Question 1

A dependent variable can still be normally distributed,even if it has only two or three values and,hence,the assumption of nonparametric tests will be violated.

Accepted Answer

A normal distribution requires a continuous variable with a range of values, whereas having only two or three values does not allow for a continuous distribution. Therefore, if the dependent variable has only two or three values, it cannot be normally distributed.

Question 2

The inference of the relationship is indexed by eta-squared for all statistical test.

Accepted Answer

Eta-squared is a measure of effect size for the relationship between variables specifically in the context of an ANOVA (Analysis of Variance) test, not for all statistical tests.

Question 3

The use of Case II statistics is appropriate when a qualitative variable has many values and there are insufficient number of cases within each level to justify the analysis of means,medians,and the like.

Accepted Answer

The answer of The use of Case II statistics is...

Question 4

The major factor to be weighed when deciding whether to use a parametric or a nonparametric test is the decision about what aspect of the distribution that you want to compare groups on.

Accepted Answer

The decision about what aspect of the distribution that you want to compare groups on is the main factor in deciding whether to use a parametric or nonparametric test. Parametric tests assume that the data follows a normal distribution and have specific requirements, while nonparametric tests do not make assumptions about the distribution of the data. Therefore, the decision about the aspect of the distribution being compared is crucial in selecting the appropriate test.

Question 5

If means (or linear relationships)are the focus,then a nonparametric test is the analysis of choice.

Accepted Answer

Nonparametric tests do not assume a linear relationship between variables and can be used when assumptions of parametric tests are not met. However, they may not be the analysis of choice if means or specifically linear relationships are the focus, as parametric tests may be more appropriate in those situations.

Question 6

When both variables under study are qualitative and between-subjects in nature,the appropriate method of data analysis is the chi square test as discussed in Chapter 15.

Accepted Answer

The chi square test is appropriate when analyzing data with two qualitative variables that are between-subjects in nature.

Question 7

If ranks are the focus,then parametric methods should be pursued.

Accepted Answer

If ranks are the focus, non-parametric methods should be pursued as they are designed to analyze ordinal (ranked) data.

Question 8

A Two-way ANOVA is used when there are two independent variables.

Accepted Answer

A Two-way ANOVA is also known as a two-factor ANOVA, and it is used when there are two independent variables. It is used to analyze the effects of both independent variables on the dependent variable.

Question 9

When one or both variables are measured on an ordinal level that seriously departs from interval level characteristics,Pearson correlation is the test of choice.

Accepted Answer

Pearson correlation assumes interval level data and can produce misleading results when used with ordinal data. Instead, Spearman or Kendall correlation would be more appropriate for ordinal data.

Question 10

In situations where the dependent measure has only two or three values,most researchers will not use a parametric test.

Accepted Answer

When the dependent measure has only two or three values, it is often categorical and not continuous. In such cases, non-parametric tests, such as chi-square or Fisher's exact test, may be more appropriate than parametric tests, which assume normality and equal variances in the populations being compared.

Question 11

Generally speaking,parametric methods are not conducive to analyzing qualitative variables.

Accepted Answer

Parametric methods assume that the data is normally distributed and have equal variances, making them less suitable for analyzing qualitative variables which do not have a numerical value or fixed range. Non-parametric methods are more appropriate in such cases.

Question 12

If the dependent variable is measured on an ordinal level that seriously departs from interval level characteristics,a parametric test is always the optimal test.

Accepted Answer

If the dependent variable is measured on an ordinal level and departs significantly from interval level characteristics, a non-parametric test would be more appropriate. Parametric tests assume normally distributed data and equal variances, which may not be valid for ordinal data. Non-parametric tests, such as the Mann-Whitney U test, do not make these assumptions and are better suited for ordinal data.

Question 13

The observed distribution may bias one's decision about a parameter to focus on.

Accepted Answer

If the observed distribution is not representative of the population distribution, it may lead to biased parameter estimates and influence decision-making.

Question 14

A non-parametric test should be pursued if there is reason to believe that the distributional assumptions of the corresponding parametric test have been violated to the extent that the parametric test is not robust.

Accepted Answer

Non-parametric tests do not make any assumptions about the distribution of the population, and therefore they can be used when the assumptions of the parametric test have been violated.

Question 15

The choice of which distributional parameter to compare across groups is often dictated by theory or the substantive question of interest.

Accepted Answer

It is true that the choice of distributional parameter to compare across groups is often based on theory or the research question being studied. For example, if researchers are studying differences between two groups in terms of their means, then they may choose to compare the group means. Alternatively, if the research question is about differences in variability or spread across groups, then researchers may choose to compare the standard deviations or variances of the two groups.

Question 16

Given two between-subject variables that are measured on a level that at least approximates interval characteristics,the most common method of analysis is the chi square test.

Accepted Answer

The most common method of analysis for two between-subject variables that are measured on a level that at least approximates interval characteristics is the independent samples t-test or ANOVA, not the chi square test. The chi square test is used for categorical data.

Question 17

With only two or three values on the dependent variable,there will be a large number of ties in ranks,and numerous ties can create serious problems for ranked-based approaches.

Accepted Answer

When there are many ties in ranks, it can be difficult to accurately rank and compare the variables using ranked-based approaches.

Question 18

The appropriate statistical methods for Case III situations are techniques called logistic regression and polychotomous logistic regression.

Accepted Answer

Logistic regression and polychotomous logistic regression are appropriate for Case III situations where the dependent variable is categorical and has more than two categories.

Question 19

An investigator should never focus on multiple parameters (means,medians,ranks).

Accepted Answer

A good investigator should consider multiple parameters while analyzing data to get a complete picture of the situation. Focusing on only one parameter may lead to inaccurate conclusions or missing important information.

Question 20

If both variables are qualitative,but one or both are within-subjects in nature,the appropriate method of data analysis is the chi square test as discussed in Chapter 15..

Accepted Answer

The chi-square test is used for analyzing categorical data from independent groups, not for within-subjects (repeated measures) designs. For within-subjects designs involving qualitative data, methods such as McNemar's test or Cochran's Q test are more appropriate.

Quiz 18: Overview and Extension: Statistical Tests for More Complex Designs