Bilingual education cognitive development and school achievement : a study of the bilingual programs for Tibetan children

Abstract: This study examines the effects of two models of school bilingual programs on Tibetan children's cognitive development and school achievement.Based on the findings of empirical studies, Cummins' threshold theory was critically evaluated. It is hypothesized that children's general IQ, language development and achievement in content subjects have different developmental frameworks and are therefore affected by different factors. The effects of bilingual programs on children's general IQ are mainly determined by the total strength of the intellectual stimulations of the programs and also by the pupils' developmental curves. Bilingual children's language development is more sensitive to the total time input in learning the particular languages and also sensitive to achieved general IQ. The achievement in content subject is affected not only by the achieved IQ level and the time input, but also by the proficiency of the working language.The present study involved 508 pupils from Grade 2 and Grade 4 in a Tibetan area of China. The bilingual group consisted of 308 pupils enrolled in two models of Tibetan-Chinese bilingual programs (BL1 and BL2), and the monolingual group consisted of 200 pupils enrolled in Han Chinese programs. Raven's Standard Progressive Matrices were administered twice to all the sampled pupils with a one-year interval as the measures of general IQ. Pupils' records from two related uniform examinations were taken as the measures of pupils' language development and achievement in content subjects. Meanwhile, some questionnaires were administered to pupils, teachers and headmasters to collect background information.Data were analyzed in three stages: (1) bivariate analyses of the test scores, (2) bivariate analyses of the background variables and (3) multivariate analyses (ANACOVA and LISREL modeling) of the total data. The results from the bivariate and multivariate analyses were finally synthesized.The basic conclusions from the present study are: (1) pupils' development in general IQ was basically not affected by bilingual programs; (2) the bilingual pupils were significantly poorer than the Chinese monolingual pupils in Chinese language, but they were learning two languages; (3) there were no substantial differences between the BL1 pupils and their monolingual peers in mathematics, particularly at Grade 2, but the BL2 pupils performed worse than the monolingual pupils in mathematics, particularly at Grade 4.