US Wage Atlas: pay-band width and price-adjusted pay by occupation and metro area

Two derived datasets on US wages, built from federal sources and released under CC BY 4.0. Pay-band width (64,511 rows): the gap between the 25th and 75th percentile of pay inside a single occupation in a single metro area, expressed as a ratio, in dollars and as a share of the median, from the BLS Occupational Employment and Wage Statistics survey (May 2025). Price-adjusted pay (47,380 rows): the same wages divided by the BEA Regional Price Parity for that metro area, expressing pay in national-average dollars, with the housing parity kept as its own column rather than averaged into a single cost-of-living figure. Cells where any two adjacent published percentiles collapse onto the same value are estimation artefacts and are removed as the OEWS slice is built (3,840 of 67,953, 5.7%), so they appear in no row and the interval-collapse column is empty throughout. "All Other" residual SOC buckets are handled the other way round: they stay in the files, carry a flag, and are excluded only from headline figures. Both source reference periods ship as columns because they differ: wages are May 2025, price parities are the 2024 vintage. The band measures wage dispersion. It is not a measure of negotiating room: a wide gap mixes employer effects, seniority, industry mix and worker characteristics, which an aggregate survey cannot separate. Dataset home, column reference and downloads: https://voicedapp.co/datasets