Ecological inference and vampires

Via Sullivan, here's a map by Goodreads in which they map out reviews of the book "Twilight" by state. Bluer states have more negative reviews of the book; redder states have more positive reviews: 

The mysteries presented are twofold:

1. Why does this map seem to mirror voting patterns so sharply? That is, why do Republicans like "Twilight" while Democrats hate it? 
2. Why has Utah generated such a disproportionately high number of reviews?

The second one is easy: the author, Stephenie Meyer, is Mormon, and a graduate of BYU. I can only assume that there's been a disproportionate share of buzz about her in Salt Lake City. Still, it's curious that Utah departs from the partisan trend; she has so many readers, but most of them dislike the book. 

The first one is trickier. What can account for Republicans liking the book so much more than Democrats do, other than, as Erik Loomis says, "Republicans having horrendous taste in literature"?

Actually, we don't know that. This is a classic case of ecological inference fallacy. It is certainly possible that Republicans like the book more than Democrats do, but the map can be deceiving. The people in these states who vote in presidential elections may not be the ones who register opinions on Goodreads. Indeed, a goodly number of "Twilight" readers are probably under 18. 

The only way we can really test the partisan connection is to examine individual level data, and Goodreads (regrettably) does not collect data on party ID. Still, there might be a way to infer party ID from Goodreads profiles, perhaps by looking for hints of political preferences in the "about me" section. If Gary King or some other enterprising individual would like to work on this, have at it.

The persistence of ideology

Name the speaker:

[The President] has lost faith in the American people. Just look at the men surrounding him. They are cynics who scoff at our simple virtues. They think that the people are too dumb to understand democracy. Their idea is that they, the intelligentsia, can govern us with catch phrases and sleight-of-hand.... Give our country back to us. We want it. We love it.

Sarah Palin? Newt Gingrich? Michelle Bachmann? Good guesses. No, those words were spoken by Wendell Willkie during his 1940 run for president against Franklin Roosevelt.

I learned this gem from John Gerring's wonderful book Party Ideologies in America, 1828-1996. Gerring demonstrates a remarkable stability in the language and imagery used by party orators over time. The themes that modern Republicans discuss in their speeches -- limited federal government, low taxes, the supremacy of small business, etc. -- were really developed back in the 1920s. Speeches delivered by George W. Bush or Ronald Reagan could just as easily have been delivered by Herbert Hoover or Calvin Coolidge.

Gerring also finds some interesting turning points in the history of American ideologies. Whigs and Republicans in the 19th and early 20th century used to have a different set of beliefs, speaking of liberty as though it were inextricably linked to personal responsibilities, and seeing the government as the means to inculcate and enforce those responsibilities. The notion of "self-government" had two meanings in Lincoln's era, both choosing one's leaders and ruling one's own passions.

I'm still in the middle of this book, but I'm learning a ton from it. I hope to share more later.

Game Change

I finally read "Game Change." In some ways, it's a bit of a disappointment. For the most part, it really doesn't tell you anything you wouldn't have known by closely following the news in 2008, which, if you bought this book, you probably did. I would say that "Game Change" is to "What it Takes" as "CSI" is to "The Wire" -- entertaining enough, to be sure, but lacking in depth, quality, and revelatory power.

That said, there are two aspects of "Game Change" that make it worth reading, and possibly assigning to undergrads. The first is the coverage of the early efforts by party leaders to groom Obama for a presidential run. You really get the impression of Obama as a made guy: he'd certainly thought about running for president, but not so early in his national political career. But party elites (broadly defined) from Harry Reid to David Geffen were looking for an alternative to Hillary Clinton and directly urged Obama to run. The book's treatment of the Republican nominations process is breezy and uninformative (although I loved the descriptions of McCain and his potty mouth), but these early sections on the Democratic process were really quite good and would fit well into a parties class. (They also suggest that 2008 was a better case for The Party Decides than it originally seemed.)

The second is the examination of how the McCain campaign managed Sarah Palin. Palin comes through in the book as an odd but somewhat sympathetic character. She was initially very eager to help out the McCain ticket and dove into memorizing world facts and developing speeches, but she was soon overwhelmed, thrown into a media firestorm without much preparation. The book makes it sound like she was clinically depressed at one point, sleeping rarely, losing weight, and going blank during conversations. Just prior to the vice presidential debate, the campaign managers decided to intervene, essentially demanding she eat better and take sleeping pills, limiting the amount of information she was expected to memorize, and reuniting her with her family for debate prep. I might use a chapter from this when discussing the role of campaign managers in my campaigns class.

What It Takes

At the repeated urging of Jonathan Bernstein, I finally read Richard Ben Cramer's What It Takes. And you know what? It's excellent. It's a very detailed behind-the-scenes look at most of the major candidates in the 1988 presidential election, including Dole, Bush, Gephardt, Biden, Hart, and Dukakis. The stories are wonderful. The candidates' backgrounds are fascinating -- I was surprised how many of them dealt with a serious illness or death of a loved one (Bush's daughter, Biden's wife, Dukakis' brother, Gephardt's son), not to mention Dole's own personal travails.

The book is very much not political science, but it contains a ton of evidence that political scientists would find useful. I got the impression that there's enough material there to write another The Party Decides, although the book just isn't organized that way. Take, for example, this discussion of Gephardt's campaign:

When you got down to it, [Gephardt] meant to run Jimmy Carter's campaign -- twelve years later, different issues, new wrinkles, but still, he meant to hike the trail that Carter blazed. He would come out of nowhere, win Iowa... get the bump... and then the hot light would hit. Dick had to be ready. He had to know how to run in the South, how to make his campaign truly national; had to know what the press would do, how to get the money while his name was hot, how to tie down the pols who meant to ride with a winner... how to build momentum until his nomination, like Carter's, could not be stopped.

So he sought Carter's advice, and he followed it. He started early. He made sure his contact with Iowans was not only broad, but deep -- and deeply personal. Thought his money was tight, he staffed not only Iowa and New Hampshire, but offices in several southern states. He picked his campaign team, and he backed it -- even in the worst times, he never second-guessed. He worked small towns, and corn boils, church picnics, county fairs... he did everything, in short, that Jimmy Carter did... did it just as hard, and much longer... and with two months before the Iowa caucus, he could see... it hadn't worked worth a damn.

Others, including the authors of The Party Decides, have pointed out how the 1970s were an atypical era in presidential politics where a bandwagoning candidate could actually put together a campaign on his own and win the nomination, and how that approach stopped working by the 1980s. And here's Cramer hitting the same point back in 1988. Pretty cool.

Now, Cramer grants a much more powerful role to the media than most political scientists do today. And the media really do not come off very well. Respected journalists like E.J. Dionne are depicted buying into pack journalistic mentalities and grilling Gary Hart about his affairs because "everybody knows" he has a problem. What's more, the candidates seem to react to this press coverage. Hart and Biden are seen as pulling out of the race because the media would not leave them alone about Donna Rice and plagiarism, respectively. You can almost see Bill Clinton reading this book in 1990 and saying, "So all I have to do is ignore them and I can still win the race? Piece of cake."

The political media clearly have their own agenda separate from that of the parties, and they can seem quite feckless at times. But these stories aren't always wrong. Bill Clinton really did have a character problem, one that ended up dominating much of his second term in office. But Cramer presents evidence of pretty good people being driven from public life by nasty, personal media coverage based on little factual content, and it's hard to defend that.

Anyway, the book is definitely worth the read.

The new party model

Denver 9News reporter Adam Schrager and former CO statehouse Rep. Rob Witwer have a new book coming out called The Blueprint: How the Democrats Won Colorado. I was lucky enough to receive an advance copy, and I recommend the book highly.

The book focuses on a new style of campaigning that's been going on in Colorado over the past few election cycles, but what the authors are really describing is a new form of political party.  Schrager and Witwer argue that a small group of wealthy liberal donors hasn't taken over the Colorado Democratic Party so much as they've built a new one. Starting in 2003, the "Gang of Four" (Tim Gill, Pat Stryker, Rutt Bridges, and Jared Polis) put together a series of 527s designed to channel tons of money into state legislative races with the express purpose of flipping control of the statehouse. Republicans basically didn't see it coming -- no one had ever devoted this level of funding or strategic targeting into statehouse races before. And because the funding wasn't flowing directly from donors or from the party to the candidates, it was difficult to perceive until it was too late. The Gang is motivated largely by policy concerns -- Gill and Polis, for example, are gay and were really irked by some legislation the state Republicans were pushing. The Gang's funding is largely credited for the Democratic takeover of both statehouses in 2004 even while Bush won the state by five points.

There's an argument advanced in there that this represents a new and better form of party organization:

Everyone knew the party had been notoriously inefficient when it came to spending its money. In Polis' view, this was a function of how people get into decision-making roles in state political parties. Party leaders "were selected because they travel the state," he observed. "They know people. They show up at every dinner. People like them. They manage the palace intruiges effectively."

But that was also a weakness. Applying a businessman's eye for organizational effectiveness, Polis identified what he believed was the main problem with political parties. "There's no reason to think [party leaders] would be good at running campaigns and making tough decisions.... In fact, to the contrary. They would have a tendency to put valuable resources into races they're probably not going to win because they want to win friends. So, if they like so and so and they're running in a very Republican district, they're going to give them help, which takes it away from a very competitive district. So it wasn't a very good way to allocate resources."

This is a topic I've been thinking about for some time but haven't been sure what to do with. Dick Daley was a strong party leader, sure, but just how efficient was his organization? If he wanted to control city council or even state legislative votes, he surely could have done a lot less and not really compromised much of his effectiveness. I'm wondering if the stuff Maxine Waters or the Orange County Lincoln Club does today is no less effective than the work done by Tammany Hall but perhaps performed with a lot less overall effort.

By the way, whether you like this style of campaigning or not, it's probably coming soon to a state legislative race near you.  Schrager and Witwer note how this model is being exported to other states, arguing that Tim Gill may currently have more influence over more state legislative races than any other person in history.

This book is a great and easy read and isn't just for Colorado political junkies. If you're interested in parties, campaigns, or state politics, or you're teaching courses on these topics, you might want to check this out.