A Preliminary Study On The Development Of A Rating Scale For ELT Certificate Lesson Observation And Report Task

The purpose of the present study is to develop a rating scale for the lesson observation and report—a task of English Language Teaching Certificate(ELT Certificate)test developed by China Language Assessment(CLA).In this task,test-takers are required to observe a video excerpt from a real English lesson and then to give an oral report in English.To score their performance on this task,a reliable rating scale is needed.This study sought to develop such a scale by investigating its score levels,scale dimensions and level descriptors.The study utilized focus group discussion and feedback content analysis methods.Performance samples from 39 English major students and 2 novice high school English teachers were collected.Two English teaching experts from universities,one high school English teaching researcher and one experienced high school English teacher participated in the focus group discussion.They listened to 15 samples and discussed the score levels and the scale dimensions.Based on the discussion,they were required to rate 8 representative samples and write comments about their reasons for assigning a certain score.From the qualitative analysis of the transcribed focus group discussion and the written comments,it was found that the raters used four score levels to distinguish the performance samples,which were rated from both oral English proficiency and the ability to understand and analyze classroom teaching.Within each dimension,various sub-dimensions were attended to.By grouping and integrating the evaluative statements coded from the comments,the researcher created descriptors for the sub-dimensions at different score levels.An analytic rating scale was developed,to which Level 0 and its descriptors were added.The rating scale provides sufficient diagnostic information for test-takers as they can clearly know their strengths and weaknesses from their scores and the corresponding descriptors.This study also provides further evidence that feedback content analysis is an effective method for rating scale development.