A Study And Case Analysis Of The Application Of Modern Educational Measurement Theory In Criterion-referenced Language Testing

Higher Education Quality Assurance is a key issue in the theory of higher education. This includes the multi-evaluation of the quality of higher education, paper design, method of testing, score description-analysis, and the essential role of examination in teaching as a main assessment-evaluation for quality education.This study which used the Criterion-referenced Language Testing (CRLT) method aimed to examine a syllabus based language teaching. It explained the principles of CRLT and its comparison to the principles of Norm-Referenced Test (NRT) to explicate the similarities and differences between the two tests; a distinction on NRT and CRLT scores that described and presented the statistical data; an analysis on the result of Test English as a Major Grade 4 (TEM4) taken by students in Jimei University to improve CRLT; a CRLT assessment principle of the quality of Higher Education; the use of A, B,Φvalues, Item Response Theory (IRT) to guide CRLT items test design and to protect the quality assurance of Higher education. The dissertation comprises four parts.Chapter one gave a brief introduction of the main reasons for choosing this topic, literature review, method of research and identification of several concept definitions. Chapter two described in details the principles of CRLT based on the traditional theory of educational measurement. This part also described the interpretation and comparison of the contents, the basic ideas and methods of CRTs and NRTs to explicate the similarities and differences between the two tests, while to put forward another method to explain the scores. The CRTs assessment principles of the quality of Higher Education, which are the theoretical foundation of the next chapter, are presents at the end of the chapter.The third chapter introduced and distinguished the test scores between NRTs and CRTs by describing and explaining the statistical values, then introduced how to use A, B andΦvalues and IRT to guide the test design. It also offered a case analysis as an example. Chapter Four was an empirical analysis by means of analyzing the TEM4 results about sophomore in Jimei University to further prove that A, B,Φvalues and a, b, c values in IRT are better guidance to improve CRLT items designing, then concluded that we must improve the quality of designing test paper in order to identify the different abilities of students, so as to protect the Higher Education Quality Assurance.The epilogue summarized the innovation of the research and the main conclusions. It also made a simple explanation of some questions needed to continue in study.