A Validation Study On Descriptor Level Of The Self-assessment Scale For Interpreting Ability Of China's Standards Of English Language Ability

Against the backdrop of establishing the foreign language ability evaluation system,China's Standards of English Language Ability(CSE)was released in 2018.CSE consists of different scales with different language abilities,including interpreting scales.Self-Assessment Scale for Interpreting Ability(SASIA)provides a tool for interpreting teaching,learning and assessing.As a newly-developed language proficiency scale,its validation is necessary before its use.In addition,some students were asked to assess themselves based on the SASIA.The survey results revealed that students felt difficult to determine which level they were in.Besides,through the literature review,researchers pointed out that there were some problems about the descriptor level of the scales.For example,the levels are interdependent or the levels cannot clearly discriminate learners from different proficiency levels.In this context,this study aims to investigate the appropriateness of descriptor level of the SASIA of the CSE.Under the research background mentioned above,the study aims to answer three research questions:(1)How well can the items of the SASIA differentiate different respondents?(2)How well can the items of the SASIA be differentiated by groups of respondents?(3)How well the descriptor level is classified in the SASIA?In order to answer these three questions,methods of questionnaire and interview are adopted in the study.Participants for questionnaire are students from BTI grade 1to grade 4,MTI grade 1 to grade 2 as well as interpreting experts.Valid questionnaires are 287.There are 4 participants for interview from BTI grade 2 and grade 4,MTI grade 2 and interpreting experts respectively.Rasch Model is used for data analysis.Research results are as follows.Firstly,person separation index is 5.01,which indicates that the items of the SASIA can well differentiate different respondents,and the respondents in this study can be classified into five different levels based on their interpreting ability.Secondly,item separation index is 12.56,which indicates that the items of the SASIA can well be differentiated by differentrespondents,and the items of the scale can be classified into twelve levels based on their difficulty.Thirdly,the descriptor level in the SASIA is not classified well enough.It shows in two aspects:(1)the difficulty level of descriptors is 12 rather than 5 in the original expectation and(2)the difficulty of some descriptors is not in the prescribed difficulty level.In addition,the person reliability index is 0.96 and item reliability index is 0.99,which proved that the reliability and validity of the SASIA are relatively high.Based on the results from Rasch Model analysis,interview is conducted about some descriptors of the scale.Qualitative data from the interview is used for content analysis including transcript,coding and summarizing.The results show that the weakness of descriptors may involve following aspects: the clarity of descriptor,the design of difficulty,the limitation of objective conditions and relevance with the interpreting ability.The study validates the appropriateness of descriptor level of the SASIA of the CSE.It is expected to provide a reference for the revision and adjustment of the scale,which is beneficial for its development as well as application.In addition,as part of validity,the study also has positive effects on interpreting teaching,learning and testing.