Pollster.com

Wow. Uh....what can one say? I'm not one to dismiss particular polling organizations out of hand, but, well....wow.

I can't wait to see this one on the pollster plots...I"ll bet I'll be able to spot the ARG poll! It's like a Where's Waldo game!

Huh?

Okay, I know ARG isn't particularly great when it comes to polling, but wow!

As much as I would like to believe Bush's approval rate fell below the mendoza line, I think instead that ARG has eschewed actually polling in favor of giving a blue monkey with a pink hat and tassled cymbals responsibility for throwing trash at a dart board

I've created a polling company where our methedology will involve guessing, in effect 'self-polling'. It should complete nicely with other polling organizations.

Yeah, I'm not a Bush fan, but I'm extremely skeptical of this poll. Aren't these the folks who had Clinton up 6 points in Wisconsin?

Hey Pollster, time to re-align that Y axis!

Did anyone notice that the interview was 29% GOP and 43% Democratic. Why would they even bother publishing such a lopsided survey?

How 34% Becomes 19%

I'd like to add to the previous comments about how bad ARG is. Here they repeat an error that they have made on and off for years, an an area that is basically polling 101. Over the years, I have used ARG polls in this regard as a text-book example of how not to conduct a presidential approval poll.

So what is it that they do wrong? Scroll down to the bottom of their page and you'll see the actual wording of the questions, and the order in which they ask them. Notice that they have a whole battery of questions about the economy first, and then, right at the end of the survey they ask about Bush's job approval.

This is the exact opposite of what a good public opinion poll does. You ask about the core "dependent variable" first (Bush approval) and THEN you ask about the independent variables (views of the economy).

When it's done right, with the approval question first, you find out what people's views are of Bush. When it's done the way ARG does it, with approval last, you find out what people think about Bush after they've just been asked a whole bunch of questions about the economy.

So, OK, from pollsters who actually have some standards and have at least a bit of a clue about what the hell they're doing (i.e., almost everyone other than ARG) we know that Bush' actual job approval is in the low-to-mid 30% range. From ARG we know that if we get people thinking about the economy, it goes even lower.

That's actually useful information, from a message-testing perspective. Democrats need to keep talking about the economy.

Cheers!

-- Joel

joe way to make a diamond out of a piece of coal... the economy huh... if only the democrats would only focus on taking down the Repungican at 1600 pennsylvania with issues like the economy, instead of who has more pseudo experience or who's more proud of their country..

ARG continues to produce howlers. Much as I'd love to see Bush's approval down around where interest rates are, this is artifactual

Well, of COURSE it must be laughable. It's too pessimistic for the Bushbots. I noticed that when the government-controlled NPR had a poll showing Bush with nearly 40% approval, there weren't any derisive howls here.

Pollster, who do you think you're fooling?