Say Anything Blog

Statistically, Republicans are less likely to respond to a poll.

 But a responsible pollster would keep asking till the numbers are within 3% of eachother. 

I wonder how many of the repsondents work at a Midland Kinkos…

I agree the numbers seem way out of line with other polls, especially those like Rasmussen that use likely voters rather than all registered voters.  The numbers I’m seeing are Rasmussen 43%, Time 40%, Hotline 45%, Cook 40%, CBS 34%.

That said, the IEA port deal probably hurt Bush short term.  The question is whether there will be another boomerang effect in the polls when people start recognizing that the liberals have once again way overplayed their hand.  If Bush didn’t suck so badly at PR, he’d be eating their lunch already…

The fact that the Democrats refuse to get their act together works in his favour.

What is the breakdown of registered voters?

Mike, I linked to the poll’s methodology in the post.  Maybe my eyes are tired, but I didn’t see a breakdown for registered voters.

Carrick, you name all those other polls...but which one do you think we’ll be hearing about the most?

The war against illegal plunder has been fought since the beginning of the world. But how is… legal plunder to be identified? Quite simply. See if the law takes from some persons what belongs to them, and gives it to other persons to whom it does not belong. See if the law benefits one citizen at the expense of another by doing what the citizen himself cannot do without committing a crime. Then abolish this law without delay … If such a law is not abolished immediately it will spread, multiply and develop into a system.

Frédéric Bastiat, The Law

Rob’s recently listened-to songs:

If its just registered voters its bogus anyways.  It needs to be Likely Voters.

I bet if I can poll the Dems up to 50%, I can get the final tally to be below 30%…

Think of the carnage if we polled the Dems at 75%!!!

I’m surprised that CBS would allow the polling numbers to be so skewed.  Biases aside, they ‘are’ a professional news organization and I am guessing here, have a few people that have taken Statistical Analysis 101.  I would like to know which direction the Independants lean, if they lean at all.  That ‘might’ make the numbers seem slightly more legit.

michael w: CBS has very little credibility, and has been shown to be biased.  This is no surprise.

The last two sentences were NOT sarcastic…

I wish I was that naive. 

Sorry.  I am unfamiliar with you, and don’t know your orientation.  There are lots of people who think CBS is a reliable, professional news organization.  Glad you know better.

michael w: To address your question, my experience is that most people who call themselves "Independent" are Dems who don’t want to admit it, or Republicans with doubts.  As a conservative, I consider both groups to be effectively Dems.  To my mind, then, I think the poll was extremely biased.  My observation is that the Prez really has over 50% approval, if an accurate poll could be conducted.  I mean equal numbers of Republicans and Democrats, few "Independents" and equal polling of red and blue areas of the country.  Also, equal numbers of men and women asking the questions, polling at all times of the day, so working people are fairly represented; you get the idea.

Bet that George Bush won’t win the next Presidential election with numbers like those.

Unfortunately for the democrats, this is not translating to increased support for them.

Look at Hillary’s numbers, her support is dropping faster than Paris Hilton’s drawers on Saturday night.

And as I keep telling people, Bush ain’t running for reelection, and he is not beholden to any political party. Republicans have stabbed him in the back often enough that I doubt he is going to expend much effort to keep them in office. All polls are lies. People need to wake up and smell the coffee. Polling agencies get paid for the results those who hire them want. Don’t give your employer what they want, they don’t hire you again.

my experience is that most people who call themselves "Independent" are Dems who don’t want to admit it, or Republicans with doubts.  As a conservative, I consider both groups to be effectively Dems.

Robert, of course that’s what you believe.  You have this seething desire to be in the minority - to be the oppressed "fight against all odds" western cowboy hero.

Now, do you have any facts to back up your assertion that Independents are Dems or is this just more wishful thinking? 

And how, rick, does this address the discussion about polls?

Frankly, 2H9;  I think there is a point to be made about Independents and that is that it is probable that the majority of Independents vote against the incumbent, or have views in opposition of the encumbent.  This would mean that right now Independents are unfavorable toward Bush (but not necessarily favorable toward the Democrats as the polls also show).  When Perot ran in ‘92, he pulled more votes from the incumbent than from Clinton even though his support came mostly from self-described independents.

That would contribute more to the analysis than this ambiguous, "everyone’s against me" rhetoric. 

Sorry, I meant to continue by saying that Nader pulled more from Gore, however, by those self-same Independents. 

So, I conclude by saying that incumbents (as Gore was perceived as VP) are hurt more by Independents regardless party affiliation. 

Margin of error: +/- 20%

Independants = Dems, or not.  Whatever!

This is what matters…

Bet that George Bush won’t win the next Presidential election with numbers like those.

Polls are what pay the bills for the Media and the NewsJunkies.  Votes are what count.  Period. 

CV Rick:

 I think there is a point to be made about Independents and that is that it is probable that the majority of Independents vote against the incumbent, or have views in opposition of the encumbent.

Balderdash.  Based on this logic, nobody would ever be reelected.

Since there are roughly equal numbers of Democrats and Republicans (at least among actual voters), without a strong (at least 50%) showing among independents, Bush would not be president right now.

That said, I think the IAE story has left a mark on Bush.  Since any reasonable person would conclude that it is mostly a nonstory, I’d have to put much of the blame for this on Bush and his inept PR staff, in their failure for not anticipating the storm and getting the facts out more quickly.  Bush needs to do the same thing that Clinton did and hire a heavy hitter to run his PR staff.

The bias & contrarian nature of the press is a fact that any president must reconcile themselves to.  This bias is something that must be managed, not simply bitched about.

Carrick,

 We may find, in twenty years, that Bush does (did) not, in fact, have to cow to the Press.  We may find, in hindsight (since none of us have the final set of particulars), that he does have a plan and it is working swimmingly.  (Not saying that’s the case, although it is my personal opinion.)

A good leader will strike out on a bold new direction and not be deterred when the masses don’t immediately follow him.  Check history, the leaders who did what the general populace wanted did not make any significant mark on history.  Only the ones who took that tough, bold step.  (Of course, sometimes those bold steps place the leader in the ditch, once all the cookies have crumbled.  Time will tell....)

The results are crap for more than one reason.  Even the weighted populations are way out of whack with reality.  Additionally, many of the questions can be answered negatively, but for differently, but for vastly different reasons.  For example, there is a question asking if the respondent approved or disapproved of the President’s handling of the economy.  Now, Democrats would answer through the nanny-state prism of taxing the rich and giving to the poor, thus they would be dissatisfied.  Republicans, however, might say no, but not for the same reasons.  For example, Republicans might say no because they think the President should be pressing to make any tax cuts larger or permanent.  Anther might be dissatisfied with the growth of government during a Republican administration and its impact on the economy.  Still others might object to an ever-increasing number of tax dollars going to pork barrel projects.

All of these are valid things to be dissatisfied about, but they create a misleading picture as they are all not of equal severity.

Rick: As I said: "In my experience..."  Independents are, in fact, liars.  When they go into the voting booth, they generally decide one way or the other, and that is what needs to be polled.  Asking them almost a year before the vote takes place is like asking them what kind of ice cream they are going to like on Election Day.  If there is a strong third party candidate, though, all bets are off.  We won’t know that until just before the election.

Here’s one analysis of the poll:

Today, we’re going to look at two of the techniques that are often used by the mainstream media to slant polls against Republicans, both of which are on display in the latest CBS poll.

Let’s start with the sexy headline and the first paragraph of the CBS piece on the poll:

Poll: Bush Ratings At All-Time Low

The latest CBS News poll finds President Bush’s approval rating has fallen to an all-time low of 34 percent, while pessimism about the Iraq war has risen to a new high.

Wow, sounds terrible doesn’t it? But, let’s take a look at a couple of key factors.

The first thing you have to understand is that there are 3 different groups of voters the media may poll: adults, registered voters, and likely voters. Out of these 3 categories, "likely voters" is the group that almost always turns out to be closest to the actual election results while "adults" is the group that slants most heavily towards Democrats. Although it’s difficult to pinpoint exactly how much polling "adults" instead of "likely voters" slants the poll results to the left (when compared to election results), it’s probably somewhere between 5-10 points. So, let’s split the difference and say 7.5 points.

So, it seems likely that Bush’s approval would probably be somewhere around 41.5% if this had been a poll of likely voters. Still, pretty bad.

But, there’s another factor we haven’t adjusted for: the percentage of Republicans, Democrats, and Independents who participated in the poll. In the 2004 election, the breakdown by party was as follows:

Democrats: 37%
Republicans: 37%
Independents: 26%

While those numbers can change and do change over time and there’s no set rule that says for a poll to be fair those percentages should match up exactly with the breakdown from the last election, the numbers should be pretty close.

So, let’s look at the weighted party breakdown from the CBS poll: 1018

Democrats: 37.4% (381)
Republicans: 28.4% (289)
Independents: 34.2% (348)

So, they undersampled the number of Republicans by more than 8.5% and over sampled Independents by more than 8%. Let’s adjust for that (in a very general way). Add in 8 more Republicans and we’ll say Bush’s favorability goes up 8 points. Take out 8 Independents and we’ll figure Bush loses 4 points of support (Independents were roughly split between Bush and Kerry in 2004) and now Bush’s approval rating, after having 4 points added onto it, is at 45.5. Of course, it’s not quite as simple as I’ve made it look here, nor is this as accurate as simply polling likely voters with a correct breakdown of party affiliation, but it’s close enough for our purposes.

Then, we consider the polls margin of error, 3 points, and Bush’s real approval rating among voters who’ll actually be going to the polls in November is probably somewhere roughly between 42.5 - 48.5. That’s not great, but it doesn’t have exactly the same sort of zing that 34% has either, does it?

Of course, you could make the argument that the CBS poll is just a poll of adults and it’s not meant to give people an accurate picture of how people will vote. But, if it’s just a garbage poll that doesn’t have any bearing on election results, why bother doing it in the first place? For political purposes, any poll that doesn’t use likely voters and doesn’t have a breakdown of party affiliation that’s at least roughly similar to the numbers from the last election isn’t very important or useful.

The data I have read indicates that only a percentage point or two separates those voters who identify themselves as Republicans and Democrats so it’s reasonable to say that in the case of this CBS poll, the Republicans have indeed been underrepresented. Whether this shows political bias is debatable but it does suggest that one should look to other polls which apparantly use different methodologies...a basket of polls should provide a clearer picture than any single poll.

Mike,

I agree with you.  We should be using different polls to evaluate all this...  They are called ELECTIONS.   :b 

MikeA: There are a lot more problems with polls than skewed sampling.  First of all, it’s self-reporting.  Are all the Republicans really Republicans, for instance?  We don’t know.  Is there equal sampling between red and blue areas?  Most cities are blue areas, and I am willing to bet that most polling is done in cities.  The worst distortion is the frequency of the polls.  What we are really doing is polling how people react to the last series of polls.  We are addicted to polls, IMO.  The only poll that counts is the one conducted in voting booths.

I agree that elections are what count and I think that too many polls ask more questions than they answer.