Question

g P-values and confidence intervals are both used in hypothesis testing. Explain three reasons why it may be preferable to report a confidence interval over a P-value. Provide a specific example to justify your reasons.

Answer 1

Explanation:

There are a few reasons why it may be preferable to report a confidence interval in place of a p-value.

(A) Using a p-value, nonsignificance does not mean absence of an effect of the intervention BUT using a confidence interval, nonsignificance implies that there's no effect of the intervention on the experimental group.

(B) Using a p-value, studies with small sample size sometimes report nonsignificance when there are real effects of the intervention on the group.

(C) On the other hand, in using a p-value, statistical significance does not necessarily imply that the effect observed is clinically important BUT confidence interval gives information about the direction and strength of the observed effect. This aids the derivation of clinical relevance from the study.

(D) The p-value does not tell us how large or small the effect of the intervention is BUT the confidence interval gives an approximation of the actual effect of the intervention.