Liberal media bias? The hell you say.Follow

Six of one, half a dozen of the other.

Your job must suck. I'm not entirely sure what the best part of my job is. The ability to literally command people to do whatever I want, the pure security in knowing I can't possibly be fired, all the movies and video games I play while getting paid, the fact you pay my paycheck and can't do anything about it, or the fact that you'd suck my **** with a bunch of false patriotism.

Sorry, I can't hear your generic internet cliches over my awesomeness. You should go watch Fox News and come back when you've got something you think is worth my time to directly respond to.

Joph's Law, represent! Thanks!


So you don't understand the numbers. That's fine. It's pretty obvious that you need this.

If it were fabricated outrage in a liberal opinion article you'd be outraged about it though, wouldn't you?

No. I generally don't read opinion columns, I stick just to news.

Heh.


Why would we?

Yes, years worth of very consistent skew, lol.

Do you ever listen to yourself sometimes? It's pathetic.

Er? We don't have to. The more likely assumption would be that a majority of those in the middle group are going to lean in favor of Obama. You do get that the "even if we assume an even split" is me giving your position the greatest benefit of the doubt possible, right?


Start with the facts:

23% Liberal
36% Moderate
38% Conservative

46% Democrat
24% dont know/don't lean
29% Republican

Now let's apply a bit of logic:

Everyone who identifies as a Republican is going to also identify as a Conservative (because that label isn't a negative to Republicans). So Republican is a subset of Conservative. On the other side, that's not true though. Not all Democrats identify as Liberal. If we take the absolute best case possible and assume that all 23% who make up Liberals are Democrats, and then the first 23% of Moderates round out the whole Democrat portion of the poll, that still leaves us with 46% Democrat and 38% Conservative, with about another 16% of Moderates (plus folks who didn't say) in the middle. Even if every single one of those people leaned against Obama (extremely unlikely), you'd still end out with a 46/54 split (about even).

Here's the problem though. That's a "best case". We can reasonably assume that not all those who identify themselves as Liberal also identify themselves as Democrats, right? Thus, every percentage point of people in that group pushes the "Democrat" portion of the polling group farther into the Moderates (since that block still takes up 46% of the whole group and has to come from somewhere). Meaning that even more of that group is going to lean towards Obama. And that further assumes that no one inside the Moderate group *also* tends to lean in favor of Obama but doesn't identify as Democrats either.

It's just that you have to have only 4% of the total respondents being non-Democrats who favor Obama for the poll to come out "even". That seems incredibly unlikely. And I'm not even addressing the issue of people who identify themselves as Democrats *and* also as Conservatives. Obviously, we can't nail down the degree of skew here, but you kinda have to be engaging in massive wishful thinking (or willful ignorance) to think that it's not there and its not significant.

Aren't you needed on set for The Hobbit?

It's not really worth coming up with a new one, plus classics are always good.

Edited, May 17th 2011 3:00pm by Xsarus

No. The poll isn't wrong. Your interpretation of the poll is wrong. You don't actually think that polls tell us things like whether or not the media is biased do you? Because that's a bit strange even for you.



I'm not looking at the other responses at all. The only question I'm interested in is whether the sample used in the poll might be skewed in favor of Obama. And based on the makeup, the odds are very very high that they are. The only way they come out even is if we assume that nearly every single person in the sample who might favor Obama also identified themselves as a Democrat. Put another way, it assumes that not a single person who identifies as a Republican, or a Conservative, or a Moderate or "don't know" leans in favor of Obama. Again, that's incredibly unlikely.



Ah... I'm "confused" because I can recognize that a poll with 46% of respondents self-identifying as Democrats might just be a bit skewed towards the leader of the Democratic party. Um... Really?

Well you certainly were unable to detect that for the last 2 years.

Don't worry, somebody will generate something new to be outraged about by Friday. You won't need to keep backpedaling on this topic once the new rage of the week gets released.



Edited, May 18th 2011 8:24pm by bsphil