What Polls Really Tell Us
Friday, July 9th, 2010
A recent Rasmussen poll tells us that 60 percent of Americans want the new health care law repealed. However, Kaiser Family Foundation tracking poll taken about about the same time said that 48 percent of Americans had a favorable opinion of the reform law, while 41 percent had an unfavorable opinion, and 10 percent had no opinion.
Nate Silver, the statistician famous for his accurate predictions of elections, also keeps track of polling companies and analyzes their results for accuracy. He nearly always put Rasmussen at the top of the error list. Rasmussen’s poll results overwhelmingly tend to err in favor of conservative candidates and causes, Nate has found.
It is unlikely that Rasmussen is deliberately mis-reporting its findings. It is more likely the company crafts its surveys in a way to achieve the desired outcome. In polling, this is called a “house effect” — a company that works on behalf of a particular cause, party, or candidate will find ways of polling to get the best (for them) possible results. The way a question is worded, or other questions in the poll, can considerably alter results.
The health care reform law has a number of provisions that will be enormously helpful to people suffering from rare and deadly diseases such as mesothelioma cancer, once they are in effect. Beside making it easier for everyone to get affordable health insurance, for example, the reform law also provides for research into the effectiveness of medical treatments. As it is now, most physicians have to rely on what pharmaceutical and medical technology companies say about their products.
And the next question is, why would a polling company try to distort public opinion? Well, for one thing, it results in headlines that say 60 percent of Americans want health care reform to be repealed. The natural reaction of most people reading that headline is, wow, there must be something really wrong with health care reform.
In other words, the purpose of such polling is not to measure public opinion, but to manipulate public opinion. Sociologists and political scientists call this the “bandwagon effect.” This is a tendency for people to believe things because many other people believe those things. The bandwagon effect has been around for a long time, but publicists and politicians have gotten very good at using mass media to amplify the effect. Publishing poll numbers favorable to your cause is one way to create a bandwagon effect.
If you go further into Rasmussen’s press release you see this paragraph –
“Supporters of the health care plan have often looked north to Canada’s nationalized system as a model for what they have in mind, but just 32% of voters nationwide say Canada has a better health care system than the United States.”
This is piling propaganda on top of propaganda. The health care reform bill that became law last spring in no way resembles the health care system of Canada, so why drag Canada into this survey? Because Americans have been subjected to many years of scare stories about how awful health care is in Canada, and so many Americans are certain Canadians have a terrible health care system. Canadians disagree, but propaganda isn’t famous for being truthful.
And one of the propaganda techniques used by opponents of health care reform is to associate U.S. health care reform with the Canadian single payer health care system. Hence, a completely irrelevant mention of the Canadian health care system is added to the release.
So whenever you hear a news story about a poll that says “Americans think” this or that, take it with a grain of salt. Just because 80 percent of adults can be persuaded to say the sky is green doesn’t mean the sky is green. All it means is that somebody wants you to think the sky is green.
