Background
The Harvard Language Diversity Project (Catherine E. Snow, Principal Investigator) is a sub-project of the New England Quality Research Center on Head Start (NEQRC). The NEQRC, directed by David K. Dickinson of the Education Development Center, Newton, MA, is a research consortium funded by the Administration for Children, Youth, and Families of the Department of Health and Human Services. The research consortium also includes Boston College (Martha Bronson, Principal Investigator) and the Massachusetts Society for the Prevention of Cruelty to Children (David Robinson, Principal Investigator). Head Start research partners have included Community Teamwork, Inc. (CTI) Head Start in Lowell, MA, Cambridge Head Start in Cambridge, MA, the Community Action Programs in the Inner City (CAPIC) Head Start in Chelsea, MA, Communities United Head Start in Waltham, MA, and Action for Boston Community Development, Inc. (ABCD) Head Start in Boston, MA.Methodology and Results
In 1996, in response to the growing diversity of the families being served by Head Start, the NEQRC developed a multi-faceted program of research with a particular focus on bilingual children and their families. One major strand of work in the NEQRC involved the development and administration of measures of preparation for literacy in English and Spanish appropriate for use with three and four year old children. These included measures of phonemic awareness (phoneme deletion, rhyme recognition), emergent literacy (understanding print, letter identification, early writing), and book concepts (for more information about these tests contact msweet@edc.org), as well as the administration of the Peabody Picture Vocabulary Test (PPVT) and the Test de Vocabulario en Imagenes Peabody (TVIP) for receptive vocabulary in English and Spanish.Another strand of the research program, carried out by the Harvard Language Diversity Team (Consuelo Aceves, Lilia Bartolomé, Linda J.Caswell, Mariela M. Páez, Catherine E. Snow, Patton O. Tabors, and Anne Wolf), involved the establishment of a Spanish-language Head Start classroom for three-year-olds from Spanish-speaking and bilingual Spanish-English homes, and the development of classroom ethnographies in that classroom and two others, an English-language classroom, where there were children from a variety of home language backgrounds, and a bilingual Spanish-English classroom where there were children who were either bilingual in Spanish and English, or were Spanish speakers who were acquiring English.
The most significant finding of these ethnographies (Tabors, P., O., Aceves, C., Bartolomé, L., Páez, M. M., Wolf, A., Language development of linguistically diverse children in Head Start classrooms: Three ethnographic portraits; available from msweet@edc.org) was the contrast between the experience of children in the English-language classroom and the experiences of children in the bilingual and the Spanish-language classroom. Throughout the year, the bilingual children in the English-language classroom had to cope with a situation in which the language they were most familiar with was rarely available to them, whereas the children in the other two classrooms always had access to teachers and/or children who could speak the language they felt most comfortable speaking. In practical terms, this meant that the bilingual children in the English-language classroom spent a lot of time listening to talk that they might not understand without being able to contribute much to the talk themselves. In the other two classrooms, children had no trouble finding teachers or children to talk with and could, therefore, proceed to communicate actively and participate fully in the classroom experience.
Further, the three classrooms delivered different messages to the children concerning their home languages. For the most part, the children in the English-language classroom were in a subtractive bilingual environment, in which their home language was being replaced with English, as English was the language of status in the classroom. In the bilingual classroom, the children were in an additive bilingual environment, in which both Spanish and English were given status as appropriate languages to be used in the classroom. In the Spanish-language classroom, the message was that Spanish was the more important language. This might be considered a pre-additive bilingual environment that strengthens the home language prior to exposure to English so that the addition of English will not result in the subtractive situation typical of the English-language classroom.
Which of these is the best situation for young bilingual children? The evidence derived from the ethnographies does not answer this question. The only available evidence comes from the receptive vocabulary scores, which represent only a small slice of the children's abilities. However, the receptive vocabulary scores point to two clear conclusions:
1. The children from homes where Spanish was spoken and the children from homes where Spanish and some English were spoken scored better in Spanish receptive vocabulary than in English receptive vocabulary after a year in a bilingual or Spanish-language Head Start classroom.
2. The bilingual children in all three classrooms scored well below average in English, whether they were in a classroom that emphasized English or not.
The number of children who were tested is very small and these conclusions can not be considered anything but suggestive; however, there is remarkable consistency in these findings across the three classrooms. Apparently, being in an English-language classroom, even one supportive of second language acquisition, did not provide the bilingual children with enough exposure to bring their receptive vocabulary in English up to the national average for their age.