English proficiency, tracking, and the mathematics achievement of Latino English learners

The mathematics scores of Latinos on the National Assessment for Educational Progress (NAEP) have been described as "pervasively, disproportionately, and persistently low" over time, relative to whites (Valencia, 2002). The scores for Latinos who are not proficient English-speakers are particularly low. Thus, the mathematics achievement of Latinos, particularly English Learners (ELs), is a critical issue in our schools and, consequently, for educational research and policy. Both individual and institutional factors have been found to impact Latinos' mathematics achievement and some studies showed that: (1) English proficiency was linked difficulties in comprehending instruction in English, and, (2) Latino ELs with low-levels of English proficiency were often placed in lower-track courses. Using the Education Longitudinal Study of 2002 (ELS:2002) dataset, I analyze a nationally representative sub-sample of 2,234 native and non-native English-speaking Latino 10 th graders. I use generalized least squares (GLS) regression analysis to fit multi-level models that describe the mathematics achievement of native and non-native English-speaking Latino students as a function of: (1) their English proficiency, (2) their placement in a general or academic track in school, and, (3) whether they are provided with native language support. My study reveals that Latino non-native English speakers with low levels of English proficiency perform at much lower levels t0han their native-English speaking peers--about one standard deviation lower. English proficiency, however, was not the only factor that accounted for this substantial difference in mathematics performance. Institutional-level factors also played a strong role in mediating their achievement. Placement in the college preparatory track (compared to general track placement), and having access to a certified teacher with a mathematics background together accounted for a difference of about one-half of a standard deviation in performance. My findings suggest that access to college preparatory courses and native language support for Latino ELs can promote higher levels of mathematics achievement. Additionally, policies aimed at increasing the achievement of the most disadvantaged students in this study--Latino ELs--need to focus on equity and access to both rigorous mathematics content and adequately trained teachers that have a mathematics background in addition to training to meet the language needs of ELs.