摘要：This experimental study tries to compare the effectiveness of corpus-informed, contextualized and decontextualized vocabulary instruction in an EFL setting. The participants were 69 high school pre-intermediate language learners in a state school. During the 8-week experimentation process, one of the experimental groups was treated with corpus-informed instruction while the other one studied the same target vocabulary through decontextualized vocabulary activities. With the control group, on the other hand, sessions where the learners were exposed to the target words in meaningful reading contexts were carried out. Before these sessions, the corpus group was trained about the rationale behind concordances and how to use them. A multiple-choice vocabulary test which was composed of the target words was administered both before and after the treatments. Because of its convenience in experimental settings, an analysis of covariance (ANCOVA) was performed. The results of the statistical analyses indicated that the group that studied the target words through decontextualized learning activities scored a significantly higher mean in the post-test when compared to the other two groups. The group with corpus-informed instruction also displayed some progress, albeit statistically insignificant. Interestingly, the control group, which was treated with contextualized vocabulary activities, made the least progress.
