Calculating the Number of Respondents You Need

The number of respondents you need depends on your survey goals and how confident you want to be in your results. The more confident you want to be, the less of a margin of error you should accept.

To calculate the number of respondents you need (known as your sample size), use our sample size calculator.

TIP! If you need a guaranteed number of respondents, buy survey responses from SurveyMonkey Audience. You specify the number of responses you need and we'll use your targeting criteria to find you respondents.

Definitions

To calculate your sample size, you'll need to know the following information:

Population Size
Your population size is the size of the entire population you wish to represent.
- Population: The entire group you're interested in making conclusions about.
- Sample: The group you're surveying.
Think about the potential size of your target population. For example, if you're sending a survey to male iPhone users in California, you may need to do some research to determine how many total men fit that criteria.
Margin of Error
Margin of error tells you how much error surrounds a measure. It's a percentage that describes how much the opinions and behavior of the sample you survey is likely to deviate from the total population. To calculate your margin of error, use our margin of error calculator.
The smaller the margin of error is, the closer you are to having the exact answer at a given confidence level.
In general, the larger your sample size, the lower the margin of error. The closer your sample is in size to your population, the more representative your results are likely to be. And that’s why you’ll notice that the recommended sample size in the table below gets smaller as your tolerance for error gets larger.
For example, let's say we asked 400 people if they have a favorable or unfavorable opinion of Barack Obama and 55% say favorable. Using a 95% confidence level and ±5% margin of error, if we repeated this survey 100 times under the same conditions, 95 out of 100 times, the response would be somewhere between 50% and 60%.
Confidence Level
A confidence level tells you how reliable a measure is. Common standards used by researchers are 90%, 95%, and 99%.
A 95% confidence level means if the same survey were to be repeated 100 times under the same conditions, 95 times out of 100 the measure would lie somewhere within the margin of error.
When calculating your sample size, you'll use the z-score for your confidence level. The z-score is the number of standard deviations a given proportion is away from the mean.
Confidence Level z-score
90% 1.65
95% 1.96
99% 2.58
Percentage Value
Sample size requirements vary based on the percentage of your sample that picks a particular answer. For example, if in a previous survey you found that 75% of your customers said yes they are satisfied with your product and you are looking to conduct that survey again, you can use p = 0.75 to calculate your needed sample size.
If you're running a survey for the first time, and since most surveys have more than one question (and therefore more than one percentage value to evaluate), we recommend using p = 0.5 to calculate your optimum sample size. This produces a sample size estimate that is neither too conservative nor too loose.

Calculating Sample Size

Calculate the number of respondents you need in seconds using our sample size calculator. If you’d like to do the sample size calculation by hand, use the following formula:

z squared times open parenthesis 1 minus p close parenthesis over e squared over 1 plus open parenthesis z squared times p open parenthesis 1 minus p close parenthesis over e squared capital n close parenthesis.

Statistic	Description
N	Population Size
e	Margin of Error (as a decimal)
z	Confidence Level (as a z-score)
p	Percentage Value (as a decimal)

Suggested Sample Sizes

Below is a table with suggested population sizes by margin of error at a 95% confidence level. We used the above formula to calculate the suggested sample sizes. In some cases, we rounded the sample sizes up to the nearest 5 or 10. For a more exact calculation, use our sample size calculator.

Population Size	Sample Size per Margin of Error	Sample Size per Margin of Error	Sample Size per Margin of Error
	±3%	±5%	±10%
500	345	220	80
1,000	525	285	90
3,000	810	350	100
5,000	910	370	100
10,000	1,000	385	100
100,000+	1,100	400	100

Example Sample Size

You're sending a survey with a Yes or No question asking if parents of children at your school are in favor of an extended school day.

The total number of parents (your population size) is 10,000 and you're comfortable with a ±10% margin of error. Using the table above, you can see you'll need at least 100 people to take your survey.

70% of the 100 parents surveyed answered that they are in favor of an extended school day. This means you can assume that if all 10,000 parents answered the survey, between 60% to 80% of people would be in favor of an extended school day.

How many people should I ask to take my survey?

Your response rate may affect the number of people you send your survey to. The higher the response rate, the fewer people you need to ask to take your survey.

For example, 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 to take your survey.

To calculate the number of people you need to invite to take your survey based on your expected response rate, use the following equation:

A response rate percentage formula with "pound sign of responses to your survey" on top divided by “expected percentage of the response rate" on the bottom times 100

Respondents needed

Take the number of people you need to answer your survey

÷ Response rate

Divide by the expected response rate (use 25 for 25%, if you use .25 skip step 3)

x 100

Then multiply by 100 (to account for the using a percentage versus a decimal in step 2)

SurveyMonkey Audience: Additional Targeting Settings

Feature Importance (MaxDiff)

Using SurveyMonkey for Market Research

Confidence Level	z-score
90%	1.65
95%	1.96
99%	2.58

SurveyMonkey