Mann-Whitney vs Chi-Square

#1
What is the difference between a Mann-Whitney U test, and a Chi-square. E.g. for testing marital status as an independent variable to predict QOL (dependent variable), which test would you use?
Is the level of measurement for the independent variable the determining factor for which test you use, or is it the level of measurement for the dependent variable? Or is it both?
 

JohnM

TS Contributor
#2
The test is determined by the level of measurement of both variables.

Mann-Whitney is for situations involving a categorical independent variable (2 levels) and an ordinal (rank) dependent variable. If the IV has >2 levels, use Kruskal-Wallis which is the nonparametric version of ANOVA.

Chi-square is for situations where both the IV and DV are categorical (with any number of levels >= 2).