Does transposing a contingency table change Pearson chi-square or Cramér's V?
StatisticsLevel: FoundationBinary and categorical inference
Does transposing a contingency table change Pearson chi-square or Cramér's V?
Hints
- Transpose expected counts too.
- Check the symmetry of the normalization.
Answer
No. Cell contributions and n stay the same, and min(r−1,c−1) is symmetric. The verbal interpretation of row versus column roles may change.
Source and licence: SciPy — contingency tables · BSD-3-Clause · answer: Quantoros editorial