Table 3.
Diagnostics of the modeling for men’s and women’s absence rate from work due to own illness or disability.
| Diagnostics | Men | Women | ||
|---|---|---|---|---|
| OLS | GWR | OLS | GWR | |
| R-Squared | 0.64 | 0.71 | 0.56 | 0.77 |
| Adjusted R-Squared | 0.58 | 0.87 | 0.48 | 0.75 |
| AICc | 141.4 | 103.5 | 107.1 | 58.7 |
| Moran’s I | 0.06 ** | 0.008 | 0.12 ** | 0.005 |
| Joint Wald Statistics | 113.4 *** | - | 59.6 *** | - |
| Koenker (BP) | 21.80 * | - | 35.62 ** | - |
| Jarque–Bera | 5.6 * | - | 1.1 | - |
Note: the Moran’s I statistic showed spatial autocorrelation in the OLS residuals (initially, it was confirmed by the ESDA results, Table A1 and Figure A1). Moreover, the Koenker (BP) statistic indicated that the OLS modeled relationships were not consistent either, due to non-stationarity or heteroscedasticity. Significance levels: *** p ≤ 0.01, ** p ≤ 0.05, * p ≤ 0.1.