Author Name	Galindo, Claudia Lucia
Email Address	cgalindo@csos.jhu.edu
Graduate Program	Educational Theory and Policy
Degree	Doctor of Philosophy
Document Type	Dissertation
Date of Defense	2005-08-17
Committee	Suet-ling Pong, Committee Co-Chair Sean Reardon, Committee Co-Chair Miryam Espinosa-Dulanto, Committee Member Salvador Oropesa, Committee Member
Availability	Open Access
Title	LATINO STUDENTS’ MATH LEARNING TRAJECTORIES IN THE EARLY SCHOOL YEARS: THE ROLE OF ENGLISH ABILITY AND SOCIOECONOMIC STATUS
Abstract	The rapid Hispanic population growth is significantly impacting the overall U.S population. In addition, Hispanics are over-represented among individuals living in poverty and are considered a high-risk group within schools. During the past twenty years, research has emphasized the relevance of the early school years for future life chances and learning experiences (Entwisle & Alexander, 1988; 1993; Cunningham & Stanovich, 1997; Farkas & Beron 2001). First grade children learn ten times more than high school students (Jencks, 1985), they show positive dispositions toward learning (Entwisle & Alexander, 1994), and the literacy and numeric skills acquired during this period are strong predictors of school achievement years later (Cunningham & Stanovich, 1997; Farkas & Beron, 2001). The early school years may be even more critical for Hispanic children who are predominantly children with foreign-born parents and are often raised in non-English speaking homes. This study, taking a longitudinal approach, uses the ECLS-K database to examine Hispanic students’ math learning trajectories between kindergarten and third grade. Specifically, this study describes Hispanic students’ math achievement patterns by Hispanic country of origin and generational status. In addition, I investigate the relative importance of English ability and SES in shaping their learning trajectories. A piecewise linear growth model is used to estimate learning growth during kindergarten, in first grade, and between first and third grade. Hierarchical Linear Modeling is used to analyze two-level models –assessments nested within students. This study finds important variability in math outcomes among Hispanic subgroups; variability that is generally masked when Hispanics of different countries of origin and generations are aggregated within one pan-ethnic denomination. Overall, Hispanic students could be clustered into two distinct groups. First and second generation Mexican students, Puerto Rican students regardless of generation, and Central American students regardless of generation are included in the first group. These students start kindergarten with significant math disadvantages that persist until third grade. The second group includes third+ generation Mexican, Cuban and South American students. This group show modest differences in average math scores at Kindergarten entry and similar growth rates between kindergarten and third grade. Moreover, SES and English ability effects on educational outcomes vary across school grades and Hispanic subgroup. For instance, English ability alone is crucial in explaining math achievement gaps between White students and Cuban and South American students. English ability is also more important in explaining math achievement gaps at kindergarten entry and during kindergarten. In contrast, SES is essential in explaining Mexican – White students’ math achievement gaps during first grade and from first to third grade.
Keywords	Longitudinal Study Hispanics Math achievement ECLS-K Immigration
Files	thesis_cgalindo.pdf - (666,828 bytes)