Skip to main content
. Author manuscript; available in PMC: 2011 Jul 13.
Published in final edited form as: Child Dev. 2011 May 11;82(4):1267–1281. doi: 10.1111/j.1467-8624.2011.01598.x

Table 5.

Model Comparisons for Multivariate Genetic Analyses for the Two Child Psychopathy Scale Factors With Reactive and Proactive Aggression, Caregiver and Twin Self-Reports

Overall fit
 Model comparison
Model −2LL df AIC BIC χ2 df p Δχ2 Δdf p Comparison
model
Caregiver reports
1 Saturated modela 11,708.90 4,644 2,420.90 −9,045.20
2 Cholesky m ≠ f 11,858.89 4,796 2,266.89 −9,457.88 149.99 152 .531 — — — —
3 One-factor IP m ≠ f 11,878.18 4,808 2,262.18 −9,486.73 169.28 164 .354 19.29 12 .08 Model 2
3a One-factor IP m = f
All (common + specifics)		 11,909.33 4,832 2,245.33 −9,548.16 200.43 188 .254 31.15 24 .15 Model 3
3b One-factor IP m = f
All (common + specifics;
  drop C in common		 11,921.60 4,836 2,249.60 −9,554.86 212.70 192 .146 12.27 4 .02 Model 3a
3c One-factor IP m = f
All (common + specifics);
  drop C in
  common + specifics 		11,923.91 4,840 2,243.91 −9,566.54 215.01 196 .167 14.58 8 .07 Model 3a
3e One-factor IP m = f
Drop A in common		 11,921.46 4,836 2,249.46 −9,554.93 212.56 192 .147 12.13 4 .02 Model 3a
3f One-factor IP m = f
Drop A in
  common + specifics		 11,970.15 4,840 2,290.15 −9,543.42 261.25 196 .001 60.82 8 <.01 Model 3a
Twin self-reports
1 Saturated modela 12,181.83 4,620 2,941.83 −8,731.74
2 Cholesky m ≠ f 12,340.69 4,772 2,796.69 −9,139.98 158.86 152 .34 177.56 164 .22 —
3 One-factor IP m ≠ f 12,359.39 4,784 2,791.39 −9,169.13 177.56 164 .22 — — — Model 1
3a One-factor IP m = f
All (common)		 12,379.87 4,796 2,787.87 −9,197.39 198.04 176 .12 20.48 12 .08 Model 3
3b One-factor IP m = f
All (common + specifics)		 12,412.47 4,808 2,796.47 −9,219.59 230.64 188 .02 230.64 188 .02 Model 1
3c One-factor IP m = f
All (common);
  drop C in common		 12,383.08 4,800 2,783.08 −9,208.62 201.25 180 .13 201.25 180 .13 Model 1
3d One-factor IP m = f
All (common); drop C in
  common + specifics 		12,389.03 4,808 2,773.03 −9,231.31 207.20 188 .16 9.16 12 .10 Model 3a
3f One-factor IP m = f
Drop A in common		 12,393.06 4,800 2,793.06 −9,203.63 211.23 180 .055 13.19 4 <.001 Model 3a

Note. Best fitting models are presented in bold. −2LL = −2(log-likelihood); AIC = Akaike’s information criterion; BIC = Bayesian information criterion; IP = independent pathway; Δχ2 = difference in log-likelihoods between nested model.

a

All models compared with Model 1 unless otherwise specified.