Skip to main content
Fig. 3 | BMC Medical Research Methodology

Fig. 3

From: Two-stage sampling in the estimation of growth parameters and percentile norms: sample weights versus auxiliary variable estimation

Fig. 3

The negative binomial auxiliary model. Boxes represent observed variables and circles represent latent variables. As in a typical growth model, the continuous variable Age affects the partially observed repeated measurements language score variable. The random intercepts factor, which represents current ability, is predicting the responses, the auxiliary variable and the random coefficients factor which represents the individual velocities. The association between the two factors is captured by allowing the random intercepts to predict the random coefficients

Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us