Page 190 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

P. 190

```
190 Chapter 8 5 4,5 4 3,5 3 2,5 Age in Months Growth GUIRAUD Growth GIS Within 80% - Within 80% - WWithiitnhi8n0%80+% + BeBtwetewenee8n0%80-% - BeBtwetewenee8n0%80+% + Figure 8.9: Graphical Representation of the General Development Model of Guiraud Index Score (162 cases) with 80% Reliability Figure 8.9 shows that of the effect of Age on GIS is significant. The variance within individuals (represented by the striped lines in Figure 8.9) does not depend on Age, however the variance between individuals (represented by the dotted lines in Figure 8.9) was a function of Age; variance in GIS score between older individuals is smaller than those between younger pupils. The analysis was continued with the new General Development Model, GISDEF. The inclusion of the percentage of time spent on Mathematical Activities (ΔΧ2 GISVI) = 6.61; df = 1; p = .01; see Table 8.14) as a fixed main effect improved the fit of the model for GIS significantly: the higher the percentage of time spent on Mathematical Activities, the lower the GIS. There was no interaction effect between Age and percentage of time spent in Mathematical Activities (GISVII). Table 8.14: Fit of Different Polynomials (-2LL) for Changes in Guiraud Index Score (162 cases) with the Addition of Mathematic Activities (MathAct) as Explanatory Variable. Comparison -2LL Models ΔΧ2 Δdf p 6.61 1 .01 1.24 1 .27ns Model GISIV: β0ijcons + β1jAge1ij 349.52 GISVI: GISIV + β2MathActij GISVII: GISVI + β3Age*MathActij 342.90 GISIV vs GISVI 341.66 GISVI vs GISVII The difference in GIS score when the percentage of time spent on Mathematical Activities was taken into account is graphically presented in Figure 8.10. There was a negative fixed main effect of the percentage of time spent on Mathematical Activities, but there was no 51 55 59 63 67 71 75 79 83 87 91 95 99 103 Guiraud Index Score
```