Introduction: The study presented here explored the relationship between computer and Internet use and the performance on standardized tests by secondary school students with visual impairments. Methods: With data retrieved from the first three waves (2001-05) of the National Longitudinal Transition Study-2, the correlational study focused on secondary school students with a primary disability of visual impairment who were capable of taking standardized tests. The unweighted sample sizes ranged from 210 to 280. The relationships between using a computer for homework and for the Internet and students' performance on standardized tests on synonyms and antonyms, comprehension of passages, calculation, applied problems, science, and social studies were investigated using multiple regression analyses, with other variables held constant, including gender, severity of vision loss, household income, multiple disability status, and race or ethnicity. Results: When gender, the severity of vision loss, household income, multiple disability status, and race or ethnicity were held constant, using a computer for the Internet had a significant positive relationship with students' performances on the passage comprehension test (p = 0.034), calculation test (p = 0.027), and science test (p = 0.000). However, using a computer for homework had no significant relationship with students' performances in any of the six tests that were examined. Discussion: Students who used a computer for the Internet scored significantly higher than did those who did not on the passage comprehension, calculation, and science tests. But students who used a computer for homework performed at the same level as those who did not on all six tests that were examined. Implications for practitioners: Internet use by secondary school students with visual impairments may be beneficial to the students' performance on passage comprehension, calculation, and science tests and therefore should be encouraged. Special assistance for students from low-income families, whose rate of Internet use was significantly lower than those from high-income families, should be considered. (Contains 3 tables.)