Skip to content

Chapter 13: Research Ethics & Best Practices

🎓 Learning Objectives

  • Understand research ethics principles
  • Learn about responsible research practices
  • Understand authorship and credit
  • Learn about data ethics and privacy
  • Master best practices for research

Why Research Ethics Matter

Research ethics ensure:

  • Integrity: Honest and reliable research
  • Trust: Public trust in science
  • Fairness: Fair treatment of all
  • Responsibility: Responsible use of research
  • Reproducibility: Reproducible and verifiable results

Ethics Importance

Ethical violations can end careers and harm the field. Always act ethically.

Core Ethical Principles

1. Honesty

Practices: - Report results accurately - Don't fabricate data - Don't falsify results - Acknowledge errors - Be transparent

Honesty

Foundation of research. Always be honest in reporting.

2. Integrity

Practices: - Follow research protocols - Maintain standards - Keep commitments - Be consistent - Act with principle

3. Objectivity

Practices: - Avoid bias - Fair evaluation - Consider alternatives - Acknowledge limitations - Critical thinking

Objectivity

Recognize your biases. Strive for objective evaluation.

4. Respect

Practices: - Respect participants - Respect colleagues - Respect intellectual property - Respect diversity - Respect privacy

5. Responsibility

Practices: - Consider consequences - Use resources wisely - Share knowledge - Mentor others - Serve society

Research Misconduct

Types of Misconduct

1. Fabrication: - Making up data - Inventing results - Creating false evidence

2. Falsification: - Manipulating data - Changing results - Omitting data - Misrepresenting findings

3. Plagiarism: - Copying without credit - Using others' work - Self-plagiarism - Inadequate citation

Misconduct Consequences

Research misconduct has serious consequences: - Career damage - Loss of credibility - Legal issues - Harm to field

Avoiding Misconduct

Best Practices: - Always cite sources - Report honestly - Don't manipulate data - Acknowledge contributions - Follow protocols

Avoiding Misconduct

When in doubt, ask. Better to be cautious than violate ethics.

Authorship and Credit

Authorship Criteria

ICMJE Criteria: 1. Substantial contributions to conception/design 2. Drafting or revising article 3. Final approval of version 4. Agreement to be accountable

Authorship

All authors must meet all criteria. Discuss authorship early.

Authorship Order

Conventions: - First author: Primary contributor - Last author: Senior/advisor - Middle authors: Alphabetical or by contribution

Best Practices: - Discuss order early - Be fair - Document contributions - Use CRediT taxonomy

Authorship

  • Discuss early and openly
  • Be fair about contributions
  • Document contributions
  • Resolve conflicts professionally

Acknowledgment

Acknowledge: - Funding sources - Data providers - Computing resources - Helpful discussions - Reviewers (if appropriate)

Acknowledgment

Proper acknowledgment shows professionalism and gratitude.

Data Ethics

Data Collection

Ethical Practices: - Informed consent - Privacy protection - Data minimization - Purpose limitation - Right to withdraw

Data Ethics

Always consider: - Privacy - Consent - Anonymization - Usage rights

Data Usage

Responsible Practices: - Use data as intended - Protect privacy - Secure storage - Limited access - Proper disposal

Bias and Fairness

Considerations: - Check for bias - Fair representation - Document limitations - Mitigate bias - Test fairness

Bias

All data has bias. Acknowledge and address it.

Reproducibility

Why Reproducibility Matters

Benefits: - Verifies results - Enables extension - Builds trust - Advances science - Required by many venues

Reproducibility

Make your research reproducible. It's ethical and good practice.

Reproducibility Practices

Code: - Share code - Document well - Version control - Clear instructions

Data: - Share when possible - Document thoroughly - Provide access - Respect privacy

Experiments: - Document setup - Report all details - Multiple runs - Statistical analysis

Responsible AI Research

Considerations

1. Impact Assessment: - Potential benefits - Potential harms - Who benefits? - Who might be harmed?

2. Fairness: - Fair to all groups - No discrimination - Equal access - Bias mitigation

3. Transparency: - Explain methods - Document limitations - Share code/data - Clear communication

4. Safety: - Test thoroughly - Consider misuse - Security measures - Risk assessment

Responsible AI

Consider broader impact. Research can have real-world consequences.

Best Practices

Research Practices

1. Planning: - Clear research questions - Feasible design - Ethical considerations - Resource planning

2. Execution: - Follow protocols - Document everything - Quality control - Regular review

3. Reporting: - Honest reporting - Complete disclosure - Acknowledge limitations - Proper citation

4. Sharing: - Share code/data - Publish results - Help others - Contribute to field

Best Practices

Follow best practices from the start. Easier than fixing later.

Collaboration

Best Practices: - Clear communication - Fair contributions - Respect differences - Resolve conflicts - Support each other

Collaboration

Good collaboration enhances research. Poor collaboration harms it.

Common Ethical Issues

Issue 1: P-hacking

Problem: Manipulating analysis to get significant results

Solution: Pre-register analysis plan, report all tests

P-hacking

Don't manipulate analysis. Report all results honestly.

Issue 2: HARKing

Problem: Hypothesizing After Results Known

Solution: Formulate hypotheses before experiments

Issue 3: Selective Reporting

Problem: Only reporting positive results

Solution: Report all results, including failures

Issue 4: Inadequate Citation

Problem: Not citing properly

Solution: Cite all sources, use proper format

Ethical Issues

Be aware of common issues. Avoid them proactively.

Resources

📚 Ethics Resources
  1. Research Ethics Guide - NIEHS
  2. Responsible AI - Partnership on AI
  3. Research Integrity - ORI
📋 Guidelines
  1. ICMJE Guidelines - Authorship
  2. COPE Guidelines - Publication ethics
  3. FAIR Principles - Data sharing
🎓 Training
  1. CITI Training - Research ethics
  2. RCR Training - Responsible conduct

Next Steps

Key Takeaways: - Research ethics are fundamental to science - Always be honest, objective, and responsible - Avoid research misconduct - Handle authorship fairly - Consider data ethics and privacy - Make research reproducible - Consider broader impact - Follow best practices