Data Library on Occupational Characteristics

          This page provides access to and explains the construction of the datasets used in "The Evolving
          U.S. Occupational Structure." These datasets were constructed from text originally published
          in the Boston Globe, New York Times, and Wall Street Journal. The datasets contain information
          on occupations' skill requirements, technology usage, work activities, work styles, and other
          job characteristics, between 1940 and 2000.

Downloadable Data

          A) By job title and year (318MB)
          B) By job title (253MB)
          C) By SOC code and year (11MB)
          D) By SOC code (623KB)


          1) Codebook/Overview of the dataset
          2) Description of the initial text cleaning
              Python Notebook related to this step.
          3) Details of the LDA procedure to classify ads as jobs ads vs. other types of ads
              LDA results
              Python Notebook related to this step.
          4) Details of the procedure to discern the job title, boundaries between ads
              A list of job title words.
              Python Notebook related to this step.           5) Mapping between occupational characteristics and words/phrases
              Mapping between job titles and SOC codes
              An explanation for these mappings
              Python Notebook: estimating the CBOW model
              Python Notebook: retrieving SOC codes from the CBOW model.

The downloadable data were last updated November 2, 2017.