Book Review. Following are the data from Exercise 16.48 on the number of pages for random samples of books in five rating groups.

1* 2* 3* 4* 5*

382 560 384 325 360 391 343 458 390 298 335 512 409 304 272 368 329 309 240 368 400 391 374 306 320 372 367 459 169 326

a. Use the Kruskal-Wallis test to decide at the 1% significance level, whether the data provide sufficient evidence to conclude that a difference exists in mean number of pages among books in the five rating groups.

b. The hypothesis test in part

(a) was done in Exercise 16.48 by using the one-way ANOVA test. The assumption there is that number of books in the five rating groups are normally distributed and have equal standard deviations. Presuming that to be true, why is performing a Kruskal–Wallis test to compare the means permissible?

In this case, is use of the one-way ANOVA test or the Kruskal–

Wallis test better? Explain your answers.