Table 2.
Coefficients of the regression model for finish time
| Effect | Coefficient | Standard error | Coef/SE | p |
|---|---|---|---|---|
| First time since 1985 (constant) | 24.060 | 0.545 | 44.131 | <0.001 |
| Previous drop (constant) | 18.490 | 1.399 | 13.216 | <0.001 |
| Finished previous start (constant) | 24.160 | 0.561 | 43.030 | <0.001 |
| Relative performance at previous start (for those who finished the previous start) | −1.725 | 0.310 | −5.566 | <0.001 |
| Female | 1.347 | 0.123 | 10.968 | <0.001 |
| One year greater age up to 38 years | 0.043 | 0.015 | 2.840 | 0.005 |
| One year greater age within 38–50-year interval | 0.167 | 0.010 | 16.375 | <0.001 |
| One year greater age above 50 years | 0.122 | 0.018 | 6.857 | <0.001 |
| Minimum ambient temperature (°C) | 0.081 | 0.039 | 2.089 | 0.037 |
| Finished WSER before 1986 | −4.302 | 1.509 | −2.851 | 0.004 |
| Proportion of previous starts finished since 1986 | −1.821 | 0.317 | −5.750 | <0.001 |
| Finished WSER before 1986 × first time since 1985 | −0.533 | 0.219 | −2.436 | 0.0148 |
| One year greater age up to 38 years × previous drop | 0.139 | 0.036 | 3.836 | <0.001 |
| Relative performance at previous start × minimum ambient temperature (°C) | 0.070 | 0.031 | 2.263 | 0.0236 |
| Relative performance at previous start × proportion of previous starts finished since 1986 | −2.198 | 0.894 | −2.459 | 0.0139 |
| Female × minimum ambient temperature (°C) | −0.047 | 0.023 | −2.047 | 0.0407 |
| One year greater age up to 38 years × finished WSER before 1986 | 0.083 | 0.040 | 2.088 | 0.0368 |
| One year greater age within 38–50-year interval × proportion of previous starts finished since 1986 | 0.099 | 0.028 | 3.522 | <0.001 |
| One year greater age above 50 years × finished WSER before 1986 | 0.080 | 0.033 | 2.408 | 0.016 |
The constants apply to the three categories into which all finishes are sorted each year. Subsequent coefficients apply equally to all finishes except where noted explicitly. Positive coefficients increase and negative coefficients decrease the average finish time. Specifically, the coefficient for a dichotomous variable (female or finished WSER before 1986) is added directly to the average finish time. On the other hand, the coefficient for a quantitative variable is multiplied by a difference in the variable’s value and this product is added to the average finish time