CSI’s Inequality Discussion Groups bring together Cornell faculty and graduate students from around campus to discuss and improve their in-progress research.

Title: Social Rigidity over the Life Course: Using Machine Learning to Predict Ultimate Socioeconomic Status

Abstract: How well can individuals' background and previous life experiences predict their ultimate socioeconomic status attainment? This question is important for stratification research, as a strong power of earlier experiences in predicting later-life outcomes signals substantial intra- or intergenerational status persistence, or put simply, social rigidity. Using the National Longitudinal Survey of Youth 1979 data and running a set of cutting-edge predictive machine learning models, we find that a large number (~4,000) of predictors commonly used in the stratification literature can improve the prediction of one's life chances in middle to late adulthood by about 30 percent, compared with a null model that uses a simple mean of the outcome variable. Grouping this comprehensive list of predictors into three unique sets that cover family background, childhood and adolescence experiences, and early labor market experiences, we find that income, employment status, and occupational characteristics during an individual's early labor market period are the strongest predictors for one's ultimate socioeconomic status attainment. Applying the same exercise across different social groups, we find higher social rigidity for Blacks than Whites and for men than women when predicting total personal income. The gender gap decreases when we shift to predict log total family income, suggesting that family income may be a better representation of women's ultimate life chances.