Assessing Classroom Quality in Settings Serving Young Dual Language Learners

• In Event: 70.051 - Assessment and Young Dual Language Learners in Pre-Kindergarten and Kindergarten: Issues to Consider

Abstract

Purpose

Increasingly federal and state governments view quality preschool education as a critical component in closing the achievement gap. A high quality preschool experience enhances early school performance and is associated with a variety of positive long-term outcomes (NICHD Early Child Care Research Network, 2006; Nores, Barnett, Belfield, & Schweinhart, 2005). Yet, studies of preschool quality suggest that the vast majority of programs are mediocre (Helburn, 1995). Thus, there is considerable interest in not only identifying what constitutes a quality preschool experience, but also how best to measure it.

Currently, there are several widely used measures of classroom quality (Ackerman, 2014). Proponents posit that their use is appropriate in classrooms with dual language learners (DLLs) (Burchinal, & Cryer, 2003; Downer, Lopez, Grimm, Hamaguri, Pianta, & Howes, 2012). However, this claim is not well grounded in either theory or research relevant to DLLs’ development. This presentation will evaluate the Environmental Rating System (ERS; Harms, Clifford, & Cryer, 2005) and the Classroom Assessment Scoring System (CLASS; Pianta, LaParo, & Hamre, 2008) as measures of classroom quality for young DLLs.

Theoretical Perspective

Much of child development theory and research posits that there are universal psychological processes. The major implication of this perspective is the belief that good quality preschool will result in positive outcomes for all children, regardless of children’s background characteristics. In contrast, developmental psychologists and early childhood education stakeholders view development as highly influenced by language and culture (Rogoff, 2003). This culturally and linguistically specific perspective is especially important for young DLLs, whose life experiences may be qualitatively different from those of U.S. monolingual English speaking children. This perspective has implications for the construct validity (Messick, 1989) and consequential validity (Welner, 2013) of classroom quality observations used in early education settings serving young DLLs, as well.

Methods and Data Sources

Data for this project was gathered as part of a collaborative discussion among experts in DLL development, with the topics discussed focusing on the cultural and linguistic appropriateness of the ERS and the CLASS. This discussion was conducted under the auspices of a statewide coalition committed to securing equal access to quality education for all children.

Results

Qualitative analysis of the issues discussed suggests four main concerns: each measure’s purpose, content, and psychometric properties; and assessor administration competence. The primary concerns regarding the ERS and the CLASS for use with DLL populations are incomplete construct validity and consequential validity. By overlooking how language and culture can affect teacher and child interactions, the overall construct of classroom quality is not captured. Assessor training and oversight is not systematically addressed in the research literature, as well. An implication of this preliminary research is that early education settings serving DLLs may need to consider a supplementary assessment that is developed specifically for this population.

Significance

Classroom quality measures are being used across a variety of early education efforts (Bryant, 2010). Yet, these measures were not developed with the unique needs of young DLLs in mind. Because of this, current measures overlook important features of the pedagogical environment salient for improved DLL outcomes.

Author

• Marlene Zepeda, California State University - Los Angeles