Notes: The ratio column in panel A measures non-White women's overrepresentation, comparing the volume of care done by non-White women versus White women. For instance, the 1.42 rate in 1965 indicates that non-White women participated in 1.42 hours of care work for every 1 hour of care work White women provided (or that non-White women provided 42% more care work time than White women). The composition column in panel A indicates the distribution of total care work time across domains (paid care, unpaid care, childcare, and adult care). For instance, the .65 composition value for unpaid childcare in 1965 indicates that 65% of total care work time was unpaid childcare time. The composition quantities are slightly different from Table 3 because this analysis includes only women. For an analysis of the overrepresentation of non-White people, including men, see Table S10. The change column calculates the difference in ratios and composition between 1965 and 2018. For instance, the −0.54 results from subtracting 0.87 from 1.42, and it indicates that the overrepresentation of non-White women in overall care work declined by −0.54 units or hours. In panel B, the 1965 and 2018 columns display simulated overrepresentation ratios for the two years, and the change column estimates the change in the overall overrepresentation ratio corresponding to each of these simulations. For instance, the −0.46 value indicates that if non-White women's representation in unpaid childcare had not changed since 1965, the decline in the overall representation of non-White women in total care work time between 1965 and 2018 would have been −0.46 instead of −0.54. The explained change and percentage contribution columns calculate how much of the change in the overrepresentation of non-White women in total care work time can be attributed to changes in the overrepresentation ratio within each care domain and to changes in the overall composition of total care work time.

Sources: 1965–2018 AHTUS and 1965–2018 CPS-ASEC.