In a recent widely-read column in The Hill (a magazine devoted to covering Congress), Markos Moulitsas warns of impending catastrophe for Democrats in the 2010 midterms. Under the headline “Brutal poll for Dems”, Moulitsas cites the results of a recent survey commissioned by his website, The Daily Kos, that reveals a huge disparity in the number of Democrats and Republicans who say they are “definitely” or “probably” likely to vote in the congressional races next year. Moulitsas writes, “Among Republican respondents, 81 percent said they were definitely or probably going to vote, versus only 14 percent who were definitely or not likely to do so. Among independent voters, it was 65-23. Among Democrats? A woeful 56-40: Two out of every five Democrats are currently unlikely to vote.” His conclusion? “If these numbers hold for the next year…If base Democratic voters don’t turn out, like what happened in New Jersey and Virginia this year, Democrats will suffer at the ballot box.”
Moulitsas trumpets these numbers as evidence that the Democratic base is turned off by Obama’s and the Democratically-controlled Congress’ tepid leadership and failure to pursue the more progressive policies that Moulitsas champions. He may be right. But the poll results he cites (which, by the way, are not actually linked to in the article) don’t necessarily support his argument. Here’s why.
To begin, the Daily Kos poll surveys all adults, not just likely voters. You’ve heard me say this before, but it bears repeating: polls of all adults are typically weighted more heavily to Democrats than are polls of likely voters. However, the Kos’ “nonpartisan” poll actually includes fewer Democrats and Republicans than many polls because it has separate categories for not only independents, but also categories labeled “other” and “not vote” (the latter presumably refers to people who didn’t vote in the last election and thus aren’t affiliated with any party.) In fact, if you look at the crosstabs for the actual poll, as I did, you’ll find that the respondents break down into 31% Democrats, 22% Republicans, 25% independents, 17% “nonvoters” and 5% other. For comparison purposes, the most recent Gallup poll breaks down the public’s partisan affiliation as 35% Democrat, 28% Republican and 35% independent.
Why is that important? Because Moulitsas bases his prediction largely by comparing turnout rates among Democrats and Republicans, who comprise barely more than half of his survey. You wouldn’t know this by reading the article which merely references the turnout percentages within each party.
However, for the sake of argument, let us assume that the percentages are an accurate predictor of the actual voting by Democrats and Republicans next November. Doesn’t that disparity in likely turnout still mean a bloodbath for Democrats as Moulitsas says? No, but again you need to look at the actual survey to see why not. Moulitsas compares two percentages – the 81% of Republicans in the poll that say they will likely vote versus the “woeful” 56% of Democrats. But it you translate that into actual numbers, that is, if you take the percentage of the actual number of Democrats and Republicans who are surveyed, you might come to a different conclusion. Eighty one percent of 526 Republicans (the number in the sample) equals 426 Republicans likely to vote in 2010 versus the 51% of the 744 Democrats sampled which comes to…. drum roll, please… 417 Democrats likely to vote. That’s a difference of 9 likely voters, or less than 1% in a survey with a margin of error of 2%. In other words, given the current partisan composition of the electorate as indicated in the survey, these turnout figures – based on Moulitsas’ own logic – suggest a dead heat in 2010 (that is, if we assume straight party voting, and ignore independent and other voters who comprise 47% of the survey).
Now, one can spin these numbers any way one wants. Certainly the lower projected turnout isn’t the best news for Democrats. But Moulitsas could have easily (and accurately) headlined his story as follows, “Survey suggests Democrats poised to break even in 2010 midterm congressional elections.” Indeed, an equally viable spin is that this survey is exceedingly good news for Democrats, since the party that holds the White House has, with two exceptions, lost seats in Congress during the first midterm dating back to every president since FDR. In 2010, however, Democrats might buck history and break even!
But that spin doesn’t serve Moulitsas’ purpose, which is to castigate Obama and Democrats for not embracing the Daily Kos worldview. And I don’t blame Moulitsas for putting the more pessimistic gloss on the survey. There is a reason liberals and progressives flock to the Daily Kos website in such great numbers (and why I lurk there to see what the left wing is saying, just as I survey conservative websites to get the opposing view.) Moulitsas and his colleagues are skilled preachers. As I noted in an interview with a local reporter, websites like Kos’ serve as the equivalent of political churches, where like-minded devotees gather together to read from the same hymnal containing familiar doctrines, and to hear sermons from ministers who share their political convictions. (And where sinners, like those who supported Hillary in 2008, are excommunicated and cast, weeping and gnashing their teeth, into darkness.) The Daily Kos website is comforting and reassuring and a reaffirmation that there is at least one place where people have the correct outlook on life, even if the rest of the world is populated by goose-stepping, bible-thumping, gun-toting fascists that hate Mom, America and Apple Pie. (Upon consideration, it’s a lot like living in Vermont. Or teaching at Middlebury College. ) And the conservatives have their own “churches of the blog” as well where actual gun-toting bible thumpers can mingle among their own fascist kind, like bankers and insurance executives.
My point is that these blogs serve an important function, but it isn’t to teach you about politics, or even give you the “truth” about political affairs (if by truth one means what can be demonstrated to be factual.) They attract many readers for the same reason that churches attract numerous worshipers. But you don’t go to church expecting to learn how to operate heavy machinery – you go to commune with like-minded people (and to save your soul). For the same reason, you shouldn’t go to the Daily Kos website, or read Andrew Sullivan, or Michelle Malkin expecting to learn the facts about presidential or national politics. That’s not their purpose. And when they do post a column purporting to give you the facts, as Moulitsas does regarding the impending Democratic catastrophe, alarms bells should sound and you should immediately begin checking the details.
Praise the Lord. Pass the Facts. End of Sermon.
CORRECTION: An earlier version of this post misspelled Moulitsas’ name – I think I’ve corrected all references now.