Question

I am doing a stats project and I'm determining confidence levels for my survey. I used a simple random sample by sending an email out to my population of 306 people and only 25 responded. When determining a confidence interval to a yes or no question, do I use the sample number for my n, which would be 25, or do I use the total population in which I was targeting for my n, which would be 306?

Answer 1

For determining the Confidence Interval for the population parameter from the sample statistic, we must use only the sample size = n = 25, the number of people who have responded to the survey. Here the sample consists of the response of 25 people only and NOT the entire 306 people. The reason is that all the characteristics such as sample mean, sample standard deviation etc. can be calculated only from 25 responses only because only that data is available with us. the rest of the population: 306 - 25 = 281, nothing can be done because no data is available about them.