Level of Measurement and Improving Associations between Early Care and Education Quality and Practice and Child Outcomes

• In Event: 3-021 - Classroom Quality in Early Childhood: Understanding the Importance of Individual Experiences

Integrative Statement

High-quality early care and education (ECE) improves opportunities for all children, especially those who are dual language learners or from low-income families (Yoshikawa et al., 2013). The quality of interactions between caregivers and children broadly define quality (Hamre, 2014), but show only modest associations with child outcomes (Burchinal, 2018). Additional ECE practices that relate to child outcomes include curriculum (Burchinal, 2018), the language exchanges between caregivers and children (Justice et al., 2008) time spent in content-related instruction (Burchinal et al., 2016), and time spent in small and whole group settings (Foorman, & Torgesen, 2001). Each practice can be measured at the child or classroom level. The purpose of this study is to relate gains in child outcomes to classroom or child-level measures of ECE quality and practices.
This study includes 351 children in 63 pre-kindergarten classrooms (49% male, 37% African American, 32% dual language learners [DLLs]). Children were assessed by trained data collectors in the fall and spring on the Woodcock-Johnson III Picture Vocabulary (WJPV), Letter-Word Identification (WJLW), and Applied Problems (WJAP); Expressive One-Word Picture Vocabulary Test (EOW); DIBELS First Sound Fluency (FSF) and Phonemic Segmentation Fluency (PSF); NIH Toolbox Flanker and Dimensional Change Card Sort; and teacher ratings of self-regulation and social skills based on Teacher-Child Relationship Scale and Learning Behaviors Scale. Quality of prekindergarten classrooms and teacher practices were assessed with teacher report of the curriculum, Classroom Assessment Scoring System (CLASS), and Language Interaction Snapshot (LISn). LISn captures at the individual child level the frequency of quality language interchanges with the teacher (teacher elaborations/elicitations and sustained conversations), reading- and math-related activities, and small and whole group settings. See Table 1.
Analyses examined children’s fall to spring gains on each of the outcomes in three sets of analyses using hierarchical linear models with random classroom intercepts. Each of the three models examined different ECE quality or practice variables. Model 1 included quality measures collected at the classroom level. LISn variables measuring ECE practices scored at the classroom level were in Model 2 and at the child level in Model 3.
Results are shown in Table 2. In Model 1, CLASS showed mixed, but limited, associations and Creative Curriculum related to smaller gains in reading skills (WJLW d=-.46, FSF d=-.43, PSF d=-.41). Contrasting the results from Model 2 and Model 3 suggests that, with as little as 20 minutes per child, prediction of child outcomes was as good or better with child than with classroom level measures of ECE practices. Child-level measures of language exchanges related to gains in language (EOW d=.12, WJPV d=.16) and prosocial skills (d=.13). Child-level measures of content-related instruction predicted larger gains in reading (FSF d=.13, PSF d=.13) and of whole group settings predicted smaller gains in language (d=-.17), math (d=-.17), and inhibitory control (d=-.13).
Findings suggest more focus on measuring individual child experiences may be warranted. This focus may improve our ability to measure quality in meaningful ways and to provide targeted professional development to caregivers that improves child outcomes.

Authors

• Peg Burchinal, University of North Carolina
  Presenting Author

• Ellen Peisner-Feinberg, University of North Carolina-Chapel Hill
  Non-Presenting Author

• Mary Bratsch-Hines, UNC-Chapel Hill
  Non-Presenting Author

• Ximena Franco, University of North Carolina at Chapel Hill
  Non-Presenting Author