This document provides an overview of chi-square distribution tests and nonparametric tests. It defines chi-square tests as non-parametric tests used to measure differences between observed and expected values. Three main applications are listed: goodness of fit tests, tests of independence, and tests of homogeneity. Nonparametric tests are described as not assuming any underlying data distribution and appropriate when normality is violated. Several common nonparametric tests are outlined, including sign tests, Wilcoxon signed-rank tests, Friedman tests, and Kruskal-Wallis tests.
This document provides an overview of chi-square distribution tests and nonparametric tests. It defines chi-square tests as non-parametric tests used to measure differences between observed and expected values. Three main applications are listed: goodness of fit tests, tests of independence, and tests of homogeneity. Nonparametric tests are described as not assuming any underlying data distribution and appropriate when normality is violated. Several common nonparametric tests are outlined, including sign tests, Wilcoxon signed-rank tests, Friedman tests, and Kruskal-Wallis tests.
This document provides an overview of chi-square distribution tests and nonparametric tests. It defines chi-square tests as non-parametric tests used to measure differences between observed and expected values. Three main applications are listed: goodness of fit tests, tests of independence, and tests of homogeneity. Nonparametric tests are described as not assuming any underlying data distribution and appropriate when normality is violated. Several common nonparametric tests are outlined, including sign tests, Wilcoxon signed-rank tests, Friedman tests, and Kruskal-Wallis tests.
CHI-SQUARE DISTRIBUTION TEST o One or more assumptions of a parametric test have been violated A test that is used to measure the differences between what o Your sample size is too small to run a parametric is observed and what is expected according to an assumed test hypothesis o Your data has outliers that cannot be removed a non-parametric test based on frequencies o You want to test for the median rather than the The test is used for testing the hypothesis and is not useful mean for estimation This test is an important non-parametric test as no rigid DATA TYPES: assumptions are necessary in regard to the type of o Rule of Thumb: population, no need of parameter values, and relatively Parametric test – interval or ratio scales less mathematical details are involve Non parametric test – nomina or ordinal Application of Chi-Square Test: scales o Goodness of Fit distribution o Main Nonparametric test: o Test of Independence 1-sample sign test – use this test to o Test of Homogeneity estimate the median of a population and compare it to a reference value or target I. Goodness of Fit distribution value This test enables us to see how well does the assumed 1-sample Wilcoxon signed rank test – theoretical distribution fit to the observed data. the test assumes that the data comes Assumptions: from a symmetric distribution o The data are obtained from a random sample. Friedman test – used to test for o The expected frequency for each category differences between groups with ordinal must be 5 or more dependent variables. It can also be used o Note: This test is a right-tailed test, since for continuous data if the one-way when the O - E values are squared, the ANOVA with repeated measures is answer will be positive or zero. inappropriate Goodman Kruska’s Gamma – a test of II. Test of Independence association for ranked variables Is used to test the independence of two variables Kruskal-Wallis Test – use this test For example, suppose a new postoperative procedure is instead of a one-way ANOVA to find out administered to a number of patients in a large hospital. if two or more medians are different. The researcher can ask the question, Do the doctors Ranks of the data points are used for feel differently about this procedure from the nurses, or calculations, rather than the data oints do they feel basically the same way? themselves. Note that the question is not whether they prefer the Mann-Kendall Trend Test – looks for procedure but whether there is a difference of opinion trends in time-series data between the two groups Mann-Whitney Test – use this test to compare differences between two III. Test of Homogeneity independent groups when dependent Samples are selected from several different variables are either ordinal or continuous populations, and the researcher is interested in Mood’s Median Test – use this test determining whether the proportions of elements that instead of sign test when you have two have a common characteristic are the same for each independent samples population. Spearman Rank Correlation – use when you want to find a correlation between two sets of data NONPARAMETRIC TEST REFERENCES: Bluman, A. G. (2012). Elementary statistics: a step by step Does not assume anything about the underlying approach (8th ed.). 1221 Avenue of the Americas,New York, distribution NY 10020: McGraw-Hill Companies It is used when the data is not normal Daniel, W. W. (2005). Biostatistics: A foundation for analysis in the health sciences (8th ed.). Philippines: Mindmover Other reasons to run nonparametric tests: Publishing. doi:570.15195/D221 Nuevo, J. M. (2019). Biostatistics and epidemiology. Powerpoint presentation.