摘 要

浙江、上海等省市分别从2016、2017年在高考英语中增加了概要写作任务；英语专业四、八级考试从2016年起也开始增设了概要写作环节，但国内基于实证研究且经过效度检验的概要写作评分量表却十分罕见。考虑到同伴互评在英语写作测评中的重要作用，非常有必要开发和研制适用于同伴互评的概要写作评分量表，服务于英语写作教学。本文首先通过回顾各类权威英语写作测试评分量表，并结合4位高校英语教师对概要写作自由撰写的评语，尝试构建概要写作测试评分量表，再根据多元线性回归分析结果确定各评分维度的分值分配。其次，本研究采用多层面Rasch模型及独立样本t检验进行数据分析，从而对评分量表在实际评分中的使用情况进行效度验证，并探究学生英语水平对同伴互评的影响。结果表明，评分量表具有较好的区分度，各维度分值分布较为合理，不存在趋中现象。但3个评分维度均有个别分值使用不够合理，不能反映学生的真实写作水平。评分员自身英语水平对评分的质量及结果影响不大，同伴互评值得信赖。总体而言，本研究研制的同伴互评评分量表具有较好的效度和一定的推广价值。

关键词：同伴互评；概要写作；评分量表；效度验证

Contents

1. Introduction…………………………………………………………………...….1

2. Literature review ……...……………...………………………………….............2

3. Research method……………………………………........................................…5

3.1. Research questions…………................................................................……..…5

3.2. Subject.................................................................................................................5

3.3. Experimental materials………………………………………………………....5

3.4. Collection and analysis of data……………………….........…………………...6

3.4.1. Construction stage of rating scale……………….................………………....6

3.4.2. Validation stage…………………………........................…………………….7

4. Results……………………………………………………........……………….....9

4.1. Construction of rating scale……………......……………………..…...………...9

4.1.1. Construction of initial rating scale…………………………..........……….......9

4.1.2. Scoring distribution of each dimension of rating scale…………........…….....12

4.2. Validation of rating scale ……......……......………………..……..……….…....13

4.2.1. Overall analysis…………………………………………………...........…......14

4.2.2. Validation of rating scale from Micro-view…………..........................…........15

4.2.2.1. Overall analysis of each level of rating scale………...........................…......15

4.2.2.2. Analysis of each level of rating scale………………………………....….....16

4.2.2.3. Bias interaction between rater and rating scale….………………….............18

4.2.3.4. Independent samples test………………………………….....................…..19

5.Discussion………………………….…...…………………………..…….….…....20

6.Conclusion………………………………………………………..………...……..22

Bibliography………………………………………………………………………....24

Appendix………………………………………………………………………..…...31.

A Study of Rating Scale for Summary Writing Tasks

in Peer Assessment in Classroom-based Context

1.Introduction

With the combination of reading and writing, summary writing is often regarded as an important item in English writing tests, which could effectively improve undergraduates’ reading and writing competence. It is an important academic writing capacity that should be cultivated and trained in higher education (Friend, 2001; Li, 2014; Radmacher amp; Latosi-Swain, 1995). There has been, in recent years, increasing importance attached to summary writing domestically. Zhejiang, Shanghai and other provinces have added summary writing in National Matriculation English Test (NMET) from 2016 and 2017 respectively. Since 2016, TEM-4 and TEM-8 also started to test students’ capacity of summary writing. At present, most studies on summary writing focus on the construction of writing tasks(Yu, 2007;Li Jiuliang, 2014), writing strategies (Hood, 2008; Yu, 2008; McDonough, 2014) and the influence of source text on writing tasks (Hidi amp; Anderson, 1986; Yang amp; Shi, 2003; Yu, 2009). The research on summary writing assessment, however, especially its rating scale is still very scarce. As a performance test judged by raters, the formulation and selection of rating scale for writing tests is very important (Chen Jianlin, 2016). Although summary writing has become one of the important items in domestic writing tests, there has been shortage of rating scales based on empirical research to evaluate the quality of summary writing. In addition, peer assessment could not only enhance the writing motivation of students effectively, but also make students know others’ opinions and ideas on their works, which could assist the students to read and revise their works critically from the perspective of readers and make them gain confidence from writing(Mitten, 1980). Therefore, in classroom-based context, peer assessment should be introduced positively in summary writing instructions as the beneficial supplement of teachers’ evaluation. However, due to the scarcity of rating scale for summary writing in peer assessment, it greatly hampers the effective implementation of peer assessment in teaching summary writing. This paper intends to construct a scientific, standard and easy-to-use rating scale for summary writing in peer assessment based on quantitative and qualitative research and conduct validation for it, thus providing important reference for teaching and evaluation of summary writing in classroom-based context.

2.Literature review