Note: This grant was awarded in FY 2009 but the project did not take place because the state made a decision not to carry out the planned expansion of the preschool program that was to be evaluated. The information included in the abstract below describes the planned scope of work that was reviewed and recommended for funding prior to the state's decision. All but $18,000 of the awarded funds were returned to IES.

Purpose: This project will evaluate the effectiveness of the New Jersey Preschool Expansion (NJPE) Program in attaining its primary objectives of enhancing the school readiness of economically disadvantaged children. It will also examine the relationship between student outcomes and characteristics of students, teachers, and districts to determine which are associated with the largest effects and thus provide guidance for program improvement. Many states have undertaken pre-k initiatives with similar expectations, so the results of this study will be of interest both to and beyond New Jersey.

Project Activities: This evaluation will use a randomized control trial to determine if the NJPE Program is improving school readiness of students entering kindergarten.

Products: The products of this project will be evidence of the effectiveness of the NJPE Program for improving the school readiness of economically disadvantaged students eligible for the program, and whether a child's participation in the program for 2 years yields greater benefits than 1 year. If positive benefits are found, further information will be provided on the cost-feasibility of NJPE, the characteristics of students who receive the greatest benefits, and the characteristics of teachers and districts associated with the benefits of the program. The evidence will be directly provided to the New Jersey Department of Education through the participation of department personnel on the project and to the research community and other education practitioners and policymakers through publications.

Setting: The evaluation will take place in 15 districts (or groups of contiguous districts) across the state of New Jersey.

Population: The New Jersey Preschool Expansion Program (NJPE) serves at-risk students. This evaluation will follow 2000 3 year olds for 2 years and 2000 4 year olds for one year.

Intervention: The New Jersey Preschool Expansion (NJPE) Program is a state initiative begun in 2008 aimed at providing access to intensive, high-quality preschool services to all low-income 3- and 4-year-old children in the state. It expands the Abbott preschool program that has been implemented in 31 high-poverty urban districts. Standards for the NJPE require such features as full-day services; a research-based curriculum; small classes; teachers with a B.A. and certification; adequate classroom space and resources; and supportive services. Services will be provided in schools, private child care centers, or Head Start. School districts are expected to implement NJPE gradually over the next 6 years using funds from the state based on the type of placement used for eligible children. The state appropriated $543.8 million for the NJPE program for 2008–2009.

Research Design and Methods: A randomized control trial will be used in districts which are oversubscribed for NJPE slots in the early years of the program's implementation. Random assignment will be used to place 2,000 3 year olds and 2,000 4 year olds either in a NJPE preschool classroom or not. To measure program impacts, all participants will take standardized assessments measuring language development, early literacy achievement, and early mathematics achievement. Data on fidelity of implementation relative to NJPE standards will be collected through classroom observations, staff interviews, and document review. Contextual information and student background data will be gathered from existing school district records. Analyses will measure children's achievement and growth, focusing separately on children entering at age 3 and at age 4. Analyses will also measure impacts for subgroups of children and for children in school districts with different characteristics.

Control Condition: Students who were not randomly chosen to attend a NJPE pre-k classroom (control students) may remain at home, be placed in child care, or attend another type of preschool depending on their families' decision. This represents a business-as-usual counterfactual based on what would be available to parents if there was no NJPE.

Key Measures: At the beginning of the study, students will take the Preschool Language Assessment Survey 2000 (Pre-LAS) to determine if they require Spanish-language assessments or cannot be assessed because of limited proficiency. Baseline data will be gathered using the Expressive One-Word Picture Vocabulary (EOWPVT), Woodcock-Johnson III Letter-Word Identification subtest (WJ3-LWI), and Woodcock-Johnson III Applied Problems subtest (WJ3-AP). After the first year in the study, 3 year olds will be assessed using these same three tests. Four year olds (and 3 year olds after 2 years in the study) will take these same three tests plus the Peabody Picture Vocabulary Test (PPVT-4), ECLS-B Preschool-Kindergarten Language and Literacy Assessment, and ECLS-B Preschool-Kindergarten Math Assessment. Each treatment classroom and a sample of control classrooms (for control students enrolled in preschool) will be assessed once using the Early Childhood Environment Rating Scale-Revised (ECERS-R Teaching and Interactions subscale, and Provisions for Learning subscale) and Classroom Assessment Scoring System (CLASS).

Data Analytic Strategy: To estimate impacts, the child outcomes measured after one year (for both 3 and 4 year olds) and after two years (for 3 year olds) will be regressed on treatment status, a baseline measure of the outcome, district indicator variables, and child and family background variables measured at baseline, The main impacts of the preschool expansion on children will be estimated using a student-level regression equation that includes school district fixed effects and other covariates. As a sensitivity analysis, the main impacts will also be estimated using a random effects model (also known as a hierarchical linear model). Additional models (single and multi-level) will be used to examine impacts for subgroups and fidelity of implementation.