How many respondents do I need?
The number of respondents (people who complete your surveys) you need depends on how confident you need to feel in your results. The more confident you want to be, the less of a margin of error you'd likely be willing to accept.
To calculate the number of respondents you will need to take your survey, use the following formula, where:
e= the percentage margin of error you are willing to accept (in decimal form)
( 1 + N*e2 )
For example, if the size of the total population you wish to represent is 5,000, and you are willing to accept a 5% margin of error, here's how you would calculate the number of respondents needed:
( 1 + N*e2 )
( 1 + 5000*.052 )
= 370 respondents needed
Here’s a basic chart for ease of use:
Respondents Needed at Margin of Error of ±3%, ±5%, ±10%
Let’s walk through a quick example on how to use this formula and table:
Say you know your population (the total number of people eligible to take your survey) is 10,000. You think about the margin of error you're willing to accept and decide that you're comfortable accepting a margin of error as great as ±10%. So let's look at the fifth row of the table (Population Size = 10,000) and the third column over (Margin of Error = ±10%), you'll need at least 100 people to take your survey.
Still confused about what the ±10% exactly means?
That's ok! Here's more information: Say, you've heard from 100 people as the table recommends, and find that 70% of those surveyed are in favor of, for example, an extended school day. Given your 10% acceptable margin of error, you can assume that if you'd asked every person in your area to take your survey (hypothetically around 10,000 people), the actual proportion in favor of the extended school day would range from 60% to 80% (70% ±10%).
But what if that range is too big? What if you need to be more precise? Well, then you're going to need to sample more people. If you're more comfortable with a range of ±3% (a lower margin of error, more confidence in the results), using the same example and based on the table above, you'd need to sample 1,000 people.
I'm clear on how many survey response I need, but how do I know how many people should I invite to take my survey?
The table above recommends the number of respondents you'll need, not how many people you should invite to take your survey. So if you need 100 respondents and you expect 25% of the people invited to take your survey will actually respond, then you need to invite 400 people (100 respondents / .25 response rate = 400 invitations).
We've seen the likeliness of people to respond to survey invitations varies, based on a variety of parameters including survey and audience type.
With SurveyMonkey Audience projects, we monitor response rates for your project and we'll take care of inviting enough participants to make sure you get the number of survey respondents you need! You just need to tell us how many respondents you want, and we'll take care of the rest!