Table 3.

OLS analysis and quantile regression estimation for model 2 (Internet usage)

Variables	Dependent variable: depression (sample n = 8853)
	OLS	Quantile regression
		0.5	0.6	0.7	0.8	0.9
	(1)	(2)	(3)	(4)	(5)	(6)
Individual socioeconomic status
Father’s education	−0.168	−0.163	−0.264	−0.138	−0.118	−0.143
	(0.132)	(0.149)	(0.163)	(0.188)	(0.238)	(0.310)
SRH-16	−0.437***	−0.443***	− 0.451***	−0.526***	− 0.640***	−0.666***
	(0.055)	(0.062)	(0.070)	(0.083)	(0.099)	(0.128)
Education	−0.734***	−0.722***	− 0.874***	−1.120***	−1.660***	−0.803*
	(0.175)	(0.238)	(0.255)	(0.376)	(0.311)	(0.421)
Income	−0.685***	−0.642***	− 0.752***	−0.894***	−1.130***	−1.300***
	(0.048)	(0.057)	(0.061)	(0.077)	(0.090)	(0.114)
Hukou	−0.137	−0.154	−0.238	− 0.266	−0.082	0.256
	(0.181)	(0.201)	(0.215)	(0.242)	(0.329)	(0.407)
Digital technology usage
Internet usage	−5.800**	−3.500*	−4.690***	−9.620***	−9.370**	−12.40***
	(2.890)	(2.070)	(1.680)	(2.910)	(4.140)	(3.890)
Other
Age	0.014*	0.008	0.011	0.009	0.013	0.033*
	(0.007)	(0.008)	(0.009)	(0.011)	(0.013)	(0.017)
Gender	−1.330***	−1.290***	−1.360***	−1.970***	−2.190***	−2.790***
	(0.133)	(0.158)	(0.175)	(0.213)	(0.250)	(0.315)
Marital	−1.410***	−1.400***	− 1.610***	−1.970***	−2.050***	− 2.520***
	(0.161)	(0.212)	(0.235)	(0.307)	(0.290)	(0.414)
Interaction effect
Internet usage * education	0.556	−1.460***	−0.142	0.765	4.570***	4.640*
	(2.400)	(0.546)	(0.719)	(1.230)	(1.450)	(2.620)
Internet usage * income	0.502**	0.472**	0.431**	0.816***	0.473	0.824*
	(0.206)	(0.211)	(0.179)	(0.271)	(0.439)	(0.449)
Internet usage * hukou	−0.760*	−0.550	−0.440	−0.369	−1.020	−2.680**
	(0.454)	(0.449)	(0.375)	(0.625)	(0.798)	(1.100)
Constant	16.500***	15.200***	18.000***	22.100***	27.200***	31.200***
	(0.753)	(0.900)	(0.951)	(1.210)	(1.400)	(1.760)
Observations	8853	8853	8853	8853	8853	8853
R2	0.110
Pseudo R2		0.595	0.596	0.595	0.607	0.601

^a standardize coefficients are reported; standard errors in parentheses

^b ***p < 0.01, **p < 0.05, *p < 0.1