Title: “Augmented CPS data on industry and occupation”

Abstract

The Current Population Survey (CPS) classifies the jobs of respondents into hundreds of detailed industry and occupation categories. The classification systems change periodically, creating breaks in time series. Standard concordances bridge the periods, but often leave empty cells or inaccurate sharp changes in time series. They also usually build in the assumption that categories from a certain period of time can be representative, on more aggregate levels, and of longer historical periods.

For each employed CPS respondent from before the year 2000 we impute post-2000 Census industry and occupation classifications and related variables. The imputations use micro data about each individual and training data sets that were classified by specialists into two industry and occupation category systems – that is, they are dual-coded.

We train a random forests classifier to handle the changes in classification between the 1990s and 2000s largely on the dual-coded data set and apply it to the full CPS and IPUMS-CPS to impute several variables, including industry and occupation. For changes in classification when an industry or occupation splits, we train the algorithms on the observations with the newly classified industry or occupation split to predict how the historical observations would have been classified. We generate an augmented CPS, with additional columns of standardized industry and occupation. This data set can serve research on many topics.

Keywords: CPS, prediction, imputation, occupation, industry, classification, employment