The Latest Stupid Poll From The New York Times Analyzed
Posted by bulldogpundit
Ankle Biting Pundits
Thursday, 15 September 2005
As we mentioned previously, the latest New York Times/CBS Poll that shows the President's low approval ratings should automatically be dismissed as a joke because it has 10% more Democrat respondents than it does GOP respondents (36-26%).
But it also appears in addition to the other problems we'll outline below, the poll purposely oversampled blacks, a group that overwhelmingly voted against President Bush (89-11%) in 2004.
Click READ MORE to see even more problems with this poll that pretty much makes it a joke in terms of gaining any insight as to any political consequences.
First of all, this poll is so unusual that the NYT had to print a very unusual explanation as to how the poll was conducted (at least, to their credit, they did that).
Quote:
<<<
Black Americans were sampled at a higher rate than normal to permit the analysis of black attitudes in greater depth. From all national polls completed by The Times and CBS News since mid-March, every household in which a black respondent had been interviewed was called again for this poll.
Of the total in the new poll, 877 said they were white and 211 said they were black. There were too few respondents of other races to show their results separately.
Blacks and nonblacks were then weighted to their proper proportion of the overall population.
>>>
Yes, the poll says it did weight for blacks as a proportion of the overall population explaining why blacks made up about 19% of the total respondents, but in the data tables shows that only 12% of the respondents were black. And perhaps I'm missing something, but I'm not sure which of the question responses are weighted, and which aren't. Some questions are broken out by race, and some aren't. It's possible that when a question isn't broken out by race the result reflects the weighted average but that's something I have to assume.
Be that as it may, however, to use this poll as a predictor of any political consequences would be useless. Why? Let us count the ways.
Ok, you need to take a look at the raw numbers.
washingtonpost.com
Good, Now we'll tell you what's wrong with it.
1. Party Leanings - The respondents tend to "think of themselves" as follows: 36% Democrat; 26% Republican, and 28% Independent and 9% "Other.. The 2004 exit poll showed R's and D's split at about 37%. So, right away, you can almost discount it when it comes to drawing conclusions about election day. And why such a high percentage of "Other" (9%)? Are they talking about the Green party loons?
2. Sample Group and Timing Of Poll - First of all, the NYT polls only "adults," not "registered" or even "likely" voters. They do break it down and say that 86% of respondents are "registered", but of that amount, 22% either didn't vote or don't know if they did. Thus, of the respondents only 735 of the 1,107 respondents even voted.
Additionally, in this poll, respondents voted 37% each for Bush and Kerry, even though President Bush won the 2004 election by 3%.
Next, 3 out of 5 of the polling nights are considered weekend nights, and weekend polling is notoriously unreliable and favorable to Democrats.
3. Age of Respondents - The poll also over samples the number of 18-29 year old voters, the age group that voted most for Kerry. In 2004, 17% of the electorate was between 18-29, and Kerry's advantage among them was +9%. In this poll that age group was 22% of the sample.
4. Income Level of Respondents - Next take into consideration the annual income of the Times's poll's respondents. In 2004, 45% of the electorate was making under $50K, and voted for Kerry 55-44%. But in the Post's poll, 48% of respondents make under $50k. However, 7% of the respondents refused to say what they earned, so that figure could go even higher..
5. Religion - Next, let's look at the religion of poll respondents. In this poll 14% of respondents had "no" religion, while in 2004, only 10% of voters had "no" religion, and they voted overwhelmingly for Kerry (+36%). Catholics are also underrepresented by 8% in this poll, another group that went for Bush in 2004.
Now let's look at some of the idiotic and misleading questions asked in this poll.
Quote:
<<<
56. If you had to choose, which is more important to you right now -- changing the Social Security system or rebuilding New Orleans?
57. If you had to choose, which is more important to you right now -- cutting taxes or rebuilding New Orleans?
>>>
That's a loaded question and a false choice. Evidence shows that the Bush tax cuts increased tax revenue to the government which would better allow New Orleans to rebuild. Same with Social Security. If it's reformed, then there would be less spent on entitlements and more spent on New Orleans. But what do you expect when liberals write the questions.
anklebitingpundits.com
anklebitingpundits.com
nytimes.com
nytimes.com
cnn.com
captainsquartersblog.com
cnn.com |
