# Margin of Error & Sample size Calculator

Margin of error is an estimate of how far true population values may be from the collected sample data. It is generally expressed in percentage points and depends on the size of your Target Market, Sample Size, and the Confidence Level.

Simply put, you want to have the lowest possible margin of error at the highest possible Confidence Level. Theoretically, if you had a chance to survey every person in the population - you'd be able to eliminate the margin of error completely and be at a 100% Confidence Level because your sample size would be equal to the population. Since in practice online surveys deal with sample sizes of several hundred to several thousand responses, margin of error calculators are used to set the right expectations and achieve balanced cost/accuracy decisions when determining how many respondents you need.

###
Population size (your target market):

Enter the number of people who theoretically fall within your target audience. For example, if your survey is targeting any adult in the US, leave it at the default 250,000,000. For a more restrictive example, if you're targeting K-5 teachers only, Google how many elementary school teachers there are in the country of your interest and enter that number here.

###
Sample size:

Enter the number of responses you have budgeted for. You can change it to run "what if" scenarios and see the margin of error shrink as you increase the sample size. As a rule of thumb, you want to keep the sample size (also referred as N) above 400 if you want to keep margin of error at or below 5% and a confidence level of 95%.

###
Confidence Level:

If you are not sure, leave the Confidence Level at the industry standard of 95%. Simply put, it tells you that if you rerun the study 100 times, 95 times out of 100, answers will fall somewhere within the corresponding margin of error and 5 times they may fluctuate to a greater extent.

###
Margin of Error:

Margin of error estimates how far real population values may fall from your sample responses. For example, if you run a survey with margin of error of 5%, and 45% of respondents said that they're interested in learning more about your product, you should interpret it as: "40% to 50% of the population was interested in learning more". As you can see, margin of error works both ways: +5% or -5% of the sample average, forming what is commonly called "Confidence Interval". You don't know for sure where within that interval the actual interest level lies for the entire population because if you rerun the survey, answers will fluctuate somewhat every time. Margin of error helps you to predict the extent of that fluctuation, though.

Finally, margin of error calculators offer a solid statistical way of predicting the accuracy and reliability of your survey results compared to the entire population. It relies on an assumption that you are using probability sampling methodology, meaning that all respondents in your survey are selected randomly. Actual fulfillment methods may affect the margin or error estimates or even render them useless. We recommend consulting a statistician or an experienced market researcher if in doubt.