So you’ve collected your data, and now you want to compare those pesky categorical variables with each other. The analysis you’re looking for is something called Pearson’s Chi Square. The chi-square test will compare frequencies that you observed (i.e., in your data) to those expected by chance. This seems easy enough, but when it comes to effect size you have two options: Phi and Cramer’s V. Which to use, you ask? According to Field (2009), Phi works well if all variables only have two categories. If any of the variables being compared have more than two categories, then Cramer’s V is preferred. Statistics.laerd.com provides a good overview, with SPSS screenshots.

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.