The result will give you the number of responses you’ll need to achieve the accuracy you enter at a 95% confidence level. We also allow you to calculate the responses needed for a 90% and a 99% confidence level. A 95% level of confidence is typically used.
With the number of responses you need you can determine to how many people you should send your invitation. For example, if the calculator says you need 200 responses, and you expect a 25% response rate, then you need to send an invitation to 200 / 0.25 = 800 people.
If your population is 1000 or less, then you’ll probably be sending the invitations to everyone.
Please note: this calculator — and every other calculator we’ve found online — uses the most conservative assumption about the variation expected in the survey results. Yes, we know for the statistically disinclined this is a head scratcher, but just know that you accuracy will be no worse than what this calculator tells you — and perhaps better.
How to Determine the Accuracy of a Completed Survey
You’ve done your survey. You know the number of responses you actually got, and you know your population. You can use our calculator to determine the survey accuracy — also known as “margin of error” or MOE.
- Enter your population
- Enter different accuracy levels until the resulting Number of Responses is close to the number you actually got
Want a better understanding of all this? Consider attending our Survey Design and Survey Data Analysis Workshops. We discuss this in detail. We also discuss this in our Customer Survey Guidebook — new edition in the works.