Obama '08?

Miller:

In my opinion and many others....it is way too soon to predict what will happen with the "approval" ratings for Hillary when the primaries really do begin. It has been my thoughts (and perhaps negative bias) that Hillary's positives have been over-inflated by inside Beltway pundits and elites.

0 Replies

Vietnamnurse, I agree. Hillary's image of today will not be the image of tomorrow, and Obama will eventually overtake Hillary as it gets closer to election day.

0 Replies

Miller wrote:

What was recently pointed out in a NYTimes article was that a very large % of Obama money was coming in from American Blacks whose average incomes were in the $30,000 range.

And that would signify... what, exactly?

0 Replies

Miller wrote:

cicerone imposter wrote:
Obama is in front of all the front-runners on raising money; beat Clinton by 50% - and money "talks." This trend does too!

If "money talks", why are 48% of registered Dems for Hillary Clinton and less than 30% for Obama?

Money may talk, but does money VOTE??

Two reasons -

First, Hillary has massive name rec, whereas Obama just doesn't.

Second, Edwards is taking up a significant chunk of the 'non-hillary' crowd. Once he drops out, Obama's numbers will go up overnight.

Cycloptichorn

0 Replies

If and when Edwards voices his support for Obama, it'll be a whole new ball game.

0 Replies

Many of Obama's base of supporters are cell-phone-only households. They do not have land lines. Pollsters do not call cell phones. Therefore all the polling results are skewed and do not reflect the current reality of modern American society.

The impact of cell phone only households on polls was just beginning to be realized in 2004. The 2008 results will help pollsters decide whether it is conclusive enough to begin including cell phone numbers in their calling efforts.

http://www.pollster.com/blogs/cell_phones_and_political_surv.php

0 Replies

Butrflynet, Good info; will help us to consider other things on future polls.

0 Replies

Butrflynet wrote:

Many of Obama's base of supporters are cell-phone-only households. They do not have land lines. Pollsters do not call cell phones. Therefore all the polling results are skewed and do not reflect the current reality of modern American society.

The impact of cell phone only households on polls was just beginning to be realized in 2004. The 2008 results will help pollsters decide whether it is conclusive enough to begin including cell phone numbers in their calling efforts.

http://www.pollster.com/blogs/cell_phones_and_political_surv.php

Cell phone base?

Most black American women are for Hillary. Does this suggest something about their use or nonuse of cell phones?

0 Replies

Miller wrote:

Butrflynet wrote:
Many of Obama's base of supporters are cell-phone-only households. They do not have land lines. Pollsters do not call cell phones. Therefore all the polling results are skewed and do not reflect the current reality of modern American society.

The impact of cell phone only households on polls was just beginning to be realized in 2004. The 2008 results will help pollsters decide whether it is conclusive enough to begin including cell phone numbers in their calling efforts.

http://www.pollster.com/blogs/cell_phones_and_political_surv.php

Cell phone base?

Most black American women are for Hillary. Does this suggest something about their use or nonuse of cell phones?

Most black women with home phone lines. Which of course equals older black women.

Did you just miss the point completely, or what?

Cycloptichorn

0 Replies

Quote:

Most black women with home phone lines. Which of course equals older black women.

Not remotely clear to me, as expressed in the above passage.

0 Replies

I'm sure the stats have changed dramatically since this survey was done, but it's interesting to see these older results - and how our perceptions could be wrong.

Detroit Ranks First in Cellular Phone Ownership, Scarborough Study Says
Business Wire, July 27, 1998 NEW YORK--(BUSINESS WIRE)--July 27, 1998--The cellular telephone capital of the U.S., based upon percentage of ownership, is Detroit, Michigan, according to a recent study conducted by Scarborough Research, the leading local market consumer ratings firm.

The study found that three of the top four cities for adults living in households owning a mobile phone were in the Southeast: Atlanta, Greensboro and Charlotte. The top four rankings are as follows: Detroit (48.0%), Greensboro (47.7%) Atlanta (47.4%) and Charlotte (45.6%).

Surprisingly, major cities such as New York and Los Angeles did not rank prominently in the survey, ranking in the second half. Out of the 60 major metros surveyed, Los Angeles ranked 38th with 33.7% of its residents living in households owning a mobile phone and New York placed 41st with only 32.8%.

Smaller metropolitan areas revealed the lowest percentages in cellular phone ownership. For example, Louisville (26.9%), Dayton (25.5%) Grand Rapids (25.4%), Buffalo (25.0%), Charleston, WV (25.0%) and Wilkes-Barre-Scranton (19.6%) placed last on the list.

This data is contained in The 1997 Scarborough Consumer, Media & Retail Report's Top 50+ Market Summary. Data is gathered from more than 163,000 interviews with adults in 60 of the country's largest markets. The reports are single-source studies that examine a variety of characteristics including demographics, socio-economic lifestyle, consumer habits and product usage as well as media behavior.

0 Replies

It is undeniable that older folks are more likely to have home phone lines then younger. I don't know a single person my age who has a landline.

Cycloptichorn

0 Replies

Miller wrote:

Quote:
Most black women with home phone lines. Which of course equals older black women.

Not remotely clear to me, as expressed in the above passage.

Miller, I must have missed something in the post I wrote. Could you tell me where I mentioned anything at all about Blacks, women, age or Hillary?

Actually, Cycloptichorn, you would be surprised by how many older Americans now have cell phones. They use them much like that "help I've fallen and can't get up" device. It gives them a sense of security and freedom of movement to know they can carry a cell phone around with them and call for help should they need it. Many are starting to give up their land lines because the cell phone companies offer a wider variety of cheaper flat-rate long distance calling plans for keeping in touch with friends and family.

0 Replies

Butrflynet wrote:

Miller wrote:
Quote:
Most black women with home phone lines. Which of course equals older black women.

Not remotely clear to me, as expressed in the above passage.

Miller, I must have missed something in the post I wrote. Could you tell me where I mentioned anything at all about Blacks, women, age or Hillary?

Actually, Cycloptichorn, you would be surprised by how many older Americans now have cell phones. They use them much like that "help I've fallen and can't get up" device. It gives them a sense of security and freedom of movement to know they can carry a cell phone around with them and call for help should they need it. Many are starting to give up their land lines because the cell phone companies offer a wider variety of cheaper flat-rate long distance calling plans for keeping in touch with friends and family.

Maybe. But it is undeniable that those who are older are far more likely to ever have had a home phone line in the first place, and less likely to rely ONLY on the cell phone.

Cycloptichorn

0 Replies

http://pewresearch.org/pubs/515/polling-cell-only-problem

0 Replies

Butrflynet wrote:

Many of Obama's base of supporters are cell-phone-only households. They do not have land lines. Pollsters do not call cell phones. Therefore all the polling results are skewed and do not reflect the current reality of modern American society. [..]

http://www.pollster.com/blogs/cell_phones_and_political_surv.php

Uhmm.. no offense, but did you actually read that article you just linked in?

Because it concludes exactly the opposite of what you assert in that sentence I made bold.

The article is about the most thorough of the analyses I've seen on the impact of the missing cellphone-only households. And it concludes that, since cellphone-only households turn out to have much the same political outlooks as their counterparts with landlines in the same age bracket - and since pollsters use a number ways to correct and weigh their samples so as not to get an undersample of individual age brackets - the effect of not reaching cellphone-only households is, for now, marginal.

In the article's own words:

Quote:

So what have we learned? For the moment, the Pew Center studies tell us that including samples of "cell-phone-only" Americans produces survey results that "are nearly identical to those from the landline sample alone."

The only thing the article concedes is that maybe, or maybe not, it might become more of an issue in the future.

For anyone interested in this subject, please read this very scrupulous analysis, before speculating any further about how the cellphone issue means polls cant be trusted anymore.

0 Replies

nimh wrote:

The article is about the most thorough of the analyses I've seen on the impact of the missing cellphone-only households. And it concludes that, since cellphone-only households turn out to have much the same political outlooks as their counterparts with landlines in the same age bracket - and since pollsters use a number ways to correct and weigh their samples so as not to get an undersample of individual age brackets - the effect of not reaching cellphone-only households is, for now, marginal.

Maybe. But then again, maybe not.

Determining if a poll is representative of the larger population by taking another poll is a form of question-begging. In essence, the second poll already assumes that polling is representative, which means that it assumes that which is to be proven. If one has no confidence in the accuracy of the first poll, there is little reason to repose more confidence in the second.

0 Replies

joefromchicago wrote:

Maybe. But then again, maybe not.

Determining if a poll is representative of the larger population by taking another poll is a form of question-begging.

That wasn't the question though. The question was whether including cellphone-only households would make the poll more representative of the general population. Comparing the two polls, and finding no difference after correcting for age, is good reason to think that the answer is "no". The argument doesn't beg the question actually asked.

0 Replies

Thomas wrote:

That wasn't the question though. The question was whether including cellphone-only households would make the poll more representative of the general population. Comparing the two polls, and finding no difference after correcting for age, is good reason to think that the answer is "no".

That's only true if one accepts that the second poll is representative of its population. But to test the representativeness of one poll by looking at another poll, the representativeness of which is merely assumed, is question-begging.

0 Replies

The compounding of the error rates need to be taken into consideration. Most survey results publish a margin of error based on the sample size of the survey. Large national polls usually publish a margin of error of +/- 3%. Smaller surveys typically have a 5% error band. These numbers indicate with 95% confidence that the result is within the margin of error. The 5% associated error rate is seldom discussed (they intend to be wrong 5% of the time). 99% confidence intervals are almost never calculated for political surveys because they require a much larger sample size to stay within the same margin or error. The cost of being wrong (within a given margin of error) is greater as the confidence of the estimate increases. Another way to look at survey results with a 3% margin of error is to say that the pollsters are 95% confident that the actual result is somewhere within 3% of a given number.

The exit poll study from 2004 indicated a 7% weight for cell-phone only users and up to a 2% difference in the weighted outcomes. These differences need to be added to the error already included in the study design. According to the graph, the gap being covered by the weighted estimate had grown to 11% by 2006 and is currently around 12%. As a given metric is determined by an ever increasing weighted estimate, the amount of error associated with the estimate increases. On-going surveys of the two groups are required to determine the delta in the bias. What was a 2% difference in 2004 could easily be a 3-4% difference today. As the gap between the two groups widens and the weighted outcomes are driven by a smaller and smaller proportion, the bias introduced by weighting alone could increase beyond the margin of error typically deemed allowable in survey statistics.

0 Replies

Related Topics

Quick Links

My Account

able2know