A simple survey sample size calculator

Do you want to get “statistically significant” results from your next survey?

But do you also not want to pay for more survey respondents than you actually need?

Sample sizes and survey results

When you see a percentage figure quoted from a survey, you should view it as a range, rather than an exact figure. 

For example, imagine a survey which showed 40% of respondents saying they have heard of your brand. Really, this means a range, such as 38%-42% (depending on the sample size), have heard of your brand. What a bigger sample size does is reduce this range.  

How it works

Try putting different sample sizes in the box below (and then pressing “Enter” each time). You will see the range for the percentage figure below then change. This range is called the “confidence interval”. In the example it is basically saying we can be 95% confident that the true percentage figure out in the real world is within the range shown.

So what is a Statistically “valid” sample size?

There's actually no such thing! All the sample size does is affect the range of the percentage scores from the survey, so what matters is what range you are happy with for these scores.

Conclusion

Is it really worth paying for a very large sample (>1000), when so many other factors will determine the quality of your survey results? Feel free to talk to us about this. We can give you an idea of the costs involved.