Phi is a measure of association used with which type of data?

• A. Two categorical variables with two categories each.
• B. Two continuous variables.
• C. A mix of categorical and continuous variable.
• D. Three categorical variables.

Answer: (A)

Phi measures the strength of association between two binary (dichotomous) variables. It’s computed from a 2x2 contingency table and essentially acts like the Pearson correlation when both variables are coded as 0/1. Because it relies on both variables being dichotomous, it’s not appropriate for two continuous variables, for a mix of categorical and continuous data, or for more than two categories. When you have larger contingency tables, a related measure called Cramer's V is used instead.