about how rampant individualism, dumb youth etc are undermining the Republic. As McDonald points out, we have good evidence from Gerber and Green inter alia that certain party GOTV tools can make a demonstrable difference, and we also know that these results are feeding back into the strategies of political professionals.2 But I’m not seeing Gerber and Green’s arguments getting extensive play in The Atlantic Monthly etc. The tentative lesson I draw is that political science research is most likely to be taken up into broader social debates when it plays to contemporary anxieties and neuroses, and least likely to be taken u when it tells us something about actual politics. But I’m likely being over-cynical here (and am certainly happy to hear about alternative views and countervailing evidence).
1 McDonald himself later notes that “there are thousands of obvious birth-date errors on voter registration files, such that if the dates are to be believed people are traveling from the future to change the course of history or are over a thousand years old.” But if McDonald is wrong and these records are factually correct, who’s going to report it except the Sci-Fi Channel???
2 My thanks to a commenter on an earlier post for pointing out in a mildly snarky fashion that I was seriously underestimating the extent to which this was the case.
Comments
Henry
I was the snark. Thanks for taking it in the mild fashion intended. I thought my internet handle was well known.
Michael has been making the rounds with this set of comments. Part of this is driven by a heated set of exchanges between him and Curtis Gans that reveals some serious disagreements. I think Michael is the reigning expert on turnout, and this does no disservice to Curtis’s long years of contributions in this field, but the level of discord was surprising.
Intake of breath … all that being said, I think Michael may be overplaying his hand a bit here. The small industry that developed DID find an empirical regularity—turnout DID decline markedly from 1960-1996 (ignoring 1992)—and the small industry that developed tested and rejected a number of explanations for this decline, eventually settling on the account you describe above.
It is far too soon, I think, to assume that these explanations were all wrong. It is very possible, as Michael himself allows, that other things have changed that simply ameliorate their effects.
Posted by: paul gronke | January 13, 2009 01:28 PM
In addition to the work of academics that have informed the turnout work of campaign managers and activists, there is a lot of research and evaluation work now being done (with varying levels of availability and quality, etc.) within the professional (nonprofit and for-profit) “voter registration and turnout” circles. In short, the whole “voter turnout” industry or social movement, however you see it, has revolutionized. Whether this explains much or they should pat themselves on the back, I don’t know, but it is possible that they’ve had some policy and management success. Heck, there may even be a “reflective” action going on here: people hear the “woe is us, civic participation is declining” message and have become more active. Possible? (Li’l bit? Maybe?)
Posted by: Doug | January 13, 2009 04:48 PM
Paul - no offence taken of course, and I hadn’t been aware of the back-and-forth with Gans.
Posted by: Henry | January 13, 2009 06:01 PM
What I find interesting as we move forward from this point is that those wedded to the view that turnout is declining view the 2008 (and perhaps 2004) elections as deviations from the depressed post-1972 levels. Even if this is the case, I think we can learn something from understanding why we are suddenly seeing an increase. And if so, why not apply these explanations to higher turnout rates in previous elections? If we move away from exploring correlations between lower turnout rates over the past 30 years (prior to 2004) and other ways in which society has changed, and move towards understanding ups and downs of turnout in a longer context, I believe we will move forward our understanding of what motivates people to vote.
If there is one thing I hope that people will take away from my article, it is that we political scientists need to revist turnout theories. The other parts of the article simply suggest some potentially fruitful directions to proceed, but I have no firm belief that there is a correct answer yet.
P.S. This article’s deadline was prior to my engagement with Curtis Gans on his initial turnout report three days following the election that indicated turnout rates were steady from 2004 (he since revised his report to correctly show an increase). The few changes that I did make were minor corrections to turnout rates as more data became available. There is nothing in this article that should be taken as part of my dialouge with Curtis, who I respect for his years of gathering and disseminating turnout data.
Posted by: Michael McDonald | January 14, 2009 01:47 PM
This remark in the paper: “ACORN, a community-based organization that has
notoriously conducted shoddy voter registration drives…” is uncalled for. Somebody should write a book about how badly the press mangled that story and about how many local clerks tossed out as bad or fake applications that were legit. I’ve seen some data that show that many drives gather as high a valid rate as government agencies.
Posted by: Doug | January 15, 2009 05:26 AM