Longitudinal effectiveness of a program of bilingual individualized instruction in math, English and Spanish reading implemented by bilingual paraprofessional teachers recruited from migrant families was assessed among Spanish dominant monolingual and bilingual migrant students. Test scores in math and English reading for children age 5 to 9 were examined for a 6 year period, 1974-79, and test scores in Spanish reading from a 4 year period, 1977-80. After 1 year of individualized bilingual instruction the superiority of project children over a baseline comparison group was significant at the .01 level in all three subjects. Longer periods of attendance produced higher scores, with children going from a pretest average at the 9th percentile to the 50th percentile after 3 years in English reading, from the 14th percentile at pretest in math to the 70th percentile after 3 years, based on national norms from standardized tests. Because of schooling interruptions resulting from being migrants and coming from homes in which the primary language was not English, subjects had more educational hurdles to overcome than average children. Despite these obstacles, by means of individualized instruction and bilingual teachers, in several school years students were well on their way toward achieving literacy in two languages. (Author/NEC)