Parity Difference

Overview

The difference in true values between privileged groups vs unprivileged groups (ex. how many men were hired - how many women were hired)

Formula

Parity Difference = Privileged Selection Rate - Unprivileged Selection Rate

Source

Where:

• Selection Rate = # of samples with positive label / total samples

Source

Usage

Manually

Python

# Calculate parity difference
parity_diff = parity_difference(df, protected_attribute, privileged_group, labels, positive_label)

print("Parity Difference:", parity_diff)

Using Fairness Object

parity_diff = (fo.compute(parity_difference))

Results

Parity Difference: -0.045454545453677664

These results are obtained by using the input data given in the Create Example Data page under Getting Started

Interpretation

The Parity Difference metric quantifies the difference in performance between two groups or populations based on a chosen performance metric. A positive Parity Difference indicates that the first group has a higher performance metric compared to the second group, while a negative value indicates the opposite.