Explain when a nonparametric test is to be used in place of the one-way ANOVA F-test, when testing the difference among several population means

What will be an ideal response?

The one-way analysis of variance test generalizes to the multi-population case for the t test comparing two population means when independent random samples are available. The test is based on an assumption that the underlying population distributions are normal. It is also possible to develop a nonparametric alternative to the one-way analysis of variance test. Such a nonparametric test that is valid for the comparison of the central locations of two populations based on independent random samples, even when the population distributions are not normal. This is the Kruskal-Wallis test, employed when an investigator has strong grounds for suspecting that the parent population distributions may be markedly different from the normal. Like the majority of the nonparametric tests we have already encountered, the Kruskal-Wallis test is based on the ranks of the sample observations.