Extended Data Table 3 Associations between Race-Specific Upward Income Mobility and Economic Connectedness in Racially Homogeneous Areas
From: Social capital I: measurement and associations with economic mobility
- This table presents correlations between race-specific measures of upward income mobility and economic connectedness across racially homogeneous counties and ZIP codes. Data on upward income mobility by race are obtained from Chetty et al.73. Upward mobility is measured as the predicted household income rank in adulthood for children in the 1978-83 birth cohorts with parents at the 25th percentile of the national income distribution. In columns 1-4, we correlate upward mobility for white individuals with economic connectedness in counties and ZIP codes where over 80% or over 90% of the population is white (based on data from the 2000 Census). In columns 5-6, we correlate upward mobility for Black individuals with economic connectedness in ZIP codes where over 80% or over 90% of the population is Black. In columns 7-8, we correlate upward mobility for Hispanic individuals with economic connectedness in ZIP codes where over 80% or over 90% of the population is Hispanic. Because the statistics on upward mobility are constructed using individuals who grew up in the US. and a large share of Hispanic individuals are immigrants, in Columns 7-8 we measure economic connectedness using only individuals who have a US hometown in the Facebook data. The bottom row of the table shows the percentage of individuals in the estimation sample of the focal racial group (e.g., percentage white in columns 1-4). For white and Black individuals, the correlations are weighted by number of children with below-median parental income as calculated in the Opportunity Atlas72 using Census data. In columns 7-8, the weights are the number of children who have a US hometown and have below-median parental SES in the Facebook data. Standard errors (reported in parentheses) below each correlation are clustered at the commuting zone level for county-level correlations and at the county level for ZIP code-level correlations. Asterisks indicate the level of significance: *5%, **1%, ***0.1%.
