A Study of Collocation Selection in Japanese Language Learners

Based on a Comparison between the DIC Method and the DIC-LP Method

Abstract

In this paper, I make a comparison between the method of using only a Japanese-English bilingual dictionary (hereinafter, the DIC method) and the method of using both a Japanese-English bilingual dictionary and a lexical profiling tool (hereinafter, the DIC-LP method), in a collocation test.

The results of the experiment reveal that the scores obtained with the DIC-LP method were significantly higher. The negative effects of the DIC method on the scores were due to: (1) insufficient appropriate translations, (2) insufficient expressions that are the answers or lead to the answers, (3) multiple translations with inadequate information to understand their differences, (4) an excessive number of translations, (5) incorrect analogy by the learners, and (6) not confirming answers with multiple dictionaries. In contrast, the DIC-LP method had positive effects on the scores as it allowed learners to (1) consider the meanings using the list of collocations and (2) confirm the presence or absence of collocations as well as the number of collocations. As a result, a significant difference was observed between the two methods. It was also revealed that an excessive number of example sentences and incorrect guesses at the use of postpositional particles were factors that created negative effects on the scores in the collocation test with the DIC method. On the other hand, incorrect guesses at the use of postpositional particles and not being able to narrow down the verb candidates using the dictionary were factors that created negative effects on the scores with the DIC-LP method.