Finally, row 3 of Table 8 estimates the regression model using only the information
on whether a person is native-born, refugee, or non-refugee to define the variables in the
vector I. Although the coefficients now have larger standard errors, the point estimates are
consistent with the overall story. The impact of the probability of receiving Medicaid on the
probability of being covered by employer-sponsored insurance is −1.35 (1.33), while the
impact of Medicaid on the probability of being covered by some type of health insurance
coverage is 0.11 (0.87).
The bottom panel of the table reports the IV coefficients from the regressions estimated in
the sample of children.25 These coefficients tend to be quite similar to those reported in the
sample of all persons. For example, the two coefficients in the specification presented in row
1, where the vector I uses information on the citizenship status of the head of the household
to classify the children in the household, are−0.98 (0.98) in the ESI coverage regression and
0.12 (0.73) in the health insurance coverage regression. The evidence, therefore, suggests
Finally, row 3 of Table 8 estimates the regression model using only the informationon whether a person is native-born, refugee, or non-refugee to define the variables in thevector I. Although the coefficients now have larger standard errors, the point estimates areconsistent with the overall story. The impact of the probability of receiving Medicaid on theprobability of being covered by employer-sponsored insurance is −1.35 (1.33), while theimpact of Medicaid on the probability of being covered by some type of health insurancecoverage is 0.11 (0.87).The bottom panel of the table reports the IV coefficients from the regressions estimated inthe sample of children.25 These coefficients tend to be quite similar to those reported in thesample of all persons. For example, the two coefficients in the specification presented in row1, where the vector I uses information on the citizenship status of the head of the householdto classify the children in the household, are−0.98 (0.98) in the ESI coverage regression and0.12 (0.73) in the health insurance coverage regression. The evidence, therefore, suggests
การแปล กรุณารอสักครู่..
