« Dan Rather on Letterman blackmailer | Main | Washington Post soft on Orly Taitz »

October 05, 2009


Without the background of a statistician, I guess I'm confused about how the second graph isn't evidence that the speech did change the game.

Prior to it, opposition to the health care plan was skyrocketing and approval was plummeting...granted, it didn't completely reverse the earlier trajectories...it did stop the decline, right? Why isn't that a game changer (or to keep up the sports analogies, a layout d by a soccer goalie or an over-the-fence home-run stopping catch)?

Proponents of Democratic health care proposals may find solace in the idea that previously approval was "plummeting" but the President's speech did "stop the decline." That's rather cold comfort, given that the proposals are still approved by a minority of respondents. It also seems to suggest that, absent the President's speech, approval would have continued to decline at a steady rate--until, presumably, the approval rate reached zero. That kind of extrapolation makes no sense at all. (Surely there's some level of hardcore support for Obama's plan, if only among the 11% of Democrats who think he may be the anti-Christ but support him anyway.)

The comments by Rob and Brandon show why political science is barely a science. Political science does use numbers and graphs and statistical formulas, but controlled experiments are generally not possible. There's no way to know what would have happened if Obama hadn't given the speach.

Also, other events were taking place at the same time as the speech. There's no way to separate the effect of Obama's speech from those other events.

Still, the kind of dispassionate, objective analysis provided by Brendan is superior to some reporter using the POOMA method.

The comments to this entry are closed.