In this study, an experiment was conducted to compare the three teaching methods: the inferring word-meaning method, the morpheme-based teaching method, and the word-based teaching method. The results showed that students taught by the inferring word-meaning method outperformed both the students taught by the morpheme-based method and the word-based method in terms of word retention. The possible reason is that the inferring word-meaning section enabled students to pay attention to and spend more time on each word. On the other hand, the disadvantage of the inferring word-meaning method is that it might not be an efficient way to teach a large amount of new vocabulary because it requires sufficient time for students to get involved with the context and the discussion of guessing words. Another drawback is that textbooks are seldom designed for the purpose of inferring word-meaning from context, which means that the teacher has to spend a substantial amount of time editing the dialogues or reading material from the textbook in order to give students more information to help them guess the meaning of each target word. For the morpheme-based teaching method and the word-based teaching method, there were no statistically significant differences observed. Students from both test groups achieved the same percentage of word retention.