L.A. Times Poll (Is the Poll Skewed?)

self | August 25, 2003 | DiamondDon1

[DiamondDon1]

Welcome to the world of the Left Angeles Times…

On August 24, 2003 the Los Angeles Times released the results of a survey which has been touted by many as a huge boost for Cruz Bustamante’s chances to become the next governor.

Problem. The poll contacted 1351 registered voters but did not achieve the appropriate number of Latino voters. So the “main sample”, whatever that is supposed to mean was supplemented by an unknown number of Latinos in order to provide 125 additional Latino “likely voters”. Even by the Times own numbers, this little bit of skewing affects the numbers by 9 percent plus or minus.

Now when I took statistics, a “random” sample was just that “random”. When a poll is taken where “supplemental voters” and added to provide additional voters in one category that is a problem, the conclusions reached cannot be trusted.

When you read the poll below, take your time. Look at what they asked, and how they theoretically asked it (because such things as voice inflection can skew results).

We can look at the results differently than the Times and arrive at completely different numbers…such as:

Removing the Times supplementing process, we have 801 – 125 or 676 likely voters. If these 125 Latino “likely voters” were removed (51 percent according to the Times support Bustamante, and 12 percent support Arnold), the numbers change thus:

Bustamante = 280 – 64 = 216 likely voters or 32 percent

Arnold = 176 – 15 = 161 likely voters or 24 percent

Now while this in itself changes the numbers from a 13 percent lead for Cruz, to 8 percent other factors should also be considered.

Review the questions shown below. Notice anything? Notice a slanted bias? Note that the questions below occur before the candidate questions, and thereby influences the participant.

“In a recall election, a replacement candidate does not have to win by a majority vote. Because there are so many candidates on the replacement ballot, it is possible a new governor could be chosen with the support of less than 20 percent of the voters-even if, for example, Davis lost the recall election by 51 percent to 49 percent. Does knowing this could possibly happen make you more likely to vote in favor of recalling Gov. Davis, or more likely to vote against recalling Gov. Davis, or won’t it change the way you are likely to vote one way or the other?”

“As you may know, 135 candidates were certified to be on the ballot as candidates to replace Governor Davis if the recall passes. Does the fact that so many candidates will be on the ballot make you more likely to vote in favor of recalling Governor Davis, or won’t it change the way you are likely to vote one way or the other?”

Down a few questions…

“Some people say that the recall election is an attempt by Republicans to overturn an election they lost in November 2002. Do you agree or disagree?”

“Some people say that recall elections like this one interfere with an elected state official’s ability to fulfill his or her duties and efficiently run state government. Do you agree or disagree?”

“Does the fact that the special recall election will cost taxpayers at least 66 million dollars enter into your decision to vote yes or no to recall Governor Davis, or does that fact not enter your decision at all?”

Read the link below for yourself…there are many more questions similar to those above that appear skewed to make the recall sound like a right-wing power grab that is ethically, and morally wrong…

http://images.latimes.com/media/acrobat/2003-08/9130937.pdf

One only knows what the actual (unsupplemented) numbers regarding the recall “yes or no” are…

DD

[deport]

Looks like a push poll also......

[DiamondDon1]

Note how the L.A. Times Poll was conducted...and if we only knew their sampling techniques...

DD

[South40]

What is your source for all this? How about a link?

[LittleJoe]

You can't take any poll conducted by the LA Times seriously.

Look at their past polls, especially the ones from the Regan era.

[South40]

Nevermind.

[DiamondDon1]

In my entire life where statistics are used, I have never seen the sample population supplemented by a non-random group (Latinos).

This poll has about as much validity as throwing darts at rotating dartboard in a hurricane.

DD

[Redcloak]

Is the Poll Skewed?

Is it by the L.A. Times?

[My2Cents]

Thanks for the analysis. Looks like the LA Times' attempt to shape public opinion, not measure it.

[KE]

Brit Hume was on Fox News talking about this. He said that other Dem polls did not show Bust. ahead, and also that the Rep. are saying their polls show them just about even. He also said that other poll takers are scratching their heads about these numbers and don't agree with them.

Karen

[KE]

Brit coming up after this commercial "Is Bust. really ahead"? I will watch and post.

Karen

[South40]

Brit Hume was just discussing this on the air. The poll started on Aug. 16th right after dipwad warren buffet stated he'd like to see Prop 13 repealed.

The whole thing is skewed; it doesn't reflect how people feel today.

[FastCoyote]

I wrote a book on nuclear waste issues where the enviro opposition push-pull polled with questions like this made up by a commie academic named Paul Slovic. Yup, the above California poll is skewed.

[SwinneySwitch]

L.A. Times Poll (Is the Poll Skewed?)

Does a bear have hair?

[kellynla]

Of course the poll was skewed! And since the poll was taken before Bust-the-bank vowed to "RAISE YOUR TAXES BY 8 BILLION DOLLARS" and BEFORE Simon withdrew from the race... the poll is BOGUS! Take the poll today and people who pay taxes in CA will not be so anxious to vote for MORE TAXES, hence Busty's number falls not to mention the 135 Hispanics that were loaded in the sample, McClintock's number will be higher since more than likely Simon's supporters would vote for McClintock and finally Mr. Hide & Seek's number would at best stay the same or continue to drop as it has since he entered the race. The debate on September 7 will separate the REAL DEAL from the liberals of both parties! Aunold & Busty will feel like two one legged men in a butt kicking contest when they have to go up against McClintock in a debate! VOTE4MCCLINTOCK http://www.tommcclintock.com

[DiamondDon1]

If you haven't yet, read the rest of the questions on the link...there are four questions before the RECALL question (and the question is long enough to choke a horse) and about a dozen questions between the RECALL question and the CANDIDATES question...apparently these participants didn't have anything better to do with their lives...

DD

[kellynla]

http://www.latimes.com/news/politics/recall/la-me-poll24aug24,1,7382768.story?coll=la-home-headlines Here is your link.

[DiamondDon1]

Hey Kelly!

The question at the top of the thread was rhetorical in nature...it is quite clear what was going on here...

DD

[DiamondDon1]

Kelly:

The data is shown in the .pdf file below...

http://images.latimes.com/media/acrobat/2003-08/9130937.pdf

The article does not have the raw data, so you can see how the sampling was done...and it makes you wonder what the editor at the L.A. Slimes (oops...) Times was thinking...

DD

[Amerigomag]

Either you are politically naive or an modal Californian (who just immigrated from Mexico).

The LA Times does not conduct polls ... they publish editorials.