Timedora
Free Online Flashcard Deck

Data Foundations and Quality Free Online Flashcards

Study Data Foundations and Quality with 12 free online flashcards. Review key terms, definitions, and concepts with this interactive flashcard deck.

12 cards
01
Front

How do primary and foreign keys differ?

Back

A primary key uniquely identifies a row. A foreign key refers to a key in another table so related records can be connected.

02
Front

How should an analyst communicate a data-quality issue?

Back

Report the issue and its consequence: for example, missing product categories may make category-level totals too low. Document the treatment rule and assess its effect.

03
Front

How do completeness and accuracy differ?

Back

Completeness concerns whether required records and fields are present; accuracy concerns whether values reflect the real entities or events. Complete data is not necessarily accurate.

04
Front

Why might purchase records be unsuitable for studying all customers?

Back

A transaction source may reliably record completed purchases but omit people who never bought. Its suitability depends on whether the question concerns purchasers or a broader population.

05
Front

How should the product code `00127` be interpreted?

Back

Treat `00127` as a text identifier, not a quantity. Storing it as a number could remove its leading zero or invite meaningless arithmetic.

06
Front

What does completeness measure in data quality?

Back

Completeness asks whether the required records and fields are present. It does not show whether the values present are accurate.

07
Front

What happens to grain when customer records join to many orders?

Back

A customer with many orders appears in multiple joined rows. The result is at the order grain rather than one row per customer.

08
Front

Why distinguish missing values from zero and empty strings?

Back

A missing value, zero, an empty string, and “not applicable” can mean different things. Do not silently treat them as equivalent.

09
Front

What does grain tell you about an order-line table?

Back

Grain is what one record represents. If each row is one order line, counting rows counts lines, not orders; an order may contain several lines.

10
Front

How should questionable data values be handled?

Back

Do not automatically delete or fill questionable values. Document the treatment rule, assess its effect, and retain information needed to explain the result.

11
Front

What should you examine to judge whether a data source fits a question?

Back

Assess the source’s scope, collection method and timing, update frequency, definitions, units, filters, transformations, and grain against the intended question.

12
Front

What does a relational table’s schema document?

Back

A schema documents table and column names, data types, and often rules such as whether values may be missing or must be unique.