Type of provision represents whether the classroom was part of gap 26 day
care center or preschool. To control for possible confounding of type of provision with structural quality characteristics, a dummy variable was created and included in the main analysis, with the values 1 = preschool and 0 = day care.
For the purpose of sample description, additional information was obtained about the age and ethnic composition of the classrooms, and about teachers’ gender, age, work experience, and ethnic background. These measures were not included in the main analysis.
The teachers’ self-reports show a comparatively high level of emotional support and much lower levels of educational support through providing pretend play and academic activities, with the reported support of children\'s self-managed play falling in-between.
Table 3 shows the bivariate correlations between all process quality indicators. The inter-correlations of the CLASS dimensions were moderate to strong. Also, the inter-correlations between the self-reported process quality indicators were moderate to strong. The correlations between
the CLASS dimensions and the self-reported activities were mostly significant and generally in the expected direction, but much smaller in magnitude. Table 4 presents the intraclass correlations for all CLASS dimensions, revealing significant within and between classrooms variance, indicating apnea
multilevel modeling is indeed required.