Evaluation of the Program for Infant/Toddler Care

• In Event: Implementing Randomized Controlled Trials in Local Education Settings: Design, Methods, Findings, and Practical Research Realities

Abstract

In recent decades, the quality of child care for children younger than three years old has been of concern nationally. Advances in research on early brain development in the 1990s underscored the critical importance of children’s early learning experiences and the potential benefits of effective early interventions. Preliminary studies, mostly nonexperimental, have identified training strategies that warrant additional research, including the use of specific curricula, on-site consultation, high intensity and/or duration interventions, and focus on practice (Fukkink and Lont 2007; Zaslow et al. 2010).
The current study used an experimental intent-to-treat design to measure the impact of an established caregiver training intervention, the Program for Infant/Toddler Care (PITC), on child development and child care program quality. The Regional Educational Laboratory West, administered by WestEd, contracted with Berkeley Policy Associates to conduct a third-party evaluation of the PITC.
The intervention under study combines direct caregiver training and on-site coaching or other tailored assistance. For center-based child care programs, the intervention is delivered on-site. Family child care providers participate in groups of 5 to 10 programs, coming together for training sessions in a provider’s home or in a convenient community center or school. The trainer also visits individual family child care homes for on-site technical assistance and support. Technical assistance includes observations, director meetings, and “reflective action planning,” group meetings in which staff reflect on progress and plan further improvements. Individual participants receive at least 56 hours of training and 8 hours of reflective action planning.
The study was implemented over 2007–2010 in six Southern California counties and four Arizona counties. The study sample of 251 child care programs included 92 child care centers and 159 licensed family child care homes, and the sample of 936 children included an average of eight children per center and between one and two children per family child care home. The primary questions focus on child outcomes:
• What is the impact of the PITC on a composite measure of children’s cognitive and language skills, at least 6 months after its full delivery to the children’s child care programs (within an average of 23 months after random assignment)?
• What is the impact of the PITC on a composite measure of children’s social and behavioral skills, at least 6 months after its full delivery to the children’s child care programs (within an average of 23 months after random assignment)?
The confirmatory research questions were addressed using hierarchical linear regression models to account for the effect of clustering observations within programs. Each impact analysis included covariates collected before random assignment to improve the estimates’ statistical precision and reduce the likelihood that random sampling variation would affect the impact estimates.
The findings for this study are currently embargoed while under final review in the Standards and Review Office of the US Department of Education. The public release of the study findings are expected in fall 2011.

Author

• Phyllis Weinstock, Berkeley Policy Associates