Page 50 - Second language development of newly arrived migrant kindergarteners - Frederike Groothoff

50 Chapter 2 sensitive, and supportive” (p. 897). To investigate process quality in education, researchers primarily use classroom observations. However, linear associations between indicators of quality and child outcomes are mostly small (Burchinal, Kainz, & Cai, 2011) and sometimes even non-significant (e.g., Weiland, Ulvestad, Sachs, & Yoshikawa, 2013). Pianta, Barnett, Burchinal, and Thornburg (2009) argue that this is due to a large variation in program designs, curriculum, staffing, and level of educational aims. According to Mashburn et al. (2008) there is on the other hand not much variation in for example observations with the ECERS-R (Harms et al., 1998) – an observation tool for overall quality of the environment in pre-kindergarten – which makes it difficult to establish effects as well. Mashburn et al. argue that due to regulation all facilities have improved the quality of the classroom reducing the variation between the classrooms and with that, attenuating predictions of the ECERS-R to child outcomes. Hatfield et al. (2016) investigated the small effects of process quality on child outcomes in more detail and found that there are certain thresholds for quality measurements: an effect is only visible when scores are above a certain threshold (see also Zaslow et al., 2010). All of these aspects must be considered when measuring classroom quality and looking for effects of quality on child development. An essential issue for this study is whether these quality measures are suitable for measuring the environment of second language learners. Peisner-Feinberg et al. (2014) reviewed ten studies to see how well measures of quality in early child care education would also reflect the needs of dual language learners. They concluded that the widely used measures for quality discussed above function similarly for dual language learners when compared to mainstream populations. However, they encourage researchers to further disentangle the aspects of quality that stand out for dual language learners, for example examining whether interactions are culturally and linguistically responsive. Furthermore, it is also important to keep in mind that even though similar methods are used for observing process quality, the context of the observation – for example the population in the classroom, regional differences, or implemented programs – make it hard to compare outcome measures (Buysse, Peisner-Feinberg, Páez, Scheffer Hammer, & Knowles, 2014; Buell, Han, & Vukelich, 2017). In the present study, process quality will be studied at two levels: (1) teacher behavior and (2) focal pupils’ experiences. The next two sections will discuss some findings from previous research regarding these two levels. 2.7.2 Focus on teacher behavior Quality of education largely depends on the processes in the classroom such as socio- emotional climate and instructional support of the teacher (La Paro, Pianta, & Stuhlman, 2004). Teachers who are sensitive towards the pupils and who create a positive climate in