It seems that Chris’ displeasure with reading is shared by many Bush supporters. To wit: a new study that finds Bush Supporters Misread Many of His Foreign Policy Positions.
Excerpt:
As the nation prepares to watch the presidential candidates debate foreign policy issues, a new PIPA-Knowledge Networks poll finds that Americans who plan to vote for President Bush have many incorrect assumptions about his foreign policy positions. Kerry supporters, on the other hand, are largely accurate in their assessments.
[Please keep any comments related to the above post limited to original analysis of the study cited above. Or alternatively, comments that attempt to defend the position that an informed electorate is not important in a democracy will also be entertained. Any tangential, generic political punditry will be moderated out. It’s good to be the king ;-)]
Add Jim to your del.icio.us network
I read the piece..It goes on to state that “The poll was conducted with a nationwide sample of 959 respondents over September 8-12. The margin of error was plus or minus 3.2-4.0%, depending on whether the question was administered to two-thirds or the entire sample”. I went onto to read that the group that conducts their polls, Knowledge networks, says that “Our surveys are conducted with probability samples of persons who are members of the web-enabled panel, which is designed to be representative of the U.S. population”.
The highest concentration of Democrats is in urban areas. Within all of those urban areas there is a high population of poor, uneducated people (I think it is very safe to say that 80-90 percent are dems) who do not have computers. Therefore a VERY large percentage of the democratic constituency was not included in this poll, thereby compromising it’s results. Everyone says that all of the Republicans are Rich, therefore polls that attept to accurately represent the views of both parties, should include both rich and poor dems, not just the rich ones with computers.
If I get what JJ’s saying, Bush would have shown better numbers in the poll if poor, rural republicans were more represented in the study. Something tells me that poor, rural republicans aren’t rushing home at 5pm to catch the News Hour with Jim Lehrer so they can bone up on their foriegn policy. Perhaps I’m missing something, I’ll go back and look at their sampling and maybe I can get Bob W., a friend of mine who knows a thing or two about sampling to comment here.
Democrats (you included) state that most republicans are rich . I also bo not think that there are anywhere near as many poor rural republicans as there are poor urban democrats. Unfortunately I do not have concrete stats to back that up, but I’m pretty sure that is common sense.
I took a couple minutes to look over the methodology (I had the distinct misfortune of earning a living by survey sampling statistician a few years ago) for the PIPA/Knowledge Networks survey and I don’t think that sampling bias is likely to be the cause of these differences. If Knowledge Networks only sampled persons with computers or internet access this would be a big problem for the obvious reason that people who own computers are pretty different from those who don’t (i.e. fewer dates in high school, less sports trophies, more Star Wars conventions, more disposable income, insert outdated stereotype here.). However, this is not what they do. Instead, they draw a random sample of all persons with a telephone in their residence (about 95% of the population) and provide internet access to the people who agree to participate so that they can complete their surveys. The demographics look pretty close to what you might get from a telephone survey. I doubt therefore that the sample of respondents is responsible for the differences.
I think that a more important source of bias might be the questionnaire design. Most of the knowledge items probed in this survey are things that Senator Kerry approves of (Kyoto, international labor standards, land mine regulation, nuclear test ban treaty, etc.) and President Bush does not. It is well known in the survey research world that respondents tend to agree/support more than disagree/reject during a survey interview. We actually have a name for this, we call it “acquiescence bias.” This source of bias tends to be greater when the knowledge or attitude being probed is something that the person doesn’t know much about. It is my hunch that that some of the reason that we see these stark differences is that a non-trivial proportion of both the Bush and Kerry supporters are merely acquiescing. In the case of Kerry supporters, on most of these questions, acquiescing makes you correct. And, in the case of Bush supporters, this makes you wrong.
Of course, this doesn’t explain why Kerry supporters appear to know the president’s positions better than his supporters do. I would suspect that this has something to do with issue salience. I am guessing that Kyoto, land mines, and international labor standards are more much important issues for Democrats than Republicans. If this is true, it might explain why they know more. Of course, it might just be that they are better informed, although I seriously doubt that the members of one party are much better informed than the members of the other party.
Very interesting.Thanks Bob