After hearing Dubner speak today about the often skewed nature of surveys, I found myself thinking about his comments all night. I found that the more I thought about the topic, the more I found his comments to be extremely truthful. With this thought in mind, I was intrigued by a comment made on page 1186 of Lit for Composition where Caldwell Titcomb references a poll taken by the Boston Globe, as part of the proof for his argument. The survey stated that 493 people would prefer to have a different national anthem while only 220 would prefer to keep the anthem the same as it is now. Using the information I learned today, I attempted to analyze the statistic to see if it was representative of the whole population or not. Since the survey seemed to be only conducted in Boston during a time of controversy, and the survey results are utilized in an essay that is heavily pushing for a a new anthem in the United States, I came to the conclusion that the survey statistic is probably biased and most likely not to be fully trusted. What does everyone else think?
- Tara
Subscribe to:
Post Comments (Atom)
You're probably right. I think most surveys are at least somewhat biased. Its almost impossible to judge opinion based on percentages, because it varies so much from area to area, and person to person.
ReplyDelete-Melissa C.
I agree. Its definitely hard to know for sure whether or not the statistics fairly portray the general consensus of America.
ReplyDeleteAlso, when I read this I thought about how easily someone's opinion on a topic like this can change. Personally, my first reaction if asked about what the national anthem should be would be "I don't know/care". Call me unpatriotic, but I'm pretty sure that most Americans would probably have a reaction similar to mine.
However, once I read the articles and became educated on the topic, I quickly found myself (heavily influenced by the anti-StarSpangledBanner bias readings) supporting a change in the anthem.
Because of this personal reaction, I feel that the results of the survey could have a great deal of variability from week to week depending on the public's knowledge of the topic (which could be heavily influenced by the newspaper,tv, etc.)
-Bryce Cody
When I took the statistics class my sophomore year, I never understood why every tests focused so much on the "randomness." All of the statistical tests had to be randomly distributed, or else, they were false and incorrect. But now I understand why the random distribution of a test is important. The staticians try their best to randomly distribute the test so that they can make the test as least biased as possible.
ReplyDelete-Joanne Park