Question 1

Which of the following is not a requirement for the Mann-Whitney U?
A)The independent variable must be between-subjects.
B)The independent variable must be qualitative.
C)The dependent variable is measured at the ordinal level.
D)The independent variable has only two levels.

Accepted Answer

The Mann-Whitney U test does not require that the independent variable be qualitative. It can be either qualitative or quantitative. The other options are requirements for the test: the independent variable must be between-subjects (A), the dependent variable must be measured at the ordinal level (C), and the independent variable must have only two levels (D).

Question 2

When one or both sample sizes are smaller than 10,the sampling distribution of the R statistic does not approximate a normal distribution,so application of the _____ is not appropriate.
A)Wilcoxon rank sum test
B)Mann-Whitney U test
C)multiple R test
D)none of these

Accepted Answer

The answer of When one or both sample sizes are...

Question 3

If one wants to compare the rank ordering of scores as a function of different groups,then a(n)_____ should be used.
A)nonparametric test
B)parametric test
C)t test
D)F test

Accepted Answer

Nonparametric tests do not make assumptions about the distribution of data and are therefore suitable for comparing rank ordering of scores across different groups. Parametric tests assume normality of the data, which may not be the case in some situations, and t-test and F-test are both examples of parametric tests. Therefore, the best choice in this scenario is A) nonparametric test.

Question 4

You are given the following data:
Subject 1 2 3 4 5
Score 10 12 8 10 2
Rank -- -- -- -- --
The ranks for the scores of subjects 1-5 are ____.
A)3.5,5,2,3.5,1
B)1,2,3.5,3.5,5
C)3,5,2,4,1
D)none of these

Accepted Answer

The ranks are assigned based on the scores from lowest to highest. The score of 2 is the lowest, so it gets a rank of 1. The scores of 8 and 10 are next, but since there are two scores of 10, they share the average rank of 3 and 4, which is 3.5. The score of 12 is the highest, so it gets the highest rank of 5. This matches option A: 3.5 for both subjects 1 and 4 (who scored 10), 5 for subject 2 (who scored 12), 2 for subject 3 (who scored 8), and 1 for subject 5 (who scored 2).

Question 5

There are two major types of nonparametric counterparts to parametric test: _____ tests and _____ tests.
A)robust;nonrobust
B)significance;nonsignificance
C)rank;signed
D)signed;unsigned

Accepted Answer

The two major types of nonparametric tests are rank tests and signed tests. Rank tests replace the numerical values in the sample with their ranks, whereas signed tests are used for paired samples where the numerical differences between the pairs are used instead of the original values. Therefore, the correct answer is (C) rank tests and signed tests.

Question 6

A benefit of many nonparametric statistical tests is their relative lack of sensitivity to:
A)standard errors
B)slightly skewed distributions
C)error scores
D)outliers

Accepted Answer

Nonparametric tests do not rely on specific assumptions about the distribution of the data, and therefore are less sensitive to outliers than parametric tests.

Question 7

The most common approach to dealing with tied scores is to rank the scores by assigning the average rank for ties and then to introduce a(n)_____ into the formula for the test statistic to adjust for the presence of ties.
A)correction term
B)error term
C)nonparametric term
D)rank-order term

Accepted Answer

The tied scores need to be ranked, but since they have the same rank, the average of those ranks is assigned to each score. This results in an adjustment in the formula for the test statistic, which requires a correction term to take into account the ties. Therefore, option A - correction term - is the best choice.

Question 8

In cases where a tie occurs,the tied scores are assigned the _____ of the ranks involved.
A)highest
B)average
C)lowest
D)mode

Accepted Answer

If two or more scores tie, their ranks are averaged and each receives that average rank. For example, if two scores tie for a rank of 3, and the next score is ranked 5, then the two scores both receive a rank of ((3+4)/2) = 3.5, and the score after them receives a rank of 5.

Question 9

Rather than comparing groups in terms of _____,the nonparametric procedures compare groups in terms of _____ or other features of a distribution.
A)medians;means
B)means;medians
C)modes;means
D)means;variances

Accepted Answer

Nonparametric procedures do not assume a specific distribution and often compare groups in terms of medians (a form of central tendency that is not as affected by outliers as the mean) rather than means, which are used in parametric procedures that assume a normal distribution.

Question 10

Which of the following is not a requirement for the Wilcoxon ranked-sum test?
A)The independent variable is within-groups or matched.
B)The independent variable has two levels.
C)The dependent variable is measured at the ordinal level.
D)The R statistic is normally distributeD.

Accepted Answer

The Wilcoxon ranked-sum test, also known as the Mann-Whitney U test, requires that the independent variable has two levels (B) and that the dependent variable is measured at least at the ordinal level (C). It does not assume the R statistic is normally distributed (D). The requirement that the independent variable is within-groups or matched (A) is incorrect for the Wilcoxon ranked-sum test; this condition applies to the Wilcoxon signed-rank test, which is used for matched or paired samples. The ranked-sum test is used for independent samples.

Question 11

Nonparametric tests can be used to analyze _____ variables that are measured on a(n)_____ level.
A)quantitative;ordinal
B)qualitative;interval
C)quantitative;nominal
D)all of these

Accepted Answer

The answer of Nonparametric tests can be used to analyze...

Question 12

The Wilcoxon rank sum test and the Mann-Whitney U test are the nonparametric counterparts of the:
A)one-sample z test
B)correlated groups t test
C)independent groups t test
D)none of these

Accepted Answer

Both the Wilcoxon rank sum test and the Mann-Whitney U test are used to compare two independent groups, just like the independent groups t-test. Therefore, the best choice is C, independent groups t-test.

Question 13

_____ statistical tests are a class of statistics that,in general,make few,if any,distributional assumptions
A)nonparametric
B)distribution-free
C)parametric
D)a and b

Accepted Answer

Nonparametric and distribution-free statistical tests are interchangeable terms used for a class of statistics that make few or no distributional assumptions. Parametric tests, on the other hand, require that the data meet certain assumptions about the underlying distribution, such as normality or homogeneity of variance. Therefore, options A and B are correct, and the answer is D (both A and B).

Question 14

Extreme values might prove to be problematic for parametric analyses,but they pose no particular problem for many nonparametric analyses.Because of this property,nonparametric tests are often said to be:
A)robust
B)corrected
C)outlier sensitive
D)outlier resistant

Accepted Answer

The answer of Extreme values might prove to be problematic...

Question 15

Statistical tests that require assumptions about the distribution of scores in the populations from which the samples are selected are called _____ statistical tests.
A)robust
B)distributional
C)parametric
D)nonparametric

Accepted Answer

Parametric statistical tests assume that the data follows a specific distribution, such as a normal distribution. Nonparametric tests do not make this assumption and are therefore more flexible but often less powerful. Robust tests are a type of parametric test that can accommodate some deviations from the assumed distribution, but they still require some level of normality. Distributional tests are not a recognized category of statistical tests.

Question 16

The _____ involves converting a set of scores on a variable to ranks and then analyzing these rank scores using the traditional parametric formulas.
A)rank order approach
B)rank transformation approach
C)parametric approach
D)traditional approach

Accepted Answer

The rank transformation approach involves converting a set of scores on a variable to ranks and then analyzing these rank scores using the traditional parametric formulas. This approach is used when the data do not meet the assumptions of normality or equal variances required for parametric tests.

Question 17

Distribution-free tests are used when _____.
A)samples are larger than 40
B)parametric assumptions are markedly violated
C)data are normally distributed
D)none of these

Accepted Answer

Distribution-free tests, also known as nonparametric tests, do not make assumptions about the underlying distribution of the data. They are typically used when the parametric assumptions (such as normality) are violated or when the sample size is not large enough for parametric tests. Therefore, the best choice is B, because it accurately describes the situations in which distribution-free tests are used.

Question 18

Two types of nonparametric tests are _____ and _____ tests.
A)frequency;rank
B)frequency;sign
C)rank;variance
D)rank;sign

Accepted Answer

The two types of nonparametric tests are rank tests, such as the Wilcoxon signed-rank test or Mann-Whitney U test, and sign tests. The other two options, frequency and variance tests, are not types of nonparametric tests.

Question 19

Most of the parametric tests that we have considered have a nonparametric counterpart in the sense that the structure of the _____ is comparable across the two methods of analysis.
A)summary table
B)underlying distribution
C)critical ratio
D)independent variable

Accepted Answer

The answer of Most of the parametric tests that we...

Question 20

A nonparametric statistical test will typically be used when data are measured at the _____ or _____ level.
A)nominal;interval
B)nominal;ordinal
C)ordinal;ratio
D)interval;ratio

Accepted Answer

Nonparametric tests are typically used when data are measured at the nominal or ordinal level. Nominal data are categorical and cannot be ranked, while ordinal data can be ranked but the differences between values are not necessarily equal. Nonparametric tests do not assume a specific distribution for the data and can be used when assumptions for parametric tests are not met, such as normality or equal variances.

Quiz 16: Nonparametric Statistics

Which of the following is not a requirement for the Mann-Whitney U?

When one or both sample sizes are smaller than 10,the sampling distribution of the R statistic does not approximate a normal distribution,so application of the _____ is not appropriate.

If one wants to compare the rank ordering of scores as a function of different groups,then a(n) _____ should be used.

You are given the following data: Subject 1 2 3 4 5 Score 10 12 8 10 2 Rank -- -- -- -- -- The ranks for the scores of subjects 1-5 are ____.

There are two major types of nonparametric counterparts to parametric test: _ tests and _ tests.

A benefit of many nonparametric statistical tests is their relative lack of sensitivity to:

The most common approach to dealing with tied scores is to rank the scores by assigning the average rank for ties and then to introduce a(n) _____ into the formula for the test statistic to adjust for the presence of ties.

In cases where a tie occurs,the tied scores are assigned the _____ of the ranks involved.

Rather than comparing groups in terms of _,the nonparametric procedures compare groups in terms of _ or other features of a distribution.

Which of the following is not a requirement for the Wilcoxon ranked-sum test?

Nonparametric tests can be used to analyze _ variables that are measured on a(n) _ level.

The Wilcoxon rank sum test and the Mann-Whitney U test are the nonparametric counterparts of the:

_____ statistical tests are a class of statistics that,in general,make few,if any,distributional assumptions

Extreme values might prove to be problematic for parametric analyses,but they pose no particular problem for many nonparametric analyses.Because of this property,nonparametric tests are often said to be:

Statistical tests that require assumptions about the distribution of scores in the populations from which the samples are selected are called _____ statistical tests.

The _____ involves converting a set of scores on a variable to ranks and then analyzing these rank scores using the traditional parametric formulas.

Distribution-free tests are used when _____.

Two types of nonparametric tests are _ and _ tests.

Most of the parametric tests that we have considered have a nonparametric counterpart in the sense that the structure of the _____ is comparable across the two methods of analysis.

A nonparametric statistical test will typically be used when data are measured at the _ or _ level.