学位论文 > 优秀研究生学位论文题录展示

英语专业四级听写测试评分的新方法

作　者: 贺娟娟
导　师: 聂建中
学　校: 山西大学
专　业: 外国语言学及应用语言学
关键词: 部分得分评分制听写测试英语专业四级考试相关系数
分类号: H319
类　型: 硕士论文
年　份: 2011年
下　载: 4次
引　用: 0次
阅　读: 论文下载

内容摘要

大学英语专四听写测试的评分采用的是比较传统的数错扣分法。数错扣分法是负分法,是将错误分为主、次两个水平,并给予不同的赋值,然后从总分中扣除。这种方法存在一些问题。首先,它重在语言离散点和产出的精确性,忽略了语篇意义的重要性。虽然听写评分法客观,听写的信度较高,但是效度较低,与考试的宗旨相矛盾。其次,数错扣分法本质上类似于二级评分,不能完全预见学生在测试中复杂的产出情况。再者,负分法的评分制对于教学的影响是消极的,负面的。本文对短文听写测试的评估提出了一种新的试验方法——部分得分评分制。部分得分评分制既采用部分得分的新评分方法(确定三个能力标准),又部分保留了原来的评分制(以语意群为评分单位)。在评分实施中,我们确立了语篇意义的完整性,语法的准确性和正确的书写方法等三个能力标准。评分单位仍为语意群,所有的语意群都使用统一的多级量表。实验数据来自2010年4月20日的专业四级模拟考试。考生是山西农业大学外语学院专业二年级的学生,共28人。这次模拟考试与TEM4测试在内容、题型、题量、权重等方面都一致,短文听写是其中一项。短文长152个单词,8个句子,15个语意群(平均长10个单词)。考试结束,将全部试卷复印,将一套试卷由评卷小组按照TEM4听写评分法评阅,得出六组不同的分数(旧分数)。另一套再由这组评分员按照新的试验评分法(部分得分评分制)进行评阅,得出六组新的分数。为了检验评分量表的完整性以及实验评分制能否体现《考试大纲》中的能力目标等问题,我们计算了新评分结果与测试总分及其其他题项的相关系数(SPSS 13.0版)。从实验评分及数据分析来看,新的评分制能够较准确地测量考生的听写能力,但还需要进一步的修改、验证,如级点描述需要完善等等。比较而言,数错扣分法的评分方法有它的长处：简化了许多复杂的问题,操作起来简便、快捷,省时、省钱、省人力,适当的培训就能使评分员信度达到良好的水平。而试验评分制的优点对测试的公平,有效这两方面更重要,能很好地保证测试的效度并且更好地体现了考试大纲中能力标准的多重性。考生的听写平均成绩提高了,这也意味着专四总成绩的提高。再者,新的评分制对测量能力较弱的学生的听写水平较少偏差。最后,实验评分制是正分制,对英语教学以及考前准备所起的作用也是积极的。

全文目录

Acknowledgements  5-6
Abstract  6-8
摘要  8-10
CONTENTS  10-13
目录  13-16
List of figures  16-17
List of tables  17-18
CHAPTER ONE INTRODUCTION  18-21
  1.1 Research background  18-19
  1.2 Objective of this study  19
  1.3 The structure of the paper  19-21
CHAPTER TWO LITERATURE REVIEW  21-41
  2.1 Early claims made about dictation  21-24
  2.2 The development of dictation in TEM4  24-27
    2.2.1 The forms of dictation in TEM4  25
    2.2.2 The significance of dictation in TEM4  25-27
  2.3 Reliability in scoring procedure  27-30
    2.3.1 Scoring procedure in language testing  27-28
    2.3.2 Reliability and validity of test  28-30
      2.3.2.1 Rater inconsistency/reliability  28-29
      2.3.2.2 Validity of scoring  29-30
  2.4 Procedures for scoring the responses  30-38
    2.4.1 The rating scales of dictation  30-31
    2.4.2 The present rating scales applied in dictation in TEM4  31-35
      2.4.2.1 Exact word scoring  32-33
      2.4.2.2 Chunk scoring  33-34
      2.4.2.3 TEM4 scoring method  34-35
      2.4.2.4 The comprehensive feature of dictation test  35
    2.4.3 The advantages of traditional scoring method  35-36
    2.4.4 The potential limitations existed in negative scoring  36-38
      2.4.4.1 Implicit mistakes in the classification and the corresponding distribution of scores  36-37
      2.4.4.2 Scoring scale was the lack of predictability  37-38
  2.5 The backwash of teaching and testing preparation  38-39
  2.6 A brief review of partial credit scoring  39-41
    2.6.1 The origin and development of the partial credit scoring  39
    2.6.2 Partial-Credit scoring methods applied in dictation  39-41
CHAPTER THREE RESEARCH METHODOLOGY  41-46
  3.1 Overview of the present study of scoring methods adopted in dictation  41
  3.2 Participant  41-42
    3.2.1 Test takers  41-42
    3.2.2 Raters  42
  3.3 Testing protocols  42-43
  3.4 Testing materials  43-44
  3.5 Testing method  44
  3.6 Data collection procedure  44-46
CHAPTER FOUR RESULTS AND DISCUSSION  46-60
  4.1 Quantitative analyses of scores of dictation item  46-50
  4.2 Comparison of scores came from different raters according to the quantitative analysis  50-51
  4.3 The qualitative judgments  51-53
    4.3.1 Rater consistency  51-52
    4.3.2 Some information of test takers  52-53
  4.4 The result of qualitative analysis  53-54
  4.5 The verification of the integrity of the new experimental method  54-60
    4.5.1 Reliability  54-55
    4.5.2 Statistics description of test scores by the six raters use of two scoring methods  55-56
    4.5.3 The verification of the integrity of partial credit scoring method  56-60
CHAPTER FIVE CONCLUSION  60-64
  5.1 Summary and implications of the major studies  60-62
  5.2 Limitation of the present studies  62
  5.3 Some advantages about partial credit scoring method  62-63
  5.4 Some directions for future study  63-64
BIBLOGAPHY  64-73
APPENDIX1  73-74
APPENDIX2  74-76
PUBLICATION  76-78

英语专业四级听写测试评分的新方法

内容摘要

全文目录

相似论文