User Research
How to calculate and interpret confidence intervals in user research
How many people need to take your research study?
Take this short quiz to find an approximate sample size for your research study–the number of participants needed to truly trust your results.
Other variables to consider when determining sample size
Factor | How it impacts research sample size |
|---|---|
Confidence interval | Typical confidence intervals lie between 90-95%. If you aim for a higher value, you'll need ↗ more participants to ensure data robustness. |
Criticality of project | Highly critical projects—like medical devices, security services, or key business platforms - will require ↗ more participants. |
Population/segment size | Larger populations need ↗ more participants. |
Population diversity | User research with more diverse populations need ↗ more participants as they are more prone to uncover a broader range of perspectives. |
Quantitative data type | When categorical data types are used, you need ↗ more participants than when dealing with numerical data types. |
Sampling | If you pick your participants at random, you'll likely capture the breadth of perspectives that live in your population, so ↘ fewer participants are needed. However, when introducing non-random sampling, you need ↗ more participants. |
Type of study | For a tactical study, conducted to make quick fixes and immediate improvements, generally ↘ fewer participants are required. On the other hand, when dealing with a strategic study in order to shape a product’s long-term direction you should aim to recruit ↗ more participants. |
It's not just about quantity
Find participants who actually matter. With 400+ available filters and 130+ countries to choose from, recruiting the exact participants you need has never been easier.
