School of Languages and Linguistics - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemIntroducing EFL speaking tests into a Japanese senior high school entrance examination( 2004)This thesis investigates the feasibility of introducing speaking tests into the existing English test of the senior high school entrance examination in Japan by employing Messick's (1989) validity framework. The study demostrates that validity investigations need to include not only psychometric analysis but also a consideration of the competing values of stakeholders. The teaching guidelines for English issued by the Japanese Ministry of Education (1998) state that speaking is one of the most important skills for junior high school students. An entry decision to senior high school is based on both school-based assessment implemented by junior high school teachers and the existing external standardized English test. Despite the emphasis on the development of speaking skills, the existing English test does not include the assessment of speaking skills. There is a clear discrepancy between the aims of the guidelines and the skills tested in the entrance examination. A way to bridge this gap could be to introduce speaking tests into the English test of senior high school entrance examination, a step that would necessitate considering the validity of such test. The major issue of test validity relates to the meaning, relevance and utility of test scores as well as the value implications of test scores and the social consequences of test use (e.g. Messick, 1989; Bachman, 1990; McNamara, 2001). A questionnaire survey of teachers and students, and interviews with government officials and academics responsible for the test, were used to ascertain stakeholders' attitudes towards the introduction of speaking tests and their view of possible washback effects on the teaching or English (Study I). In order to respond concerns expressed by stakeholders in Study 1 about the reliability, validity and practicality of tests assessing oral skills, a possible oral skills component in the existing test was developed, and trialled and test scores were analysed, focusing on the practicality of the administration and psychometric adequacy of investigating student ability, raters, tasks and items via Rasch measurement (Study 2). Study 1 revealed that while most stakeholders were positive about the introduction of speaking tests, two stakeholders groups—the Education Board and senior high school teachers were not. The former, the test developers, took a conservative approach in wanting to maintain the status quo, and the latter, the test administrators, were resistant to the introduction of speaking tests for complex reasons, both internal and external. The views held by these two stakeholder groups are major obstacles to introducing such a test. Preliminary findings from Study 2 showed that the speaking tests developed were psychometrically adequate to measure junior high school students' oral skills. This study demonstrates that careful consideration needs to be given to the possible psychological fear aroused in stakeholders by the changes that would occur if speaking tests were included in the senior high school entrance examination. These changes can also challenge the values that underpin the existing educational system, both at the institutional and individual level, with different groups of stakeholders holding competing values. Clearly, taking these values into account is important in investigating the feasibility of introducing speaking tests into the entrance examinations in that any future component oral skills component in the entrance examination challenge the existing examination embodying ideological, political, and educational values - for as Messick argues, validity needs to be viewed in terms that go beyond psychometric rigour. The thesis concludes with a discussion on the implications for validity theory and the development of language assessment policy.