If you want to ensure that your survey is statistically significant without having to survey your entire population or database, you need to work out how many people you need to send your survey to.
You need to get enough completed surveys back so that it accurately represents your population as a whole.
This can seem confusing and a little complicated, but it doesn’t need to be. Here is a simple guide to how to calculate your sample size and calculate how many people you need to send your survey to.
1. Population Size
Determine what your whole population size is. Your population is the entire set of people you want to study with your survey.
2. Confidence Interval (margin of error)
This is how sure you want to be that the responses that you receive will accurately represent the views of your population. This will be your margin of error.
A standard confidence interval is 5%. You can choose a confidence of up to 10% depending on your survey, but it is not advised to go above 10%.
3. Confidence Level (accuracy)
Confidence level refers to how accurately your sample represents your population.
So, a 95% confidence level means that you would get the same results 95% of the time even if you chose different samples from your population.
A standard confidence interval is 95%. You can choose a confidence level of up to 99% depending on your survey, but it is not advised to go below 90%.
4. Sample Size
Your sample size is the number of completed responses you need to get back. Note that this is different from the number of people you need to invite to complete your survey as not everyone that you send your survey to will necessarily complete it.
5. Response Rate
This is a bit of a tricky one to work out as you won’t know your response rate until after you send out your survey. However, to calculate the number of people you need to send your survey to, you will need to use a little bit of guesswork here.
If you have not sent any survey to your population before, a good recommendation would be to have a response rate estimate of 10-15% as this is a conservative percentage.
So, how many people do you need to send your own survey to?
And how many survey responses is required to be statistically significant.
To get the work out numbers for the sections outlined above you can use the calculator at the end of this article to help you to get your numbers.
And now for the easy part! To work out how many people you need to send your survey to, here is the magic formula:
“sample size divided by response rate”
For example, if you have a sample size of 200 and an estimated response rate of 10%, this would be your calculation:
200 / 0.10 = 2000
To break it down, this formula is saying that you are expecting 10% of your respondents to complete your survey. and you need 200 completed responses to be received.
Therefore, you need to send enough surveys invitations out so that when you receive your 10% of completed surveys, this equates to 200 responses. So if 200 responses is 10% of your total invitations sent, you need to send out 90% MORE survey invitations. In other words: 10% of 2000 = 200 (which is your sample size required).