I85PREP
⌘ K
DEA Level 3 study materials
DEA Level 3 study resources

DEA Level 3 practice questions

One question at a time. Build your DEA Level 3 understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Data Management and Governance

Which document is the primary output of a data governance program that defines decision rights and accountabilities for data assets?

  1. A data dictionary
  2. An entity relationship diagram
  3. A data governance charter Correct answer.
  4. A data flow diagram

Explanation

A governance charter establishes the mandate, scope, decision rights, and accountability structure for the program. Technical artifacts like dictionaries and diagrams describe data rather than assign authority over it.

Question 2Data Management and Governance

A data steward role is best described as which of the following?

  1. An external auditor who certifies financial statements
  2. A business-side role accountable for data quality and policy adherence within a domain Correct answer.
  3. A vendor role that supplies third-party reference data
  4. An IT role that provisions database servers and storage

Explanation

Stewards are typically business or domain roles responsible for day-to-day data quality, definitions, and compliance with governance policy. Infrastructure provisioning and external audit are separate functions.

Question 3Data Management and Governance

Which characteristic distinguishes a data owner from a data custodian?

  1. The owner performs backups, while the custodian approves access requests
  2. The owner holds accountability and decision rights, while the custodian implements technical controls Correct answer.
  3. The owner is always in IT, while the custodian is always in finance
  4. The owner defines schemas, while the custodian defines business glossary terms

Explanation

Ownership is an accountability role with authority over access, use, and quality decisions. Custodianship is the operational role that executes controls such as backup, access enforcement, and storage.

Question 4Data Management and Governance

A retail firm wants a single agreed definition of 'active customer' used across marketing, finance, and operations. Which governance artifact addresses this need?

  1. A capacity plan
  2. A backup retention schedule
  3. A network topology map
  4. A business glossary entry Correct answer.

Explanation

A business glossary captures approved business term definitions and their owners so that terms are used consistently. The other artifacts concern operations and infrastructure rather than shared semantics.

Question 5Data Management and Governance

In a data quality dimension model, which dimension measures whether values conform to a defined format, range, or list of allowed values?

  1. Timeliness
  2. Uniqueness
  3. Completeness
  4. Validity Correct answer.

Explanation

Validity assesses conformance to defined rules such as formats, domains, or ranges. Uniqueness addresses duplication, timeliness addresses freshness, and completeness addresses missing values.

Question 6Data Management and Governance

A customer table has 10,000 rows, and 400 rows have a null value in the postal code column. What is the completeness of that column?

  1. 4 percent
  2. 96 percent Correct answer.
  3. Cannot be determined
  4. 40 percent

Explanation

Completeness is the proportion of rows with a populated value: (10,000 - 400) / 10,000 = 0.96, or 96 percent. The 4 percent figure is the null rate rather than completeness.

Question 7Data Management and Governance

Which practice most directly supports the principle of data minimization under privacy regulation?

  1. Replicating production data into every development environment
  2. Broadcasting data access to all staff to increase transparency
  3. Retaining all source data indefinitely to support future analytics
  4. Collecting only the attributes needed for a stated purpose and deleting them when no longer required Correct answer.

Explanation

Data minimization limits collection and retention to what is necessary for a defined purpose. Indefinite retention and broad replication expand exposure and conflict with the principle.

Question 8Data Management and Governance

A data classification scheme uses Public, Internal, Confidential, and Restricted. What is the primary purpose of assigning these labels?

  1. To specify the physical storage medium for each dataset
  2. To determine the controls required for handling, access, and sharing Correct answer.
  3. To rank datasets by query performance
  4. To define the ETL schedule for each dataset

Explanation

Classification drives proportionate controls such as access restrictions, encryption, and sharing rules. It does not determine storage media, load schedules, or performance.

Question 9Data Management and Governance

Which element is essential in a data retention policy?

  1. A list of all employees with database accounts
  2. The physical location of each backup tape
  3. The vendor contract for the reporting tool
  4. Defined retention periods tied to legal, regulatory, and business requirements Correct answer.

Explanation

Retention policies must specify how long each class of data is kept and the basis for that period, plus disposition rules. The other items are operational or procurement details.

Question 10Data Management and Governance

A regulation requires that personal data be erased upon customer request. Which capability is most important to enable compliance?

  1. The ability to archive data to cold storage
  2. The ability to locate and delete all copies of a subject's data across systems Correct answer.
  3. The ability to generate ad hoc reports quickly
  4. The ability to compress backups to save storage

Explanation

Erasure obligations require knowing where subject data resides, including copies, so it can be located and removed. Compression, reporting speed, and archiving do not address erasure.

Question 11Data Management and Governance

Which statement best describes a data lineage capability?

  1. It traces the origin and transformation path of data from source to consumption Correct answer.
  2. It schedules batch jobs across servers
  3. It measures query latency across clusters
  4. It enforces password complexity for database users

Explanation

Lineage records where data came from and how it was transformed, supporting impact analysis and audit. Security, performance, and scheduling are separate concerns.

Question 12Data Management and Governance

An analyst must assess which downstream reports would be affected if a source column is renamed. Which governance capability is most directly relevant?

  1. Impact analysis through data lineage Correct answer.
  2. Master data consolidation
  3. Row-level security policy testing
  4. Storage tiering optimization

Explanation

Impact analysis uses lineage to identify downstream consumers of a changed object. Security testing, tiering, and consolidation address different problems.

Question 13Data Management and Governance

What is the central goal of master data management?

  1. Maximizing the compression ratio of archival storage
  2. Reducing the number of ETL developers required
  3. Replacing all transactional systems with a single platform
  4. Providing a single, trusted, and consistent view of core business entities across systems Correct answer.

Explanation

MDM establishes authoritative, consistent records for entities such as customer, product, and supplier across the organization. Cost, staffing, and platform replacement are not its objectives.

Question 14Data Management and Governance

Two source systems store the same customer with different spellings of the name and different identifiers. Which MDM activity resolves this?

  1. Query optimization
  2. Capacity planning
  3. Entity resolution and matching Correct answer.
  4. Schema versioning

Explanation

Entity resolution uses matching rules and survivorship to determine when records refer to the same real-world entity and to produce a golden record. The other activities are unrelated to identity reconciliation.

Question 15Data Management and Governance

In a data quality control process, what is the role of a data quality rule?

  1. It assigns user access privileges to a schema
  2. It determines the backup frequency for a database
  3. It specifies a testable condition that data must satisfy and the action to take on failure Correct answer.
  4. It defines the physical data model for a warehouse

Explanation

A quality rule encodes an expectation, such as a range or format check, plus handling for violations. Modeling, access control, and backup are separate functions.

Question 16Data Management and Governance

A governance council is deciding who may approve a new use of sensitive customer data. Which mechanism is most appropriate?

  1. Deferring all decisions to the data vendor
  2. Automatic approval for any request from IT
  3. An informal email thread among analysts
  4. A documented approval workflow with defined decision rights Correct answer.

Explanation

Governance requires explicit, documented decision rights and an auditable approval path for sensitive uses. Informal or automatic approval bypasses accountability.

Question 17Data Management and Governance

Which metric best indicates whether a data governance program is being adopted by the business?

  1. The average CPU utilization of the warehouse
  2. The number of servers in the data center
  3. The count of ETL jobs executed per night
  4. The proportion of critical data elements with assigned, active stewards Correct answer.

Explanation

Adoption is reflected in business participation, such as stewardship coverage of critical data elements. Infrastructure and job metrics measure operations rather than governance engagement.

Question 18Data Management and Governance

What distinguishes a data domain from a data product in a modern data management model?

  1. A domain is a database schema, while a product is a table within it
  2. A domain is a backup set, while a product is a restore point
  3. A domain is an organizational boundary of accountability, while a product is an owned, consumable data asset with defined quality commitments Correct answer.
  4. A domain is a vendor contract, while a product is a license

Explanation

Domains organize accountability by business area, while data products are treated as assets with owners, interfaces, and service commitments. Schema and backup concepts are technical, not organizational.

Question 19Data Management and Governance

Which control best enforces least privilege for analysts querying a sensitive table?

  1. Granting access only to the specific columns and rows each analyst requires Correct answer.
  2. Disabling logging to reduce overhead
  3. Granting full table access to the entire analytics group
  4. Sharing a single service account among all analysts

Explanation

Least privilege restricts access to the minimum data needed, often via column- and row-level controls. Broad group access, shared accounts, and disabled logging all increase risk.

Question 20Data Management and Governance

A regulator asks an organization to demonstrate who accessed a patient record and when. Which capability answers this?

  1. Data compression ratios
  2. Schema drift reports
  3. ETL throughput metrics
  4. Audit logging of data access events Correct answer.

Explanation

Audit logs record access events with identity and timestamps, providing the evidence required. The other capabilities concern performance, storage, or structure.

Question 21Data Management and Governance

Which scenario best illustrates a metadata management failure rather than a data quality failure?

  1. A product record is duplicated across two systems
  2. A customer record contains an invalid email address
  3. A report column labeled 'revenue' is later found to mean gross sales, not net sales Correct answer.
  4. A transaction amount is recorded as negative

Explanation

Metadata failures concern the meaning, definition, or documentation of data rather than the correctness of values. Invalid emails, wrong signs, and duplicates are data quality defects.

Question 22Data Management and Governance

What is the purpose of a data quality scorecard presented to executive leadership?

  1. To list every database table in the enterprise
  2. To document the source code of ETL pipelines
  3. To record employee training completion
  4. To summarize quality against defined dimensions and targets for prioritized data Correct answer.

Explanation

A scorecard communicates quality status against targets so leaders can prioritize remediation. Table inventories, code, and training records do not serve that purpose.

Question 23Data Management and Governance

An organization must keep records for seven years for legal reasons but wants to delete operational data after two years. What is the most appropriate policy approach?

  1. Apply the shortest period uniformly to all data
  2. Retain all data indefinitely to avoid risk
  3. Apply differentiated retention schedules by data category and legal obligation Correct answer.
  4. Delete all data after two years regardless of obligation

Explanation

Retention schedules should reflect the specific legal, regulatory, and business requirements for each data category. Uniform short retention risks noncompliance, and indefinite retention increases exposure.

Question 24Data Management and Governance

Which activity is part of data stewardship rather than data engineering?

  1. Configuring cluster autoscaling
  2. Managing disk partitions
  3. Reviewing and approving business definitions for critical data elements Correct answer.
  4. Tuning SQL queries for performance

Explanation

Stewardship involves business-side decisions about definitions, quality expectations, and policy adherence. Performance tuning, autoscaling, and partitioning are engineering tasks.

Question 25Data Management and Governance

A data governance program wants to measure whether data quality issues are being resolved promptly. Which metric is most appropriate?

  1. Total storage consumed by the warehouse
  2. Average time from issue identification to verified resolution Correct answer.
  3. Count of dashboards published per quarter
  4. Number of data sources onboarded to the catalog

Explanation

Resolution cycle time directly measures responsiveness to identified quality issues. Onboarding counts, storage, and dashboard counts measure activity volume rather than remediation effectiveness.

Question 26Data Management and Governance

Which role is typically accountable for the day-to-day quality and proper use of data within a specific business domain?

  1. The external auditor
  2. The data steward Correct answer.
  3. The chief information security officer
  4. The data engineer

Explanation

Data stewardship is a business-side accountability for the meaning, quality, and appropriate use of data within an assigned domain. Stewards resolve definitional conflicts and monitor quality, while technical roles build pipelines and security roles set access policy.

Question 27Data Management and Governance

An organization defines a metric as 'total revenue' in three separate reports, and each report computes it differently. Which governance artifact most directly prevents this inconsistency?

  1. A business glossary with approved metric definitions Correct answer.
  2. A network diagram of the reporting servers
  3. A backup and recovery schedule for the reporting database
  4. A list of report owners sorted by department

Explanation

A business glossary records agreed business terms and their approved definitions so that reports reuse one meaning. It addresses semantic consistency, which is the root cause when the same metric is calculated differently across reports.

Question 28Data Management and Governance

Which statement describes the function of a data owner in a governance operating model?

  1. They design the physical schema for operational databases.
  2. They monitor server uptime for the analytics platform.
  3. They approve access requests and accept risk for an assigned data domain. Correct answer.
  4. They write the ETL code that populates the warehouse.

Explanation

A data owner is an accountable business role that decides who may use a data domain and accepts the associated risk. This accountability is distinct from the technical work of building pipelines or schemas.

Question 29Data Management and Governance

A data catalog records, for each dataset, its owner, its classification, and the systems that produce it. Which governance objective does this most directly serve?

  1. Reducing storage costs through compression
  2. Making data assets discoverable and accountable Correct answer.
  3. Increasing query throughput on the warehouse
  4. Enforcing referential integrity in source systems

Explanation

A data catalog is a metadata inventory that helps users find data and identifies who is accountable for it. Discoverability and accountability are governance outcomes, whereas performance and integrity are technical concerns.

Question 30Data Management and Governance

An organization wants to track the origin and transformation history of every field in a regulatory report. Which capability must it implement?

  1. Column-level lineage Correct answer.
  2. Schema versioning only
  3. Table-level row counts
  4. Data masking at rest

Explanation

Column-level lineage records how each field is derived from sources through transformations, which is what tracing a report field to its origin requires. Table-level information is too coarse to show field derivation.

Question 31Data Management and Governance

A governance team needs to decide whether a proposed analytics use of sensitive data is acceptable. Which mechanism is designed to make that decision consistently?

  1. An automated schema migration script
  2. A data use review board with documented criteria Correct answer.
  3. A nightly data quality batch job
  4. A dashboard of storage utilization

Explanation

A review board applies documented criteria to proposed uses so that decisions are consistent and defensible. Automated jobs and dashboards provide information but do not adjudicate acceptable use.

Question 32Data Management and Governance

Which control most directly limits a contractor to seeing only the customer records for their assigned region?

  1. Row-level security policies based on region Correct answer.
  2. Anonymizing the customer identifier column
  3. A quarterly access recertification campaign
  4. Encrypting the entire customer table with one key

Explanation

Row-level security filters which records a user can retrieve according to attributes such as region. It enforces the restriction at query time rather than relying on periodic review.

Question 33Data Management and Governance

A privacy regulation requires that personal data be kept only as long as necessary for its stated purpose. Which governance practice best demonstrates compliance?

  1. Retaining all data indefinitely in case it is needed
  2. Copying personal data into a separate archive without rules
  3. Defining purpose-based retention periods and automating deletion Correct answer.
  4. Relying on employees to delete records manually when convenient

Explanation

Storage limitation requires that retention be tied to a defined purpose and enforced, which purpose-based schedules with automated deletion achieve. Indefinite retention or ad hoc manual deletion cannot demonstrate compliance.

Question 34Data Management and Governance

A data quality dimension measures whether values conform to an expected format, such as a two-letter state code. Which dimension is this?

  1. Uniqueness
  2. Timeliness
  3. Completeness
  4. Validity Correct answer.

Explanation

Validity assesses conformance to a defined format, type, or domain of allowed values. Uniqueness concerns duplicates, timeliness concerns availability when needed, and completeness concerns missing values.

Question 35Data Management and Governance

An organization wants a single trusted record for each product across four business units. Which discipline addresses this requirement?

  1. Capacity planning
  2. Application performance monitoring
  3. Master data management Correct answer.
  4. Data loss prevention

Explanation

Master data management creates and maintains a consistent, authoritative record for shared entities such as products, customers, and suppliers across units. The other options concern operational monitoring or security, not shared reference entities.

Question 36Data Management and Governance

Which practice best supports the principle that data should be collected only for specified, legitimate purposes?

  1. Storing a copy of all data in a shared analytics sandbox
  2. Collecting every available field to avoid future requests
  3. Allowing any team to repurpose data without review
  4. Documenting the purpose for each data element at collection time Correct answer.

Explanation

Purpose specification requires that the reason for collecting each element be defined and recorded so that use can be limited to it. Broad collection or unreviewed repurposing undermines that principle.

Question 37Data Management and Governance

A governance program reports that 70 percent of critical data elements have an assigned steward. Which type of measure is this?

  1. An adoption or coverage metric for the governance program Correct answer.
  2. A data quality score for a specific dataset
  3. A technical latency measurement
  4. A financial return-on-investment calculation

Explanation

Coverage of stewardship assignments measures how widely the governance operating model has been adopted. It indicates program reach rather than the quality of any particular dataset.

Question 38Data Management and Governance

A data domain team discovers that two systems assign different identifiers to the same supplier. Which action belongs to master data management rather than to pipeline engineering?

  1. Adding an index to speed up the join query
  2. Increasing the parallel worker count in the ingestion job
  3. Defining survivorship rules to choose the trusted attribute values Correct answer.
  4. Compressing the staging files to save space

Explanation

Survivorship rules determine which attribute values from competing sources are trusted in the master record, which is a core MDM design decision. Performance and storage tuning are engineering tasks that do not resolve entity identity.

Question 39Data Management and Governance

A regulator asks an organization to prove that a specific analyst viewed a patient record on a given date. Which combination of capabilities is required?

  1. Data masking plus tokenization
  2. Data lineage plus schema versioning
  3. Backup plus disaster recovery
  4. Access logging plus identity management Correct answer.

Explanation

Answering who accessed what and when requires audit logs of access events tied to authenticated identities. Lineage, masking, and recovery capabilities serve different purposes and do not record individual access events.

Question 40Data Management and Governance

A report shows correct totals, but analysts cannot tell which source system supplied a column or when it was last refreshed. Which governance weakness does this illustrate?

  1. Insufficient metadata management Correct answer.
  2. Weak encryption standards
  3. Excessive data duplication
  4. Poor network throughput

Explanation

Metadata management provides descriptive and operational information such as source, ownership, and refresh status. When values are correct but their context is undocumented, the gap is metadata rather than data quality.

Question 41Data Architecture and Modeling

In relational theory, which property must a candidate key satisfy in addition to uniqueness?

  1. It must be non-null (no component may be null) Correct answer.
  2. It must be a single attribute
  3. It must be referenced by a foreign key
  4. It must be a surrogate integer

Explanation

A candidate key is a minimal set of attributes whose values uniquely identify each tuple, and key attributes cannot be null because nulls do not identify anything. Minimality means no proper subset is itself a key, which does not require a single attribute.

Question 42Data Architecture and Modeling

A table stores order lines with columns (order_id, line_no, product_id, qty). The combination (order_id, line_no) is unique and neither column alone is unique. What is this combination called?

  1. A non-identifying attribute
  2. A foreign key
  3. A composite candidate key Correct answer.
  4. A superkey that is not a key

Explanation

A candidate key made of two or more attributes is a composite (compound) key. Because the pair is unique and minimal, it qualifies as a candidate key and could serve as the primary key.

Question 43Data Architecture and Modeling

Which normal form eliminates partial dependencies of non-key attributes on part of a composite primary key?

  1. First normal form
  2. Boyce-Codd normal form
  3. Third normal form
  4. Second normal form Correct answer.

Explanation

Second normal form requires that every non-key attribute depend on the whole of every candidate key, removing partial dependencies. Third normal form then removes transitive dependencies on non-key attributes.

Question 44Data Architecture and Modeling

A table has columns (employee_id, skill_code, skill_name) where skill_name depends only on skill_code. Which normalization problem does this illustrate?

  1. A transitive dependency on a non-key attribute Correct answer.
  2. A lossless-join violation
  3. A partial dependency on part of a composite key
  4. A multivalued dependency

Explanation

Because skill_name is determined by skill_code, which is not a candidate key, the table has a transitive dependency and violates third normal form. Resolving it means moving skill_name to a separate skill table.

Question 45Data Architecture and Modeling

In dimensional modeling, which statement best distinguishes a fact table from a dimension table?

  1. Fact tables store numeric measures and foreign keys; dimension tables hold descriptive attributes Correct answer.
  2. Fact tables contain only surrogate keys and no measures
  3. Fact tables hold descriptive text; dimension tables hold numeric measures
  4. Fact tables must be normalized to 3NF; dimension tables are denormalized

Explanation

A fact table captures measurable events with foreign keys to dimensions, while dimension tables provide the descriptive context used for filtering and grouping. This star-shaped arrangement is the core of dimensional modeling.

Question 46Data Architecture and Modeling

A data warehouse team needs to track the full history of changes to a customer's address so reports can show the address as of any past date. Which technique fits best?

  1. A junk dimension combining address attributes
  2. Type 2 slowly changing dimension with validity dates Correct answer.
  3. Type 1 slowly changing dimension overwrite
  4. A factless fact table

Explanation

A Type 2 dimension preserves history by inserting a new row per change with effective and expiration dates and a current flag. This allows point-in-time reporting on prior attribute values.

Question 47Data Architecture and Modeling

What is the primary purpose of a surrogate key in a dimension table?

  1. To encode the natural business meaning of the row
  2. To reduce the number of dimension attributes
  3. To enforce referential integrity to the source system
  4. To insulate the warehouse from changes in source system keys Correct answer.

Explanation

Surrogate keys are meaningless integers assigned by the warehouse so that changes or overlaps in source natural keys do not ripple through the model. They also enable Type 2 versioning where one natural key maps to many rows.

Question 48Data Architecture and Modeling

A team needs a schema optimized for ad hoc analytical queries across many dimensions with minimal join complexity. Which design is most appropriate?

  1. Third normal form enterprise model
  2. Star schema with denormalized dimensions Correct answer.
  3. Entity-attribute-value model
  4. Snowflake schema with fully normalized dimensions

Explanation

A star schema keeps dimensions denormalized so queries join facts to a small number of dimension tables, simplifying and speeding analytical access. Snowflaking normalizes dimensions and adds joins.

Question 49Data Architecture and Modeling

Which modeling approach stores each entity as a set of time-stamped rows and preserves the full audit trail of all attribute values?

  1. Flat wide-table modeling
  2. Anchor modeling
  3. Data vault modeling Correct answer.
  4. Dimensional star modeling

Explanation

Data vault organizes data into hubs (business keys), links (relationships), and satellites (time-stamped descriptive attributes), preserving full history and auditability. It is designed for integration and traceability rather than direct end-user querying.

Question 50Data Architecture and Modeling

In data vault, what does a link table primarily represent?

  1. A pre-aggregated summary of measures
  2. A relationship between two or more hubs Correct answer.
  3. A time-stamped set of descriptive attributes
  4. A unique list of business keys

Explanation

A link captures the association between hubs (for example, customer and order) and holds the keys that connect them. Descriptive context and history belong in satellites attached to hubs or links.