Skip to main content
. 2018 Jul 16;9:775. doi: 10.3389/fphar.2018.00775

Table 3.

Relationship between medical staffing and methicillin resistance of S. aureus.

Variable Static Dynamic
Pooled OLS (1) FE(2) GMM-SYS (3) GMM-SYS (4)
LAGS OF DEPENDENT VARIABLE
lnSTW(t−1) 0.307 0.287
(0.254) (0.270)
lnSTW(t−2) 0.140
(0.133)
LABOR INPUT
lnINF −0.036 −0.165 −0.184 −0.405
(0.339) (0.275) (0.292) (0.318)
lnPHR −0.147 −0.602 −3.5E−4 0.106
(0.398) (0.630) (0.510) (0.465)
lnMCB −0.282 −0.422 −0.463* −0.561
(0.367) (0.318) (0.259) (0.350)
lnNRS 0.512 −0.350 0.394 0.648**
(0.468) (0.571) (0.385) (0.322)
CAPITAL INPUT
lnBED −0.371 1.256** 0.272 0.337
(0.255) (0.598) (0.399) (0.346)
TECHNOLOGY
T −0.084* −0.074* −0.088 −0.105*
(0.046) (0.041) (0.054) (0.059)
Constant −0.677 −4.699 −4.158 −6.096**
(3.47) (3.084) (3.237) (2.947)
No. obs. 100 100 78 64
F 14.74*** 5.70***
Wald χ2 46.49*** 104.86***
Arellano-Bond test
 AR(1) (p-Value) 0.039** 0.009***
 AR(2) (p-Value) 0.029** 0.360
Sargan test (p-Value) 0.436 0.168

Significance:

***

p < 0.01,

**

p < 0.05,

*

p < 0.10.

Robust standard errors in parentheses. Column (1) employs pooled ordinary least squares (OLS) model, while column (2) employs fixed–effect (FE) model. Column (3) and (4) employ dynamic models: Column (3) enrolls one lag of the dependent variable, while column (4) enrolls two lags. AR (1) and AR (2) examine the serial correlations of residuals in first and second differences, respectively.