SIP

Tests

Name of Test	Assumptions	When to Use
Pearson’s Correlation Coefficient	Variables should be continuous. Data from both variables follow normal distributions. Your data have no outliers. Your data is from a random or representative sample. You expect a linear relationship between the two variables.	When you want to know the relationship between variables when both variables are continuous when you have no covariates
Spearman’s Rank Correlation Coefficient	Variables are at least ordinal or continuous. There is a monotonic relationship between the variables.	When your variables of interest are continuous with outliers or ordinal When you only have two variables
Cochran-Hanzel-Mantel	The odds ratio is the same in the different repeats of the experiment.	Used for repeated tests of independence.
Kendall-Tau	The variables should be measured on an ordinal or continuous scale. There is a monotonic relationship between your variables.	When you only have two variables. When you want to know the relationship between two variables. When your variables of interest are continuous with outliers or ordinal.
Point Biserial	One of your two variables should be measured on a continuous scale. Your other variable should be dichotomous. Your continuous variable should be approximately normally distributed for each category of the dichotomous variable. Your continuous variable should have equal variances for each category of the dichotomous variable.	When your variables of interest include one continuous and one binary variable. When you only have two variables. When you want to know the relationship between the variables.
Chi-Square Test of Independence	Your two variables should be measured at an ordinal or nominal level. Your two variables should consist of two or more categorical, independent groups.	When you want to test the difference between two variables. When your variable of interest is proportional or categorical. when you have two or more options. When you have independent samples.
Analysis of Variance (ANOVA)	The dependent variable should be measured at the interval or ratio level. The independent variable should consist of two or more categorical, independent groups. There is independence of observations, which means that there is no relationship between the observations in each group or between the group themselves. There should be no significant outliers. The dependent variable should be approximately normally distributed for each category of the independent variable. There needs to be homogeneity of variances.	When you want to test a hypothesis between groups.

Source