When designing a study, determining the sample size is a critical decision.
In order to estimate the population average from a sample of units, by means of a confidence interval at the 1-alpha level, that provides a particular precission; we can use the following formula to compute the amount of units that would be needed:
In order to evaluate the mean difference between two groups in a given population from a sample of units, when the comparison will be performed by means of a Z-test with a significance level (alpha); the following formula will provide the number of units per group to achieve a given statistical power (1-beta):
Ene 3.0 is a software aimed to solve the sample size problem when designing studies for a wide range of experimental designs and objectives. You can obtain a free copy writing an email to: s.estadistica@uab.cat.