Skip to content

What do the statistical columns in group comparison tables mean?

1 min readAnalysis

Overview

When a code-by-group table reports a test, it compares how a code is spread across the groups you chose. It uses chi-square, or Fisher’s exact test for two-group tables when an expected count is below 5. “adj. p” is the Benjamini-Hochberg adjusted p-value across the codes shown, and V is Cramér’s V, an effect size. A dagger († ) marks a cell where an expected count is below 5. These figures are marked exploratory.

What to do

  1. Read the group sizes first; small groups make any test weak.
  2. Look at the effect size (V) as well as adj. p.
  3. Treat a dagger as a warning and report the counts.
  4. Open the evidence behind any contrast before describing it.

Terms in this answer