move aggregation_level (6.14) to its own section

labordynamicsinstitute / qwi_schemas

Unofficial LEHD Schema files

https://lehd.ces.census.gov/data/schema/

Creative Commons Zero v1.0 Universal

1 stars 6 forks source link

move aggregation_level (6.14) to its own section #115

Open srt1 opened 5 years ago

srt1 commented 5 years ago

The aggregation_level is not really a categorical variable on its own - it is conceptually a combination of categorical variables. Plus, with PSEO, we will be adding a second aggregation_level_pseo table, since it would be too much of a mess to try to shoehorn it into the QWI/J2J version. I suggest something like:

adding one more nesting level to the existing 6.1.x-6.13.x, to make it 6.1.1.x-6.1.13.x
Add section 6.2 to describe aggregation levels
- 6.2.1 - aggregation_level.csv (J2J/QWI)
- 6.2.2 - aggregation_level_pseo.csv

heathhayward commented 5 years ago

I like this idea. Again, should be reletively simple to restructure this as needed. We will need some text to accompany the PSEO aggregation levels section.

srt1 commented 5 years ago

On second thought, maybe not; the section is "Identifiers," not "Characteristics - which is a bit more generic. We could potentially subdivide this into different types of identifiers (characteristics (establishment/worker/etc), aggregation level, timing, other info), but that is not really necessary. I think I was partly getting tied up in how to think of some of the other informational fields proposed for PSEO. We can probably close this, unless anyone else has strong feelings about it.

srt1 commented 5 years ago

Darn - simultaneous comments after 3 days. Well, let's keep it under consideration.

heathhayward commented 5 years ago

I was thinking that the identifiers can refer to things that are categorical variables and things that are not (i.e. year and quarter and agg_levels). Seems like we will need a space for the PSEO agg levels and I'm ok with it being either 6.2.2 or 6.14.2 (with j2j/qwi as 6.14.1).