Solutions Manual For Categorical and Nonparametric Data
Analysis Choosing the Best Statistical Technique 1st Edition By
E. Michael Nussbaum 9781848726031 ALL Chapters .
Interval Estimate - ANSWER: A range of plausible values for a population parameter.
Confidence Level - ANSWER: The probability that a confidence interval contains the population
parameter.
Interpretation of 95% Confidence Interval - ANSWER: 95% of confidence intervals from repeated
samples will contain the population parameter.
Effect Size Estimate - ANSWER: A confidence interval that provides an estimate of effect size.
Interpretation of Confidence Interval for t-test - ANSWER: If the CI contains zero, null hypothesis is not
rejected; if not, null hypothesis is rejected.
Eta Squared (η²) Formula - ANSWER: η² =
Meaning of Eta Squared Statistic - ANSWER: Percentage of total variation due to the effect of the
Independent Variable.
Cohen's Small Effect (η²) - ANSWER: 0.1
Cohen's Medium Effect (η²) - ANSWER: 0.6
Cohen's Large Effect (η²) - ANSWER: 0.14
Standard Assumptions for Parametric Tests - ANSWER: (A) Normally distributed dependent variable,
(B) Interval or ratio data
Use of Non-Parametric Test - ANSWER: Small population size or ordinal/nominal data
Mann-Whitney Test Design - ANSWER: Between-Subject, One Factor, Two Level Design
Null Hypothesis for Mann-Whitney Test - ANSWER: Median for Group 1 = Median for Group 2
Logic/Theory of Mann-Whitney Test - ANSWER: Rank scores, check if higher ranks in one group and
lower ranks in another
Wilcoxon Signed Ranks Test Design - ANSWER: Repeated Measures, One-Factor, Two-Level Design
Null Hypothesis for Wilcoxon Test - ANSWER: Median Condition 1 = Median Condition 2
Logic/Theory of Wilcoxon Test - ANSWER: Rank differences, compare ranks of positive vs negative
differences
Kruskal-Wallis Test Design - ANSWER: Between-Subjects One-Factor, Multiple Level Design
Null Hypothesis for Kruskal-Wallis Test - ANSWER: Before Ranks = After Ranks
Logic/Theory of Kruskal-Wallis Test - ANSWER: Rank scores, check if ranks in one group and low ranks
in another
Analysis Choosing the Best Statistical Technique 1st Edition By
E. Michael Nussbaum 9781848726031 ALL Chapters .
Interval Estimate - ANSWER: A range of plausible values for a population parameter.
Confidence Level - ANSWER: The probability that a confidence interval contains the population
parameter.
Interpretation of 95% Confidence Interval - ANSWER: 95% of confidence intervals from repeated
samples will contain the population parameter.
Effect Size Estimate - ANSWER: A confidence interval that provides an estimate of effect size.
Interpretation of Confidence Interval for t-test - ANSWER: If the CI contains zero, null hypothesis is not
rejected; if not, null hypothesis is rejected.
Eta Squared (η²) Formula - ANSWER: η² =
Meaning of Eta Squared Statistic - ANSWER: Percentage of total variation due to the effect of the
Independent Variable.
Cohen's Small Effect (η²) - ANSWER: 0.1
Cohen's Medium Effect (η²) - ANSWER: 0.6
Cohen's Large Effect (η²) - ANSWER: 0.14
Standard Assumptions for Parametric Tests - ANSWER: (A) Normally distributed dependent variable,
(B) Interval or ratio data
Use of Non-Parametric Test - ANSWER: Small population size or ordinal/nominal data
Mann-Whitney Test Design - ANSWER: Between-Subject, One Factor, Two Level Design
Null Hypothesis for Mann-Whitney Test - ANSWER: Median for Group 1 = Median for Group 2
Logic/Theory of Mann-Whitney Test - ANSWER: Rank scores, check if higher ranks in one group and
lower ranks in another
Wilcoxon Signed Ranks Test Design - ANSWER: Repeated Measures, One-Factor, Two-Level Design
Null Hypothesis for Wilcoxon Test - ANSWER: Median Condition 1 = Median Condition 2
Logic/Theory of Wilcoxon Test - ANSWER: Rank differences, compare ranks of positive vs negative
differences
Kruskal-Wallis Test Design - ANSWER: Between-Subjects One-Factor, Multiple Level Design
Null Hypothesis for Kruskal-Wallis Test - ANSWER: Before Ranks = After Ranks
Logic/Theory of Kruskal-Wallis Test - ANSWER: Rank scores, check if ranks in one group and low ranks
in another
