Question

What is wrong with relying on P values to interpret findings in a study.

Answer 1

Answer-

A P-value or Probability value refers to the probability of acquiring results of a statistical hypothesis test while assuming that the null hypothesis is accurate. A p-value is used as a substitute for rejection points by rejecting the null hypothesis by offering the smallest level of significance. Therefore, the smaller the p-value, the stronger is the evidence in favor of the alternative hypothesis.

Why p-values cannot be relied upon-

i) P-values cannot measure the probability of the truth of the studied hypothesis. In addition to this, p-values cannot determine the probability of whether the data was produced by a random chance. In some cases, researchers often incorrectly assume that smaller p-values indicate that the null hypothesis is false. However, p-values can only predict the probability of acquiring results at least as great as those seen in a case where the null hypothesis was true.

ii) Business policies and scientific decisions- A p-value is merely a statistic. For example- P>0.5 does not necessarily mean that the hypothesis is true. Also, p-values can be influenced by several aspects, like sample size. The sample size of the study greatly affects the p-value. Therefore, practical decisions can not be made based on p-values alone.

iii) P-values cannot measure the magnitude of an effect/ significance of a result- P-values cannot determine the practical or clinical importance of a result. Some researchers may consider a very low p-value as "highly significant", however, it does not necessarily show the relationship between the variables of the study.

Therefore, a p-value, by itself, cannot be relied upon to interpret findings in a study.