Number Nonsense

"You might be an out-of-touch DC elite if: you hire Mark Penn (!!) to do a poll on out-of-touch DC elites," polling wizard Nate Silver tweeted this morning. The object of his ire? A Politico package about differences of opinion between people who live in the Washington metro area and other parts of the country that drew its conclusions from a survey conducted by Penn Schoen Berland, Mark Penn's company, and from Bureau of Labor Statistics data that shows Washington with the lowest unemployment of any major metropolitan area in the country. Politico isn't the first publication to try to find hard evidence for Washington elitism, but the paper might want to take a closer, harder look at the numbers they're touting as proof.

First, let's look at the poll. Penn's company talked to just 227 "Washington elites," defined as people who live in the metropolitan area, make more than $75,000 annually, graduated from college, and work "in the political process or work on key political issues or policy decisions." Total ambiguity about what that last phrase means aside, it's a very small sample size, especially because the factors used to filter out so-called elites aren't particularly limiting. Penn's company talked to 1,011 people who live outside of the Washington area, and could have easily found a comparable number in DC. So why the smaller size?

Especially because it produced a margin of error of 6.53 percent in the figures on "Washington elite" opinions. That's a big potential fluctuation to hang a series of articles on, and it's a figure that appears only at the very bottom of one story in the package and that doesn't appear at all in the other. The margin of error may not have been enough to flip Politico's conclusions, but it bears noting, and high up in each piece.

Then there's the economic data, which features prominently in the story in the package written by Jim VandeHei, Politico's executive editor, and Zachary Abrahamson. In that piece, they note that "Since the most recent recession began in December 2007, metropolitan Washington has shed about 71,000 employees on nonfarm payrolls, the fewest number among the nation’s 15 largest metropolitan areas." That's true, but it's not necessarily the case that the recession has created opportunities for Washington while leaving the rest of the country behind.

In the Bureau of Labor Statistics' measurements of unemployment in major metropolitan areas, the Washington area has had lower unemployment than any metro area for 7 of the last 11 years. In two other years, it finished 2nd, and then 7th and 5th. In other words, Washington may be doing better economically than its counterpart regions, but this is hardly a new phenomenon, and hardly due to the recession.

And it's interesting to read VandeHei and Abrahamson argue that "Washington has been largely shielded from the economic downturn, even in 2009, when most states and cities were hit the hardest." There's no question that unemployment is bad in Flint, Mich., one of the areas the reporters cite to argue that Washington isn't feeling the rest of the country's pain. Flint's unemployment rate rose from 4.4 percent in 2000 to 14.7 in May 2010. But it's not as if the Washington area has gone untouched. During that same time, and also according to the Bureau of Labor Statistics, unemployment in the Washington metropolitan area rose from 2.7 percent to 6.0 percent. It's not as big an increase as Flint's, but it's not nothing. And it means that job creation isn't keeping up with job loss.

The difference between Washingtonians and the rest of the country in economic security and in outlook is important, and well worth exploring. But that means it's worth doing right, and worth examining as a persistent phenomenon—not just as a way to win this or any other morning.

Subscribe to Washingtonian
Follow Washingtonian on Twitter

More>> Capital Comment Blog | News & Politics | Party Photos