Table 7. Linear and nonlinear...

Table 7.

Linear and nonlinear estimations of the corruption model with individual attributes of the employees (with untransformed independent variables)

Dependent variable: Transparency (TRANS)
Independent variables	Linear probability model (OLS)^a	Probit model (ML)	Logit model (ML)
PREV	0.0447(0.422)	0.1837(0.3721)	0.269(0.4351)
AACOR	0.0672*(0.0742)	0.2433*(0.0798)	0.4223*(0.0789)
TE	0.0629 (0.217)	0.2059(0.2897)	0.3795(0.2619)
GEND	0.0617(0.226)	0.2336(0.2176)	0.4016(0.2247)
AGE	0.0315(0.484)	0.1458(0.427)	0.2107(0.4981)
EXP	−0.1089***(0.0057)	−0.3897***(0.0098)	−0.6380**(0.0118)
Intercept	0.3158(0.312)	−0.9169(0.3971)	−1.5639(0.3987)
R²/pseudo R	0.0851	0.0831	0.0811

Dependent variable: Transparency (TRANS)
Independent variables	Linear probability model (OLS)^a	Probit model (ML)	Logit model (ML)
PREV	0.0447(0.422)	0.1837(0.3721)	0.269(0.4351)
AACOR	0.0672*(0.0742)	0.2433*(0.0798)	0.4223*(0.0789)
TE	0.0629 (0.217)	0.2059(0.2897)	0.3795(0.2619)
GEND	0.0617(0.226)	0.2336(0.2176)	0.4016(0.2247)
AGE	0.0315(0.484)	0.1458(0.427)	0.2107(0.4981)
EXP	−0.1089***(0.0057)	−0.3897***(0.0098)	−0.6380**(0.0118)
Intercept	0.3158(0.312)	−0.9169(0.3971)	−1.5639(0.3987)
R²/pseudo R	0.0851	0.0831	0.0811

Note(s):

***represents significance at the 1% level; **represents significance at 5% level; *represents significance at 10% level; ^aEstimated using heteroscedasticity-robust standard errors and p-values are given in parentheses. #These variables are used in their original discrete values from 1 to 5. TRANS is the presence of transparency, PREV is the preventive measures, AACOR is the actions against corruption and TE is the training and ethics, GEND is gender, AGE is age group and EXP is experience

Source(s): Authors’ own work

[ViewLarge]

Sharing Unavailable