Log in

No account? Create an account

Previous Entry | Next Entry

Nov. 19th, 2004


For 11% of voters, the most important topics were Domestic Issues like Social Security and Health Care. Ten percent (10%) named cultural issues such as same-sex marriage and abortion, while 4% cite fiscal issues such as taxes and government spending.

The number saying National Security issues and Cultural issues increased from a survey conducted two weeks before Election Day.

Exit polls conducted by the networks created quite a stir by reporting that 22% of voters said "Moral Values" were most important. Our survey framed the question differently and drew attention specifically to same-sex marriage and abortion.

In the week leading up to the Election, we interviewed 652 Likely Voters who named cultural issues as most important to them. They were Bush voters by a 73% to 26% margin.

Fifty-two percent (52%) of Cultural Issues Voters are under 40 and 61% are women. Those over 65 were least likely to name cultural issues as important.

Fifty-three percent (53%) of Cultural Issues Voters are Investors and 83% are White.

Overall, on Election Day, 56% of Bush voters named National Security Issues most important. Fourteen percent (14%) said Cultural Issues, 13% Economic Issues, 6% Domestic Issues, and 3% Fiscal Issues.

Kerry voters had a different focus--40% named Economic Issues most important, 24% National Security Issues, 15% Domestic Issues, 6% Cultural Issues, and 4% Fiscal Issues.





Nov. 22nd, 2004 01:38 pm (UTC)
where are the data?
that's fine, bxiie, but it's all analysis. i don't like seeing "we framed the question differently" without seeing precisely how the questions WERE framed. as presented, there's no way of knowing what folks were responding to--i.e., what counts as "national security" (and this is important in that many far-right Federal mouthpieces have gone on record citing "protection of marriage" as a national security issue) and what counts as "cultural issues"--and here, tempering "gay marriage" with "funding for the arts" or "public television" (=Sesame Street in many folks' minds, and they're in favor thereof) may have toned down the importance of the category as a whole to many respondents.

i'm not defending the CNN polls as ideal or precise. however, CNN does at least post every question they asked, just as they asked it. see http://www.cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.html. the problem here is that the questions WERE very vague, leaving the definition of national security and moral values up to the individual--so analysis is de facto speculative. as has been noted here and elsewhere, the decision to go to war entails moral values; conversely, see above re protection of marriage being a matter of national security.

rasmussen surveyed 1,000 people on election night. CNN surveyed 13,660. perhaps 1,000 is a statistically significant sample, but it's certainly not as robust as CNN's.

i'm a little mistrustful of a source that posts its high-level responses FROM ELECTION NIGHT under the question, "When thinking of how you will vote this November, which type of issues will be most important?"

more troubling, rasmussen shows the totals of a survey from election night and one from October 19, two weeks previous. each sample was 1,000 respondents. on October 19, 8 percent (or 80 persons) cited "cultural issues" as most important to them; yet "In the week leading up to the Election, we interviewed *652* Likely Voters who named cultural issues as most important to them." (emphasis added.) where did the extra 572 interviewees come from? i'm not saying they made them up, but they DON'T TELL US, and i find that troubling.

in fact, the whole site's very vague, beginning with the fact that there's nothing on the rasmussen homepage identifying the organization that publishes the site, telling you anything about contributors, etc. "Learn more about RR" (on individual report pages, not on the home page) takes us to a sales pitch for premium membership. there's certainly no easily-found information about how their sampling is done... "1,000 likely voters" from where, exactly? granted, CNN doesn't provide their exit poll selection methodology either.

from a different rasmussen report:

"Monday November 08, 2004--Fifty-five percent (55%) of voters say that the same sex marriage issue was somewhat or very important in their voting decision last Tuesday. This includes 34% who deemed the issue 'very important.'

"Among those who said the issue was very important, 84% support the traditional definition of marriage as a union between one man and one woman."

the calculation of the second percentage (34) is wholly ambiguous: is it 34 percent of the 55 percent, or 34 percent of all respondents? we're talking about either 13.2 percent or 28.6 percent of total respondents, respectively, who support the "one man and one woman" definition--a big difference. i can't swear to either one, but the opening of the second paragraph shows that the author (not attributed *here or anywhere on the rasmussen site*) understands the necessity of being clear about percents of percents. so it's either sloppy or deceitful, and i can't tell which.



Bjamexza Q. Pyndejo / James O. Payne, Jr.
Bxiie Q. Pyndejo

Latest Month

May 2013
Powered by LiveJournal.com
Designed by Keri Maijala