Nonparametric Methods
Parametric Methods
Statistical methods that begin with an assumptions about the probability distribution of the population which is often that the population has a normal distribution. A sampling distribution for the test statistic can then be derived and used to make an inference about one or more parameters of the population such as the population mean or the population standard deviation.
Distribution-free Methods
Statistical methods that make no assumption about the probability distribution of the population
Nonparametric Methods
Statistical methods that require no assumptions about the form of the probability distribution of the population and are often refereed to as distribution free methods. Several of the methods can be applied with categorical as well as quantitative data.
Spearman Rank-Correlation Coefficient
A correlation measure of the association between two variables based on rank-ordered data
Sign Test
A nonparametric test for a hypothesis about a population median or for identifying differences between two populations based on matched samples. The data are summarized in tow categories, denoted by a plus sign or a minus sign, and the binomial distribution with p =0.5 provides the sampling distribution for the test statistic
Wilcoxon signed-rank Test
A nonparametric test for the difference between the medians of two populations based on matched samples. The procedure uses quantitative data and is based on the assumption that the distribution of the differences is symmetric. The paired-difference data are used to make an inference about the medians of the two populations. This test can also be used to make inferences about the median of a symmetric population.
Mann-Whitney-Wilcoxon (MWW) Test
A nonparametric test for the difference between two populations based on an independent sample from each population. The null hypothesis is that the two populations are identical. If the assumptions can be made that the populations have the same shape, this test provides an inference about the differences among the medians of the populations
Kruskal-Wallis Test
A nonparametric test for the differences among three or more populations based on the analysis of an independent sample from each population.The null hypothesis is that the two populations are identical. If the assumptions can be made that the populations have the same shape, this test provides an inference about the differences among the medians of the populations.
