Bradley Effect: R.I.P.

by David Epstein

OK, so I stayed up entering all the data by hand just to answer one question: what was the final word on the Bradley Effect I've had to discuss about a million times in the past few weeks.

The answer is: nada. I took the last polling projections from 538.com and compared them to the final results from the election, state by state. These numbers don't include Alaska or Hawaii, which aren't in yet, or DC, which is once again such an outlier that it throws everything off (95% of the vote??). Here's the graph, which pretty much tells the story:

The projections were more or less spot on. There's a slight upward tilt to the actual results, meaning that Obama did better than he polled in the states he won, and worse than he polled in the states he lost. On average, for all states, he underperfomed his polling by about 0.8%.

But we're not necessarily that interested in whether he won New York by 25 or 26 points, or whether he lost Wyoming by 33 or 34 points. In the states where the last polls had Obama within 10% of McCain, in either direction, Obama actually outperformed his polling by about 0.5%.

Of course, this doesn't mean that some people didn't lie to pollsters because of being embarrassed to admit they were voting against Obama. It means that, overall, the polls did a great job at estimating outcomes, and there's no evidence at all of a massive divergence between polled results and actual outcomes. Chalk one up for the guys in green eyeshades, and a bit strike against the theory of hidden racism in American elections.

Comments

[Tony Facade]

Outliers are people too. Eliminating an outlier because it skews the story line is the reason why folks get nervous around stats.