Multilingualism, mathematics achievement and instructional language policy

A significant and growing proportion of students in the United States speak primarily a non-English language at home. This dissertation contributes to the understanding of academic achievement patterns among language minority students in the United States.;The first essay uses data from the Early Childhood Longitudinal Survey Kindergarten Class of 1998 (ECLS-K) to characterize the development of mathematical skills of multilingual children. Although descriptively children from households with a predominant non-English language are disadvantaged, previous work has suggested their relative strength in acquisition of math skills. My study focuses on three questions: (1) Do bilingual children learn math faster than monolingual children?, (2) Does the bilingual benefit accrue only when mathematical content becomes more complex, and (3) Does bilingualism associate with heightened attentional abilities? Empirical results suggest that bilingual children do indeed increase in math skills at a faster rate, and that the rate of growth associates positively with the reported amount of non-English language spoken at the household. There is inconclusive evidence to associate higher growth rates with the introduction of more complex mathematical material. Last, while bilingualism is found to positively associate with teacher reports of the child's attentional behavior, this does not mediate the impact of bilingualism on math learning.;The second essay then looks at the adoption of state level English immersion policy. After California passed Proposition 227 in 1998 to essentially cease ESL and bilingual instruction, Arizona and Massachusetts passed similar measures in 2000 and 2002, respectively. I use both the ECLS-K and the National Assessment of Educational Progress (NAEP) data to examine state level achievement patterns with respect to these English immersion policies, taking into account differential effects by state and child's language background. Results generally indicate either no policy impact or a negative impact on achievement for language minority students, while some positive effects are found among language majority students, thus calling into question the effectiveness of the policy in its stated goals.