Monday, May 01, 2006

Unhealthy Obsessions

I have a couple. The one I feel like discussing today is, well, Today. I am fascinated yet repulsed by the morning schlock-fest that is NBC's morning "news" show. Why NBC? 'Cause that's the channel that comes in most clearly for me. I suppose this could just as easily be about half-wit Charlie Gibson and the revoltingly unctuous Diane Sawyer, or about Harry Smith and the other four empty suits on CBS, but it's about NBC's Today Show.

Last week, I almost pissed myself when Katie Couric put on her reading glasses (so you know she's serious) and talked to Tim Russert about President Bush's falling approval ratings. You see, according to an NBC News/Wall Street Journal poll of 1109 adults, the President's job approval rating was 36%, down one percentage point from the previous month. They give a +/- 3.1% margin of error; I calculated that this MoE means that they're using a 97% confidence interval. If you don't know what confidence intervals are, just know this: We have absolutely no idea what the actual value (of President's job approval) for the population of all Americans is. However, thanks to the magic of the Central Limit Theorem, we know that all possible sample proportions are approximately Normally distributed around the true population proportion, with an estimated standard deviation of 0.0144; therefore, the probability of getting a sample proportion that radically deviates from the true population proportion is pretty small. For those of you who know better: This description is only a rough sketch of how these things actually work. Please do not send nasty e-mails or post snotty comments about how Pierre-Simon LaPlace needs to come back to life and kick my ass. If you want a more detailed and accurate explanation, attend my class some day.

ANYWAY, what I'm saying is that, if done properly, this poll is a pretty good estimate of the actual population proportion for the President's job approval. And it's a really, really crappy number for Bush. Well, Katie Couric, in an astonishing display of ignorance or stupidity or god only knows what, actually had the following exchange with Tim Russert:
COURIC: Let's start with the poll numbers, Tim. This has been -- no doubt about it -- a rough patch for the administration. His -- President Bush's approval rating is down just one point. Do you think in a strange way, the White House is breathing a sigh of relief?

RUSSERT: No. They understand, Katie, that they have to start rebounding, and rebounding quickly. When you look inside these numbers, Katie, the mood of the country is so unsettled. Two out of three Americans say we're simply on the wrong track. The president has to be -- address that, starting with the high prices of gasoline. And that's why you saw his rhetoric change rather dramatically this week.

What the hell? "Down just one point"? One percent of the adult population represents 2.1 million people, dumbass. That's another two million people who are, for whatever reason, dissatisfied with the President's job (again; this is just an approximation, so those of you who are statistically inclined can just keep your electronic traps shut). I'll let you think about how fucking stupid Katie Couric sounded for just a minute.

Done thinking? Good. Today's Today atrocity involved Stephen Colbert. In that they didn't even mention him. Once. They mentioned the Correspondents' Association Dinner. They mentioned Bush's dumbass dual-Bush bit. They showed the powerful and their sycophants yukking it up. But they didn't once mention Stephen Colbert. Not once. He must have really made some press types unhappy. Are they, like the Boy King, whiny-ass titty babies? Yes. Yes, they are. Peter Daou at the Huffington Post has noted that Colbert's routine is being scrubbed from the Official Record of the evening.

Down the memory hole!

Billmon, as usual, has more and better stuff than I could ever come up with. Although, if he keeps going to see dreck like American Dreamz, he might lose enough brain function so that he and I sound like equals.

Labels: , , , ,