joseph-data/daioe_dataset: AI-Econ Lab, Örebro University

DAIOE datasets This repository hosts the DAIOE (Direct AI Occupational Exposure) index in multiple occupational classifications. Each dataset folder includes three file formats for convenience: .csv, .dta (Stata), and .xlsx (Excel). Flavours (occupational classifications) daioe_isco08: ISCO-08 (international) daioe_onetsoc2010: O*NET-SOC 2010 (US) daioe_soc2010: SOC 2010 (US) daioe_ssyk96: SSYK 96 (Sweden) daioe_ssyk2012: SSYK 2012 (Sweden) File formats per folder daioe_<flavour>.csv daioe_<flavour>.dta daioe_<flavour>.xlsx Column layout (all files) Col A: occupational code Col B: occupational title Col C: year Cols D–N: DAIOE index scores (overall + nine sub-domains + generative AI) Cols O–Y: DAIOE percentile rankings (overall + nine sub-domains + generative AI) Index structure Col D (O): overall DAIOE index (ranking) Cols E–M (P–X): nine sub-domain DAIOE indices (rankings) Col N (Y): generative AI DAIOE index (ranking) Notes Early rows may be empty for armed forces occupations or legislators where DAIOE cannot be computed due to missing O*NET observations. Scores allow comparison of AI exposure across occupations and over time; a one-unit change has no standalone interpretation.