Skip to main content

Table 7 Estimated coefficients of the spatial SEM, SAR, SDM models

From: Spatiotemporal evolution and influencing factors of the allocation of social care resources for the older adults in China

Variables

OLS

SAR

SEM

SDM1

SDM2

Rho

 

-0.237***

 

-0.402***

-1.487***

 

(0.091)

 

(0.074)

(0.252)

Lambda

  

-0.372***

  
  

(0.099)

  

Ln (the per capita GDP)

0.610***

0.387**

0.369**

0.516***

0.408**

(0.069)

(0.171)

(0.159)

(0.177)

(0.160)

Ln (the proportion of social welfare expenditure in GDP)

0.225***

0.103

0.138

0.155**

0.177**

(0.058)

(0.086)

(0.091)

(0.079)

(0.088)

Ln (the proportion of the tertiary industry in GDP)

0.642***

0.802***

0.720***

0.797***

0.554**

(0.163)

(0.265)

(0.253)

(0.235)

(0.231)

Ln (the proportion of the older adults aged 65 +)

0.0309

-0.227

-0.300*

-0.287*

-0.429***

(0.076)

(0.166)

(0.172)

(0.149)

(0.160)

Ln (the old age dependency ratio)

-0.415***

-0.635***

-0.723***

-0.616***

-0.663***

(0.108)

(0.225)

(0.202)

(0.170)

(0.156)

Ln (the park green space area per capita)

-0.030

0.152

0.185

0.170

0.354***

(0.099)

(0.106)

(0.114)

(0.122)

(0.135)

R-square

0.585

0.513

0.380

0.348

0.314

AIC

99.340

-248.387

-253.223

-256.775

-265.920

BIC

123.000

-221.348

-226.184

-209.457

-218.601

Log-likelihood

-42.670

132.194

134.612

142.388

146.960

  1. *, ** and *** indicate significance at the level of 0.1, 0.05, 0.01, respectively. Rho, Spatial Autoregressive Parameters; Lambda, Residual lag parameter; Ln, natural logarithm; the SDM1 model is based on binary adjacency matrix, the SDM2 model is based on the geographical distance matrix. The value in parentheses represents the standard deviation