Head Start (HS) programs serve approximately one million children growing up in poverty. Among HS children nationally, one-third are dual language learners (DLLs) and 80% of HS DLLs speak Spanish. To date, Spanish-speaking DLL children in HS programs have generally been considered a homogeneous group, however they are likely developmentally heterogeneous given variability in language proficiency in English and Spanish and other developmental skills. Although the within-group variability of DLLs enrolled in HS programs is increasingly recognized, there remains a need for understanding variation in HS DLL children’s longitudinal academic skill development after HS. In addition, very little is known about factors such as child/family characteristics and how HS program, school, and neighborhood environments are associated with the variations in the longitudinal academic skills of HS DLL children.
To address this gap, the proposed project will integrate multiple datasets including local HS evaluation data, public school administrative data, public school rating data, and census data to create a longitudinal and multi-layered dataset. One of the main goals of HS is to increase the school readiness of children from low-income families and enhance their long-term success in life. However, conducting a longitudinal study with a large study sample is expensive and requires tremendous effort and energy to follow the study sample over time. Data integration provides an alternative strategy and becomes a workable solution by using reliable public data that measures real-life outcomes in society. Using the integrated data, the following questions will be addressed:
- Among Head Start dual language learners, are there distinct developmental patterns of academic skills from kindergarten to third grade?
- Which child/family characteristics are associated with HS DLL children’s profiles?
- Which HS program, public school, and neighborhood environmental factors are associated with HS DLL children’s profiles? (author abstract)