GUARD Focused Lab · Student Card
Bias and Fairness Explorer
Observe how thresholds and unequal opportunity affect people.
GUARD focus: R / D · 12–15 minutes · Classes 9–10
Situation
A school uses a score to shortlist students for a scholarship interview. The score rewards grades, attendance and extracurricular participation. Some students have had fewer opportunities to join paid clubs or travel to competitions. The tool does not make the final award, but students below the threshold are not interviewed.
Learning objective
Recognise that changing a threshold changes outcomes but may not repair biased inputs, and identify stakeholders, remedies and appeal mechanisms.
1. Explore the threshold
Move the threshold and observe selection rates and qualified students missed. Then choose the strongest conclusion.
- Choose whichever threshold selects the same number from each group; that proves the system is fair.
- Threshold changes affect outcomes, but they cannot by themselves fix a score that may encode unequal access; the features and missed students need review.
- Use the threshold that selects the most students because a higher selection rate is always fairer.
Reasoning:
2. Identify affected people
Who must be considered when reviewing this system?
- □ Students selected and not selected, including those with fewer opportunities.
- □ Families who may need to understand or challenge the process.
- □ Teachers and the scholarship committee accountable for the decision process.
- □ The tool provider, which should explain design assumptions and limitations.
- □ Only the vendor, because the score is a technical product.
Reasoning:
3. Choose proportionate remedies
Which controls could improve the process before use?
- □ Review whether extracurricular participation is relevant and whether opportunity differences require removal or adjustment.
- □ Test selection and error patterns by relevant groups, with a clear purpose and safeguards.
- □ Require trained human review of borderline and adverse cases.
- □ Give students a clear explanation and a way to provide missing context or appeal.
- □ Keep the criteria secret so students cannot question the system.
- □ Collect as much sensitive personal data as possible, even without a defined need.
Reasoning:
4. Make the deployment decision
What is the strongest current decision?
- Use the score as the automatic shortlist because a human makes the final award later.
- Pause automatic use, revise and test the model, then use it only as one input with human review, explanation and appeal.
- Delete every use of data in scholarship decisions and select students randomly.
Reasoning:
Final reflection
What evidence or stakeholder consideration most affected your decision?