Web-based language learning materials have been among the preferred materials to be used by English language instructors since they are plentiful, straightforwardly accessed, user-friendly, and most importantly, are provided for free. This study investigates the effects of learning the simple perfect tense translation in three different English translation classes which applied three different modes of teaching, which were the traditional face-to-face method, integrative method, and web-based learning method. Furthermore, it is aimed to examine the effects of proficiency level in each mode of teaching and also to determine the best method for learning translation. The research is a quantitative study and the design is a pre-and post-test quasi-experimental design. The pretest-posttest study involved 87, third-year undergraduate students, that is 52 advanced and 35 intermediate students. All groups were exposed to one mode of teaching only for 4 weeks. The results indicated that there is a significant difference in both tests for all modes used. Keywords: Web-based language learning, translation proficiency, traditional method, integrative method, web-based learning method;