Developing a Measurement of Sentence Complexity and Its Application in Japanese Reading Comprehension

Abstract

This paper aimed to determine the complexity of Japanese sentences that are difficult for learners to understand by considering the factors that contribute to their complexity. Based on these factors, an index was developed to evaluate the complexity of sentences. The sentence complexity index was divided into two categories comprising a syntactic complexity index and a contextual complexity index, each of which can be summarized as follows. The syntactic complexity index consists of five indicators including (1) number of clauses, (2) number of dependent clauses, (3) number of phrases called Bunsetsu, (4) number of modifying components, and (5) number of components inside the predicate. The contextual complexity index consists of two indices including (1) number of indicatives and (2) number of abbreviated phrases. By evaluating Japanese sentences using these indices, the author believes it is possible to distinguish which sentences are difficult for learners to understand. Finally, a post-reading translation and retelling activity was suggested to assess the accuracy of the reading as well as the extent to which content that had already been read was retained in memory. The sentence complexity index presented in this paper is only a trial draft. It is believed that further improvement will be required in the future by utilizing this draft and modifying it as necessary.