Measurement of preschool quality within the national educational panel study—results of a methodological study

Abstract

It has been argued that for the assessment of educational quality—especially in preschool—observational methods are the silver bullet. However, in large-scale assessments like the National Educational Panel Study (NEPS), observations can hardly be conducted. Against this background, we carried out a study organized around the question of the extent to which preschool quality can be assessed using teacher questionnaires. Therefore, a standardized observation of 60 preschool groups, using the German versions of the Early Childhood Environment Rating Scale (ECERS; KES-R and KES-R-E)—a well-established rating instrument for (process) quality in preschool environments—was conducted. Moreover, teachers filled out a questionnaire on preschool quality from pilot studies of the NEPS. In this paper, we present main results from the comparative analyses of observations and surveys of preschool quality using regression analytical methods. It can be shown that on a global level, preschool quality can be reproduced quite well by means of questionnaire data. Conclusions concerning the questionnaire design of the NEPS main study in Starting Cohort 2—Kindergarten are drawn. Finally, some cautionary notes on the use of single indicators of preschool quality and on causal inferences are given.