Skip to main content

Table 8 Robustness checks

From: Does the immediate reimbursement of medical insurance reduce the socioeconomic inequality in health among the floating population? Evidence from China

 

A: Parallel trend test

(N = 59,562)

B: Multi-period DID

(N = 122,061)

C: Generalized linear model

(N = 122,061)

D: PSM-DID with kernel matching

(N = 102,427)

E: PSM-DID with radius matching

(N = 102,963)

OR (95% CI)

P value

OR (95% CI)

P value

OR (95% CI)

P value

OR (95% CI)

P value

OR (95% CI)

P value

Policy change

0.263 (0.050–1.370)

0.113

1.1149 (1.013–1.227)

0.026

0.004 (0.001–0.007)

0.004

1.036 (1.010–1.062)

0.007

1.037 (1.010–1.063)

0.006

Income

1.120 (1.030–1.218)

0.008

1.395 (1.286–1.512)

0.000

0.033 (0.028–0.038)

0.000

1.256 (1.195–1.321)

0.000

1.257 (1.196–1.322)

0.000

Policy change × Income

1.060 (0.936–1.200)

0.360

0.829 (0.758–0.907)

0.000

-0.006 (-0.008–0.004)

0.000

0.959 (0.943–0.975)

0.000

0.959 (0.943–0.975)

0.000

Age

0.940 (0.937–0.942)

0.000

0.941 (0.939–0.943)

0.000

-0.010 (-0.010–0.010)

0.000

0.941 (0.939–0.943)

0.000

0.941 (0.939–0.943)

0.000

Gender (Male = 1)

1.208 (1.142–1.278)

0.000

1.220 (1.175–1.265)

0.000

0.027 (0.022–0.032)

0.000

1.197 (1.150–1.246)

0.000

1.198 (1.151–1.246)

0.000

Hukou type (Rural = 1)

1.048 (0.965–1.138)

0.268

1.069 (1.011–1.129)

0.019

0.008 (0.001–0.015)

0.029

1.066 (1.004–1.131)

0.036

1.064 (1.002–1.129)

0.042

Marriage (Unmarried = 1)

1.143 (0.974–1.342)

0.102

1.181 (1.062–1.314)

0.002

0.011 (-0.003–0.026)

0.131

1.152 (1.026–1.292)

0.016

1.147 (1.022–1.286)

0.020

Marriage (Married = 1)

1.210 (1.073–1.366)

0.002

1.170 (1.078–1.269)

0.000

0.052 (0.037–0.066)

0.000

1.144 (1.046–1.250)

0.003

1.141 (1.044–1.246)

0.004

Education (Primary school = 1)

1.343 (1.166–1.546)

0.000

1.300 (1.185–1.426)

0.000

0.105 (0.081–0.130)

0.000

1.280 (1.155–1.418)

0.000

1.281 (1.156–1.419)

0.000

Education (Junior high school = 1)

1.689 (1.470–1.940)

0.000

1.700 (1.552–1.861)

0.000

0.158 (0.135–0.182)

0.000

1.689 (1.527–1.869)

0.000

1.688 (1.526–1.868)

0.000

Education (High school = 1)

1.813 (1.560–2.108)

0.000

1.757 (1.591–1.939)

0.000

0.158 (0.135–0.182)

0.000

1.723 (1.544–1.922)

0.000

1.722 (1.543–1.921)

0.000

Education (College or above = 1)

1.842 (1.557–2.179)

0.000

1.769 (1.584–1.976)

0.000

0.159 (0.135–0.183)

0.000

1.743 (1.544–1.968)

0.000

1.729 (1.531–1.953)

0.000

Migrating reasons (Work = 1)

1.649 (1.449–1.877)

0.000

1.475 (1.349–1.613)

0.000

0.090 (0.075–0.106)

0.000

1.465 (1.331–1.613)

0.000

1.466 (1.331–1.613)

0.000

Migrating reasons (Business = 1)

1.981 (1.725–2.275)

0.000

1.797 (1.635–1.976)

0.000

0.118 (0.102–0.135)

0.000

1.790 (1.616–1.983)

0.000

1.793 (1.619–1.987)

0.000

Migrating reasons (Family = 1)

0.976 (0.848–1.123)

0.734

0.975 (0.885–1.074)

0.608

-0.009 (-0.027–0.009)

0.350

0.932 (0.840–1.034)

0.186

0.932 (0.840–1.034)

0.186

Willingness to settle (Yes = 1)

0.906 (0.873–0.940)

0.000

1.169 (1.118–1.221)

0.000

0.019 (0.013–0.026)

0.000

1.183 (1.128–1.242)

0.000

1.183 (1.128–1.242)

0.000

Year fixed effects

control

control

control

control

control

District fixed effects

control

control

control

control

control

  1. In Panel A, we only used the data in 2018. In Panel B, we measured the policy change as a binary variable taking a value of 1 if the city implemented the policy change, and a value of 0 otherwise. In Panel C, we employed the generalized linear model. In Panel D and Panel E, we conducted the PSM-DID method. In Panel C, we used a with replacement kernel matching at the bandwidth of 0.06, and the kernel type is epan kernel. After matching, we obtained 50106 individuals from 50 cities which implemented the policy change before May 2017 and 52321 individuals from 163 cities which implemented the policy change after May 2017. In Panel D, we used a with replacement radius matching at the caliper value of 0.1. After matching, we obtained 50642 individuals from 52 cities which implemented the policy change before May 2017 and 52321 individuals from 163 cities which implemented the policy change after May 2017. Except those, other specifications are the same as that in Table 2. The 95% CI are shown in brackets. *** p < .01. ** p < .05. * p < .10