| Summary: | Statistical inference is an essential element within the experimental scientific method. Inference processes can be developed through the application of parametric quantitative methods or using non-parametric statistical models. The former is used when strict assumptions are met about the shape of the probability distribution of the population from which the random sample is to be selected as the basis for estimating one or more unknown population parameters, constructing confidence intervals for such parameters or determining critical regions to reject the null hypothesis. Non-parametric statistical models are statistical techniques that require more flexible assumptions or use distribution-free test statistics, which are generally defined by the ranges associated with the data of the variable or variables to be investigated, whether it is an inference problem associated with one random sample, two samples, or three or more samples.
This research work focuses on the way in which some university textbooks address the topics of non-parametric statistics associated with hypothesis testing. For this purpose, a methodology based on a mixed approach is used, since it is supported by both quantitative and qualitative research methods (content analysis and documentary research). In addition, it considers several concrete problem situations, susceptible to promote scientific research from university cloisters. This book is aimed at students, professors, professionals and researchers interested in applying classical non-parametric statistical models in hypothesis testing in their research work or in increasing their knowledge about such models. To approach it, basic knowledge of descriptive and inferential statistics from a parametric perspective is needed.
|
|---|
