In testing educational achievement, it sometimes seems that testing is part of education. More frequently, however, education seems subordinate to testing. In particular, high-stakes tests, those that can have a great influence on individuals and society, can reach far beyond the mere evaluation of educational content and can provide future guidance for both students and teachers.
Currently, two of the most influential tests used for evaluating the communication abilities of Korean students are TOPIK (Test of Proficiency in Korean) and KLPT (Korean Language Proficiency Test). The two tests have the same objective: evaluating a student’s fluency in Korean. These tests are used as the basis for evaluating an applicant's college admission qualifications, are required for graduation and sometimes for employment. The tests are also used as a tool when teaching Korean and in assessing people's Korean proficiency. The handbook for both tests indicates that the test results are used for college admission. However, some educators feel that the tests are not adequate for evaluating an applicant's Korean learning ability. Increasingly more schools are developing their own independent admission tests. As high-stake tests, TOPIK, KLPT and college entrance exams have a great influence on the direction of Korean learning; they can be critical tools in setting a course for the applicant's life. Therefore, the tests need to be closely examined as evaluation tools.
Among the four sensory aspects of communication, listening is the most widely used in daily life; people first learn a language through hearing it spoken. Thus, an evaluation tool for listening comprehension needs to be improved for more efficient learning. To evaluate listening comprehension the most effectively, it is important to evaluate what needs to be measured in a test; that is, to determine the validity of a test.
In general, an evaluation tool should be valid, reliable and practical. Validity is the most significant factor that determines the quality of a test. Even if a test is reliable and practical the results cannot be trusted without validity of the test. The results of an invalid test will have a negative impact on both the applicants and the institutions that uses it.
Therefore, it is necessary to examine whether or not high-stakes Korean tests are based on accurate expectations and whether or not they measure adequate items. This study identifies factors that affect the validity of a test. TOPIK, KLPT and a Korean ability test developed by P University will be analyzed. Based on the identified factors, the study aims to suggest how to ensure the validity of high-stakes Korean tests.
The is organized as follows.
Chapter Ⅰ illustrates the research objectives and needs, describes the ’s scope and methodology and examines the results of previous studies.
Chapter Ⅱ provides the theoretical background. It adopts Weir's concepts of the validity of a test, as provided in Language Testing and Validation (2005), as the theoretical basis. The examines the current condition of listening comprehension tests, including TOPIK, KLPT and a Korean ability test developed by P University.
Chapter Ⅲ analyzes the validity of each test in terms of its theoretical background and context, but also discusses how to design a valid test. The validity of the theoretical background is based on the language capability model, as suggested by Bachman & Palmer (1990). The validity of the context is divided into seven items for analysis: instruction, test objective, form of response, the order of evaluation items, materials, function and dialogist variables (speed, gender). Based on the results, the study provides suggestions as to what should be considered when designing a high-stakes Korean test.
The study identifies problems in current Korean tests and provides suggestions to ensure the validity of a test. Validity is essential and fundamental to designing a robust language test.
，韩语毕业论文，韩语毕业论文 |
