The first randomized controlled trial in Head Start’s history found that its effects on children largely faded out by the end of first grade (U.S. Department of Health and Human Services, 2010).  Notably, there was considerable variation in program quality.  For example, fewer than 1 in 20 4-year-olds in the treatment group were in centers with an “excellent” quality rating and only about half were in centers with recommended pupil/staff ratios.

These findings illustrate a common and vexing problem in early childhood interventions - how to maintain program quality and positive impacts on children in an at-scale program.  One particularly successful public prekindergarten program in the Boston Public Schools (BPS) that has had success in both regards offers one possible model for doing so.  A rigorous, large-scale study found that the BPS program has moderate-to-large impacts on children’s language, literacy, and mathematics skills and small effects on children’s emotional development and executive function skills.  Impacts on children’s language and mathematics skills are the largest of any similarly evaluated public prekindergarten program (Weiland & Yoshikawa, 2013).  A second study found that the program scores well on instructional quality, compared to a national sample of public prekindergarten programs (Weiland, Ulvestad, Sachs, & Yoshikawa, 2013). 

We will present findings from a pilot effort to expand the BPS prekindergarten model to community-based preschools in low-income Boston neighborhoods.  The expansion addresses questions regarding the external validity of the Boston results, as the community-based preschools are more structurally similar to Head Start than to BPS (e.g. staff are paid less and are less educated on average than BPS teachers). Teachers in 13 classrooms in 10 community-based preschools are being provided with key quality supports of the Boston prekindergarten model: training on the language, literacy, and mathematics curricula used in the BPS prekindergarten program, bi-weekly coaching from experienced BPS coaches, and curricular materials, along with a $5-$15K increase in teachers’ salaries.  Data on language, literacy, and mathematics instructional quality were collected in each classroom before the intervention began and will also be collected at the end of this pilot phase, as well as in a random sample of 10 BPS prekindergarten classrooms in the same neighborhoods as the community-based programs.  Specifically, we are using the Early Language and Literacy Classroom Observation (Smith, Dickinson, Sangeorge, & Anastasopoulos, 2002) to measure language/literacy quality and the Classroom Observation of Early Mathematics Environment and Teaching (Clements & Sarama, 2008) to measure mathematics quality.

At baseline, we found that the community-based classrooms on average scored below the “basic/adequate” level for quality of language/literacy instruction and at the “basic/adequate” level for quality of mathematics instructions. We will present findings on the change in quality after intervention in the community-based programs, as well as comparisons to average classroom quality in the BPS prekindergarten sample.  We will describe what these findings suggest about the scalability of the BPS program.  Relevant to the current policy context for Head Start, findings will also highlight the importance of systems-building in supporting quality improvement.