Page 132 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

132 Chapter 6 States. It is important that an instrument such as the CLASS has validity worldwide because it allows for comparison between countries. The use of the CLASS as an observational instrument for classroom quality seems to be also justifiable in the present study with newly arrived migrant pupils in the Netherlands since Slot, Boom, Verhagen, and Leseman (2017) showed that the CLASS was also an adequate measurement of early care and education in the Netherlands. The results of the CLASS in the different classes can be used in this specific context for evaluation of differences and similarities between schools. The CLASS will be used in the present study to see whether DL2-schools differ from Mainstream schools with respect to pedagogical practices focusing on teacher behavior. 6.2.2 Dutch studies using the CLASS In the Netherlands the CLASS is used in the national Cohort study Pre-COOL, since 2011 (N = 3000 pupils in 250-300 classes; e.g., Veen et al., 2012b). In this cohort study the CLASS was used to determine quality of day care, preschool, and kindergarten education17, and to relate the CLASS scores to child development in the domains of language, mathematics, and executive functions. Several publications show relations between the CLASS scores and developmental progress. For example, Henrichs and Leseman (2016) found that Dutch kindergarten teachers (N = 96) had a mean score of 5 for emotional support, which is at the high end of the medium range on the CLASS scale of maximum 7 points. The mean for instructional support was lower: 3, which is at the low end of the medium range. Within the instructional support domain Henrichs and Leseman report a large variation between the teachers. This pattern of good emotional support, but medium instructional support is comparable with what is described in international research using the CLASS. According to Henrichs and Leseman the medium score for Instructional support shows that on that point the quality of kindergarten education in the Netherlands is higher compared to international studies (compare NICHD ECCRN, 2005; Pakarinen et al., 2010; von Suchodeltz, Fäsche, Gunzenhauser, & Hamre, 2014). Furthermore, the CLASS scores were related to child outcomes: Higher instructional support had a significant, although small to medium, positive effect on vocabulary and mathematical skills. Veen, van der Veen, van Schaik, and Leseman (2017) found similar patterns in quality as Henrichs and Leseman (2016) with medium scores on emotional support but lower scores on instructional support in Dutch kindergarten classes. Furthermore, they found a small positive effect of behavior regulation and productivity in kindergarten on attention development. For vocabulary they only found a very small effect. For kindergarteners with a non-low-educated mother higher behavior regulation relates to lower 17 See Section 1.4 in Chapter 1 for an explanation of the Dutch educational system.