Table 1.
Type III sums of squares for rain-fed soybean yield
Model parameter | Degrees of freedom | Sums of squares | F value | P value† |
Intercept | 1 | 42,427,808 | 377.0 | 0.002 |
Year | 1 | 97,894,399 | 869.8 | 0.002 |
County | 253 | 51,990,756 | 1.8 | 0.002 |
Tmax | 1 | 6,534,829 | 58.1 | 0.002 |
P/PET | 1 | 160,418 | 1.4 | 0.226 |
Tmax2 | 1 | 6,465,343 | 57.4 | 0.002 |
P/PET2 | 1 | 1,086,772 | 9.7 | 0.01 |
W126 | 1 | 2,028,804 | 18.0 | 0.002 |
Year*county | 253 | 52,677,918 | 1.9 | 0.002 |
Tmax*P/PET | 1 | 3,579 | 0.0 | 0.832 |
Tmax*P/PET2 | 1 | 512,755 | 4.6 | 0.032 |
P/PET*Tmax2 | 1 | 20,533 | 0.2 | 0.652 |
W126*P/PET | 1 | 2,834,395 | 25.2 | 0.002 |
W126*Tmax | 1 | 2,534,975 | 22.5 | 0.002 |
W126*P/PET2 | 1 | 2,019,730 | 17.9 | 0.002 |
W126*Tmax2 | 1 | 2,688,000 | 23.9 | 0.002 |
Residuals | 6,210 | 698,922,185 |
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.