The Usefulness Validation Of Tape-Mediated Oral Presentation Through Pictures In EFL Oral Testing

Oral presentation through pictures has been widely used in large-scale tests, especially abroad, but few researchers have explored this task type. The present research is inclined to probe into it both quantitatively and qualitatively with the aim to verify its usefulness. It is also hoped that teachers in similar settings may benefit from both the research findings and procedures.The thesis first reviews the basic theory and ways of validity and validation. The advantages and disadvantages of different oral types are then explored, and among them, interviews are advocated by many researchers and most of the performance-based oral testing research literature has focused on the traditional face-to-face interview or simulated (semi-direct) tape-mediated interview. However, oral presentation through pictures has its unique appeals, and if it is attached to the semi-direct channel, the reliability of the task will be improved.30 students who had just participated in NMET were randomly selected to fulfill two picture tasks and one topic task. All of their responses were recorded. SPSS 11.0 was used to analyze the reliability and validity of the test results. And what's different from former validation studies are that, the thesis discusses results from six aspects—reliability, validity, authenticity, interactiveness, impact and practicality—on the basis of Bachman and Palmer's theory concerning usefulness validation. As 4 trained raters were invited and tape-recorder media was adopted, the reliability of the experiment was comparatively high, and it implies that it is favorable to use more than 2 raters. Its validity was confirmed through factor analysis and correlation calculation. And the thesis provides a characteristic analysis about the TLU test task, which can be used as a basis for developing language test tasks. The present study then discusses the results of questionnaires and retrospective reports respectively. According to ANOVA analysis, the means of the responses to the picture task and the topic task are not significantly different, so the thesis compares the varied responses in accordance with the students' self-rated oral levels. It first explores the issue of task type and test-related anxiety in an attempt to discover whether or not certain tasks may cause more stress than others. Secondly, it deals with student perceptions of task validity. Then it moves on to their enjoyment of the testing experience and their preferred tasks.The semi-direct oral presentation through pictures is proved to be a feasible way to assess students' oral competence both from the quantitative and qualitative perspectives, so it is recommendable to adopt the task type in large-scale exams.