Data Governance and Stewardship
Data governance is the system of decision rights, roles, policies, and controls that determine how data is managed and used. Candidates should understand roles (data owners, data stewards, data custodians, controllers, processors), accountability structures (governance councils, review boards, DPOs), and the data lifecycle - acquisition, storage, use, sharing, retention, and disposal. Core concepts include data quality (accuracy, completeness, timeliness, consistency), metadata and lineage (knowing where data came from and how it changed), access controls, classification (public/internal/confidential/sensitive), and stewardship as the ethical duty to manage data in others' interest.
Traps: (1) confusing ownership (accountability) with custodianship (operational control) - the owner is answerable, the custodian executes; (2) treating governance as purely technical when it is socio-technical (policies, incentives, culture); (3) assuming a policy on paper equals effective control - questions reward evidence of enforcement and monitoring; (4) neglecting lineage and provenance, which are essential to auditing and to detecting bias or unauthorized use; (5) ignoring disposal - retention beyond need is a governance failure.
Rule of thumb: for any dataset, be able to answer who owns it, who may access it, where it came from, how it is protected, and when it is deleted. If any answer is missing, governance is incomplete.
Worked example: A team wants to reuse customer data for research. Good governance requires confirming the owner's approval, checking the original consent/purpose, documenting lineage, applying classification-based access controls, and setting a retention and deletion schedule - not just copying the table.