WICH TEST SHOULD I USE (Example in R) Modified from: http://www.ats.ucla.edu/stat/mult_pkg/whatstat/choosestat.html (2016)
Independent Variables | Dependent Variables |
| |
| Number | Nature | Test(s) |
|
1 | interval & normal | one-sample t-test |
| ordinal or interval | one-sample median | |
| categorical (2 categories) | binomial test | |
0 IVs (1 population) | categorical | Chi-square goodness-of-fit | |
| interval & normal | 2 independent sample t-test | |
| ordinal or interval | Wilcoxon-Mann Whitney test | |
1 IV with 2 levels (independent |
| Chi-square test | |
groups) | categorical | Fisher's exact test | |
| interval & normal | one-way ANOVA | |
1 IV with 2 or more levels | ordinal or interval | Kruskal Wallis | |
(independent groups) | categorical | Chi-square test | |
| interval & normal | paired t-test | |
1 IV with 2 levels | ordinal or interval | Wilcoxon signed ranks test | |
(dependent/matched groups) | categorical | McNema | |
| interval & normal | one-way repeated measures ANOVA | |
1 IV with 2 or more levels | ordinal or interval | Friedman test | |
(dependent/matched groups) | categorical | repeated measures logistic regression | |
| interval & normal | factorial ANOVA | |
2 or more IVs (independent | ordinal or interval | ordered logistic regression | |
groups) | categorical | factorial logistic regression | |
| interval & normal | correlation | |
| interval & normal | simple linear regression | |
| ordinal or interval | non-parametric correlation | |
1 interval IV | categorical | simple logistic regression | |
|
| multiple regression | |
| interval & normal | analysis of covariance | |
1 or more interval IVs and/or 1 | categorical | multiple logistic regression | |
or more categorical IVs | discriminant analysis | ||
1 IV with 2 or more levels |
2+ |
|
|
(independent groups) | interval & normal | one-way MANOVA | |
2+ | interval & normal | multivariate multiple linear regression | |
0 | interval & normal | factor analysis | |
0 | 2 sets of 2+ | interval & normal | canonical correlation |