Non-parametric and related methods
The t‐tests and their extensions ANOVA, ANCOVA and regression all make assumptions about the distribution of the data in the background populations. The assumptions are essentially of two kinds: homogeneity of variance and normality. This chapter focuses on the development of unpaired t‐test. Checki...
Gespeichert in:
1. Verfasser: | |
---|---|
Format: | Buchkapitel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The t‐tests and their extensions ANOVA, ANCOVA and regression all make assumptions about the distribution of the data in the background populations. The assumptions are essentially of two kinds: homogeneity of variance and normality. This chapter focuses on the development of unpaired t‐test. Checking the assumption of normality can be undertaken in one of two ways. Firstly, the chapter shows graphical method, such as a quantile–quantile plot, where normal data displays itself as a straight line. Secondly, it shows statistical test, the Shapiro–Wilk test, that gives a p‐value. Non‐parametric tests include Mann–Whitney U‐test and Wilcoxon signed rank test. The chapter shows advantages and disadvantages of non‐parametric methods. An outlier is an unusual data point well away from most of the data. |
---|---|
DOI: | 10.1002/9781118470961.ch11 |