This is a password protected decision tree
Is your dependent variable a category (described with percentages instead of mean/SD) or a continuous variable (uses range/mean/SD instead of percentages)?
- Use chi-square (if your independent variable is a continuous variable, you will likely need to convert it into a category variable with a smaller number of ordered sublevels).
Is your independent variable a category or a continuous variable?
- Use correlation (Pearson r).
How many category levels?
Two (e.g., gender). More than two (e.g., religions).
- Use group mean difference (t-test).
- Use group mean difference (ANOVA).