Our chapter in Logic and Contemporary reasoning talked a lot about how statistics can be used for good but can also easily be misconstrued. The book did not however talk about all the kinds of polls that are used today to manipulate the numbers into a party’s favor. I’d like to through a few of the ways polls are biased.
Surveys are one of the worst methods of polling. They are extremely inaccurate. A good example of this is in the 1936 Presidential Election between Alfred Landon and Franklin D. Roosevelt. The Literary Digest put a survey on the back of their magazine asking their readers who they were going to vote for in the election. They got over 2 million sent back and the results concluded that Landon was going to win by a landslide. Do you remember President Landon? I didn’t think so. The same year George Gallup did his first ever representative poll by asking 1,500 people who they were going to vote for. He predicted F.D.R. was going to win. Gallup was correct because his method of polling did not have a response bias. Think about the kinds of people who had subscriptions to magazines in the 1930’s- rich, white, men. The survey was extremely biased towards that demographic. This is why surveys are not very good indicators of the actual results.
Push polls are a commonly used technique in political campaigns. They are not even polls at all, but are a technique used by the interviewer to insight an opinion in people.
Double-barreled questions can make polls biased. The interviewer will ask a question such as, “Do you think President Obama has done a good job with the economy and has done right by our troops in Afghanistan?” Those were two questions. A person could answer with a yes, but may not agree with both statements. The chapter talked about the importance of wording in questions. Indeed, it is very important that the interviewer ask the question in unbiased manner, and make sure that the interviewee understands the question.
The last thing I would like to touch on is the margin of error that the book talks about. I don’t think the author does a great job of explaining to the reader that the margin of error is not an indicator of a poll’s usefulness. Margin of Error gives the “possible” error of a poll. It is important to look at the fine print of polls like we had in the months leading up to the 2012 Presidential Election. Polls that have a margin of error of +/- 5 are probably not the best indicators. Polls with a margin of error at +/- are more accurate.
Remember…. Do not think all polls are incorrect just because you have been presented with a bunch of information about the biases and inaccurate nature of polling. Polling can be very useful if you understand all aspects of the poll. Do not take for granted that the numbers are accurate in the way they are presented. Read the fine print and do background research to determine the poll is important. For more information on push polls, click on the link below.