Why should a poll be scientific rather than informal?

Expert Answers

An illustration of the letter 'A' in a speech bubbles

It is important to understand that even a scientific poll is subject to error, and the errors in question may be serious. The key features that make a poll scientific are (1) a large, demographically balanced and random sample, (2) unbiased questions, and (3) rigorous analysis of the data.

The first factor here is obviously subject to a continuum. The larger and more random the sample of those polled, the more accurate the results will be. If you ask three friends, all of whom are the same age, sex, and social class for their opinions about a film or a politician, you may well obtain 100% agreement. However, this informal poll is statistically valueless. A poll of 100 or 1,000 people will be more valuable, but may still be misleading if there is no balance between the ages, sexes, races, and social classes of respondents. A poll of 100,000 people which is balanced in all these departments and covers urban and rural areas throughout the nation will obviously be much more valuable.

The questions must also be clear, and not contain inherent bias or emotive language. The answers then need to be rigorously reviewed by statisticians. All these precautions are necessary for one reason: they make it more likely that the results will be accurate and reliable. Polls should be scientific rather than informal because the more scientific the poll, the more reliable the data, though it is worth emphasizing the point that no poll, however scientific, is completely reliable.

Approved by eNotes Editorial Team