Effect sizes
Cramér's V
Implemented
effect-size.cramers_vDefinition
Strength of association between two categorical variables.
Formula
V = √(χ² / (N · (min(r, c) − 1)))
Range: [0, 1]
Inputs and outputs
- contingency table
Returns: EffectSizeResult
Assumptions
No assumptions beyond valid, aligned inputs of the documented types.
Limitations
No metric-specific limitations are documented yet. Interpret the value alongside the task, data, and other metrics.
Python API
import evalsuite as es
es.cramers_v(table)References
- Cramér, H. (1946). Mathematical Methods of Statistics. Princeton University Press.