Coastal Services Center

Determining Sample Size

If you don't have the time, experience or expert assistance to determine the ideal sample size, you can use the following common sense guidelines:

	The size of sample you need depends upon two things: How accurate you want the summary data to be. If you want no sampling error at all for example, you would need to measure each entire population. How variable the data are. If you started measuring something and found that every measure was the same, you obviously wouldn't go on very long repeating the measurements to increase your accuracy. On the other hand, the more the data vary, the more data you need to collect to get a reasonably accurate measure of the mean.

A third consideration is acceptability to participants and audiences. For example, in collecting the views of staff, it might not be acceptable to sample, even though a sample would appear to be statistically adequate. Every staff member might need to be heard so that no one feels left out and there is no suspicion of bias.