Descriptives next steps

rebeccajohnson88 commented 3 years ago

Questions when reading in fuzzy matching outputs to construct outcomes variables:

[x] Correct to assume that jobs_group_id is the final identifier for a unique employer (and safe to drop jobs_group_id_part1 and jobs_group_id_part2 from the df if those were intermediate created for the cross-state stuff)?
[ ] For @eunice30718 when I read in the ACS merged file, it has the original variable names and not the ones renamed from the codebook. Is there an easy fix somewhere to get the renamed versions? If not we can maybe skip but the renamed ones are important/know that was a big focus of the earlier scripts. maybe just do the renaming in R by melting and reading in the codebook?

acs_pred = fread("intermediate/job_combined_acs_premerging.csv")

Next steps:

[ ] Ideally fix ACS var naming (can just do in R by reading in codebook if needed)
[ ] Line 144: work on deduplication of ACS-jobs merge
[ ] Write two outputs
[ ] Visualizations using ACS characteristics of three groups: (1) no investigation, (2) investigation but no violation, (3) investigation and violation (TRLA and WHD)
[ ] Visualizations of things like naics codes and attorney agent between the three groups

camguage commented 3 years ago

Correct!

rebeccajohnson88 commented 3 years ago

Some descriptive statistics before ACS is ready:

[ ] By year or by month-year, plotting the # of unique employers with jobs, # of unique employers with WHD investigations, # of unique employers with violations (using the overlap version of the outcome)
[ ] By year or by month-year, plotting the # of unique employers with jobs, # of unique employers with TRLA investigations
[ ] Something with overlap of those two
[ ] Overrepresentation of certain attorney agents in entities investigated or with violations
[ ] Overrepresentation of certain SOC codes
[ ] Overrepresentation of certain naics codes

Maybe shading post covid (post March 2020)

rebeccajohnson88 commented 3 years ago

[ ] Adding a new id column to the matched_data_WHD (eg the rownumber)
[ ] Subsetting the matched_data_WHD to: the new ID number and ACS columns (can probably identify those using starts_with or contains within dplyr select)
[ ] Melting to long format using case_number and employer_fulladdress as the id cols
[ ] Merging on the codebook using the variable name
[ ] Constructing the variable name as a combination of concept + name
[ ] Reshaping to wide format using the new id number as the id col
[ ] Merging back with the other employer attributes (eg case_number, soc codes, etc)

rebeccajohnson88 / qss20_s21_proj