I keep seeing Rasmussen polls cited in the usual Republican outlets. Maybe their selection bias really does represent the people likely to show up in next November's mid-terms better than other polls. But the discrepancy between their findings and everyone else's seems to be widening, as the GOP keeps up its campaign to bring down this presidency. Here's Rasmussen's Obama approval chart in the last half of 2009:

Now, here's the same graph, as reported by all the other polling outfits included in Pollster's poll of polls:

See what I mean?

We want to hear what you think. Submit a letter to the editor or write to letters@theatlantic.com.