Testing the Hypothesis

In order to determine differences between samples and populations, null and alternative hypothesises (I know, not a word) are formed.

Normally, the null hypothesis is mean_E - mean_C = zero.
A conservative alternative hypothesis simply states that the experimental mean will be different than control mean, without predicting a direction. As such, a z score or two-tailed t-test is used.

Type I error: finding a difference when there is none

Type II error: failing to find a difference that exists; Power is the capacity to find a difference, and is therefore 1- β