The proposed research analyzes the development of reading skills in a sample of low SES, Latino, Spanish speaking English language learners. This population is at risk for low literacy attainment. A preliminary longitudinal study conducted in the same school district (1999 - 2002) found children with limited English proficiency (LEP) participating in a multisensory phonics-based early-exit bilingual program had mean reading scores in the average range on nationally normed reading tests in English through grade 2, but were below average on normed measures of English language skill. Funding is sought to follow these children (n=225) through grade 8. Funding is also sought for a new longitudinal study. Better outcomes might be attained in a phonics-based program that emphasizes oral language to a greater extent (dual language program). The proposed study will follow five groups of children, three LEP and two non-LEP groups. Group 1 (LEP) (n=200) will be in the same early transition bilingual program utilizing multisensory phonics, group 2 (n=200) will be in a dual language program utilizing similar teaching methods, and group 3 (n=200) will be in English immersion (by parental request). Group 4 (non-LEP) (n=100) will be in the dual language program and group 5 (non-LEP) (n=100) will be in regular English instruction. A predictive test battery in Spanish, with measures of oral language skill, phonological skill, nonverbal ability and initial literacy, will be given in Spring of kindergarten. Family demographics and literacy practices in the home will be assessed. Follow-up measures of oral language, phonological skill, word reading, reading fluency, reading comprehension, and academic content knowledge will be given in both English and Spanish at the end of each grade (1 through 4th). School achievement test data will also be collected. Non-random assignment precludes drawing causal inferences about relative instructional effectiveness, but the results will indicate the extent to which English language learners in the three models of instruction attain normative levels of oral language and literacy, and how they compare to local English-speaking children. Correlational data from grades K-4 will be used to determine the characteristics of children who fail to attain average levels of English literacy within and across the 5 groups under study (e.g., based on type of program, phonological skill and exposure to literacy practices in the home).
