Skip to main content
. Author manuscript; available in PMC: 2016 Jul 1.
Published in final edited form as: Drug Alcohol Depend. 2015 Apr 27;152:185–193. doi: 10.1016/j.drugalcdep.2015.03.040

Table 2.

Goodness of fit measures comparing confirmatory factor analysis models of Nicotine Dependence: Unidimensional, three multi-symptom domain bifactor, and five multi-symptom domain bifactor models are presented.

Name Structure Multi-Symptom Sub-Domains Df AIC BIC logLik X2 df p
Cigarette
Model A Uni 0 1159 98393.51 98579.10 −49170.75
Model B Bifactor 3 1153 97236.38 97464.80 −48586.19 1169.12 6.00 <0.01
Model E Bifactor 5 1149 97034.38 97291.36 −48481.19 210.00 4.00 <0.01

Cigarette+Cigar
Model A Uni 0 204 6290.57 6404.06 −3119.29
Model B Bifactor 3 198 6260.30 6399.98 −3098.15 42.27 6.00 <0.01
Model E Bifactor 5 194 6242.19 6399.32 −3085.09 26.12 4.00 <0.01

Cigar
Model A Uni 0 45 2383.32 2494.80 −1165.66
Model B Bifactor 3 39 2361.70 2498.91 −1148.85 33.62 6.00 <0.01
Model E Bifactor 5 35 2371.20 2525.56 −1149.60 −1.49 4.00 >0.99

Smokeless
Model A Uni 0 129 5110.52 5225.48 −2529.26
Model B Bifactor 3 123 5073.77 5215.26 −2504.89 48.75 6.00 <0.01
Model E Bifactor 5 119 5071.06 5230.24 −2499.53 10.71 4.00 0.03

Note: Available degrees of freedom (Df) in unidimensional model reflect the number of unique patterns of responses minus 1.