Monday, November 5, 2012

Hating on Numbers

In the latest effort by pundits to grab Nate Silver's hands and hit him in the face with them while asking "Why are you hitting yourself?", we see Michael Gerson's column that begins with a dismissal of Silver and morphs into an attack on political science. There are a few points in here on which I feel compelled to comment, so I'll take them one by one. Here's Gerson:
Silver’s prediction is not an innovation; it is trend taken to its absurd extreme. He is doing little more than weighting and aggregating state polls and combining them with various historical assumptions to project a future outcome with exaggerated, attention-grabbing exactitude. His work is better summarized as an 86.3 percent confidence that the state polls are correct.
Well, yes, that's more or less accurate. Why pundits get so offended by this is beyond me. Silver is hardly making the argument that everything important in politics can be captured by polling models. What he is doing is trying to combine current polling and historical trends into the most accurate prediction of the election results we can come up with. Political journalists are often criticized for focusing too much on the horse race. We regularly observe a media approach in which any new poll becomes a news story, and in which Sunday talk shows focus on who's ahead and who's behind, and what the person in second place needs to do to get into first place. So suddenly some guy comes along and says, "Well, if you're going to talk about the horse race, we might as well at least know which horse is ahead and by how much." And then he gets slammed for making it all about the numbers? Poll-watcher, heal thyself.

More Gerson:
An election is not a mathematical equation; it is a nation making a decision. People are weighing the priorities of their society and the quality of their leaders. Those views, at any given moment, can be roughly measured. But spreadsheets don’t add up to a political community.
No one's arguing that they do, least of all Silver. Sure, there's plenty more to understanding an election than just following the polls. Just how would the candidates govern should they win? Who would form their administration? How much slack will they be granted by the ideological activists who backed them in the primaries? How much power do the candidates have to follow through on the things they have promised? Could they work with Congress? These and lots of other questions are terribly important, but they don't get a lot of attention in the thick of a presidential campaign. Blaming this on Nate Silver is like blaming the guy who built the scoreboard at Dodger Stadium for society's ignoring the poetry of modern baseball.

Yet more Gerson:
The current mania for measurement is a pale reflection of modern political science. Crack open most political science journals and you’ll find a profusion of numbers and formulas more suited to the study of physics.
Uh oh, here it comes.
Politics can be studied by methods informed by science. But it remains a division of the humanities. It is mainly the realm of ethics — the study of justice, human nature, moral philosophy and the common good. Those who emphasize “objective” political facts at the expense of “subjective” values have strained out the soul and significance of politics.
And there it is. Okay, let me see if I can lay this out for Gerson. Obviously, the study of justice and the common good has an important role to play within political science. Indeed, we have a whole subfield that deals with such issues. We call it political theory. But there's a lot more to study! What forms of government produce the greatest equality and economic advancement? How productive are divided governments? To what extent are democratic governments accountable to and representative of their citizens? What explains the outcomes of elections? What are the costs and benefits of partisan versus nonpartisan elections? Why do people vote, or choose not to? These and many, many other questions are not only important, but also empirically answerable! We can find data and derive answers, sometimes using the "profusion of numbers and formulas" Gerson derides. This does not strain the soul and significance from politics; indeed, it provides answers to some of the questions that have vexed us ever since two people started making decisions on behalf of a community.

Now, Gerson took a serious detour from polling to political science. Silver is not a political scientist and, as far as I know, has never claimed to be one. His focus for right now is forecasting elections, and he's taken a pretty thoughtful and data-driven approach to the task. If you don't want elections to be about the horse race, don't talk about them that way. But I'll take Silver's numbers over all the media references to "momentum," "narrative," and "relatability" eight days a week and twice on Sundays.

Update: See also John Sides.

Further update: Peggy Noonan calls the election for Romney because "all the vibrations are right." Seriously. I wish she'd be more transparent about the frequency and amplitude data.

No comments:

Post a Comment