The study took place in a Head Start program in upstate New York.
The study began with 139 four-year-old low-income children. During the course of the study, 10 children left the study, resulting in a final sample of 129 children from 10
classes. Eighty-three percent of the treatment and comparison children came from English-speaking families, and 17% of the children resided with non-English-speaking
families. Twenty-nine percent of the children were black, 22% were Hispanic or Latino, 42% were white, and 7% were Asian (primarily Afghan). Forty-eight percent of the
children were female. Classrooms were first matched on half-day or full-day status, and nine of the classrooms were randomly assigned to either the intervention or comparison conditions. The 10th classroom was placed in the comparison group because the intervention materials were not accessible to the teacher. Because the 10th classroom was assigned by convenience, the design for this study is quasi-experimental.
The children in the intervention group participated in 14 weeks of early literacy instruction designed to supplement the existing Head Start curriculum. The early literacy instruction focused on teaching phonological awareness skills and letter knowledge, such as letter names, letter sounds, and phonemic composition of words. Twenty pre-school books were introduced in large and small groups (about six children per small group), and children were encouraged to participate in discussions and to read. Children were also exposed to explicit instruction in letter names and letter sounds, and each lesson plan included phoneme awareness training using game-like activities.
The children in the comparison group participated in their regular Head Start curriculum that consisted of few emergent literacy activities and varied book reading activities.
The primary outcome domains assessed were oral language, print knowledge, phonological processing, early reading/writing, and cognition. Oral language was measured with
the Peabody Picture Vocabulary Test (PPVT-III), a standardized measure. Print knowledge was assessed with four nonstandardized measures—alphabet knowledge, letter
identification, letter-sound correspondence, and print conventions. Phonological processing was assessed with three nonstandardized measures—alliteration, rhyming, and
phoneme blending. Early reading/writing was assessed with two nonstandardized measures—invented spelling and word identification. The cognition domain was assessed
with two nonstandardized measures—sentence memory and word memory. (See Appendices A2.1–A2.5 for more detailed descriptions of outcome measures.)
Support for implementation
Teachers received an orientation packet, participated in one session of training prior to the intervention, and received weekly training once the intervention began. During the
weekly training, teachers reviewed lesson plans, and trainers addressed teacher concerns and suggestions and answered teacher questions. Because teachers implemented
the intervention in their respective program and groups, they were familiar with the children in the intervention and comparison conditions.