Course contentsShow
← Data Science
Lesson 1888 of 2,14544. Data Ethics and PrivacyFree lesson

Protected Classes and Sensitive Attributes

Identify legally protected characteristics (race, gender, age) and understand why they require special consideration in analysis.

Protected Classes and Sensitive Attributes

You'll learn: Which personal characteristics have legal protections and why data scientists must handle them with extreme care.

The Legal and Ethical Foundation

Protected classes are groups of people shielded by law from discrimination. In the United States, federal laws like the Civil Rights Act, Age Discrimination in Employment Act, and Americans with Disabilities Act protect characteristics including:

  • Race and ethnicity
  • Gender and sex
  • Age (typically 40+)
  • Religion
  • Disability status
  • National origin
  • Pregnancy status

Many countries have similar or broader protections (e.g., sexual orientation, marital status).

Why Special Consideration Matters

Think of these attributes like radioactive materials—they require special handling protocols. Using protected attributes in predictive models can:

  1. Violate anti-discrimination laws (e.g., using age to deny credit)
  2. Create disparate impact even without intent (your model systematically disadvantages a protected group)
  3. Amplify historical inequities baked into training data
  4. Damage vulnerable populations through automated decisions

The Data Science Challenge

You're not just avoiding explicitly using gender or race columns. Sensitive attributes leak through proxy variables—attributes correlated with protected classes. ZIP code proxies for race; first names proxy for gender and ethnicity; browsing history proxies for health conditions.

This is why after detecting bias (lesson 1884), you must identify both the protected classes explicitly present in your data and potential proxies hiding them. Fair modeling requires understanding which attributes demand scrutiny, documentation, and potential exclusion or special treatment.

Key Takeaway: Protected classes are legally shielded characteristics (race, gender, age, etc.) that require special consideration because using them—or their proxies—in analysis can perpetuate discrimination, violate laws, and cause real harm to vulnerable groups.