668 CHAPTER 13 Nonparametric Tests Wilcoxon Rank-Sum Test Access tech supplements, videos, and data sets at www.TriolaStats.com TECH CENTER Statdisk 1. Click Analysis in the top menu. 2. Select Wilcoxon Tests from the dropdown menu and Wilcoxon (Independent Samples) from the submenu. 3. Enter a significance level and select the two data columns to include. 4. Click Evaluate. StatCrunch 1. Click Stat in the top menu. 2. Select Nonparametrics from the dropdown menu and Mann-Whitney from the submenu. 3. Select the columns to be used for the two samples. 4. Select Hypothesis test for m1-m2 and for H0 enter the value of the claimed difference. For HA select the desired format. 5. Click Compute! Minitab 1. Enter the two sets of sample data into columns C1 and C2. 2. Click Stat in the top menu. 3. Select Nonparametrics from the dropdown menu and select Mann-Whitney from the submenu. 4. For First Sample select C1 and for Second Sample select C2. 5. Enter the confidence level (95.0 corresponds to a significance level of a = 0.05). 6. For Alternative, choose the format of the alternative hypothesis (not equal corresponds to a two-tailed hypothesis test). 7. Click OK. TI-83, 83 84 Plus Calculator 8 us Ca cu ato Requires programs RSTEST and ZZRANK (available at TriolaStats. com) 1. Enter the two sets of sample data in list L1 and L2. 2. Press N, select RSTEST, and press [ twice. 3. For GROUP A = enter L1 and press [. For GROUP B = enter L2 and press [. 4. The rank sum R, mean, standard deviation and test statistic z will be calculated based on the sample with the fewer number of values. Press [ again to get the test statistic z. Find the critical value by referring to Table A-2 or using the normalcdf function as described in the Tech Center in Section 6-1. TIP: The list name L1 (and L2 . . . L6) can be quickly entered by pressing F1. R R command: wilcox.test(x, y, paired=FALSE) where x, y include independent data. A complete list of R statistical commands is available at TriolaStats.com Excel XLSTAT Add-In (Required) 1. Click on the XLSTAT tab in the Ribbon and then click Nonparametric tests. 2. Select Comparison of two samples from the dropdown menu. 3. Enter the data range for each sample in the Sample 1 & 2 boxes. Check the Column labels box if the data range includes labels. 4. Select One column per sample under Data format. 5. Check the Mann-Whitney test option only. 6. Click the Options tab. 7. Under Alternative hypothesis select Sample 1 – Sample 2 3 D. Confirm Hypothesized difference (D) is 0. 8. Enter a significance level and check the Exact p-value box. 9. Click OK. TIP: Because XLSTAT uses a different procedure than the one described in this section, results may be somewhat different, especially for small samples. Statistical Literacy and Critical Thinking 1.Heights of Females from ANSUR I and ANSUR II Example 1 in this section used samples of heights of males from Data Set 1 “ANSUR I 1988” and Data Set 2 “ANSUR II 2012.” Listed below are samples of heights (mm) of females from those same data sets. Are the requirements for using the Wilcoxon rank-sum test satisfied? Why or why not? 13-4 Basic Skills and Concepts
RkJQdWJsaXNoZXIy NjM5ODQ=