Table 2.
Type III sums of squares for rain-fed maize yield
Model parameter | Degrees of freedom | Sums of squares | F value | P value† |
Intercept | 1 | 567,895,174 | 600.4 | 0.002 |
Year | 1 | 1,720,503,651 | 1,818.9 | 0.002 |
County | 283 | 888,816,636 | 3.3 | 0.002 |
Tmax | 1 | 54,345,824 | 57.5 | 0.002 |
P/PET | 1 | 380,084 | 0.4 | 0.546 |
Tmax2 | 1 | 63,688,564 | 67.3 | 0.002 |
P/PET2 | 1 | 5,869,113 | 6.2 | 0.016 |
W126 | 1 | 44,226,186 | 46.8 | 0.002 |
Year*county | 283 | 900,103,436 | 3.4 | 0.002 |
Tmax*P/PET | 1 | 1,638,230 | 1.7 | 0.21 |
Tmax*P/PET2 | 1 | 1,192,597 | 1.3 | 0.258 |
P/PET*Tmax2 | 1 | 2,259,738 | 2.4 | 0.12 |
W126*P/PET | 1 | 1,943,883 | 2.1 | 0.134 |
W126*Tmax | 1 | 51,755,969 | 54.7 | 0.002 |
W126*P/PET2 | 1 | 1,068,479 | 1.1 | 0.234 |
W126*Tmax2 | 1 | 54,124,209 | 57.2 | 0.002 |
Residuals | 6,932 | 6,556,987,586 |
P values were determined by comparing the F value in the table to a null F distribution that was generated by a double bootstrap procedure that accounted for spatial correlation. Thus, the P value in the table is different from that of a standard F test.