Research Tips

Sample Sizes

Many people think there’s a magic formula for determining market research sample sizes, but unfortunately, there isn’t. The sample size appropriate for a particular study depends on the expected response rate, the level of precision required for the results, and the types of analysis that will be performed on the data. For example, to get the same level of precision, analysis that involves dividing respondents into categories and making comparisons across categories (e.g., analysis of variance, cross-tabulations, etc.) requires a larger sample than does analysis that generates descriptive statistics (e.g., mean, standard deviation, etc.). Therefore, it’s helpful to base sample sizes on the number of responses that will be required to complete the most fine-grained analysis to the desired level of precision, and then adjust for expected response rates. It’s also helpful to add some buffer in case the distribution of the data or the response rate differ from expectations.