Data Qualities
I'm noticing that data shows up as a key principal in many topics (makes sense, we all love data don't we?!)
I'm wondering if there's a reason that different terminology is used to describe what we're looking for in data.
Examples:
Earthquake - Data Management: Must address data (integrity, verification, limitation)
Earthquake - Modeling best practices: Provide evidence that granularity and quality of data is appropriate.
MfAD - Claims Development: pick low if data is homogeneous, stable, credible.
Models - Validation: Data should be Reliable and Sufficient
So to describe good data we have:
integrity, verification, limitation, granularity, quality, homogeneous, stable, credible, reliable, sufficient.
I guess my question is does CAS really care that we use the specific adjectives listed in each section? Open to comments/ideas on this.
Comments
That's a good point. It's hard to say for certain what would be considered an acceptable answer in a given situation. Part of the issue is that all these readings were written by different groups of people, and the exams are often graded by different people from one cycle to the next.
But exam questions are often plucked directly from the text and if you notice this, the safest strategy is to phrase your answer exactly like in the source text. If you don't immediately recognize which reading the question comes from or if it is Bloom's Taxonomy question, then you have more leeway and can mix & match any of those data characteristics as appropriate for the given question.
Overall, the best description of what we want data to be is probably reliable and sufficient. All the other adjectives are just more specific ways of describing this how to achieve this.