Abstract
In recent English language teaching in Japan, which tends to emphasize practicality, there have been claims that literature is of no use. In accordance with this, textbooks seem to have few literary texts. There is, however, no empirical evidence justifying this situation. In order to consider whether this current situation is appropriate or not, models of literary English competence and general English competence, and the differences between them, need to be described. As a first step towards this, in the present study, multiple regression analyses were conducted by regarding the scores of a literary English competence test and the Center Test as the dependent variables, and those of five tests (English vocabulary level test, English productive vocabulary test, English grammar test, contemporary Japa- nese test, and questionnaire for motivation to study English) as independent variables. The results showed a significant difference between the R^2 values. Discussions based on R^2s and correlation coef- ficients between variables are presented as one piece of empirical evidence for the consideration of the utility of literature in ELT from now on.
