Question

How many of the following statements are true?
-The Wilcoxon signed-rank of differences requires that the data be at least ordinal scaled and that the two samples be independent.
-Nonparametric tests may be used with variables that are measured with a nominal, ordinal, interval, or ratio scale.
-The test statistic for the null hypothesis that the correlation among the ranks is equal to zero is a t statistic
-We can apply parametric tests, such as the t-test, to ordinal or ranked level measurement.
-The wilcoxon rank-sum test computes a x^ 2

A) 1
B )2
C )3
D )4

Answer 1

We can use Wilcoxon signed-rank for paired sample i.e. two samples are not independent. Hence the statement "The Wilcoxon signed-rank of differences requires that the data be at least ordinal scaled and that the two samples be independent" is false.

Since non-parametric methods are concerned with the treatment of standard statistical problems when the assumption of normality is replaced by general assumptions concerning the distribution function i.e. the variables are just assumed to come from a continuous distribution (i.e. the variables are measured with ratio scale). Hence the statement " The Wilcoxon signed-rank of differences requires that the data be at least ordinal scaled and that the two samples be independent" is false.

We know for large n i.e. n is 10 or larger, the statistic

Hence the statement "The test statistic for the null hypothesis that the correlation among the ranks is equal to zero is a t statistic" is correct.

Using above argument, the statement "We can apply parametric tests, such as the t-test, to ordinal or ranked level measurement." is correct.

"The wilcoxon rank-sum test computes a x^ 2" is false.

Hence Option (B): 2.