Data Quality interview questions

1. A.A named collection of Great Expectations assertions about a dataset that together define its quality contract.
2. B.A check that a table's row count falls within an expected range or relative change, catching partial loads and runaway duplications.
3. C.completeness, accuracy, consistency, timeliness, uniqueness, validity
4. D.A check confirming a column or key has no duplicate values, protecting primary-key and grain assumptions.

Answer + AI explanation with Pro

1. A.anomaly detection test
2. B.Great Expectations checkpoint
3. C.reconciliation
4. D.data quality dimensions

Answer + AI explanation with Pro

1. A.data quality dimensions — A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.
2. B.data quality dimensions — completeness, accuracy, consistency, timeliness, uniqueness, validity
3. C.data quality dimensions — A check verifying that foreign-key values in one table all exist in the referenced table, catching orphaned records.
4. D.data quality dimensions — A data-quality check that flags metrics deviating from their historical baseline rather than failing a fixed threshold, catching unexpected drifts.

Answer + AI explanation with Pro

1. A.Monitoring that alerts when upstream columns are added, removed, or change type so downstream transforms do not silently break.
2. B.an unexpected change in source schema that can break pipelines
3. C.A data-quality check that flags metrics deviating from their historical baseline rather than failing a fixed threshold, catching unexpected drifts.
4. D.A check that a table's row count falls within an expected range or relative change, catching partial loads and runaway duplications.

Answer + AI explanation with Pro

1. A.schema drift
2. B.anomaly detection test
3. C.freshness check
4. D.freshness SLA

Answer + AI explanation with Pro

1. A.schema drift — Automated monitoring that flags metrics (row counts, nulls, distributions) deviating from learned historical patterns, a pillar of data observability.
2. B.schema drift — An open-source Python framework for declaring data expectations (assertions) and validating datasets, producing data docs and validation results.
3. C.schema drift — an unexpected change in source schema that can break pipelines
4. D.schema drift — A data-quality check that flags metrics deviating from their historical baseline rather than failing a fixed threshold, catching unexpected drifts.

Answer + AI explanation with Pro

1. A.The practice of monitoring data health across freshness, volume, schema, distribution, and lineage to detect and resolve issues quickly.
2. B.an agreed schema/SLA between data producers and consumers
3. C.A documented maximum acceptable staleness for a dataset, breached when its last update falls outside the agreed window.
4. D.Asserting that incoming data matches an expected structure (columns, types, nullability) before it is accepted into a pipeline.

Answer + AI explanation with Pro

1. A.anomaly detection test
2. B.data observability
3. C.data contract
4. D.Soda

Answer + AI explanation with Pro

1. A.data contract — Comparing record counts or aggregates between source and target after a load to confirm no data was lost or duplicated in transit.
2. B.data contract — Monitoring that alerts when upstream columns are added, removed, or change type so downstream transforms do not silently break.
3. C.data contract — an agreed schema/SLA between data producers and consumers
4. D.data contract — A pipeline safeguard that halts downstream loads when quality checks fail, preventing bad data from propagating to consumers.

Answer + AI explanation with Pro

1. A.An open-source Python framework for declaring data expectations (assertions) and validating datasets, producing data docs and validation results.
2. B.Asserting that incoming data matches an expected structure (columns, types, nullability) before it is accepted into a pipeline.
3. C.an unexpected change in source schema that can break pipelines
4. D.The practice of monitoring data health across freshness, volume, schema, distribution, and lineage to detect and resolve issues quickly.

Answer + AI explanation with Pro

1. A.freshness SLA
2. B.Great Expectations
3. C.referential integrity test
4. D.anomaly detection test

Answer + AI explanation with Pro

1. A.Great Expectations — an agreed schema/SLA between data producers and consumers
2. B.Great Expectations — An open-source Python framework for declaring data expectations (assertions) and validating datasets, producing data docs and validation results.
3. C.Great Expectations — The practice of monitoring data health across freshness, volume, schema, distribution, and lineage to detect and resolve issues quickly.
4. D.Great Expectations — A documented maximum acceptable staleness for a dataset, breached when its last update falls outside the agreed window.

Answer + AI explanation with Pro

1. A.A reusable bundle pairing data batches with an expectation suite and actions, run to validate data and emit results in a pipeline.
2. B.A pipeline safeguard that halts downstream loads when quality checks fail, preventing bad data from propagating to consumers.
3. C.A declarative data-quality assertion written in SodaCL (such as row count, missing percent, or freshness) that runs against a dataset.
4. D.A data-quality tool using a declarative checks language (SodaCL) to scan datasets for anomalies, freshness, and rule violations in pipelines.

Answer + AI explanation with Pro

1. A.reconciliation
2. B.Soda
3. C.schema validation
4. D.data observability

Answer + AI explanation with Pro

1. A.Soda — A data-quality tool using a declarative checks language (SodaCL) to scan datasets for anomalies, freshness, and rule violations in pipelines.
2. B.Soda — A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.
3. C.Soda — A pipeline safeguard that halts downstream loads when quality checks fail, preventing bad data from propagating to consumers.
4. D.Soda — A reusable bundle pairing data batches with an expectation suite and actions, run to validate data and emit results in a pipeline.

Answer + AI explanation with Pro

1. A.A check confirming a column or key has no duplicate values, protecting primary-key and grain assumptions.
2. B.A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.
3. C.A formal, versioned agreement between data producers and consumers specifying schema, semantics, and quality guarantees, enforced in CI or at ingestion.
4. D.an agreed schema/SLA between data producers and consumers

Answer + AI explanation with Pro

1. A.dbt unit test
2. B.data contract
3. C.schema validation
4. D.Soda

Answer + AI explanation with Pro

1. A.data contract — Asserting that incoming data matches an expected structure (columns, types, nullability) before it is accepted into a pipeline.
2. B.data contract — Comparing aggregate totals between source and target (counts, sums) to confirm a pipeline neither dropped nor duplicated data.
3. C.data contract — A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.
4. D.data contract — A formal, versioned agreement between data producers and consumers specifying schema, semantics, and quality guarantees, enforced in CI or at ingestion.

Answer + AI explanation with Pro

1. A.A test asserting that the newest data in a table is within an acceptable age threshold, catching stalled or delayed pipelines.
2. B.An open-source Python framework for declaring data expectations (assertions) and validating datasets, producing data docs and validation results.
3. C.Monitoring that alerts when upstream columns are added, removed, or change type so downstream transforms do not silently break.
4. D.Comparing record counts or aggregates between source and target after a load to confirm no data was lost or duplicated in transit.

Answer + AI explanation with Pro

1. A.schema validation
2. B.reconciliation
3. C.freshness check
4. D.anomaly detection test

Answer + AI explanation with Pro

1. A.freshness check — A pipeline safeguard that halts downstream loads when quality checks fail, preventing bad data from propagating to consumers.
2. B.freshness check — A test asserting that the newest data in a table is within an acceptable age threshold, catching stalled or delayed pipelines.
3. C.freshness check — A check that a table's row count falls within an expected range or relative change, catching partial loads and runaway duplications.
4. D.freshness check — A declarative data-quality assertion written in SodaCL (such as row count, missing percent, or freshness) that runs against a dataset.

Answer + AI explanation with Pro

1. A.A reusable bundle pairing data batches with an expectation suite and actions, run to validate data and emit results in a pipeline.
2. B.an unexpected change in source schema that can break pipelines
3. C.Automated monitoring that flags metrics (row counts, nulls, distributions) deviating from learned historical patterns, a pillar of data observability.
4. D.A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.

Answer + AI explanation with Pro

1. A.schema drift
2. B.anomaly detection
3. C.freshness check
4. D.freshness SLA

Answer + AI explanation with Pro

1. A.anomaly detection — completeness, accuracy, consistency, timeliness, uniqueness, validity
2. B.anomaly detection — A named collection of Great Expectations assertions about a dataset that together define its quality contract.
3. C.anomaly detection — Automated monitoring that flags metrics (row counts, nulls, distributions) deviating from learned historical patterns, a pillar of data observability.
4. D.anomaly detection — A documented maximum acceptable staleness for a dataset, breached when its last update falls outside the agreed window.

Answer + AI explanation with Pro

1. A.an unexpected change in source schema that can break pipelines
2. B.A check that a table's row count falls within an expected range or relative change, catching partial loads and runaway duplications.
3. C.Automated monitoring that flags metrics (row counts, nulls, distributions) deviating from learned historical patterns, a pillar of data observability.
4. D.A check verifying that foreign-key values in one table all exist in the referenced table, catching orphaned records.

Answer + AI explanation with Pro

1. A.schema drift detection
2. B.referential integrity test
3. C.Soda
4. D.volume test

Answer + AI explanation with Pro

1. A.referential integrity test — A check that a table's row count falls within an expected range or relative change, catching partial loads and runaway duplications.
2. B.referential integrity test — A check verifying that foreign-key values in one table all exist in the referenced table, catching orphaned records.
3. C.referential integrity test — A named collection of Great Expectations assertions about a dataset that together define its quality contract.
4. D.referential integrity test — Automated monitoring that flags metrics (row counts, nulls, distributions) deviating from learned historical patterns, a pillar of data observability.

Answer + AI explanation with Pro

1. A.an unexpected change in source schema that can break pipelines
2. B.A data-quality tool using a declarative checks language (SodaCL) to scan datasets for anomalies, freshness, and rule violations in pipelines.
3. C.The practice of monitoring data health across freshness, volume, schema, distribution, and lineage to detect and resolve issues quickly.
4. D.Asserting that incoming data matches an expected structure (columns, types, nullability) before it is accepted into a pipeline.

Answer + AI explanation with Pro

1. A.freshness check
2. B.schema drift
3. C.data observability
4. D.Soda check

Answer + AI explanation with Pro

1. A.data observability — A formal, versioned agreement between data producers and consumers specifying schema, semantics, and quality guarantees, enforced in CI or at ingestion.
2. B.data observability — A dbt test that feeds a model fixed mock rows and asserts the transformed output, validating logic independently of warehouse data.
3. C.data observability — A named collection of Great Expectations assertions about a dataset that together define its quality contract.
4. D.data observability — The practice of monitoring data health across freshness, volume, schema, distribution, and lineage to detect and resolve issues quickly.

Answer + AI explanation with Pro