This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Blackboard Readings Sense and Nonsense About Surveys, Howard Schuman • Surveys draw on two human propensities- gathering info by asking questions AND learning info about one’s environment by examining a small part of it • Within the past century, major improvements allowed sampling process to develop o Value of a sample comes not only from size but from the way it is obtained o Ex. Literary Digest poll predicted that Roosevelt would lose 1936 presidential election- 10 million questionnaires sent out and over 2 million sent back BUT George Gallup predicted FDR would win because he used less cases but quotas in choosing respondents to represent diff economic strata • Gallup did not use probability sampling methods which calls for each person in the population to have an equal chance of being selected • When targeted members of a population are not interviewed or don’t respond to particular questions then the omissions are a serious problem if they are numerous and if those missed differ from those who are interviewed • Survey results are AS dependent on the form, wording, and context of the questions asked as they are on the sample of people who answer them • So, responses to survey questions must be viewed with skepticism • However, if question wording is kept constant, then differences overtime, across edu levels, and most other comparisons are not seriously affected by wording • To interpret surveys accurately, it is important to use a framework of comparative data in evaluating the results • Overall, surveys remain essential for learning about large populations...
View Full Document
- Spring '06
- Sociology, probability sampling methods, 1936 presidential election, Havasupai tribe bc, diff economic strata, Blackboard Readings Sense