A data scientist evaluates a loan approval model and finds the approval rate for Group A is 70% while Group B's approval rate is 50%. The resulting value of 0.20 is used to flag the model for a fairness review. Which fairness metric did the data scientist calculate?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Think of it this way: in real-world AI governance, statistical parity difference is exactly what teams reach for when they need to handle this scenario. Statistical parity difference subtracts the positive outcome rate of one demographic group from another. On the exam, remember that this falls squarely under the 1.0 AI Governance Overview domain.
Full explanation below image
Full Explanation
Statistical parity difference subtracts the positive outcome rate of one demographic group from another. A result of 0 indicates perfect parity. The value of 0.20 calculated here indicates Group A receives favorable outcomes 20 percentage points more often than Group B. The correct answer, "Statistical parity difference", directly addresses the scenario described because it aligns with the specific governance requirement in question. The incorrect options ("Disparate impact ratio", "Equalized odds", "Demographic parity") may seem plausible but do not satisfy the core requirement. Understanding the distinction between these concepts is critical for IBM watsonx.governance implementations and is frequently tested in the 1.0 AI Governance Overview section of the certification exam.