ACCT20007 Chap.6 Data Ethics and Privacy
Data Ethics and Privacy
Ethical data use requires a defensible purpose, proportionate collection, secure access, fair consequences and accountable review. Technical permission is not the same as ethical justification. This chapter analyses secondary use, indefinite retention, opaque scoring and automated exceptions, then turns criticism into operational controls and remedy paths.
Purpose should be defined before fields are collected.
Gathering data because it may become useful expands exposure and makes later use difficult to explain. Necessity narrows the information to what the stated task needs.
Retention follows the same logic: once the business and legal reason ends, deletion or irreversible de-identification should occur under a monitored schedule.
Security protects confidentiality and integrity through access, encryption, monitoring and incident response. Ethics asks additional questions. Would people reasonably expect the use? Do proxy variables create unequal effects?
Can an affected person understand and challenge the result? A secure system can still make an intrusive or unfair decision.
Secondary use deserves fresh review when the purpose, variables, recipient or consequence changes. Consent does not answer every issue where power is unequal or language is unclear. Compare less intrusive alternatives, record the reasoning and assign a decision owner.
Human review must include access to evidence, authority to override and a documented outcome.
Measures should extend beyond breach count. Monitor unusual sensitive-data access, overdue deletion, complaints, decision reversals, outcome differences and unresolved governance findings. State both the potential harm and the control failure that permits it.
The response should repair the data practice, preserve accountability and provide a route for correction or remedy.
Proxy risk arises when an apparently neutral field stands in for a sensitive characteristic or disadvantage.
Test outcomes across meaningful groups rather than assuming that removing a protected label guarantees fairness.
Transparency should help the affected person understand the decision, important data and challenge route. A long policy is weak evidence when neither the person nor reviewer can inspect and correct the relevant record.
Ethical governance is iterative.
New recipients, joined datasets, longer retention or a more serious decision can change risk even when original collection was acceptable. Define triggers for renewed review.
Record alternatives considered and why the chosen design is proportionate.
That evidence lets future reviewers challenge the decision when technology, purpose or community expectations change.
Data lineage strengthens accountability by showing origin, transformation, recipient and retention. Without lineage, an organisation cannot explain a decision or remove information when the approved purpose ends.
Remedy should be realistic.
Correction, reconsideration, deletion, notification and compensation address different harms. State who has authority to provide the remedy and what evidence closes the complaint.
For revision, examine a proposed data use through purpose, necessity, access, retention, fairness, transparency and remedy. Identify the affected person and the concrete harm before naming a control.
Then test whether the reviewer can see relevant evidence, challenge the result and authorise correction. This keeps ethical analysis connected to operational responsibility rather than relying on consent or security alone.
What this chapter covers
- 01
Purpose limitation
- 02
Data minimisation
- 03
Privacy expectations
- 04
Security controls
- 05
Fairness and proxy risk
- 06
Contestability
- 07
Retention governance
Worked example · free
Challenge a hidden customer credit score
- 1Credit ranking differs materially from the stated promotions purpose.
- 1Inferred traits may be intrusive and unnecessary proxies.
- 1Opaque scoring prevents understanding and challenge.
- 3Pause deployment, reduce fields, test outcomes, require approval and create human review.
Key terms
- Purpose Limitation
- Purpose limitation ties collection and use to a clear stated business need.
- Data Minimisation
- Data minimisation limits fields and retention to what the purpose requires.
- Contestability
- Contestability gives affected people a route to understand, challenge and correct a decision.
Data Ethics and Privacy FAQ
Is legal compliance enough for ethical data use?
Compliance establishes important boundaries but does not answer every question about expectation, proportionality, fairness or harm. Ethical analysis also considers whether the use is necessary, whether affected people can understand and challenge it, and whether less intrusive alternatives achieve the purpose.
Why is indefinite retention risky?
Keeping data for speculative future use increases exposure, weakens purpose limitation and makes access harder to justify. Define business and legal retention needs, schedule deletion or irreversible de-identification, monitor overdue jobs and require new approval before materially different reuse.
What should a human review route contain?
It needs a qualified owner, accessible evidence, authority to change the decision, recorded reasons and feedback into the rule or model. Human review is ineffective when the reviewer simply accepts the automated output or cannot see the basis for it.
Exam move
Use four headings for every ethics case: purpose, necessity, consequence and accountability. Add the stakeholders who bear benefits and harms, then specify a practical alternative and remedy route. Measure access, retention, complaints, overrides and outcome differences rather than relying only on breach counts.
Working through Data Ethics and Privacy in ACCT20007? Sia is AskSia’s AI Accounting tutor — ask any ACCT20007 Data Ethics and Privacy question and get a clear, step-by-step explanation grounded in how ACCT20007 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
