| This thesis aims to build a common reference scale (CRS) of syntactical analysis ability (SAA) for National Matriculation English Examination (NMEE). Up to now, the empirical study in China and foreign countries focuses on two aspects:"can do" statements and common standards of ability levels. The English proficiency standards in China lag behind those in the western countries. On the one hand, the current "can do" statements at home and abroad are mainly based on the Communicative Language Ability (CLA), which places extra emphasis on the composition of language ability and lacks a quantitative analysis. On the other hand, there are many descriptive problems:such as different descriptive indexes, vague definition, unclear level and so on, which brings difficulties for National Matriculation English Examination. In order to make up for the shortcomings of the empirical study, this thesis takes grammar items of NMEE volumes as research subject. And this study attempts to answer the following two questions:(1) what are the can-do statements of SAA of NMEE?(2) How many dimensionalities can the SAA be divided into?Theoretically based on the CLA and Cognitive Processing Ability of Discourse Information, this paper can make up for the shortage of failing to describe language ability from the perspective of "quantity". It is mainly divided into two parts:the collection and analysis of descriptors. The former is to explore what the SAA is and its composition, and finally extract its descriptors. The latter is aimed to divide the dimensionalities by some data analysis tools, such as SPSS. This research is divided into two phases:the construction of the scale and the validation of the scale. And to solve the above questions, the following experiment is designed: first of all, this study takes the grammar items of NMEE volumes as research subject, and then these items are classified through frequency statistics. Additionally, syntactical descriptors are collected to build the primary syntactical descriptor repertoire. Secondly, the author invites ten experts to judge the collection results and the syntactical descriptors in order to make sure that all the above data is reliable. The results show that the raters’consistency is very high. Finally, the questionnaire which consists of syntactical descriptors is then made to collect data. The dimensionalities of SAA are divided by using factor analysis.The results are as follows:1) SAA can be described from38perspectives.2) SAA can be divided into ten dimensionalities. Building a CRS of SAA for NMEE can make up for the shortage of the current syllabus and improve the quality of item-writing autonomy. Given the subjective and objective conditions, the research fails to verify the validity of the initial reference scale and is also unable to divide the level of it, so there still remain a few limitations. This study is expected to inspire others in the similar researches. |
