# Clean Data Directory for Replication Archive

This directory contains only the data files that are actually used by the replication code.

## Directory Structure

### data_clean/in/
Input data files used by the analysis:

- **sipp_raw/**: Raw SIPP data files (12 zip files)
  - Contains Survey of Income and Program Participation data from 2018-2023 only
  - Files follow pattern: `pu{year}_dta.zip` and `rw{year}_dta.zip`

- **sipp_unzipped/**: Pre-unzipped SIPP data files (6 .dta files + 1 metadata file)
  - Contains Stata format files ready for processing (2018-2023 only)

- **SIPP_data_dictionaries/**: Documentation for SIPP data
  - PDF files with metadata and variable definitions for each year

- **eora_io_data_{year}/**: EORA Input-Output data by year (2010-2021)
  - Each year folder contains:
    - `Eora26_{year}_bp_T.txt`: Trade matrix
    - `Eora26_{year}_bp_Q.txt`: Satellite accounts
    - `Eora26_{year}_bp_FD.txt`: Final demand
    - `Eora26_{year}_bp_VA.txt`: Value added
    - `labels_*.txt`: Label files for the matrices

- **Eora26Structure.xlsx**: EORA data structure documentation

- **oecd_io_data/**: OECD Input-Output data
  - `.rdata` files containing pre-processed OECD I-O tables

- **getOECD_IO.R**: Script for fetching OECD data (included for reference)

### data_clean/out/
Output data files created by the analysis:

- **sipp_job_panel.csv**: Processed SIPP job panel data (created by processSIPP.R)

## Notes

1. The EORA data for years 2010-2021 is included as these years are potentially used in the analysis, even though the main configuration uses 2015-2017.

2. SIPP data has been limited to years 2018-2023 as these are the only years used in the paper's analysis.