Leadership, Ethics, and Professional Practice
What to know
This domain covers the human and ethical side of data work. Leadership concepts: stakeholder management, communication, change management, team building, and the ability to translate data into business value. Candidates should understand the difference between management (process) and leadership (direction and influence), and the role of the data leader in building a data-driven culture.
Ethics covers fairness, transparency, accountability, privacy, and bias. Know the concepts of algorithmic bias (data bias, sampling bias, label bias), explainability, and the tension between model performance and fairness. Understand professional codes of conduct (integrity, confidentiality, competence, avoiding conflicts of interest) and the concept of a data ethics review. Be familiar with responsible AI principles: human oversight, transparency, non-maleficence, and accountability.
Traps
- Treating ethics as a compliance checkbox rather than a design consideration.
- Assuming a model is unbiased because it uses no protected attributes - proxies can encode bias.
- Confusing explainability (can we understand the model?) with interpretability (is the model inherently understandable?).
- Forgetting that change management is about people, not tools.
Worked rule of thumb
For any ethical scenario, run the stakeholder impact test: Who is affected, who is not in the room, and what could go wrong for them? Then ask whether the decision is transparent and accountable. This surfaces bias and fairness concerns that pure compliance checks miss.
Professional practice: document decisions, disclose limitations, and escalate conflicts of interest rather than resolving them silently.