0

You are working for an organization that receives data from other organizations in different data formats (geospatial files, spreadsheets, text files, etc.)

Is there specific things you would ask these organizations to make sure that the data you are receiving from them is of good quality data?

  • 2
    Is there a metadata statement? Is it accurate? with geospatial data there is capture date, capture source and accuracy that needs to be addressed. Equally important (but for different reasons) is data ownership and use restrictions. All of this is in good metadata. Spatial data with no metadata (missing) or unpopulated is suspect and unknown quality. – Michael Stimson Apr 30 '15 at 23:30
  • 1
    Quality is in the eye of the beholder. A dataset gathered for one purpose may be good for that purpose, but poor for another; so really you need to know for what purpose the data was gathered, how it was gathered and if there are any known limitations on usage. – nmtoken Mar 31 '16 at 17:49
1

Pretty broad question. Yes is the simple answer. What specific things depends on the data and organizational relationship.

As a customer or authority, I'd be insisting on certain minimum integrity checks before it was submitted to me. That might require working out schema before hand and/or writing custom tools. I might also QC/QA spot check their submissions with some of the same tools or checks, or even some of my own that I don't necessarily give them.

If we're 'equal' or cooperative departments, then there's less I can do from a demand standpoint and more about working together and understanding each other's process to improve both sides with minimal extra effort. I wouldn't expect them to do some extra involved process that only benefits me and my use of their data.

And of course, as Michael mentions, good/accurate metadata/documentation is always good.

0

First of all you have to define what properties good data is expected to have. You may define this in terms of completeness, spatial accuracy, topological correctness, conformance with given data scheme, and others.

All of this is described in the DIN EN ISO 19000 documents.

As a customer you may request a formal accomplishment of conformance with the quality elements, eg shown by logs of some software (eg FME or OpenJump workflow) or an independent checker service.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.