Skip to main content
. Author manuscript; available in PMC: 2018 Sep 1.
Published in final edited form as: J Public Health Dent. 2017 May 18;77(4):360–371. doi: 10.1111/jphd.12220

Table 4.

AIC Values for Three Methods of Representing Longitudinal Beverage Intake Data in the Iowa Fluoride Study (n = 344)

Method* # Dietary parameters estimated (k) −2log (likelihood) AIC Difference from minimum AIC Level of empirical support for method compared to lowest AIC model
Method 1 8 729.0 745.0 0.0
Method 2 16 718.6 750.6 5.6 Less support
Method 3 24 714.3 762.3 17.3 Substantially less support
*

Method 1 represents beverage intake variables through an average of questionnaires between 9.0 and 17.0 years of age; Method 2 represents beverage intake variables through questionnaires around 9 and the change of beverage intake between 9 and 13 years of age; Method 3 represents beverage intake variables through three specific time points around 9, 13, and 17 years of age.