Data Collection · Data quality · Component
The causes
Where quality problems originate, which is usually one step in collection rather than many small failures.
The deliverable
What it is
Quality problems concentrate. A small number of collection points produce most of the errors, usually because they involve manual entry or an ambiguous field.
Systematic errors are correctable and random ones mostly are not, which makes distinguishing them the first useful step.
One level in
What it is made of
Each element is a constituent part of the component. Follow one to see the attributes it carries.
The origin
Which collection step produces the error.
3 attributes: Collection point · Type · Share of errors
LearnThe pattern
Whether the error is systematic or random.
3 attributes: Pattern · Direction · Magnitude
LearnThe incentive
Whether anyone benefits from the data being wrong.
3 attributes: Incentive · Feeds measure · Addressable
Learn
Fields feeding a target are entered differently from fields that do not. That is not carelessness.
The other components in data quality
The measures
Completeness, accuracy, timeliness and consistency, per data set — measured rather than assumed.
LearnThe fixes
What was changed at source, rather than corrected downstream — because a cleaning step fixes one report and source fixes all of them.
Learn