. Author manuscript; available in PMC: 2023 Jun 6.

Published in final edited form as: Psychol Addict Behav. 2022 Jun 23;37(1):166–176. doi: 10.1037/adb0000863

Table 6.

AIC for Model Comparisons—$200 Magnitude

			AUDIT		Alcohol consumption
Class	Model	Predictors	AIC	AUDIT rank	AIC	AC rank
Single predictors	1	MM200	89.96	4^a	139.95	4^a
	2	AA200	89.01	2^a	135.91	2^a
	3	AM200	84.05	1	132.58	1
	4	MA200	89.04	3^a	136.68	3^a

Double predictors	5	MM200, AA200	91.72	4^a	138.30	4
	6	MM200, AM200	87.11	1	135.66	1
	7	MM200, MA200	93.01	5^a	138.53	5^a
	8	AA200, AM200	89.46	2^a	137.79	2^a
	9	AA200, MA200	95.08	6^a	141.30	6^a
	10	AM200, MA200	90.32	3^a	138.08	3^a

Triple predictors	11	MM200, AA200, AM200	92.84	1	140.68	1
	12	MM200, AA200, MA200	97.89	4^a	143.77	4^a
	13	MM200, AM200, MA200	93.35	2	141.13	2
	14	AA200, AM200, MA200	95.64	3^a	143.24	3

Quadruple predictors	15	MM200, AA200, AM200, MA200	98.98		146.11

Note. AIC = Akaike information criterion; AUDIT = alcohol use disorders identification test; AC = alcohol consumption. For each alcohol use measure, the best interclass AIC scores are in bold.

Denotes an unpreferred model relative to the best-performing model.