- use the z-score value to decide about the hypothesis according to criteria (a), two possibilities:
- z-score outside criterion boundaries (ɭ.96 or <-1.96 for a=0.05, in critical region): extremely unlikely if H0 is true > reject null hypothesis! (treatment has an effect)
- z-score inside boundaries (ə.96 and >-1.96 for a=0.05, not in critical region): no evidence that H0 is not true > fail to reject null hypothesis! (treatment may have no effect)
… but : note that hypothesis testing is an inferential process, providing limited information for general conclusions !
(remember: z = obtained difference / difference due to chance)