Manual de estadística no paramétrica aplicada a los negocios

This document raises the application of non-parametric statistics in quantitative business research, for which a presentation is made of each technique and / or limited statistics, its corresponding applications and limitations; its main purpose is to provide an alternative to researchers when they have data that do not satisfy the assumptions of parametric statistics. The present research work, in the light of what is proposed by Siegel (1957), takes into account this circumstance precisely, and deals with the presentation of the techniques and / or statistical tests not paramé referring to data that are measured on a scale. nominal or ordinal: - Case of a sample: binomial test, chi-square test, Kolmogorov-Smirnov test, runs test. - Case of two independent samples: chi-square test, Kolmogorov-Smirnov test, Mann-Whitney U test, Moses extreme reactions test, Wald-Wolfowitz streak test, Fischer exact test. - Case of two related samples: McNemar test, Wilcoxon test, sign test. - Case of k independent samples: chi-square test, Kruskal-Wallis test, median test, Jonckheere test. - Case of k related samples: Cochran's Q test, Friedman's test, Kendall's W test. - Non-parametric measures of correlation: contingency coefficient, Spearman rank correlation coefficient, Kendall rank correlation coefficient.