Improving Corruption Risk Management in the Slovak Republic

Annex G. Test for heteroskedasticity

Figure A ‎G.1. Plotted residuals versus fitted (predicted) values. Dependent variable: likelihood of communicating a risk

Note: OLS, dependent variable: likelihood of communicating a risk.

Figure A ‎G.2. White's test to test for the heteroskedasticity of the error terms. Dependent variable: likelihood of communicating a risk

Note: OLS, dependent variable: likelihood of communicating a risk.

Figure A ‎G.3. Breusch-Pagan test for heteroskedasticity of the error terms. Dependent variable: likelihood of communicating a risk

Note: OLS, dependent variable: likelihood of communicating a risk.

Figure A ‎G.4. Plotted residuals versus fitted (predicted) values. Dependent variable: general feeling of safety

Note: OLS with robust standard errors, dependent variable: General feeling of safety.

Figure A ‎G.5. White's test and Breusch-Pagan test for heteroskedasticity of the error terms. Dependent variable: general feeling of safety

Note: OLS with robust standard errors, dependent variable: General feeling of safety.

Figure A ‎G.6. Plotted residuals versus fitted (predicted) values. Dependent variable: Feelings of safety when reporting to preferred stakeholder

Note: OLS with robust standard errors, dependent variable: Feelings of safety when reporting to preferred stakeholder.

Figure A ‎G.7. White's test and Breusch-Pagan test for heteroskedasticity of the error terms. Dependent variable: Feelings of safety when reporting to preferred stakeholder

Note: OLS with robust standard errors, dependent variable: Feelings of safety when reporting to preferred stakeholder.