Categorical and Association

Use this overview to decide when a categorical association test is the right choice, choose between chi-square and Fisher's exact approaches, and understand what belongs in a different methods category.

Use this section when your question is whether categorical variables are associated and you need to test a contingency table rather than model a continuous or time-to-event outcome.

This section is for categorical association tests on count tables, not for mean comparisons, regression modeling, or survival analysis. If your main question is whether group averages differ, how predictors relate to a continuous or binary outcome, or when an event occurs, start with Group Comparison, Regression and Modeling, or Survival Analysis instead.

Categorical methods here cover two related but distinct questions: whether two categorical variables are associated, and whether one categorical variable follows a specified expected distribution. Licklider helps you choose the leaf page that matches that question first, then the table size or count pattern.

Child pages

Start here

  1. Start with Chi-Square Test when expected counts are adequate and you need a general test of association or independence in a contingency table.
  2. Start with Chi-Square Test for categorical association in most cases. When Licklider detects a 2x2 table, it runs Fisher's exact test automatically alongside chi-square and shows both results in the Inspector. You do not need to select Fisher's exact test separately.
  3. Start with Chi-Square Goodness-of-Fit Test when you have one categorical variable and want to compare it against a specified expected distribution rather than test association between two variables.
  • Use related links below when your categorical question also needs setup, checks, figures, or reporting guidance.