Table 3.
The genetic modelling results showing the changes in model fit (Δχ2) and degrees of freedom (Δdf) when the specified parameters are dropped from the full model.
| Model | Δχ2 | Δdf | p-Value |
|---|---|---|---|
| Full bivariate moderator model | |||
| Drop moderator effect all paths | 45.13 | 6 | <0.001 |
| Drop moderator effect all crosspaths | 34.83 | 3 | <0.001 |
| Drop moderator effect all unique paths | 3.87 | 3 | 0.28 |
| Drop moderator on A (A1 and A2) | 1.73 | 2 | 0.42 |
| Drop moderator on C (C1 and C2) | 0.94 | 2 | 0.62 |
| Drop moderator on E (E1 and E2) | 7.54 | 2 | 0.02 |
| Drop moderator on crosspath A1 | 1.25 | 1 | 0.26 |
| Drop moderator on crosspath C1 | 0.19 | 1 | 0.66 |
| Drop moderator on crosspath E1 | 2.84 | 1 | 0.09 |
| Drop moderator on unique path A2 | 0.35 | 1 | 0.55 |
| Drop moderator on unique path C2 | 0.92 | 1 | 0.34 |
| Drop moderator on unique path E2 | 3.66 | 1 | 0.06 |
A1 and A2 refer to the first and second latent genetic components, C1 and C2 refer to the first and second latent shared environmental components, and E1 and E2 refer to the first and second latent residual components (see Figure 1).