search

GSS-Cogs / family-covid-19

0 stars 0 forks source link

NRS-Deaths-involving-coronavirus-COVID-19-in-Scotland #8

Open ajtucker opened 4 years ago

LPerryman commented 4 years ago

Stage 2 of Spec completed. Did not change NHS board and Council Area to Geography codes but this is still an option if needed

https://github.com/GSS-Cogs/family-covid-19/blob/master/datasets/NRS-Deaths-involving-coronavirus-COVID-19-in-Scotland/spec.md

Tracey-B commented 4 years ago

BA Tech Spec review 09/07/2020 should covid-19 deaths include sex as this is identified seperately Table 4 not included in the Spec, however thi may have been added to the dataset after the spec was created consider adding fig 1, 2 & 8 as these include daily counts

LPerryman commented 4 years ago

Removed 'NRS' from main title

JasonHowell commented 4 years ago

BA comments:

On CMD and Open Data Format on landing page (National Records Scotland) Nothing critical but some general observations: Unit of Measure still uri. Airtable title doesn't include NRS (just “Deaths involving COVID-19 in Scotland”) Metadata page has multiple links which seem to link to reference data on PMD. No doubt good info for technical people but might be confusing for general User, may be worth getting User Research views. Landing page link returns an error. Transform Spec - quite complex, can’t fully follow due to knowledge gap but assume ok. Don’t really know why spec required as already in open data format, obviously my knowledge gap again.

RedWalters commented 4 years ago

Script added to pull newest release of dataset 'NRS All Deaths' updated to work 'NRS All Deaths by location' in progress 'NRS COVID Deaths' updated to work

rossbowen commented 3 years ago

Commentary from Bill from Swirrl:

Deaths involving COVID-19 in Scotland.

1) it looks like S04000001 has been chosen as the identifier for Scotland - that isn't wrong but it's defined as a 'Regeneration Outcome Agreement Area'. A more normal identifier for all-Scotland would be S92000003.

2) observations seem to be defined with both a health board dimension and a council area dimension. But each observation is actually about either Scotland, a health board area (S08) or a council area (S12). The observation is never about both a health board area and a council area. Perhaps a more general 'refArea' dimension would be best if the data is to be combined into a single dataset?

This is how the Scottish gov represent it as linked data: http://statistics.gov.scot/data/deaths-involving-coronavirus-covid-19