Thursday, May 24, 2007

Rudy #1 With Gun Owners?

Rudy Giuliani, a supporter of gun control, still comes out on top with gun owners, a new survey has found.




The above ought to be all a gun owner concerned about his rights needs to see.

The Daily News and the Gallup organization are hardly reliable and unbiased source for gun rights-related information.

Poll results and methodology are here. Of the "2,013 adults, aged 18 and older" surveyed, the sample contained "670 gun owners." Left unsaid is whether these were registered voters, if so, whether they were high propensity voters, whether the survey participants were a representative cross section, what questions were asked, and if the gun owners were highly motivated, indicative of a level of commitment. Also, gun ownership crosses party lines, and the report admits "[i]n the Democratic primaries, gun owners seem just as inclined as non-owners to support Clinton."

I hardly hold myself out to be a polling expert, but this appears to be a case "of sound and fury signifying nothing." If anyone with credentials would like to weigh in and either confirm or challenge my assumptions, or if you just want to say your piece about this, well, that's why there's a "Comments" section below.

[More on Rudy from WarOnGuns]

1 comment:

Anonymous said...

They're still attempting to repeat the fiction that a majority of gun owners want their civil rights violated. I half-believed it for a while, until I started talking and listening to the locals in surrounding gun shops. Most are way to the right of the NRA's positions. They'd be voting for Giuliani (D-NY) as a last resort.

The Yellow-stream Media is assuming that what the country needs is another "bridge" candidate. They did their best to undermine the right, including altering or falsifying information with such regularity that it finally hurt their business. They create discord, then accuse the other side of causing it.

It time we take to task the statisticians who do these polls. If the methods go unchallenged, other people will assume this is acceptable practice. It's not.

Two-thousand is a workable but small sample size. I have no information about the distribution of the sample. We don't know the questions that were asked. How does one "randomly select" a sample from the U.S. population? I challenge this methodology. I even challenge the use of the word "random". They chose an age group, they had to chose a limited number of people, they had limited time to perform the poll, and we don't know what time of the day it was conducted. Even so, are they comfortable with the fact that their sample suggests that one-third of the population are "gun owners"?

Perhaps I'm naive: It's easy to cherry pick the respondents in a small, 2,000 person sample so that it reflects a known demographic, therefore lending plausibility to the results.

A "gun owner" may also describe a prison guard, a police officer, or a federal agent who is armed as a necessity of the job. The poll is also among adults, not registered voters that are likely to vote.


"In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls."

Okay, this disclaimer is a good thing. But let's take it a step further. Let's see the questions, the distribution (exact locations, please), and the method for choosing the distribution. (It better represent "random".) I'd also like to know how they define a gun owner.

In short, it's time for some peer review. Will their methods mismatch their characterization? Will we be able to approximate their results?