Table 2.

Maximum likelihood estimates of extended sample selection model

Variables	(i) Unconstrained				(ii) Constrained
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	HIi	HIni	PV for HIi	PV for HIni	HIi	HIni	PV for HIi	PV for HIni
HIi*			− 0.034				0.042
			(0.045)				(0.060)
HIni*				0.187**				0.247***
				(0.084)				(0.040)
INTERNET	0.006***				0.006***
	(0.002)				(0.002)
DENSITY			0.312**	0.320**			0.309***	0.309***
			(0.126)	(0.163)			(0.103)	(0.103)
AGE	− 0.023***	− 0.010***	− 0.003***	0.002	− 0.023***	− 0.010***	0.000	0.000
	(0.001)	(0.001)	(0.001)	(0.002)	(0.001)	(0.001)	(0.001)	(0.001)
FEMALE	0.220***	0.345***	0.222***	0.112***	0.222***	0.345***	0.156***	0.156***
	(0.021)	(0.027)	(0.022)	(0.030)	(0.021)	(0.025)	(0.019)	(0.019)
ONEADULT	− 0.190***	− 0.140***	0.010	− 0.056**	− 0.188***	− 0.142***	− 0.009	− 0.009
	(0.021)	(0.021)	(0.026)	(0.023)	(0.021)	(0.021)	(0.018)	(0.018)
Employment (reference is "inactive")
EMPLOYED	0.227***	0.095***	− 0.203***	− 0.187***	0.228***	0.101***	− 0.225***	− 0.225***
	(0.020)	(0.027)	(0.027)	(0.030)	(0.020)	(0.026)	(0.023)	(0.023)
UNEMPLOY	0.105**	0.090**	− 0.131**	− 0.102*	0.107**	0.097**	− 0.139***	− 0.139***
	(0.044)	(0.042)	(0.052)	(0.061)	(0.045)	(0.042)	(0.043)	(0.043)
Area (reference is "rural/village")
SMALL	0.101***	0.074***	0.019	0.017	0.102***	0.075***	0.011	0.011
	(0.019)	(0.023)	(0.024)	(0.024)	(0.019)	(0.023)	(0.018)	(0.018)
LARGE	0.152***	0.092***	0.052*	0.098***	0.151***	0.093***	0.065***	0.065***
	(0.021)	(0.026)	(0.028)	(0.031)	(0.021)	(0.026)	(0.022)	(0.022)
Health (reference is "very bad")
FBAD	0.203***	0.177***	− 0.363***	− 0.212***	0.206***	0.188***	− 0.256***	− 0.256***
	(0.063)	(0.058)	(0.123)	(0.058)	(0.063)	(0.057)	(0.056)	(0.056)
FGOOD	0.329***	0.208***	− 0.986***	− 0.736***	0.331***	0.219***	− 0.829***	− 0.829***
	(0.060)	(0.058)	(0.116)	(0.061)	(0.060)	(0.057)	(0.060)	(0.060)
VGOOD	0.230***	0.068	− 1.327***	− 1.041***	0.232***	0.081	− 1.146***	− 1.146***
	(0.059)	(0.061)	(0.119)	(0.064)	(0.059)	(0.060)	(0.060)	(0.060)
LONGILL	0.235***	0.378***	0.800***	0.803***	0.234***	0.381***	0.782***	0.782***
	(0.016)	(0.023)	(0.023)	(0.039)	(0.017)	(0.023)	(0.021)	(0.021)
Sport (reference is "none")
MONTH	0.247***	0.230***	0.056	0.094*	0.247***	0.232***	0.058*	0.058*
	(0.030)	(0.037)	(0.044)	(0.049)	(0.030)	(0.037)	(0.032)	(0.032)
1 − 3MONTH	0.247***	0.285***	− 0.007	0.041	0.247***	0.288***	− 0.001	− 0.001
	(0.034)	(0.042)	(0.054)	(0.053)	(0.034)	(0.042)	(0.036)	(0.036)
1 − 2WEEK	0.322***	0.304***	0.013	0.082*	0.321***	0.307***	0.018	0.018
	(0.025)	(0.039)	(0.035)	(0.046)	(0.024)	(0.039)	(0.029)	(0.029)
3 − 4WEEK	0.371***	0.396***	0.037	0.099**	0.369***	0.399***	0.035	0.035
	(0.027)	(0.037)	(0.040)	(0.043)	(0.027)	(0.037)	(0.029)	(0.029)
5 WEEK	0.259***	0.316***	0.004	0.031	0.259***	0.317***	− 0.001	− 0.001
	(0.029)	(0.030)	(0.041)	(0.045)	(0.028)	(0.030)	(0.032)	(0.032)
Education (reference is "no full-time education")
EDU15	0.209*	0.173*			0.222*	0.215**
	(0.126)	(0.099)			(0.128)	(0.092)
EDU19	0.709***	0.518***			0.724***	0.561***
	(0.125)	(0.104)			(0.126)	(0.097)
EDU20	1.050***	0.800***			1.066***	0.841***
	(0.125)	(0.107)			(0.126)	(0.098)
STUDY	0.769***	0.710***			0.789***	0.765***
	(0.135)	(0.119)			(0.135)	(0.110)
Thresholds
Ck,1	0.788***	0.433***	− 1.939***	− 1.299***	0.802***	0.527***	− 1.546***	− 1.546***
	(0.184)	(0.155)	(0.138)	(0.147)	(0.188)	(0.118)	(0.112)	(0.112)
Ck,2	1.344***	0.705***	− 0.552***	− 0.056	1.358***	0.799***	− 0.223**	− 0.223**
	(0.183)	(0.161)	(0.140)	(0.149)	(0.188)	(0.122)	(0.114)	(0.114)
Ck,3	1.491***	0.775***	0.338**	0.735***	1.505***	0.869***	0.621***	0.621***
	(0.183)	(0.164)	(0.142)	(0.155)	(0.188)	(0.124)	(0.117)	(0.117)
Ck,4	1.944***	0.975***			1.958***	1.069***
	(0.183)	(0.173)			(0.188)	(0.129)
Ck,5	2.369***	1.147***			2.383***	1.242***
	(0.184)	(0.185)			(0.189)	(0.140)
Correlations
		ρ12	ρ13	ρ14		ρ12	ρ13	ρ14
		0.885***	0.095**	0.222***		0.883***	0.009	0.184***
		(0.035)	(0.042)	(0.061)		(0.029)	(0.063)	(0.045)
				ρ24				ρ24
				0.134**				0.090***
				(0.052)				(0.031)
Observations	24,971				24,971

Note: The table reports the results from estimating the extended sample selection model defined in Eqs. (1)-(4) by the method of maximum likelihood. Columns (1)–(4) show the estimates of health information and physician visitation equations for seekers and non-seekers of online health information. HIi (HIni) refers to the equation for online (offline) health information seeking, and PV for HIi (PV for HIni) refers to the equation for physician visitation for online health information seekers (non-seekers). Columns (5)–(8) report the estimates of these equations, constraining the coefficients to be identical for the common variables between the two physician visitation equations. Robust standard errors are in parentheses. Country dummies are included in the estimations and their coefficient estimates are reported in Table 2A in the Appendix

^***p < 0.01, **p < 0.05, *p < 0.1