Page 125 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

Development of narrative ability 125 From the comparison between the consecutive models (see Table 5.17) it is apparent that a model with a fixed linear component – allowing for differences in Age – fit to the data better than a model with only an intercept (ΔΧ2 (IST1) = 20.13; df = 1; p < .001). The variance within individuals did not depend on the age of participants (IST2). The variance between individuals is a function of Age (ΔΧ2 (IST3) = 6.10; df = 2; p = .05). However, the inclusion of Age2 (IST4) did not improve the model. Hence, in the final model (ISTDEF) a fixed effect of Age, as well as a variance between pupils component which depends on Age needs to be included, and with this model we continued the analysis. Table 5.17: Fit of Different Models (-2LL) for Changes in Internal State Terms (168 cases) as well as the Comparison of Consecutive Models. Comparison Model IST0: β0ijcons a 10.19.84 IST1: IST0 + β1Age1ij IST2: IST1 + e1ijAge1ij IST3: IST2 + u1ijAge1ij IST4: IST3 + β2Age2ij ISTDEF: IST0 + β1jAge1ijc 991.35 a In addition to the intercept, variance components for differences within and between individuals are estimated b Only the covariance-coefficient between the intercept- and the age-residuals was significant. c In this model only the between participants covariance-coefficient between the intercept- and the age-residuals was included. Based on this General Development Model we constructed Figure 5.12, in which both the average development as well as the differences within and between individuals are represented (see Table 2.5 in Appendix 2 for the parameter estimates). The average IST score at an age of 73 months was estimated as 5.45. Each month a pupil grew older his receptive vocabulary increased by 0.14. -2LL Models ΔΧ2 Δdf p 999.71 IST0 vs IST1 995.98 IST1 vs IST2 989.88 IST2 vs IST3 989.21 IST3 vs IST4 20.13 1 3.73 1 6.10 2b 0.68 1 <.001 .05ns .05 .41ns