In statistics, a confidence interval, abbreviated as CI,[1] is a tool that people use when they collect data in order to estimate a certain parameter, such as the mean (average) of a population.[2] A confidence interval gives a range of values that tells others how good we think our estimate is.
Each confidence interval depends on certain properties of the sample(s) that we use to make it. For example, a confidence interval that is made from a sample of 5000 people is going to be much better than a confidence interval made from a sample of only 5 people.
We usually give confidence intervals as a percentage, such as 95%. These percentages are called confidence levels.