Sample Size and Confidence Interval

This Sample Size Calculator provided by ISG is a convenient tool designed to assist you determining the required number of people you must survey in order to reflect your desired results as well as a way to find your level of accuracy from an existing sample.

In using the calculator, it is helpful if you understand the meaning of the following terms.

Confidence Interval

The confidence interval, commonly called the margin of error, is a plus or minus figure that you might experience when asking a particular research question of every member of your target survey population and receive the same answer back that the members of the sample gave in the survey. Illustration: If you used a confidence interval of 3 and 70% of the participants in the survey answered "Yes” to the question “Would you buy our product again?," you could be sure that between 73% (70+3) and 67% (70-3) of the individuals within the entire target population would also say " Yes, I would buy your product again" when asked the same question. The confidence interval in this illustration is +/- 3.

Confidence Level

A confidence level tells you how confident you can be of the data obtained from your sample. The confidence level shows how often that percentage of the target population would give an answer that falls within the confidence interval. The confidence level reflects your level of certainty that those responding to your survey will fall inside your confidence interval. A confidence level of 95% is commonly used by researchers.

Based upon our previous illustration, we found that 70% of our survey participants answered "Yes” to the question: “Would buy our product again?" resulting in a confidence interval of plus or minus 3%.

Sample Size

How many observations are contained within your sample? Larger samples deliver data that more closely reflects the target population. The lower the confidence interval desired, the higher sample size is required. If you are interviewing 1,000 people, your results may be accurate to within plus or minus 4% of your findings. If you want to lower the confidence interval to plus or minus 1%, then you will naturally need to interview more people, which will require a larger sample size.

It is important that the sample is selected carefully, thus giving every member of the population an equal chance of being chosen from the overall population. Randomly selecting your sample enables you to successfully generalize your survey results to your target audience of customer, prospects, etc. In addition, it is important that respondents are chosen to participate rather than given a choice to take the survey themselves. In short, when dealing with individuals, you need to make sure that there is an equal probability of selecting any individual participant in a survey.

Population Size (Optional)

How many people in the group (population) you are surveying meet the profile of the individual you wish to survey. That is, how many customers, club members, dog owners within your city, etc. are you surveying?

Percentage

What percentage of your respondents answered “yes” to your survey question. If a high percentage, of your respondents answered “yes,” e.g., 85% vs. 15%, the possibility of error is relatively small. If the “yes” and “no” responses were more evenly split, e.g., 45% vs. 55%, the chances of error are significantly higher.