This study investigated the effects of two types of bilingual programs (two-way and transitional) on the academic performance and attitudes of fifth-grade students who entered kindergarten or first grade with different levels of English proficiency. A mixed methods design with both quantitative and qualitative data collection and analyses phases was employed. Quantitative data analyses indicated no significant differences in standardized measures of English achievement, although significant differences were found in other measures, including measures of oral language acquisition in English, Spanish-reading ability, students' attitudes, and perceived levels of proficiency in English and Spanish. Qualitative data analysis indicated that the students in two-way bilingual education programs were more likely to express positive attitudes towards bilingualism. Based on the mixed data, it is concluded that despite some similarity in the effects, each of the bilingual programs also has unique effects. Policy decisions should be made on the basis of relative importance, value, and the costs of these unique advantages and disadvantages. (Contains 7 tables.)