When you carry out a survey, you try to interview a relatively small group of people relative to the whole population size. Nonetheless, you want to estimate what percentage of the whole population would give a particular answer to the question you asked. The approximated 95% confidence interval gives you an idea of the reliability of your survey. If you did this survey many times, each time interviewing the same total number N of people, but choosing those people at random from the population every time, then you would find that the ratio (number of people who gave answer A)/ (number N of people interviewed) would not be the same every time. The 95% confidence interval gives you a central value and upper and lower bounds within which the ratio would fall 95% of the times you conducted the interviewing experiment.

Here you can calculate the 95% confidence interval of a survey. Enter the appropriate values into the entry windows, and then click the button to calculate the approximated 95% confidence interval.

The sample size is the number of people who were interviewed.

The sample outcome is the percentage of people, among all those interviewed, who gave one particular answer.