AP CSP Practice: Algorithmic Bias & Data Ethics
Share
Practice Question
Why B is Correct
This is a classic example of algorithmic bias. When an algorithm learns from historical data that contains bias, it can:
- Learn the bias: If most successful candidates came from certain universities, the algorithm will favor those universities
- Perpetuate the bias: New candidates from other qualified universities may be unfairly screened out
- Amplify the bias: Over time, the company hires even more from the favored universities, reinforcing the pattern
The Feedback Loop Problem
Biased Historical Data
↓
Algorithm Learns Bias
↓
Algorithm Makes Biased Decisions
↓
New Biased Data is Created
↓
(Cycle Repeats - Bias Amplifies)
Common Mistakes
Speed and storage are technical constraints, not the PRIMARY concern here. The question specifically mentions the data source (past hiring decisions), which points to a data quality/bias issue.
Nothing in the scenario suggests the algorithm can't handle international applications. The issue is about WHICH candidates the algorithm favors, not which ones it can process.
Whenever a question mentions an algorithm being "trained on data" or "learning from past decisions," immediately consider whether that data might contain biases. This is a frequently tested concept on the AP CSP exam.
Want More AP CSP Practice?
Get personalized help from an experienced AP CS teacher
AP CSP Study Guide Schedule 1-on-1 Tutoring