Inferential Statistics
mbedded.ninja Author
warning
This page is in notes format, and may not be of the same quality as other pages on this site.
Overview
Inferential statistics are ways of analyzing and interpreting data. Inferential statistics consist of parametric and non-parametric tests. Parametric tests can be used when the data's distribution is sufficiently close to a normal distribution, and if this is not the case, then non-parameteric tests must be used. Non-parametric tests can also be used upon normally-distributed data. So why bother with parametric tests at all? Parametric tests should be preferred over non-parametric tests whenever possible due to their increased statistical "power"...they are better at finding (or proving) relationships between variables.
Examples of parametric tests:
- One-way analysis of variance (ANOVA)
- Paried t-test
- Pearson coefficient of correlation
Examples of non-parametric tests:
- Wilcoxon signed-rank test
- Kruskal-Wallis test
- Spearman's rank correlation