22,823 views
What are parametric and nonparametric tests, what are the conditions and how to test them? Parametric statistics: A set of methods that require specific assumptions about the community from which the sample is drawn, including: normal distribution of individuals in the community or sample, independence, homogeneity of variance. Examples: one-sample t-test, and Pearson's correlation coefficient. Nonparametric statistics: A set of alternative methods used in cases where assumptions about the community from which the sample is drawn are not met, or in the case of nominal and ordinal data. Nonparametric statistics are more flexible than parametric statistics, but their confidence level is lower than parametric ones. Examples: chi-square, Smirnov, Spearman and Kendall tests. Parametric tests require following the normal distribution, while nonparametric tests do not. #spss #parametric_tests