Voting and children’s ages.

Note: This figure shows the coefficient estimates on having a child of each age based on estimates of Equation (1) as shown in Table 2, columns (2) and (4). All regressions include state-by-year fixed effects, controls for race (White, Black, Latino, Asian), educational attainment (high school graduate, some college, college graduate, post-college), whether the individual is a naturalized citizen, whether the individual is married, whether voting behavior is reported by self or proxy, duration at current residence (indicators for less than 1 year, 1–2 years, and 3–4 years), own age dummies, and controls for whether the individual is in the CPS sample the year following a given election and whether the individual is in the CPS for two consecutive years. The bands represent the 95% confidence intervals. Observations are weighted using the CPS voter supplement weights. Standard errors are clustered at the state level.