Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
78 lines (77 sloc) 4.51 KB
title: Data Science Ethics Checklist
sections:
- title: Data Collection
section_id: A
lines:
- line_id: A.1
line_summary: Informed consent
line: If there are human subjects, have they given informed consent, where subjects affirmatively opt-in and have a clear understanding of the data uses to which they consent?
- line_id: A.2
line_summary: Collection bias
line: Have we considered sources of bias that could be introduced during data collection and survey design and taken steps to mitigate those?
- line_id: A.3
line_summary: Limit PII exposure
line: Have we considered ways to minimize exposure of personally identifiable information (PII) for example through anonymization or not collecting information that isn't relevant for analysis?
- title: Data Storage
section_id: B
lines:
- line_id: B.1
line_summary: Data security
line: Do we have a plan to protect and secure data (e.g., encryption at rest and in transit, access controls on internal users and third parties, access logs, and up-to-date software)?
- line_id: B.2
line_summary: Right to be forgotten
line: Do we have a mechanism through which an individual can request their personal information be removed?
- line_id: B.3
line_summary: Data retention plan
line: Is there a schedule or plan to delete the data after it is no longer needed?
- title: Analysis
section_id: C
lines:
- line_id: C.1
line_summary: Missing perspectives
line: Have we sought to address blindspots in the analysis through engagement with relevant stakeholders (e.g., checking assumptions and discussing implications with affected communities and subject matter experts)?
- line_id: C.2
line_summary: Dataset bias
line: Have we examined the data for possible sources of bias and taken steps to mitigate or address these biases (e.g., stereotype perpetuation, confirmation bias, imbalanced classes, or omitted confounding variables)?
- line_id: C.3
line_summary: Honest representation
line: Are our visualizations, summary statistics, and reports designed to honestly represent the underlying data?
- line_id: C.4
line_summary: Privacy in analysis
line: Have we ensured that data with PII are not used or displayed unless necessary for the analysis?
- line_id: C.5
line_summary: Auditability
line: Is the process of generating the analysis well documented and reproducible if we discover issues in the future?
- title: Modeling
section_id: D
lines:
- line_id: D.1
line_summary: Proxy discrimination
line: Have we ensured that the model does not rely on variables or proxies for variables that are unfairly discriminatory?
- line_id: D.2
line_summary: Fairness across groups
line: Have we tested model results for fairness with respect to different affected groups (e.g., tested for disparate error rates)?
- line_id: D.3
line_summary: Metric selection
line: Have we considered the effects of optimizing for our defined metrics and considered additional metrics?
- line_id: D.4
line_summary: Explainability
line: Can we explain in understandable terms a decision the model made in cases where a justification is needed?
- line_id: D.5
line_summary: Communicate bias
line: Have we communicated the shortcomings, limitations, and biases of the model to relevant stakeholders in ways that can be generally understood?
- title: Deployment
section_id: E
lines:
- line_id: E.1
line_summary: Redress
line: Have we discussed with our organization a plan for response if users are harmed by the results (e.g., how does the data science team evaluate these cases and update analysis and models to prevent future harm)?
- line_id: E.2
line_summary: Roll back
line: Is there a way to turn off or roll back the model in production if necessary?
- line_id: E.3
line_summary: Concept drift
line: Do we test and monitor for concept drift to ensure the model remains fair over time?
- line_id: E.4
line_summary: Unintended use
line: Have we taken steps to identify and prevent unintended uses and abuse of the model and do we have a plan to monitor these once the model is deployed?