Download Full Version DQ-1220 DAMA Exam

Data transformation includes operations such as format changes, structural changes, semantic conversion, de-duplication, and re-ordering. These activities modify source information so that it conforms to the syntactic, structural, or semantic requirements of a target environment.

A format transformation may convert dates from DD/MM/YYYY to an ISO representation. Structural transformation may split, combine, flatten, or restructure attributes. Semantic conversion changes representation while preserving intended meaning—for example, translating a source status code into the standardized enterprise code. De-duplication identifies multiple records representing the same real-world entity, while re-ordering changes the sequence or organization of records or attributes. The listed combination aligns directly with DAMA-oriented transformation guidance.

The distinction from the distractors is important. Organizational change, infrastructure replacement, or application modernization may trigger data transformation, but they are not themselves data-transformation techniques. “Re-duplication” is also inconsistent with the objective of improving integrated datasets.

Transformation is strongly connected to Data Quality. Poorly specified conversion rules can create invalid values, truncate data, introduce semantic inconsistencies, or produce duplicate entities. Consequently, transformations should be documented through mappings and metadata, tested against quality rules, reconciled with source totals, and monitored for exceptions.

Reference Topics: DAMA-DMBOK2 Chapter 8 — Transformation and Mapping; ETL/ELT; Chapter 13 — Standardization, Cleansing, De-duplication and Validation.

===============

The values should first be treated as potential exceptions requiring investigation. Profiling identifies anomalies; it does not automatically prove that every unusual value is incorrect or specify the appropriate remediation.

Negative age and 263 years are highly improbable for a conventional person-age attribute, suggesting a validity or reasonableness defect. However, the Data Quality team should confirm the business definition, unit of measure, derivation logic, source mappings, special-value conventions, and intended population before changing the records.

Blindly replacing values with an average introduces fabricated data and destroys traceability. Deleting the records may also remove legitimate information elsewhere in the record. Increasing the datatype range addresses technical storage rather than semantic correctness.

A disciplined Data Quality process moves from profiling to rule validation, issue assessment, root-cause analysis, remediation, and monitoring. The business rule may ultimately specify an acceptable age range, but that rule should be governed and documented rather than inferred from individual anomalies.

Metadata is essential because it should describe whether Age is stored directly, calculated from Date of Birth, or encoded using another convention.

Reference Topics: DAMA-DMBOK2 Chapter 13 — Profiling; Validity; Reasonableness; Root-Cause Analysis; Data Quality Rules; Metadata.

===============

The correct definition is the fundamental organisation of a system, and the principles governing its design and evolution. DAMA-DMBOK2 incorporates the ISO/IEC 42010 architectural concept when establishing the theoretical basis for Data Architecture.

The fuller ISO formulation describes architecture in terms of the fundamental organization of a system embodied in its components, their relationships to one another and to the environment, together with the principles governing design and evolution. DAMA uses this concept to distinguish true architecture from a collection of diagrams or implementation specifications.

Option E is therefore incomplete: architectural artifacts document architecture, but the artifacts themselves are not the architecture. Option C focuses too narrowly on implementation and maintenance, while options B and D describe potential objectives or constraints rather than the definition.

Applied to enterprise data, this means architecture establishes the high-level organization of data assets, information flows, platforms, relationships, and guiding principles by which the data environment evolves.

Data Quality benefits because architectural principles establish where authoritative information originates, how it moves, how duplication is controlled, and where governance and quality controls must operate.

Reference Topics: DAMA-DMBOK2 Chapter 4 — Architecture Definition; ISO/IEC 42010; Enterprise Data Architecture; Architectural Principles; Chapter 13 — Quality by Design.

===============

A meaningful Data Quality metric begins with the business requirement and intended use. Quality is fundamentally contextual: data is considered fit for purpose only relative to the activity, decision, report, or process that depends on it.

For example, a delivery address used for same-day logistics may require stricter timeliness and completeness thresholds than an address retained solely for historical analysis. Without understanding the business use, a numerical quality target becomes arbitrary.

The business requirement should identify what failure matters, which dimension applies, how quality will be measured, the acceptable threshold, who owns the requirement, and what response is expected when the threshold is missed.

DAMA-aligned public guidance follows this same logic by recommending that quality rules and dimensions be prioritized according to user needs and purpose.

Technology selection comes later. Profiling tools help measure data, but they cannot determine the business meaning of an acceptable result.

Governance and stewardship should approve the requirement, while Metadata Management should preserve the definition, rule, owner, and lineage.

Reference Topics: DAMA-DMBOK2 Chapter 13 — Define Data Quality Requirements; Fitness for Purpose; Metrics; Thresholds; Business Rules.

===============