A LIBRARY BASED REVIEW OF THE PRINCIPLES OF SPEAKING TEST CONSTRUCTION

Abstract

The purpose of this research is to investigate the principals involved in the construction of speaking test. The focus of the research is to: review test takers’ characteristics, mode of the test, task type, gender of the raters, holistic versus analytical assessment, length of the raters’ working hours and lastly experienced versus novice raters. This research is qualitative as well as descriptive in nature. The data was collected from international journals: Springer, Francis and Taylor, Elsevier and relevant journals of speaking skill, Google and Google Scholar. The findings of the research reveal that attitude, anxiety, motivation and topic familiarity found over all positive relationship with oral assessment. In addition, without considering test takers’ individual characteristics, a test will not measure what it supposes to measure. Automated test deems to offer a valid measure of oral ability in English, and it is an apt method for the assessment of oral skills. Although in one study, the test takers felt anxiety with automated tests. Paired and group performed better than individual test taker in the interview. The gender of the rater did not affect test taker’s performance whether it was the same gender or opposite gender, yet, the test takers showed their intent for female raters. As for holistic and analytical, they influence the performance of test takers. Finally, it was observed that experienced raters were very severe in their rating than the novice raters. This study will be effective for class room assessment, as well as for those who are working in the field of testing and evaluation in second language