Variations on this question pop up quite a bit, so I’m going to explain how random samples of 500 or 1,000 can be useful. I’ll blog an explanation in a series three posts over the next week.
- What is sampling error? (click here)
- How do you interpret the margin of error? (click here)
- Is a sample of 500 or 1,000 really enough? (click here)