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

Table 5 Growth parameter estimates from the growth model that uses weights and from the negative binomial auxiliary model (NbAux) that utilises the auxiliary variable CCC-S

	Weighted model coef. (st.error)	NbAux coef. (st.error)
Fixed intercept	33.3329 (0.5661)	29.8971 (0.4674)
Fixed Age (months)	0.7078 (0.0161)	0.6751 (0.0119)
Variance of current ability	65.1521 (6.8122)	70.5858 (6.5967)
Variance of individual velocities	0.0368 (0.0059)	0.0336 (0.0042)
Current ability and velocities covariance	0.4432 (0.1753)	0.5431 (0.1192)
Level-1 residual	35.2039 (4.2362)	38.2550 (2.6831)

Both models were fitted to 1,289 observations from the Expressive Vocabulary test. The NbAux model utilised an additional 6,411 observations from the CCC-S test

ISSN: 1471-2288