Academic Achievement in Limited English Proficient Students: A Multivariate Latent Growth Modeling Analysis of Predictors, Mediators, and Moderators
Purpose: The number of limited-English proficient (LEP) students is growing at a rapid rate across the nation. The achievement gap between LEP and non-LEP students has been repeatedly documented and is widening. In order to narrow and eliminate this gap, a systematic evaluation of institutional, pedagogical, sociocultural, and intrapersonal factors that interfere with the academic achievement of LEP students is needed. The current study will address this need by identifying the mechanisms through which native language proficiency relates to the mathematics and science achievement growth of LEP students.
Project Activities: Data from the National Education Longitudinal Study (NELS:88/2000) will be analyzed to identify the mechanisms through which native language proficiency relates to the achievement growth of LEP students.
Products: The expected outcomes of this research include published reports on the mechanisms through which native language proficiency relates to the achievement growth of LEP students.
Purpose: The purpose of this study is to identify mechanisms through which native language proficiency relates to the mathematics and science achievement growth of limited-English language students.
Setting: The current study will use data from the National Education Longitudinal Study (NELS:88/2000).
Population: The National Education Longitudinal Study includes a nationally representative sample of over 25,000 students beginning in 1988 when the students were in 8th grade. Students were surveyed four more times over a 12-year period as they progressed through high school and joined the workforce or went on to pursue postsecondary education. The study sample includes approximately 1,400 students indentified as limited-English proficient (LEP) who participated in the study at grades 8, 10, and 12, and for whom school transcript data were available.
Research Design and Methods: The research will focus on a bilingual education model in which proficiency in students' native language (L1) and exposure to their heritage culture promote academic achievement by facilitating the development of English language (L2) skills and bolstering students' self-esteem. Multiple mediator models will be used to (a) evaluate the bilingual education assumption that proficiency in the first language (L1) predict students' self-evaluations (i.e., self-esteem and/or academic self concept) which, in turn, predicts achievement; (b) identify the pathways through which self-esteem might influence achievement (i.e., facilitation of second language acquisition, enhancement of academic self-concept, or both); and (c) determine the extent to which race and ethnicity moderate the self-perception and achievement relation. Finally, the inclusion of both general self-esteem and academic self-concept measures within the same model will shed light on both the nature of the relation between these two constructs and their respective effects on achievement.
Key Measures: Measures in the analyses include L1 Native Language Proficiency (i.e., ability to understand, speak, read, and write in the native language), the Rosenberg Self-Esteem scale, English self-concept, and math self-concept. Proficiency in English, math, and science will be measured using standardized tests, along with student's grades in English, math, and science. Family and school covariates (e.g., family socioeconomic status, mother's/father's aspirations, percentage of students receiving free or reduced price lunch at school, percentage of minority students at school, and teacher's attitudes and morale) will also be included in the analyses.
Data Analytic Strategy: Multivariate latent growth curve modeling will be used to analyze the data. Competing nested models will also be tested and the moderating influence of ethnicity will be assessed by means of multi-group invariance analysis.
Journal article, monograph, or newsletter
Guglielmi, R.S. (2012). Math and Science Achievement in English Language Learners: Multivariate Latent Growth Modeling of Predictors, Mediators, and Moderators. Journal of Educational Psychology, 104(3): 580–602.