


Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Information on nonparametric tests and their parametric equivalents. It explains the differences between parametric and nonparametric tests, when to use Mann-Whitney/Wilcoxon rank-sum tests, and how to conduct Wilcoxon rank-sum and Mann-Whitney tests in SPSS. It also covers tests of normality and homogeneity of variance, as well as effect size and write-up of results for these tests. useful for students studying statistics or research methods.
Typology: Exams
1 / 4
This page cannot be seen from the preview
Don't miss anything!



Parametric vs. nonparametric tests - Correct answer -parametric tests use sample statistics to estimate population parameters requiring underlying assumptions be met -nonparametric tests don't have the same stringent assumptions and can be used when assumptions of parametric tests are not met Parametric equivalent of Mann-Whitney/Wilcoxon rank-sum tests
-tied ranks are given the same ranks: the average of potential ranks Tests of normality - Correct answer -Kolmogorov-Smirnov -Shapiro-Wilk Test of homogeneity of variance - Correct answer -Levene's test How to conduct Wilcoxon rank-sum and Mann-Whitney tests in SPSS - Correct answer -Analyze-->nonparametric tests--
independent samples -Fields-->use custom field assignments -move DVs over to test fields -groups = IV -Settings-->customize tests-->Mann-Whitney -double click on hypothesis test summary-->Model Viewer, gives you Wilcoxon results and statistical information Effect size for Wilcoxon rank-sum and Mann-Whitney tests - Correct answer -r = Z / sqrt (N) -Z = z-score that SPSS produces ("standardized test statistic") -N = sample size (number of total observations) Write-up of Mann-Whitney - Correct answer -use median (Mdn) as central tendency instead of mean because distribution is not normal -U = #, z = #, p < #, r = # Write-up of Wilcoxon rank-sum test - Correct answer -use median (Mdn) as central tendency instead of mean because distribution is not normal -Ws = #, z = #, p < #, r = #
-two way--determine whether there's a relationship between 2 categorical variables; known as a goodness-of-fit test or Pearson's chi-square -for two-way, df = (# of rows - 1)(# of columns - 1)