The Effects of Program Models on Language Acquisition by Spanish Speaking Children

This longitudinal study investigates the effects of five different program models on both acquisition and maintenance of Spanish by native Spanishspeaking kindergarten children. The five program models are: 1) Traditional or regular kindergarten, taught in English, with no formal English as a second language (ESL) training; 2) Traditional, with daily ESL; 3A) Bilingual, using the concurrent translation approach, no ESL; 3B) Bilingual, using the alternate immersion approach, no ESL; and 4) Bilingual, using the concurrent translation approach, with daily ESL. Interaction analysis data gathered in the bilingual classrooms indicated that balanced language use (50% Spanish, 50% English) occurred only in the alternate immersion approach, while unbalanced language use (28% Spanish, 72% English) occurred in the groups using the concurrent translation approach. Criterion language measures were administered on a preand posttest basis in both Spanish and English by peer testers. An instrument to measure communicative competence, designed by the investigator, was also given. Children were equated for cognitive ability and Spanish dominance. Using planned comparisons on the multivariate analysis of gain scores, it appears that bilingual treatments ((3A, 3B and 4) produced significantly greater gains in English oral comprehension than did the traditional all English treatments (1 and 2). In addition, the bilingual balanced treatment (3B) produced the greatest gains in English oral comprehension and communicative competence in Spanish and English. Finally, ESL training (2 and 4) did not facilitate English communicative competence, but may facilitate English comprehension at initial stages. As expected, treatments without ESL (1, 3A, 3B) showed significantly higher gains in Spanish vocabulary. It appears that bilingual program models with balanced language input are most facilitative for both Spanish and English acquisition by children.