第二十一章 無母數統計法 無母數 ... - pu.edu.tw

1 T-test F-test ANOVA Non_para 1 t- nominal or categorical data ordinal or rank data measurement or interval data ratio data Non_para 2 Chi-square test, =3:1 Kolmogorov-Smirnov test Shapiro-Wilk, Cramer-von Mises, Anderson-Darling SPSS SAS Non_para 3 1 2x2 table Fisher exact test ( ) rxc table Chi-square test ( ) Likehood Ratio Chi-square test SAS SPSS Non_para 4 a SAS Table of beat by weight beat(beat) weight(weight)

2 Total N O med 64 65 129 over 50 67 117 under 51 43 94 Total 165 175 340 Statistic DF Value Prob Chi-Square 2 Likelihood Ratio Chi-Square 2 Mantel-Haenszel Chi-Square 1 Phi Coefficient Contingency Coefficient Cramer's V Frequency Col Pct Non_para 5 2 Mann-Whitney test ( ) ( ) SPSS SAS ANOVA Non_para 6 Wilcoxon Two-Sample Test Statistic (S) Normal Approximation Z One-Sided Pr < Z Two-Sided Pr > |Z| t Approximation One-Sided Pr < Z Two-Sided Pr > |Z| Exact Test One-Sided Pr <= S Two-Sided Pr >= |S - Mean| SAS Wilcoxon Non_para 7 T-Tests Variable Method Variances DF t Value Pr > |t| act Pooled Equal 34 act Satterthwaite Unequal 33 Equality of Variances Variable Method Num DF Den DF F Value Pr > F act Folded F 19 15 SAS T-test Tests for Normality Test Statistic p Value Shapiro-Wilk W Pr < W Kolmogorov-Smirnov D Pr > D > Cramer-von Mises W-Sq Pr > W-Sq > Anderson-Darling A-Sq Pr > A-Sq > Non_para 8 1

3 McNenar Change test (p613) A B 2 Wilcoxon signed rank test ( ) SPSS Non_para 9 1 Chi-square test ( ) 2 CRD Kruskal-Wallis test ANOVA Dunn test ( ) SPSS K SAS ANOVA SPSS SAS Non_para 10 1 2 CRD Non_para 11 11 a friends and social life b Job or primary activity c Health and physical condition (well-being) SingleMarriedWindowed ordivorceda414942b275033c122125 Non_para 12 12 2 p-value= = Statistic DF Value Prob Chi-Square 4 Likelihood Ratio Ci-Square 4 Mantel-Haenszel Chi-Square 1 Phi Coefficient Contingency Coefficient Cramer's V SAS Chi-square test Wilcoxon Scores (Rank Sums)

4 For Variable weight Classified by Variable baby Sum of Expected Std Dev Mean baby N Scores Under H0 Under H0 Score ---------------------------------------- ---------------------------- 1 5 2 6 3 5 Average scores were used for ties. Kruskal-Wallis Test Chi-Square DF 2 Pr > Chi-Square SAS K-W Non_para 13 p-value= = 1 Chi-square test 2 Spearmann coef.

5 Of correlation ( ) Kendall s coefficient of concordance ( ) SPSS SAS Non_para 14 Non_para 15 15 Job-training program PoorFairGoodBelow average676425 Average427656 Above average102337 Non_para 16 p-value < = Statistic DF Value Prob Chi-Square 4 <.0001 Likelihood Ratio Chi-Square 4 <.0001 Mantel-Haenszel Chi-Square 1 Phi Coefficient Contingency Coefficient Cramer's V SAS Chi-square test Non_para 17 17 5 < 5 2 ( ) Non_para 18 0 10 20 30 40 50 60 single married wid_div a b c 0 10 20 30 40 50 60 70 80 poor fair good avg- avg avg+ Paired T-test Wilcoxon test T-test Mann-Whitney test (Wilcoxon test) CRD 1-way ANOVA F-test Kruskal-Wallis H test Pearson c.

6 C. Spearman Non_para 19