| Model 1 | Model 2 | Model 3a
|
---|
Phenotypic model | Quasi-causal model | Quasi-causal model |
---|
Male | Female | Male | Female | Male | Female |
---|
b
A
| | |
0.80 (0.31)
|
2.18 (0.27)
|
0.82 (0.24)
|
2.13 (0.23)
|
b
P
|
0.70 (0.10)
|
1.53 (0.10)
| 0.22 (0.14) | 0.16 (0.13) | 0.20 (0.10) | 0.20 (0.10) |
Goodness of fit |
RMSEA [90% CI] | 0.03 [0.02, 0.04] | 0.02 [0.01, 0.03] | 0.02 [0.01, 0.03] |
CFI | 0.988 | 0.996 | 0.996 |
TLI | 0.990 | 0.996 | 0.996 |
- Standard errors are presented within parentheses. The phenotypic model does not include controls for between-pair confounds, whereas quasi-causal models include controls for between-pair confounds. Bolded parameter estimates are statistically significant at p < 0.05
-
b
A
amount of variance in body mass index attributable to additive genetic influences; b
P
phenotypic association between predictor and outcome; RMSEA root mean square error of approximation; CFI comparative fit index; TLI Tucker-Lewis index
-
a
b
P
is constrained to be equal for males and females