Sample Size Calculator
Calculate the ideal number of participants for surveys, user interviews, usability tests, and other research methods.
Total number of people in your target group
How certain you want to be that results reflect the true population value
Acceptable range of error (1%–10%)
How to Calculate the Right Sample Size for Your Research
Getting your sample size right is one of the most important decisions in any research project. Too few participants and your results may not be reliable. Too many and you waste time and budget without meaningful gains in accuracy.
This calculator supports two fundamentally different approaches to determining sample size, because quantitative and qualitative research follow different rules.
Quantitative Sample Size (Surveys & Polls)
For surveys and quantitative studies, this calculator uses Cochran's formula, the standard statistical method for determining sample sizes. The formula accounts for three key variables:
- Population size — the total number of people in your target group. For large populations (100,000+), the required sample size plateaus, which is why national polls can survey just 1,000-2,000 people.
- Confidence level — typically 95% for most research. This means you can be 95% confident that the true population value falls within your margin of error.
- Margin of error — how much variation you can tolerate. A 5% margin is standard for most surveys; tighter margins like 1-3% require significantly larger samples.
The calculator also applies a finite population correction, which reduces the required sample size when your population is relatively small. This matters most when your population is under 10,000.
Qualitative Sample Size (Interviews, Usability Tests & More)
Qualitative research does not follow the same statistical rules. Instead of statistical power, qualitative sample sizes are based on the concept of thematic saturation — the point at which additional participants stop revealing new insights.
The qualitative recommendations in this calculator are grounded in peer-reviewed research:
- User interviews: 12–30 participants based on Guest, Bunce & Johnson's (2006) saturation study and Creswell & Poth's (2018) guidelines.
- Usability tests: 5 participants per round based on Nielsen & Landauer's (1993) mathematical model of usability problem discovery.
- Card sorts: 15–30 participants based on Tullis & Wood's (2004) research on correlation stability.
- Diary studies: 10–15 participants based on established longitudinal qualitative methods.
- Focus groups: 3–5 groups of 6–10 participants based on Krueger & Casey's (2015) framework.
Tips for Choosing Your Sample Size
- Account for non-response. If you expect a 30% survey response rate, you need to invite roughly 3.3x your target sample size.
- Consider subgroup analysis. If you plan to compare results across segments (e.g., free vs. paid users), each segment needs its own adequate sample size.
- Start qualitative, then go quantitative. Run interviews with 12–15 users to generate hypotheses, then validate with a statistically powered survey.
- Budget for dropouts. In longitudinal studies like diary studies, plan for 10–20% participant attrition.
Frequently Asked Questions
How does the sample size calculator work?
For quantitative research like surveys, the calculator uses Cochran's formula: n = (Z² × p × (1−p)) / e², adjusted for finite populations. You provide your total population size, desired confidence level (90%, 95%, or 99%), and acceptable margin of error (1–10%). The calculator determines how many responses you need for statistically valid results. For qualitative research, it provides research-backed participant recommendations based on the specific method you plan to use.
What sample size do I need for user interviews?
Research by Guest, Bunce, and Johnson (2006) found that 12 interviews typically uncover 92% of themes in a homogeneous sample, while Creswell and Poth (2018) recommend 20–30 participants for maximum variation sampling. Start with 12 participants for a focused study with similar users, and aim for 20–30 if your user base is diverse or spans multiple segments.
Why does Nielsen recommend only 5 users for usability testing?
Jakob Nielsen and Tom Landauer's mathematical model shows that 5 users uncover roughly 85% of usability problems in a single interface. The formula P(i) = 1-(1-L)¹ with L averaging about 31% demonstrates steep diminishing returns after the fifth participant. The key insight is that running three rounds of 5 users with design fixes between rounds catches more issues than testing 15 users once, because you fix problems iteratively.
What is the difference between confidence level and margin of error?
Confidence level is how sure you can be that your results reflect the true population value — a 95% confidence level means if you repeated the survey 100 times, 95 would produce results within the margin of error. Margin of error is the range of possible error in your results — a 5% margin means your true result could be up to 5 percentage points higher or lower than what the survey shows. Higher confidence or lower margin of error both require larger sample sizes.
When should I use qualitative vs. quantitative sample sizes?
Use quantitative sample sizes when you need to measure how many or how much — for example, survey response rates, NPS scores, or A/B test conversion rates. Use qualitative sample sizes when you need to understand why or how — for example, user interviews to discover unmet needs, usability tests to find interaction problems, or card sorts to inform information architecture. Many research projects combine both: qualitative studies to generate hypotheses, followed by quantitative studies to validate them at scale.