Sample Size and Confidence Interval Calculator
This calculator will help you answer two questions.
- How many completed surveys do I need to have a reasonably accurate view of the entire population?
- How confident can I be that the information I collected is representative?
Before you conduct a survey
Follow these steps to determine how many completed surveys are necessary.
- Determine your population of interest (e.g., all customers, all female customers, all female customers in the eastern region).
- Estimate how many individuals are in this population; e.g., 20, 500, or 1,000,000.
- Determine how confident you need to be that your results are representative. A common rule of thumb is 95% confidence so that the results are accurate to within +/- 3%.
- Enter the above two numbers in the calculator.
- The result tells you how many completed surveys you need for each population. The "Uncorrected" confidence interval is for data that is not a continuous, normal distribution.
To determine how many surveys need to be distributed, divide the above number by the anticipated response rate. For example, if you need 483 completed surveys and you anticipate a 30% response rate, you need to mail 1,610 surveys (483/.3).
Notes:
p=.5 is the most conservative assumption.
The formulas and information on this page are intended for general information and educational purposes only and do not constitute investment or legal advice. Visitors should not act upon this information without consulting with a professional advisor.