Page 117 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

Development of narrative ability 117 The difference in MLR score when Exposure to Dutch at School was taken into account is graphically presented in Figure 5.9. The curve in Figure 5.9 is caused by the estimated parameters of Age and Age2 (see Table 2.3 in Appendix 2) Since these estimates are unreliable due to large standard errors, the curve itself should not be taken into account by interpreting this graph. What Figure 5.9 does show is the effect of Exposure to Dutch at School on the MLR score: every increase in Exposure increases the MLR score, no matter how old the pupils are. 5 4,5 4 3,5 3 Exposure = 0 2,5 Exposure = mean - 1 sd 2 1,5 1 0,5 0 = 5 months Exposure = mean = 14 months Exposure = mean + 1 sd = 24 months Age in Months Figure 5.9: Development of MLR with Exposure to Dutch at School as Explanatory Variable. The addition of type of Educational Facility did not improve the model fit of the General Development Model of Measure of Lexical Richness. This means we could not show that whether the pupil was at a DL2-school or a Mainstream school influenced the MLR score, the use of less frequent words. 5.4.2 Macrostructure The three elements of macrostructure will now be presented: Story Structure (SS), Structural Complexity (SC), and Internal State Terms (IST). For all three elements General Development Models will be built. Story Structure The Story Structure is the score for the included parts of story grammar in the total narrative. Example (7) and (8) illustrate this SS score variable. The stories in Examples (7) 51 55 59 63 67 71 75 79 83 87 91 95 99 103 Measure of Lexical Richness Score