English Deutsch Français Italiano Español Português 繁體中文 Bahasa Indonesia Tiếng Việt ภาษาไทย
All categories

3 answers

If the issue is whether or not you get asked then the simple answer is that the laws of probability have ruled you out so far.

To get relatively accurate polling data for the whole nation one only needs to survey about 2500 people. The numbers involved in statistical analysis seem odd but they do work- and generally work extremely well as long as some other rules involving HOW you pick the people who you ask the question(s).

This means that an accurate whole nation survey only needs to ask 1 in a million people. These odds are like winning a mid-level prize in the lottery. Most of us haven't done THAT either.

2006-12-14 01:05:19 · answer #1 · answered by ethicsprof 3 · 2 0

Statistics. Nationwide surveys and polls are done by research firms that use large databases to randomly select, by computer, a number of people to poll. The selection may be made by address or by telephone number. Telephone surveys are the easiest for pollsters to get results, so the computer does the dialing.

Using a formula, pollsters can find out how many people need to be polled, based on the total number of people in the population information is wanted for. In most cases, pollsters like a 95 percent confidence interval, which means that, on average, 95 percent of the time the results of the poll will fall within the margin of error. Again, statistical formulas are used to figure this stuff out.

2006-12-14 14:19:12 · answer #2 · answered by Ryan R 6 · 1 0

Most Polles are rigged to the part of the country that most believes how the issue wants them to prove. For example....If the poll takers want to have an out come good for republicans they will *independently* find people or a state or class of people that is predominately republican. Don`t even pay attention to MOST POLLS

2006-12-14 08:29:49 · answer #3 · answered by hardhead 3 · 1 2

fedest.com, questions and answers