Open stfran22 opened 3 months ago
@sjmelly here is a template for taking a BEDDN database multiyear tract-level query result (including Tract10, Year, and any number of category columns) and joining the data to 2010 tract spatial boundaries with the year appended to the end of the column (wide). I created this template to create Moran's I z-scores for Stephen Dickinson's cannabis project.
@stfran22 This looks helpful. I've been struggling with when we want to use wide datasets vs long datasets. A related issue is when do we want to do QC checks on data - Shufang has been doing this with data linked to MESA participants in a wide format, but we've been working on putting all the Food & Communities data into long format. Also I guess we are using tract10 as the tract ID in NETS instead of GEOID10. Did this come from Walls & Associates or the RECVD practices? Usually when the census refers to tract ids they mean just the tract portion of the tracts FIPS without the state and county part.
Work on a draft of guidelines for extracting and working with BEDDN data. Include relevant code snippets and templates.