A STUDY OF ORAL COMMUNICATION SKILLS, TASK DIFFICULTIES AND STUDENTS’ SATISFACTION

To speak English as a foreign language, it is necessary to know a certain amount of grammar which includes vocabulary and pronunciation (Bygate, 1987; Ur, 1996; Harmer, 2001). Form-focused practice, however, limits students' choices of variation to talk. In addition, paying attention to accuracy sometimes discourages students from using the target language. As many of the Japanese students have anxiety about communicating orally in English, it is important to gain their confidence through speaking tasks. Having successful experiences can make students gain confidence (Locke, 1996), but the difficulty of achieving goals can also encourage students even though they fail to attain challenging goals (Mikami, 2017). To have a clear view of an appropriately difficult task level is a problem for language teachers. This study aims to examine both appropriate levels for English speaking tasks and students' satisfaction in a regular 4-skill English course where only written exams were carried out. In order to reduce students' unwillingness to communicate in English, goal-relevant speaking tasks were used as a pair work activity for freshmen in each class over a 15-week term in a Japanese university. Students in the test-group used slightly more difficult materials than the control-group. The students in the test group also invariably had different partners, whereas the ones in the control group usually had the same partners. To quantitatively investigate students' improvement of speaking skills, the number of words they uttered within speeches about a predefined topic was counted. Also, satisfaction surveys were conducted to measure the qualitative data. In conclusion, students in the test group found the tasks more difficult, but they produced richer speech and they were also more satisfied with the course. More importantly, the percentage of the students who gained more confidence in using English was higher in the test-group than the control-group.