Data Analytics: Foundations to Practice · Types and Sources of Data
Data Quality Dimensions
Data quality is not a single property but a set of distinct dimensions, and a dataset can score well on one while failing badly on another. This chapter covers the core dimensions worth checking before trusting any dataset.
Accuracy measures how well data reflects the real-world thing it's meant to represent, such as whether a recorded sale amount matches what was actually charged, while completeness measures whether all the data that should exist for a given record actually does, such as whether every customer record includes a genuinely filled-in email field rather than a blank one masquerading as complete data [2].
Key Takeaways
- Accuracy measures how well data reflects reality; completeness measures whether all expected data for a record actually exists.
- Consistency measures uniformity across systems; validity measures whether data conforms to expected format and rules.
- Timeliness measures how current data is relative to when it's needed; uniqueness measures whether each entity is represented exactly once.
- A dataset can score well on one quality dimension while failing badly on another, so a thorough check examines multiple dimensions, not just one.