Test on Nonparametric Statistical Methods
Nonparametric Statistical Methods: A Student's Guide
Test: Nonparametric rank tests
14 questions
Question 1: The Jonckheere-Terpstra Test is typically employed when the alternative hypothesis assumes no particular ordering among the s treatments.
A. Ano
B. Ne
Explanation: The Jonckheere-Terpstra Test is designed for ordered alternatives, where treatments are hypothesized to have a definite order, such as the second treatment being more effective than the first, and the third better than the second.
Question 2: The Kruskal-Wallis test is an appropriate statistical method for evaluating a null hypothesis of no difference between treatments against an alternative hypothesis where treatments are expected to follow a definite order, such as when comparing different dose levels.
A. Ano
B. Ne
Explanation: The study materials explicitly state that the Kruskal-Wallis test is not appropriate for ordered alternatives because it rejects the null hypothesis when the average ranks differ substantially, rather than specifically testing for a definite order among treatments. The Jonckheere-Terpstra Test is presented as the appropriate method for evaluating ordered alternatives.
Question 3: The paired t-test requires an assumption that the differences between paired observations are normally distributed for the test statistic to follow a t-distribution.
A. Ano
B. Ne
Explanation: The study materials explicitly state: "The statistic follows a t -distribution if the differences are normally distributed ⇒ t -test is a parametric method" (Slides 6 and 7). This confirms that normality of differences is an assumption for the paired t-test.
Question 4: The Terpstra W test statistic is symmetrically distributed around zero.
A. Ano
B. Ne
Explanation: The study materials state that the Jonckheere test statistic J is symmetrically distributed around zero, not the Terpstra W test statistic.
Question 5: The Jonckheere J test statistic is symmetrically distributed around zero.
A. Ano
B. Ne
Explanation: The Jonckheere J test statistic is symmetrically distributed around zero.