Certified Data Steward (CDS) — Questions and Answers
Question 1: A regulated industry requires that all reports used in compliance filings be traceable to their source data. Which capability is essential?
- End-to-end data lineage tracking (Correct answer)
- Self-service report creation for business users
- Automated report distribution via email
- Interactive data visualization
Correct answer: End-to-end data lineage tracking
End-to-end data lineage maps the journey from source system to final report, satisfying regulatory traceability requirements.
Question 2: What is 'data standardization' in the context of data quality improvement?
- Encrypting data according to industry security standards
- Removing all records that do not conform to the primary schema
- Converting all data to ISO 8601 date format
- Transforming data into consistent formats, codes, and representations across systems (Correct answer)
Correct answer: Transforming data into consistent formats, codes, and representations across systems
Data standardization transforms varied representations (e.g., 'NY', 'New York', 'N.Y.') into a single canonical form to enable accurate comparison and integration.
Question 3: Under CCPA, which right allows California consumers to prevent businesses from selling their personal information to third parties?
- Right to Erasure
- Right to Rectification
- Right to Portability
- Right to Opt-Out (Correct answer)
Correct answer: Right to Opt-Out
CCPA grants consumers the Right to Opt-Out, allowing them to direct businesses not to sell their personal information to third parties.
Question 4: In the context of data integration, what is 'schema mapping'?
- Automatically generating database indexes
- Defining the correspondence between fields in different data schemas (Correct answer)
- Encrypting schema metadata for security
- Creating visual diagrams of database server hardware
Correct answer: Defining the correspondence between fields in different data schemas
Schema mapping identifies how data fields in a source schema correspond to fields in a target schema to enable accurate data transformation.
Question 5: A data steward wants to prioritize which data quality issues to fix first when resources are limited. The BEST framework for prioritization is to:
- Fix issues in alphabetical order by affected table name
- Always fix the most technically complex issues first
- Assess each issue by its business impact, frequency, and cost to remediate (Correct answer)
- Let end users vote on which issues matter most
Correct answer: Assess each issue by its business impact, frequency, and cost to remediate
Prioritizing by business impact, frequency, and remediation cost ensures limited resources are applied where they deliver the greatest organizational value.
Question 6: COPPA applies to operators of websites or online services directed at children under which age?
- 13 (Correct answer)
- 21
- 16
- 18
Correct answer: 13
COPPA (Children's Online Privacy Protection Act) requires verifiable parental consent before collecting personal information from children under 13.
Question 7: What is the role of a 'data contract' in monitoring pipelines?
- A legal agreement between data vendors and buyers
- A budget document for data infrastructure costs
- A formal specification of expected schema, quality thresholds, and SLAs that producers must meet (Correct answer)
- An access control policy for pipeline consumers
Correct answer: A formal specification of expected schema, quality thresholds, and SLAs that producers must meet
A data contract codifies the agreed-upon structure, quality expectations, and delivery SLAs between data producers and consumers.
Question 8: Why is standardization important in Data Quality Management?
- It removes the need for metadata.
- It ensures uniformity in data format and values (Correct answer)
- It slows down data retrieval.
- It increases data storage needs.
Correct answer: It ensures uniformity in data format and values
Standardization in Data Quality Management is vital because it establishes consistent rules and formats for data across an organization. This ensures uniformity in how data is captured, stored, and interpreted, preventing inconsistencies and errors that arise from varied representations. By standardizing data, organizations improve data accuracy, comparability, and usability for analysis and decision-making.
Question 9: What is reclassification in data management, and when is it necessary?
- Reclassification converts data from one file format to another
- Reclassification refers to assigning initial classification to newly created data
- Reclassification updates a data asset's category when its sensitivity or business context has materially changed (Correct answer)
- Reclassification deletes all existing classification tags and restarts the process annually
Correct answer: Reclassification updates a data asset's category when its sensitivity or business context has materially changed
Reclassification is the process of updating a data asset's label when its sensitivity, regulatory status, or business value has changed—for example, when previously confidential merger plans become public after an announcement.
Question 10: A data pipeline experiences high latency due to repeated full table scans on a staging table. What is the most effective structural remedy?
- Switching from batch to streaming ingestion
- Rewriting the pipeline in a different programming language
- Adding a surrogate key and indexing the columns used in join and filter conditions (Correct answer)
- Increasing the pipeline's batch size
Correct answer: Adding a surrogate key and indexing the columns used in join and filter conditions
Adding targeted indexes on join and filter columns converts expensive full scans into efficient index seeks on the staging table.
Question 11: Which NIST Privacy Framework function focuses on identifying individuals, data, systems, and organizational units involved in data processing?
- Protect-P
- Govern-P
- Identify-P (Correct answer)
- Respond-P
Correct answer: Identify-P
The Identify-P function helps organizations understand privacy risks by inventorying data processing activities, systems, and the individuals whose data is processed.
Question 12: What role does a Data Protection Officer (DPO) play?
- Designs marketing strategies.
- Handles payroll.
- Manages social media campaigns.
- Ensures compliance with data protection laws (Correct answer)
Correct answer: Ensures compliance with data protection laws
A Data Protection Officer (DPO) is a designated role, often mandated by regulations like GDPR, responsible for overseeing an organization's data protection strategy and ensuring compliance with data protection laws. The DPO acts as an independent advisor, monitoring internal compliance, informing and advising on data protection obligations, and serving as a contact point for supervisory authorities and data subjects.
Question 13: What is the primary ongoing responsibility of data stewards regarding a classification scheme?
- Stewards are solely responsible for data backup and disaster recovery
- Stewards define, enforce, and periodically review classification policies and ensure data assets are correctly labeled (Correct answer)
- Stewards only approve IT infrastructure and hardware purchases
- Stewards handle only technical database administration tasks
Correct answer: Stewards define, enforce, and periodically review classification policies and ensure data assets are correctly labeled
Data stewards are responsible for defining classification policies, verifying data assets carry accurate labels, and conducting periodic reviews to ensure the classification scheme remains current and properly applied.
Question 14: Which action supports data privacy best practices?
- Sharing credentials widely.
- Leaving files unprotected.
- Encrypting sensitive data (Correct answer)
- Disabling antivirus software.
Correct answer: Encrypting sensitive data
Encrypting sensitive data is a fundamental data privacy best practice. Encryption transforms data into an unreadable format, making it inaccessible to unauthorized parties even if a data breach occurs. This protective measure significantly reduces the risk of sensitive information being compromised, thereby safeguarding personal data and helping organizations comply with privacy regulations.
Question 15: Under US SEC Rule 17a-4, broker-dealers must retain electronic records in what format to ensure authenticity?
- Encrypted ZIP archives
- Cloud-synced spreadsheets
- Non-rewriteable, non-erasable (WORM) format (Correct answer)
- Editable relational database format
Correct answer: Non-rewriteable, non-erasable (WORM) format
SEC Rule 17a-4 requires that broker-dealer electronic records be stored in a WORM format to prevent alteration or deletion, ensuring integrity for audits.
Question 16: What makes a report effective for stakeholder communication?
- Clear visualization of relevant metrics aligned with business objectives (Correct answer)
- Only positive trends and metrics
- Maximum amount of raw data
- Technical jargon demonstrating expertise
Correct answer: Clear visualization of relevant metrics aligned with business objectives
Effective reports use clear visualizations of relevant metrics aligned with business objectives, enabling stakeholders to understand status and make informed decisions.
Question 17: What is the benefit of having standardized data definitions?
- They limit data access.
- They slow down data entry.
- They improve data clarity and consistency (Correct answer)
- They complicate analytics.
Correct answer: They improve data clarity and consistency
Standardized data definitions ensure that all stakeholders within an organization share a common understanding of data elements. This consistency eliminates ambiguity, improves communication, and enhances the reliability of data analysis and reporting. It is crucial for accurate decision-making and data integration.
Question 18: Which data quality technique involves comparing data against a trusted external reference source to verify accuracy?
- Data reconciliation with a golden record (Correct answer)
- Data certification
- Data benchmarking
- Data normalization
Correct answer: Data reconciliation with a golden record
Reconciling data against a golden record or authoritative external source validates that internal data accurately reflects the real-world values it represents.
Question 19: Under CCPA/CPRA, what is the maximum civil penalty per intentional violation?
- $25,000
- $2,500
- $1,000
- $7,500 (Correct answer)
Correct answer: $7,500
CCPA/CPRA allows civil penalties of up to $2,500 per unintentional violation and up to $7,500 per intentional violation.
Question 20: When implementing a 'right-first-time' data quality strategy, what is the primary focus?
- Automating downstream error correction after ingestion
- Archiving low-quality records for future remediation
- Preventing defects at the point of data entry or creation (Correct answer)
- Retroactively cleansing historical data in bulk
Correct answer: Preventing defects at the point of data entry or creation
Right-first-time focuses on defect prevention at the source rather than downstream correction, reducing the cost of poor data quality.
Question 21: A company implements a data lake to store raw data from many sources. A key governance concern with data lakes is the risk of becoming a:
- Data dictionary
- Data silo
- Data mart
- Data swamp (Correct answer)
Correct answer: Data swamp
Without proper metadata management, cataloging, and governance, a data lake can become a 'data swamp'—where data is stored but cannot be found, trusted, or used.
Question 22: When a monitoring system sends an alert for a data anomaly, what should the escalation path include?
- Immediate shutdown of the affected pipeline until investigation completes
- Notification to the data steward with context, followed by escalation to the data owner if unresolved within the SLA (Correct answer)
- Automatic deletion of the anomalous records
- Public notification to all report consumers about the potential issue
Correct answer: Notification to the data steward with context, followed by escalation to the data owner if unresolved within the SLA
A structured escalation path ensures anomalies are acknowledged and resolved within defined timeframes while keeping appropriate stakeholders informed.
Question 23: When configuring a monitoring rule for null values, which threshold strategy is most appropriate for a mandatory identifier field?
- Alert only when nulls exceed 25% of total records
- Set a zero-tolerance threshold — any null triggers an alert (Correct answer)
- Allow up to 5% nulls as acceptable tolerance
- Monitor nulls weekly rather than in real time
Correct answer: Set a zero-tolerance threshold — any null triggers an alert
Mandatory identifier fields must never be null, so zero-tolerance thresholds are required to catch any violation immediately.
Question 24: Which communication strategy is most effective when introducing a new data governance policy to business unit stakeholders who are resistant to change?
- Engage stakeholders early with workshops that explain the business value and gather their input (Correct answer)
- Mandate compliance through executive directives without explanation
- Implement the policy silently and address complaints reactively
- Distribute the policy document via email and require acknowledgment signatures
Correct answer: Engage stakeholders early with workshops that explain the business value and gather their input
Early engagement through collaborative workshops builds stakeholder buy-in by demonstrating business value and incorporating their concerns into the policy design.
Question 25: What is capacity planning?
- Only planning for current usage levels
- Predicting future resource needs based on growth trends and requirements (Correct answer)
- Buying maximum resources upfront
- Waiting until resources are exhausted to add more
Correct answer: Predicting future resource needs based on growth trends and requirements
Capacity planning uses growth trends and requirements to predict future needs, ensuring adequate resources are available without over-provisioning.
Question 26: Which data quality dimension measures the degree to which data values fall within acceptable ranges or conform to defined domain rules?
- Timeliness
- Uniqueness
- Completeness
- Validity (Correct answer)
Correct answer: Validity
Validity measures whether data values conform to defined business rules, formats, or acceptable value domains.
Question 27: When a data steward escalates a data quality issue to the data governance council, the steward should provide:
- The IT ticket number and expected fix date
- Only a verbal description of the problem
- A documented business impact assessment with supporting data quality metrics (Correct answer)
- A list of users who reported the issue
Correct answer: A documented business impact assessment with supporting data quality metrics
Governance councils make prioritization decisions based on business impact; a documented assessment with metrics provides the evidence needed.
Question 28: What is the relationship between master data and transactional data?
- Master data is deleted after each transaction
- Transactional data overrides master data
- They are the same type of data
- Master data supports and contextualizes transactional data (Correct answer)
Correct answer: Master data supports and contextualizes transactional data
Master data and transactional data have a fundamental relationship where master data provides the context for transactional data. Transactional data describes events or activities (e.g., a sale, a payment) and typically references master data entities (e.g., which customer bought which product). Thus, master data supports and contextualizes these transactions, ensuring their accuracy and meaningful interpretation.
Question 29: Which of the following best describes metadata management?
- Controlling access to software licenses
- Maintaining and organizing metadata for effective use (Correct answer)
- Managing large data files for backups
- Analyzing internet traffic logs
Correct answer: Maintaining and organizing metadata for effective use
Metadata management is the discipline of maintaining and organizing metadata to ensure its accuracy, accessibility, and utility throughout its lifecycle. This involves defining, storing, and governing metadata to help users understand, locate, and effectively use the organization's data assets. Proper metadata management is essential for data governance, data quality, and data integration.
Question 30: What does 'data literacy' mean in the context of stakeholder enablement for data governance?
- Certification in a data management framework such as DAMA-DMBOK
- Proficiency in operating ETL and BI tools
- The ability to read, understand, question, and work effectively with data in one's role (Correct answer)
- The ability to read and write data using SQL
Correct answer: The ability to read, understand, question, and work effectively with data in one's role
Data literacy enables stakeholders to critically interpret data, understand quality issues, and make evidence-based decisions without requiring technical expertise.
Question 31: Under GDPR, a data breach involving encrypted data where the key is not compromised must be:
- Reported to the supervisory authority within 72 hours
- Reported to data subjects without undue delay
- Documented internally but does not trigger notification obligations (Correct answer)
- Reported to both the authority and data subjects within 30 days
Correct answer: Documented internally but does not trigger notification obligations
If encrypted data is breached but the encryption key is secure, the data remains unintelligible and poses no risk to data subjects, so notification obligations do not apply.
Question 32: A data steward is writing a script to validate referential integrity between two tables. Which SQL approach is most direct?
- Full table scan with Python dictionary lookup
- COUNT(*) on each table separately
- Compare table names in the schema
- LEFT JOIN with a WHERE clause filtering for unmatched foreign keys (Correct answer)
Correct answer: LEFT JOIN with a WHERE clause filtering for unmatched foreign keys
A LEFT JOIN filtering for NULL matches on the right side efficiently identifies orphaned foreign key values.
Question 33: Why is post-implementation review important?
- It is only required by auditors
- It is optional if the implementation was successful
- It delays the next project
- It identifies lessons learned and confirms objectives were met (Correct answer)
Correct answer: It identifies lessons learned and confirms objectives were met
Post-implementation reviews confirm that objectives were met, identify areas for improvement, and capture lessons learned for future projects.
Question 34: SOX Section 404 places primary responsibility for assessing and reporting on internal controls over financial reporting on:
- The SEC's enforcement division
- The company's management (Correct answer)
- The audit committee of the board
- External auditors
Correct answer: The company's management
SOX Section 404(a) requires management to assess and report on the effectiveness of internal controls over financial reporting, with external auditors attesting to that assessment.
Question 35: Which monitoring tool feature allows a data steward to automatically quarantine records that fail quality checks?
- Schema registry service
- Data masking engine
- Role-based access control module
- Quarantine or reject routing in a data quality rule engine (Correct answer)
Correct answer: Quarantine or reject routing in a data quality rule engine
A quarantine or reject routing feature isolates non-conforming records so they can be reviewed and corrected without polluting downstream systems.
Question 36: What role does a data steward play in the end-of-life stage of the data lifecycle?
- Encrypting the data before transfer to a new owner
- Ensuring data is disposed of per retention schedules, legal holds, and disposal procedures (Correct answer)
- Designing the database schema for the replacement system
- Rebuilding the dataset from raw source logs
Correct answer: Ensuring data is disposed of per retention schedules, legal holds, and disposal procedures
At end-of-life, the data steward coordinates compliant disposal including checking for legal holds, obtaining destruction certificates, and updating the asset register.
Question 37: What is the key benefit of using connection pooling in a high-throughput data application?
- It automatically balances load across multiple database servers
- It encrypts database connections for better security
- It caches query results to avoid repeated execution
- It reuses existing connections, reducing the overhead of establishing new ones (Correct answer)
Correct answer: It reuses existing connections, reducing the overhead of establishing new ones
Connection pooling maintains a set of open connections that can be reused, eliminating the latency and resource cost of creating a new connection for each request.
Question 38: What is the benefit of using a layered architecture approach?
- Elimination of all dependencies
- Faster execution speed
- Reduced hardware requirements
- Separation of concerns enabling independent development and testing (Correct answer)
Correct answer: Separation of concerns enabling independent development and testing
Layered architecture separates concerns so each layer can be developed, tested, and modified independently, improving maintainability and team productivity.
Question 39: How should a root cause analysis be conducted?
- Systematically investigate underlying causes rather than just symptoms (Correct answer)
- Fix the immediate symptom and close the ticket
- Assign blame to the responsible party
- Only analyze problems that recur three or more times
Correct answer: Systematically investigate underlying causes rather than just symptoms
Root cause analysis investigates underlying causes systematically, preventing recurrence rather than just addressing symptoms temporarily.
Question 40: What is data cleansing?
- Creating data backups.
- Encrypting data at rest.
- Sorting data alphabetically.
- Correcting or deleting inaccurate records (Correct answer)
Correct answer: Correcting or deleting inaccurate records
Data cleansing, also known as data scrubbing, is the process of detecting and correcting (or removing) corrupt, inaccurate, or irrelevant records from a dataset. This involves identifying errors such as typos, missing values, or inconsistent formatting and then rectifying them to improve the overall quality and reliability of the data. It's a fundamental step in ensuring data is fit for analysis and operational use.
Question 41: Which technique involves replacing sensitive data elements with a non-sensitive substitute that maintains a consistent reference, allowing data to be tracked across systems without exposing the original value?
- Data aggregation
- Data anonymization
- Data encryption
- Tokenization (Correct answer)
Correct answer: Tokenization
Tokenization replaces sensitive data with a token (surrogate value) that can be mapped back to the original in a secure token vault, unlike encryption it cannot be reversed without vault access.
Question 42: A data steward must choose between a hub-and-spoke and a point-to-point integration topology. What is the main advantage of hub-and-spoke?
- Elimination of the need for data transformation between systems
- Lower latency because data does not pass through an intermediary
- Reduced number of integration interfaces to maintain by centralizing routing logic (Correct answer)
- Stronger data encryption because all messages pass through a secure hub
Correct answer: Reduced number of integration interfaces to maintain by centralizing routing logic
Hub-and-spoke centralizes integration logic so that N systems require N interfaces rather than NĂ—(N-1)/2 point-to-point connections, reducing complexity.
Question 43: A data steward wants to prevent unauthorized changes to a critical master data attribute. Which governance control is MOST appropriate?
- Implementing a change approval workflow with steward sign-off (Correct answer)
- Archiving old values to a history table nightly
- Encrypting the attribute at rest
- Adding a database index to the attribute column
Correct answer: Implementing a change approval workflow with steward sign-off
Change approval workflows enforce that any modification to critical master data attributes is reviewed and authorized before being applied.
Question 44: A steward must demonstrate that a critical business report meets its defined SLA of being refreshed by 7:00 AM daily. Which monitoring artifact provides this evidence?
- A storage capacity report for the data warehouse
- A list of all users who accessed the report that day
- A user satisfaction survey about the report
- A pipeline run log with timestamped completion records (Correct answer)
Correct answer: A pipeline run log with timestamped completion records
Timestamped pipeline run logs provide objective evidence of when a data refresh completed, enabling SLA compliance verification.
Question 45: Metadata lineage tracking is most critical for which compliance requirement?
- Network throughput auditing
- Software license management
- Response time SLAs
- Data subject access requests under GDPR (Correct answer)
Correct answer: Data subject access requests under GDPR
GDPR data subject access requests require organizations to trace exactly where personal data originated, how it transformed, and where it flows.
Question 46: Which MDM KPI best measures the effectiveness of duplicate record elimination over time?
- Size of the golden record store in GB
- Number of source systems integrated
- Duplicate rate: percentage of duplicate entities remaining in the hub (Correct answer)
- Average query response time
Correct answer: Duplicate rate: percentage of duplicate entities remaining in the hub
The duplicate rate directly measures MDM matching quality by tracking what fraction of entity records remain unresolved duplicates.
Question 47: What is a key challenge organizations face when implementing enterprise-wide data classification?
- Having too few data assets to make classification worthwhile
- All enterprise data is already pre-classified by default in modern databases
- The high licensing cost of data classification tools
- Achieving consistent classification across diverse systems, formats, and business units (Correct answer)
Correct answer: Achieving consistent classification across diverse systems, formats, and business units
Achieving consistent classification across heterogeneous systems, data formats, and organizational units is a major challenge because different groups may interpret and apply classification criteria differently.
Question 48: A data steward discovers that a vendor processes EU resident data without a valid data transfer mechanism. Which action is the MOST appropriate immediate step?
- Terminate the vendor contract immediately
- File a complaint with the supervisory authority
- Suspend data transfers until a lawful mechanism is established (Correct answer)
- Notify affected data subjects
Correct answer: Suspend data transfers until a lawful mechanism is established
Suspending transfers until a valid mechanism (SCCs, adequacy decision, BCRs) is in place prevents ongoing GDPR violations while allowing time to remediate.
Question 49: A hospital shares de-identified patient data with a research university. Under HIPAA, which method allows sharing data with NO residual risk threshold requirement?
- Safe Harbor method (Correct answer)
- Limited Data Set method
- Expert Determination method
- Hybrid entity method
Correct answer: Safe Harbor method
The Safe Harbor method requires removal of 18 specific identifiers and provides a defined standard, while Expert Determination requires a statistical expert to certify very small re-identification risk.
Question 50: Which report distribution method best reduces the risk of stale data being acted upon?
- Providing live dashboard links with automatic refresh intervals (Correct answer)
- Printing reports and distributing them at monthly meetings
- Storing exported CSVs on a shared file drive
- Emailing static PDF snapshots on a weekly schedule
Correct answer: Providing live dashboard links with automatic refresh intervals
Live dashboards with automatic refresh ensure consumers always view the most current data rather than a point-in-time snapshot.
Question 51: When automating data catalog updates via API, which HTTP status code indicates that a new metadata entry was successfully created?
- 201 Created (Correct answer)
- 200 OK
- 204 No Content
- 400 Bad Request
Correct answer: 201 Created
201 Created is the standard HTTP response indicating a new resource was successfully created on the server.
Question 52: When implementing a data governance program in a highly decentralized organization, which approach is most appropriate?
- Outsourcing governance to a third-party vendor
- Federated model with enterprise standards and domain autonomy (Correct answer)
- Centralized command-and-control model enforced by IT
- No governance — decentralization precludes it
Correct answer: Federated model with enterprise standards and domain autonomy
A federated model balances local domain autonomy with enterprise-wide standards, making it suitable for decentralized organizations.
Question 53: An MDM hub uses 'deterministic matching.' This means records are linked when:
- A human data steward manually approves each link
- Exact matches are found on defined key fields like SSN or account number (Correct answer)
- More than 50% of attributes share similar values
- A machine learning model scores them above a threshold
Correct answer: Exact matches are found on defined key fields like SSN or account number
Deterministic matching links records only when specific designated key fields match exactly, providing high precision with lower recall.
Question 54: A data steward discovers that two business units use different definitions for 'active customer' in their separate reports. Which governance action resolves this?
- Merge both reports into a single shared dashboard
- Replace both definitions with a vendor-supplied default
- Allow each unit to keep its own definition as a local standard
- Establish a canonical business glossary definition and align both reports to it (Correct answer)
Correct answer: Establish a canonical business glossary definition and align both reports to it
A canonical business glossary provides an agreed-upon definition that eliminates conflicting interpretations across reports.
Question 55: In enterprise data governance, what does a business glossary primarily provide?
- Standardized business definitions and context for data terms across the organization (Correct answer)
- A vendor-approved list of licensed software applications
- Technical specifications and schemas for database design
- Step-by-step instructions for data backup and recovery procedures
Correct answer: Standardized business definitions and context for data terms across the organization
A business glossary provides standardized, business-context definitions for data terms, ensuring consistent understanding and usage of terminology across different departments and systems.
Question 56: Which of the following best describes 'data profiling' in a data quality context?
- Defining business rules for acceptable data values
- Publishing data to a data marketplace
- Encrypting sensitive fields before storage
- Analyzing data to understand its structure, content, and quality characteristics (Correct answer)
Correct answer: Analyzing data to understand its structure, content, and quality characteristics
Data profiling examines datasets to discover metadata, patterns, anomalies, and quality issues such as nulls, duplicates, and format inconsistencies.
Question 57: During a data governance tool implementation, the project team finds that business users resist entering metadata. What configuration change most directly addresses adoption friction?
- Reduce the number of governance tiers
- Generate automated email reminders to stewards
- Make all metadata fields mandatory before save
- Integrate metadata entry into existing workflow tools users already use (Correct answer)
Correct answer: Integrate metadata entry into existing workflow tools users already use
Embedding metadata capture within existing workflows (e.g., Jira, Confluence, or data pipelines) reduces context-switching and lowers resistance.
Question 58: How should security incidents be handled?
- Blaming the affected user
- Waiting to see if it happens again
- Fixing the issue and not reporting it
- Following an established incident response plan with documentation (Correct answer)
Correct answer: Following an established incident response plan with documentation
Following an established incident response plan ensures consistent, effective handling while documentation supports analysis, compliance, and future prevention.
Question 59: A data steward must automate the enforcement of naming conventions across newly created database tables. Which approach is most scalable?
- Manually review table names weekly
- Send naming convention documentation via email
- Rely on developers to self-enforce conventions
- Write a script that queries the information schema and flags non-compliant names on a schedule (Correct answer)
Correct answer: Write a script that queries the information schema and flags non-compliant names on a schedule
Automated scheduled scanning of the information schema provides continuous, scalable enforcement without manual effort.
Question 60: Which of the following best supports the sustainability of a data governance program over time?
- Centralizing all governance decisions in a single committee
- Relying solely on executive mandate to enforce compliance
- Embedding governance activities into existing business processes and workflows (Correct answer)
- Conducting annual governance training sessions only
Correct answer: Embedding governance activities into existing business processes and workflows
Embedding governance into day-to-day business workflows makes compliance natural and sustainable rather than a separate burden imposed top-down.
Question 61: What is the recommended approach for testing automation scripts?
- Skip testing for simple scripts
- Test directly in production with live data
- Test in a non-production environment with realistic data before deployment (Correct answer)
- Test only when major changes are made
Correct answer: Test in a non-production environment with realistic data before deployment
Testing in non-production environments with realistic data catches issues before they affect live systems, preventing potential data loss or service disruption.
Question 62: What is the key characteristic of a 'write-once, read-many' (WORM) storage architecture from a data governance perspective?
- It reduces storage costs by compressing data after the first read
- It automatically archives data after a single read access
- It enforces column-level encryption on all stored data
- It prevents modification or deletion of records, supporting audit and compliance requirements (Correct answer)
Correct answer: It prevents modification or deletion of records, supporting audit and compliance requirements
WORM storage ensures immutability after the initial write, creating tamper-proof records critical for regulatory compliance and audit trails.
Question 63: Which of the following best describes 'data lineage' in the context of metadata management?
- The age of a dataset measured in years
- The hierarchical relationship between parent and child tables
- The list of users who have accessed a dataset
- The documented path data travels from origin through transformations to consumption (Correct answer)
Correct answer: The documented path data travels from origin through transformations to consumption
Data lineage tracks the full lifecycle of data—its source, all transformations applied, and where it ends up—enabling impact analysis and auditability.
Question 64: Which HIPAA rule establishes standards for protecting the confidentiality, integrity, and availability of electronic protected health information (ePHI)?
- Enforcement Rule
- Privacy Rule
- Breach Notification Rule
- Security Rule (Correct answer)
Correct answer: Security Rule
The HIPAA Security Rule specifically addresses ePHI and requires administrative, physical, and technical safeguards to protect its confidentiality, integrity, and availability.
Question 65: What is a metadata repository?
- A centralized store that captures, manages, and provides access to metadata across systems (Correct answer)
- A data lake containing raw unstructured files
- A production database holding master data golden records
- A version control system for ETL scripts
Correct answer: A centralized store that captures, manages, and provides access to metadata across systems
A metadata repository is a purpose-built system for storing and serving metadata including technical definitions, business glossary terms, and lineage information.
Question 66: Which of the following is a best practice in Data Quality Management?
- Manual data checks only.
- Collecting data without validation.
- Involving stakeholders and using automated tools (Correct answer)
- Ignoring data anomalies.
Correct answer: Involving stakeholders and using automated tools
Effective Data Quality Management requires a comprehensive approach that goes beyond simple manual checks. Involving stakeholders ensures that data quality efforts align with business needs and priorities, fostering a data-driven culture. Leveraging automated tools, such as data profiling and cleansing software, allows for efficient and consistent identification and resolution of data quality issues across large datasets.
Question 67: A data steward wants to ensure that a newly deployed automation script does not break existing data pipelines. Which practice should be followed?
- Disable existing pipelines before testing the new script
- Run the new script in parallel with the old one indefinitely
- Test in a staging environment that mirrors production, then deploy with rollback capability (Correct answer)
- Deploy directly to production and monitor for issues
Correct answer: Test in a staging environment that mirrors production, then deploy with rollback capability
Staging environment testing with a defined rollback plan minimizes production risk while validating real-world behavior.
Question 68: Which element of a metadata record helps data consumers understand how current or fresh a dataset is?
- Last updated timestamp and refresh frequency (Correct answer)
- Physical storage location path
- Data owner name
- Number of table partitions
Correct answer: Last updated timestamp and refresh frequency
The last updated timestamp combined with the documented refresh frequency allows consumers to assess data currency and fitness for time-sensitive use cases.
Question 69: A data steward is configuring matching rules in an MDM system to identify duplicate customer records. Which combination of attributes produces the most reliable probabilistic match?
- Name and email address only
- Name, date of birth, address, and phone number with weighted scoring (Correct answer)
- IP address and browser fingerprint
- Customer ID across source systems only
Correct answer: Name, date of birth, address, and phone number with weighted scoring
Probabilistic matching using multiple weighted attributes reduces false positives and false negatives better than any single identifier.
Question 70: When migrating data from a legacy system, a team finds that many records have NULL values in a field that is mandatory in the target system. Which action represents the BEST data stewardship practice?
- Skip all records with NULL values to meet the load deadline
- Randomly assign values to fill the NULL fields
- Cancel the migration until the legacy system is fixed
- Document the issue, apply a default value per business rules, and flag records for review (Correct answer)
Correct answer: Document the issue, apply a default value per business rules, and flag records for review
Good data stewardship requires documenting data quality issues, applying agreed business rules for handling gaps, and flagging impacted records for validation.
Question 71: Which metadata management practice ensures that a data element's meaning is consistent across different departments and systems?
- Schema normalization
- Business glossary maintenance (Correct answer)
- Data archiving policy
- Index optimization
Correct answer: Business glossary maintenance
A business glossary standardizes definitions and ensures all stakeholders share a common understanding of each data element.
Question 72: What is the purpose of data lineage tracking?
- To create backup schedules
- To measure data storage costs
- To track which users access data
- To trace data origin, transformations, and movement through systems (Correct answer)
Correct answer: To trace data origin, transformations, and movement through systems
Data lineage tracking traces data from origin through all transformations and movements, supporting audit requirements, troubleshooting, and impact analysis.
Question 73: What is the primary risk of managing master data without a defined data stewardship ownership model?
- Inability to run batch ETL jobs
- Higher storage costs
- Inconsistent, conflicting master records across domains (Correct answer)
- Slower query performance
Correct answer: Inconsistent, conflicting master records across domains
Without clear ownership, different teams may update master data independently, creating conflicting versions and eroding trust in the data.
Question 74: A data steward must configure a workflow for handling data exception tickets. Which configuration feature is essential for ensuring timely resolution?
- A public dashboard showing open tickets
- Daily PDF reports emailed to stakeholders
- Automatic escalation rules with SLA timers (Correct answer)
- Color-coded priority labels
Correct answer: Automatic escalation rules with SLA timers
Automatic escalation with SLA timers ensures that unresolved exceptions are escalated to the appropriate owner before deadlines are missed.
Question 75: What is the role of encryption in data security?
- Protecting data confidentiality during storage and transmission (Correct answer)
- Only required for financial data
- Replacing the need for access controls
- Making data permanently inaccessible
Correct answer: Protecting data confidentiality during storage and transmission
Encryption protects data confidentiality by making information unreadable without proper decryption keys, whether data is stored (at rest) or being transmitted (in transit).
Question 76: A report shows a sudden 40% drop in daily transaction records. Which first step should the data steward take?
- Check pipeline run logs and source system feeds to determine if the drop reflects a real business change or a data delivery failure (Correct answer)
- Delete and regenerate the report from scratch
- Notify all report consumers that the data is incorrect
- Escalate immediately to executive leadership
Correct answer: Check pipeline run logs and source system feeds to determine if the drop reflects a real business change or a data delivery failure
Investigating pipeline logs and source feeds first determines whether the drop is a data delivery issue or a genuine business event before any communication.
Question 77: When writing an automation script that processes sensitive PII, which practice is most critical for compliance?
- Mask or tokenize PII before logging or displaying (Correct answer)
- Log all PII values for debugging
- Skip validation for PII fields to avoid exposure
- Store PII in plain-text log files
Correct answer: Mask or tokenize PII before logging or displaying
Masking or tokenizing PII in logs and outputs prevents accidental exposure while still allowing debugging.
Question 78: Which GDPR lawful basis would a company MOST likely rely on when processing employee payroll data?
- Consent
- Legitimate interests
- Legal obligation (Correct answer)
- Vital interests
Correct answer: Legal obligation
Processing payroll data is required to comply with tax and employment laws, making legal obligation the most appropriate lawful basis.
Question 79: What is the role of a reference data set in Master Data Management?
- It acts as the MDM hub's backup copy
- It stores transactional history for analytics
- It defines the ETL transformation logic
- It provides standardized code lists used to classify and validate master data (Correct answer)
Correct answer: It provides standardized code lists used to classify and validate master data
Reference data (e.g., country codes, product categories, currency codes) provides the standardized values that classify and constrain master data attributes.
Question 80: A data steward receives competing requests from two business units, each claiming ownership of the same data element definition. The BEST way to resolve this conflict is to:
- Defer resolution indefinitely until consensus is reached naturally
- Let IT decide which definition to use
- Escalate to the Data Governance Council or steering committee for arbitration (Correct answer)
- Allow each unit to maintain its own separate definition
Correct answer: Escalate to the Data Governance Council or steering committee for arbitration
Data governance councils exist specifically to arbitrate disputes over data definitions, ownership, and standards.
Question 81: What is the purpose of a 'data destruction certificate' in a data lifecycle program?
- To provide auditable proof that data was disposed of in accordance with policy (Correct answer)
- To authorize a data steward to create new datasets
- To grant third-party access to archived records
- To certify that a dataset has passed quality validation
Correct answer: To provide auditable proof that data was disposed of in accordance with policy
A data destruction certificate documents the method, date, and scope of disposal, providing a compliance record that data was properly destroyed.
Question 82: When should architectural decisions be reviewed and potentially revised?
- When requirements change significantly or performance targets are not met (Correct answer)
- Never, once architecture is set it should not change
- Only when migrating to new hardware
- Only during annual reviews
Correct answer: When requirements change significantly or performance targets are not met
Architectural decisions should be reviewed when requirements change significantly or performance issues arise, ensuring the system continues to meet evolving needs.
Question 83: What is the primary purpose of a data quality scorecard in a monitoring program?
- To provide a consolidated view of data quality dimensions across critical datasets (Correct answer)
- To replace manual data audits with automated approval workflows
- To automate remediation of data quality issues
- To rank database administrators by performance
Correct answer: To provide a consolidated view of data quality dimensions across critical datasets
A data quality scorecard aggregates metrics across dimensions like completeness, accuracy, and timeliness into a single governance view.
Question 84: Which scenario demonstrates a violation of the 'data minimization' principle?
- Deleting records after the retention period expires
- Storing full credit card numbers for completed, non-recurring transactions (Correct answer)
- Collecting only name and email for a newsletter signup
- Anonymizing data before sharing with analytics teams
Correct answer: Storing full credit card numbers for completed, non-recurring transactions
Retaining full payment data beyond what is needed for completed transactions violates data minimization by keeping more data than necessary.
Question 85: What is the primary goal of data management?
- Making all data publicly accessible
- Reducing storage costs only
- Storing as much data as possible
- Ensuring data quality, availability, and security throughout its lifecycle (Correct answer)
Correct answer: Ensuring data quality, availability, and security throughout its lifecycle
Data management ensures data quality, availability, and security throughout its lifecycle, supporting informed decision-making and regulatory compliance.
Question 86: What is the COST OF POOR DATA QUALITY (CPDQ) model primarily used for?
- Calculating the number of FTE hours spent on manual data entry
- Benchmarking data quality against industry peers
- Estimating storage costs for duplicate records
- Quantifying the financial impact of data defects to justify DQ investments (Correct answer)
Correct answer: Quantifying the financial impact of data defects to justify DQ investments
The CPDQ model quantifies costs such as rework, failed decisions, compliance fines, and lost revenue to build a business case for data quality programs.
Question 87: Which integration approach is BEST suited for near-real-time data synchronization between an operational system and an analytics platform?
- Manual data entry
- Nightly batch ETL
- Change Data Capture (CDC) (Correct answer)
- Full table replacement
Correct answer: Change Data Capture (CDC)
CDC captures and propagates only the changed data rows as they occur, enabling near-real-time synchronization with minimal load.
Question 88: Why is metadata important in data governance?
- It enhances data understanding and traceability (Correct answer)
- It replaces data backups.
- It enables data visualization.
- It hides data from users.
Correct answer: It enhances data understanding and traceability
Metadata, often described as 'data about data,' is crucial in data governance because it provides context, definitions, and lineage for data assets. It helps users understand what data means, where it originated, how it's transformed, and its quality. This enhanced understanding and traceability are vital for effective data management, analysis, and compliance.
Question 89: The primary difference between a data policy and a data standard is:
- Policies are technical specifications; standards are business rules
- Policies are optional guidance; standards are mandatory requirements
- Policies state what must be done; standards specify how it must be done (Correct answer)
- Policies apply only to sensitive data; standards apply to all data
Correct answer: Policies state what must be done; standards specify how it must be done
Policies define mandatory principles and requirements (the 'what'), while standards provide specific, measurable specifications for implementing those policies (the 'how').
Question 90: Which survivorship rule would a data steward apply when multiple source systems disagree on a customer's email address?
- Leave the field blank until all sources agree
- Apply a predefined trust score to select the most reliable source's value (Correct answer)
- Always pick the value from the oldest source system
- Concatenate all values with a delimiter
Correct answer: Apply a predefined trust score to select the most reliable source's value
Survivorship rules use source system trust scores, recency, or completeness metrics to elect the best value for the golden record.
Question 91: What is the first step in performance optimization?
- Add more resources without analysis
- Upgrade all hardware immediately
- Establish baseline measurements and identify bottlenecks (Correct answer)
- Optimize everything simultaneously
Correct answer: Establish baseline measurements and identify bottlenecks
Establishing baselines and identifying bottlenecks ensures optimization efforts target actual performance issues rather than assumed problems.
Question 92: Which type of data loss prevention (DLP) control monitors and restricts data leaving an organization through email or web uploads?
- Network DLP (Correct answer)
- Endpoint DLP
- Storage DLP
- Application DLP
Correct answer: Network DLP
Network DLP inspects data in motion across the network perimeter (email gateways, web proxies) to detect and block unauthorized exfiltration.
Question 93: In MDM governance, a data domain steward is MOST responsible for:
- Writing ETL code for master data integration
- Approving IT budget for MDM tools
- Managing server infrastructure for the MDM hub
- Defining and enforcing data standards within an assigned data domain (Correct answer)
Correct answer: Defining and enforcing data standards within an assigned data domain
A domain data steward owns the quality standards, definitions, and governance rules for their specific master data domain (e.g., customer, product).
Question 94: Under DAMA-DMBOK, which role is primarily accountable for the quality of data within a specific business domain?
- IT Database Administrator
- Business Data Owner (Correct answer)
- Data Architect
- Chief Data Officer
Correct answer: Business Data Owner
The Business Data Owner is accountable for data quality, access decisions, and policy compliance within their domain.
Question 95: Which metadata type describes the rules, constraints, and definitions that govern how data should be structured and used?
- Technical metadata
- Operational metadata
- Business metadata (Correct answer)
- Administrative metadata
Correct answer: Business metadata
Business metadata captures the rules, definitions, and governance constraints that describe data from a business perspective.
Question 96: Which concept describes the practice of storing a copy of data from operational systems in a staging area before loading it into the data warehouse?
- Data staging (Correct answer)
- Data masking
- Data virtualization
- Data federation
Correct answer: Data staging
A data staging area holds extracted data temporarily, decoupling the extraction from transformation and protecting source systems from heavy query loads.
Question 97: In data management, 'data lineage' is MOST useful for which of the following tasks?
- Scheduling ETL jobs in a workflow orchestrator
- Setting access control permissions on datasets
- Determining the physical storage location of a file
- Tracing how data has flowed and transformed from source to consumption (Correct answer)
Correct answer: Tracing how data has flowed and transformed from source to consumption
Data lineage maps the journey of data from its origin through all transformations and systems to its final use, enabling impact analysis and regulatory compliance.
Question 98: Which approach best sustains long-term adoption of data governance practices after initial rollout?
- A single intensive training bootcamp at program launch
- Outsourcing governance activities to a third-party vendor
- Continuous reinforcement through regular communication, success stories, metrics dashboards, and refresher training (Correct answer)
- Strict enforcement with penalties for any policy deviation
Correct answer: Continuous reinforcement through regular communication, success stories, metrics dashboards, and refresher training
Sustained adoption requires ongoing reinforcement through visible successes, updated training, and regular communication that keeps governance relevant and top of mind.
Question 99: What is the primary purpose of system monitoring?
- To justify IT department headcount
- To generate reports for management only
- To track individual employee productivity
- To detect issues proactively and maintain system health (Correct answer)
Correct answer: To detect issues proactively and maintain system health
System monitoring enables proactive issue detection and health maintenance, reducing downtime and ensuring consistent service delivery.
Question 100: Under the GDPR accountability principle, which document is the PRIMARY tool demonstrating compliance with data processing activities?
- Data Protection Impact Assessment (DPIA)
- Records of Processing Activities (RoPA) (Correct answer)
- Data Processing Agreement (DPA) with processors
- Privacy policy published on the company website
Correct answer: Records of Processing Activities (RoPA)
Article 30 requires controllers and processors to maintain Records of Processing Activities (RoPA), which serve as the foundational accountability document under GDPR.
Certified Data Steward (CDS)
The Certified Data Steward (CDS) certification validates professional competency in data stewardship across 15 topic areas including data governance, data quality, metadata and master data management, architecture, automation, and regulatory compliance. Offered by eLearningCurve, it is recognized as an industry-standard credential formalizing the Data Steward role.
Exam Rules
- You can skip questions and return to them later
- Flag questions for review before submitting
- No feedback shown until you submit the entire exam
- Unanswered questions count as wrong — answer everything
- 10 pretest questions are mixed in and don't affect your score
- Timer auto-submits when time runs out
- Your progress is auto-saved every 30 seconds