Validity of poll "margin of error" number

Submitted by Maurice Shurman (United States), Mar 10, 2003 at 22:09

The margin of error in percent is simply 100 times the reciprocal of the square root of the number of people polled (If you don't believe this, check it out on the next few polls whose margins of error and numbers of people polled are cited!). Confidence level has nothing to do with it. It is impossible for a poll of 1000 to accurately represent the entire population's true percentages of such important factors as: age, sex, religion, income, education, political effiliation, and ethnicity. the quoted "margins of error" are valid only if the population fromwhich the sample is drawn is totally homogeneous!

I think that one reson the media cite polls is that they (they media themselves) are the people who commission the polls!

---

Note: Opinions expressed in comments are those of the authors alone and not necessarily those of Daniel Pipes. Original writing only, please. Comments are screened and in some cases edited before posting. Reasoned disagreement is welcome but not comments that are scurrilous, off-topic, commercial, disparaging religions, or otherwise inappropriate. For complete regulations, see the "Guidelines for Reader Comments".