Skip to main content
. Author manuscript; available in PMC: 2009 Mar 18.
Published in final edited form as: J Consult Clin Psychol. 2009 Feb;77(1):74–87. doi: 10.1037/a0014281

Figure 2.

Figure 2

Piecewise latent growth model, which through proper specification of the factor loadings estimates three distinct components of change: presanction drinking, the change in alcohol use from the presanction baseline to the 1-month follow-up (intervention effect), and the change in alcohol use across the 1-, 6-, and 12-month follow-ups (postintervention change). Omitted factor loadings are fixed at 0, and factor scores are interpreted just as with standard regression. The Presanction Drinking factor scores estimate drinking when all other factor loadings are 0. The remaining two factors are interpreted as slopes (i.e., expected change). Therefore, the intervention effect factor scores estimate slopes from 0 (presanction) to 1 (1 month), and the Postintervention Change factor scores estimate slopes from 0 (now 1 month) through the 6-month and 12-month follow-ups. Intervention status is used to predict each of these three aspects of the fitted growth function. BMI = brief motivational intervention; Alcohol 101 = Alcohol 101 Plus.