Question

Null hypothesis significance testing (NHST) can provide critical information, but NHST is inadequate as the sole tool for statistical analysis. NHST has also been subject to many criticisms, but is still almost universally used by statisticians and others.

One of the criticisms of NHST is that "Successfully rejecting the null hypothesis may offer no support for the alternative hypothesis."

Discuss this criticism as well as the idea of NHST generally.

Answer 1

Null hypothesis Significance Testing(NHST) is a statistical method of inference and it is used to provide an evidence for an effect.Under the null hypothesis significance testing an observation is tested against a hypothesis of no effect or no relation.There are mainly two types of hypothesis null hypothesis and alternate hypothesis.Under null hypothesis we assume that there no statistically significant relationship between the variables used under the study.Alternative hypothesis is just opposite to the null hypothesis .Alternative hypothesis states that there is statistically significant relationship between the variables under the study.We reject or accept the null hypothesis based on the p value.P value acts as a reference point to identify the results.In general if the P value is less than or equal to the level of significance we reject the null hypothesis and accept the alternative hypothesis and if the p value is greater than the level of significance then we accept the null hypothesis.Usually the level of significance is set at 0.05.

Some of the main criticisms regarding the null hypothesis significant testing is that sensitivity to the sample size and it there can be chances that the wrong hypothesis can be accepted .Another criticism is that the as the results are based on the probability testing it is not necessary that the results are always true.There are chances of type 1 error and type 2 errors to occur.Type 1 error occurs when we reject the true null hypothesis and type II error happens when we accept the false hypothesis