1.The estimates of Bush's lead in the polls? varied from 12% lead to 4% lead from Al Gore. This broad difference was recorded because of sampling error. In the real world? "random sampling error" or the chances that a pure probability sample would generate responses within a certain band of percentages only because of the sample size is one of the least of measurement problems (Sampling Error, Retrieved from www.socialresearchmethods.com). The main problems of polling measurement? or more accurately mismeasurement? include:
The sample design (for telephone surveys? how the telephone numbers were selected and ...