An Argument-based Validation Study Of CSE Self-assessment Scale For Written Expression Ability

In 2018,China’s Standards of English Language Ability(CSE)was officially published.As an English language proficiency standard,CSE defines levels of English users’ language ability and describes the characteristics of these levels specifically in the Chinese context.Self-Assessment Scale for Written Expression Ability(SASWA),a vital constituent of CSE,serves as a functional tool for English writing teaching,learning,and assessing.In order to promote the application of SASWA,it is of great significance to examine whether the interpretations and uses of self-assessment results are plausible and accurate.However,validation of the SASWA remains a striking research lacuna to be filled.Drawing on Kane’s validity framework of interpretation/use argument(IUA),the present study specified and supported two vital claims of SASWA score interpretation,namely the scoring inference and the explanation inference.Correspondingly,two research questions were proposed:(1)to what extent can the SASWA help students to self-assess their written expression ability accurately?(2)to what extent are the SASWA results consistent across different students who are at the same writing ability level?In response to the questions,a mixed-methods research design combining questionnaires and interviews was adopted in the study.A total of 347 undergraduates from Shanghai International Studies University(SISU)and Xianda College of Economics and Humanities,Shanghai International Studies University(XDSISU)were involved in the questionnaire survey.Rasch analysis was employed to analyze the questionnaire data,and based on the results of the analysis,16 volunteers were invited to participate in the semi-structured interviews.Overall,the results lent support to the validity of the SASWA,whereas some contradictory evidence did exist: the SASWA can well differentiate the written expression ability of the students at different proficiency levels;most of the descriptors exhibited difficulty in line with the expectation;the five-point category functioned as intended;all the descriptors functioned consistently across gender groups;a few SASWA descriptors were identified as showing unexpected difficulty levels due to three causes,namely relevance to writing that students have not experienced,a reference to a stressful situation,and a dearth of definiteness,and corresponding suggestions on how to address these issues were provided.The present study exploring the validity of the SASWA gives suggestions for both revision and application of the SASWA,expands the scope of the application of the argument-based approach to validation,as well as provides new perspectives on how self-assessment can be used as an effective instrument to promote English writing learning.