Page 178 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

178 Chapter 8 Figure 8.3 shows that the effect of Age is significant. The differences within individuals (represented by the striped lines in Figure 8.3) and the differences between individuals (represented by the dotted lines in Figure 8.3) do not depend on Age. We cannot give more precise estimates of the NDW development for older pupils or younger pupils, nor can we say that the variation between older pupils is different from that of younger pupils. Likelihood ratio tests did not show that main effects of Positive Climate, or Regard for Student Perspectives, contributed significantly to the fit of the model to the observed data, nor did we find an interaction between Age and Positive Climate or Regard for Student Perspectives and NDW. This means that we could not show that the fact that teachers at Mainstream schools were more likely to provide a higher Positive Climate and take into account student perspectives compared to teacher at the DL2-schools had any significant impact on the pupils’ development of lexical diversity as measured with the NDW . Guiraud Index Score (GIS) From the comparison between the consecutive models (see Table 8.5) it is apparent that a model for GIS with a fixed linear component – allowing for differences in Age – fit the data better than a model with only an intercept (ΔΧ2 (GISb) = 41.83; df = 1; p < .001). The variance within individuals did not depend on the age of participants (GISc). However, the variance between individuals is a (linear) function of age (GISd). Adding Age2 to the model did not improve the fit significantly (GISe). Hence, in the final model (GISDEF) 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 8.5: Fit of Different Polynomials (-2LL) for Changes in Guiraud Index Score (114 cases) as well as the Comparison of Consecutive Models. ΔΧ2 Δdf p Comparison -2LL Models Model GISa: β0ijcons a 262.07 GISb: GISa + β1Age1ij GISc: GISb + e1ijAge1ij GISd: GISc + u10jAge1ij GISe: GISd + β2Age2ij GISDEF: GISa + β1j * Age1ij 214.81 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 estimated. 220.25 GISa vs GISb 219.00 GISb vs GISc 214.75 GISc vs GISd 212.15 GISd vs GISe 41.83 1 1.25 1 b 4,25 1 b 2.60 1 <.001 .26ns .04 .11ns