英语论文网

The ratereffects thus need to be investigated (Myford&Wolfe, 2004a). The study on rater effectsin analytic scoring in large-scale writing tests is not so sufficient.There are lots of English tests in our university every year, including Englishmajored students’ tests and non-English majored ones, either in small scale or largescale. Every test should serve as not only a tool to check the teaching and learning, butalso feedback of each student’s strength and weakness.Especially, since 2007, Hunan University has launched the CEPT (College EnglishPlacement Test) to place the freshmen to different level classes. It is a scientific way toteach English according to different competence levels of students. And it is hoped thatone day the score of the students will be marked in an analytic way, so that thefollowing teaching will be benefited from the scoring information of everyone.Moreover, the analytic score may provide clearer evidence for each student’s finalscore, which is more retrievable in that someone doubts the results.Although it seems difficult to rate so many students’ writing in analyt

Chapter 2 Literature Review

This chapter will introduce the theoretical background of the study and theprevious studies related to this field.

2.1 Introduction to holistic scoring and analytic scoring
Holistic scoring method, which can also be called global or impressionisticscoring, has been widely used in various large scale writing tests both at home andabroad, such as domestic tests like the College Entrance Exam, the College EnglishTest, the Test for English Majors, and international tests like the Test of English as aForeign Language (TOEFL), Graduate Record Examinations (GRE), and GraduateManagement Admission Test (GMAT).For holistic scoring scales, elaborate score descriptors are usually developed forseveral score steps, and the writing ability of a writer is usually represented by a single,overall holistic rating result, which is easy to deal with and time-saving. However,there is one obvious drawback of holistic scoring that it is unable to reflect examinees’specific weaknesses and strengths in writing (Wiegle, 2002). Especially for thosesecond language learners who are still developing their writing skills and who are thuslikely to show uneven profiles across different micro-skills of writing, thisdisadvantage can be especially true, because for language learners with such irregularpatterns of abilities across different aspects of writing skills, the holistic scoringresults do little help in reflecting their problems and improving writing abilityeffectively.

Chapter 3 Research Methodology and Design.......... 24-28
    3.1 Participants......... 24-25
    3.2 Materials .........25-26
        3.2.1 Writing scripts......... 25
        3.2.2 Rating scales......... 25-26
    3.3 Rating experiment design .........26-27
    3.4 Data collection for FACETS......... 27-28
Chapter 4 Data Analyses and Results......... 28-49
    4.1 Primary multi-faceted Rasch analysi......... 28-36
    4.2 Rater-category bias analyses......... 36-43
    4.3 Rater-writer bias analyses .........43-47
  &nbs本论文由英语论文网提供整理，提供论文代写，英语论文代写，代写论文，代写英语论文，代写留学生论文，代写英文论文，留学生论文代写相关核心关键词搜索。