Here is a quick reference focused primarily on quantitative survey research. Qualitative research tends to focus on smaller sample sizes and is less concerned with the ability to generalize findings to a larger population. Here is a simplified table for some common scenarios.
Sample Size Table
|95% Confidence||99% Confidence|
|Margin of Error||Margin of Error|
Confidence Level, Margin of Error and Population Size
We need to understand these terms in order to be able to correctly use the table above.
Confidence Level. The most common level used in business is 95%, which means if the survey were conducted 100 times, the data would fall within the margin of error (see below) 95 out of 100 times.
Margin of Error. The most common level you will see if +/-5%, which means the responses on your survey need to be given from 5% leeway in either direction. For example, if 60% of people said “somewhat agree”, it’s really between 55 and 65%.
Population Size. How many people are you trying to generalize to? How many people are in your customer or prospect base? That’s your population size.
All That Being Said…
There are a number of other things that can dictate sample size:
- Need to drill down into sub-populations
- Corporate culture (what’s an “acceptable” or “usable” sample size in the eyes of stakeholders). Are you in an industry where you can easily reach thousands of respondents? Or is it a chore to get 100? Do executives insist on having a magic number of 500 (or something else) to “believe” that the research credible? These issues often outweigh statistics.
This article should get you started in selecting the appropriate sample size for your project. There are some finer points, such as considering the consequences of the decisions that will be made from the research. Contact us at 877-672-8100 or firstname.lastname@example.org and we’ll be happy to walk through it with you.