1. 研究目的与意义（文献综述包含参考文献）

1. introduction as a significant language output, writing accounts for the largest proportion of tem-4, which covers 25% of the total exam points. judging from the statistical result of the previous tem-4, the writing scores are generally unsatisfactory. even for a famous school like fudan university, the scores of the candidates are relatively low. this illustrates that english major students#8217; english writing level has yet to be further improved. zheng xiaoxia (2006) found that the weakness of students#8217; writing included a large number of grammatical errors, lack of diversion in sentence patterns, being short of consistency, unclear main idea and loss of logic. solving these problems will cost students a lot of time and energy with little effect. annett (1969) proposed that timely and effective feedback played an important role in improving writing level. in process approach, feedback can efficiently help to improve the quality of students writing because students can make several revisions with the guidance of feedback. however, there is no sufficient time for english major students to practice writing and they cannot get immediate feedback. the current situation in china is that a writing teacher teaches several writing classes. it is difficult to respond in time and effectively. with the development of technology, the automated writing evaluation system tries to solve this problem. modern network technology provides new platform called automated writing evaluation system for improve the situation. automated writing evaluation system make up for the teacher#8217;s limited energy that giving timely revision and feedback for students to modify their english composition. students can revise their english composition in accordance with the feedback given by automated writing evaluation for several times. according to college english curriculum requirements, english teaching in university should adopt a novel teaching pattern based on modern information technology especially network technique (sun xianhong, 2012). in previous studies, automated writing evaluation system can provide students with immediate feedback, individual report of language mistakes, which contains not only the accuracy of the use of word, but also the analysis of structure and content. therefore, it is urgent to seek for a novel and modern solution to improve situation of english major students#8217; writing. in this thesis, the author would explore how automated writing evaluation system performs in english writing. 1.1need for the study writing is both a cognitive and an affective activity. it is influenced by many factors, including writing anxiety, motivation, belief and attitude (mcleod, 1987). more scholars focused on emotional domain in writing process gradually, some of them found student#8217;s writing motivation played a vital role in their writing and had a positive impact on revising their compositions. motivation has been widely accepted by teachers and researchers as one of key influencing factors in esl learning. it is a psychology concept prompting people to devote to a long and tedious l2 learning process (dornyei, 1998). motivation not only improves the initiative of students to learn english, but also helps students overcome difficulties in order to achieve the purpose of learning. it is easier for students to master english knowledge and improve their writing skills with motivation (liu jingyun, 2005). therefore, motivation is the key factor in the success of english writing (pintrich schunk, 2002). apart from writing motivation, there is another factor influencing english writing-----self-efficacy. the notion of writing self-efficacy is a branch of self- efficacy proposed by bandura (1977). according to bandura (1993), it is a common belief for students that self-efficacy can regulate their learning, master academic activity and determine their aspirations, level of motivation and academic accomplishments. in some degree, self-efficacy can predict people#8217;s performance in some given task. mccarthy, meier rinderer (1985) defined self-evaluation as self-efficacy and it was closely related to writing performance, and they found different self-efficacy could result in different level of writing quality, students with high writing self-efficacy were regarded as good writers, experience less anxiety in time limited writing, and their essays were much longer and their writing grades affirmatively higher. in a word, higher writing self-efficacy indicates higher writing proficiency. hence, the function of writing self-efficacy cannot be ignored when it comes to english writing. in the past several decades, a lot of researchers have been carried out to investigate the effect of writing motivation or writing self-efficacy on writing performance, there are fewer researchers explore the combined effect of the three factors. automated writing evaluation system as a newly-developing thing for english teaching is controversial and affective factors like writing motivation and self-efficacy in english writing has been paid close attention in several years, it is necessary to explore their relationship and solve the problem of english teaching. 1.2 research purpose in this study, the author will probe into how english major#8217;s writing motivation and writing self-efficacy will affect their writing performance based on automated writing evaluation system. 2. literature review 2.1 theoretical foundations with self-efficacy theory getting increasing extensive attention, self-efficacy gradually reaches to other fields, such as the field of education. researchers at home use the concept in english teaching and learning, especially in english writing teaching. english writing self-efficacy was derived from bandura (1993)#8217;s ”self-efficacy”. mccarthy rinderer (1985) defined writing self-efficacy as the students#8217; self- evaluation which showed the belief in dealing with writing skills. other scholars (graham harris, 1989) pointed out that writing self-efficacy was students#8217; judgment about their competence of evaluating whether they could finish the writing assignment perfectly. every phrase of writing process was affected by individual#8217;s emotional factor, so it was naturally influenced by the self-efficacy mechanism. the belief in dealing with writing skills can be called writing skill self-efficacy. self-efficacy involves the usage of vocabulary, tense, punctuation and sentence patterns and so on. the definition ”the judgment of students#8217; confidence in accomplishing certain tasks” means the writing task self-efficacy. writing tasks mainly refer to the literary forms of writing such as expository, argumentation, description, e-mail, letter or a report. this thesis adopts shell's viewpoint which states that writing self-efficacy should be measured from two aspects, writing skills and writing tasks. dornyei introduced the term of motivation from social psychology into language teaching. without sufficient motivation, even individuals with remarkable abilities couldn#8217;t accomplish long-term goals. motivation was probably the most frequently used catch-all term for explaining the success or failure of any complex task (dornyei, 1998: 117-135). among various definitions of motivation, the one which was accepted by most researches was that motivation was the process whereby goal-directed activity is instigated and sustained (pintrich schunk, 2002). 2.2 related empirical studies 2.2.1 related empirical studies of automated writing evaluation system the development of computer-assisted writing technology changes the way of evaluation and feedback in writing practice. the automated writing evaluation system is developed to evaluate written works and provide grade, content, structure, expression and grammar plus personal feedback. the study on writing roadmap conducted by tang jinlan wu yian (2012） revealed that the system had positive influence on writing performance for student#8217;s writing scores increased more than those who hadn#8217;t written on the system. wang (2004) found that students who wrote better on knowledge forum, had better writing attitude and would like to discuss with peers. you qiliang zhan jianfen (2004) pointed out that both teachers and parents were satisfied with the system for its fairness on grading and accuracy on feedback. studies on juku automated writing evaluation system ( fu zhuqin, 2012, gu chenghua and wang li, 2012) displayed that students#8217; scores and writing abilities such as writing fluency had improved a lot after the constant writing through juku automated writing evaluation system. at the same time, he xuliang (2013) also found that student#8217;s learning autonomy and initiative were significantly improved because of the constant revision and instant scoring from juku automated writing evaluation system. many studies have affirmed the high validity of automated writing evaluation system as well as its positive influence on student#8217;s writing while researcher (cheville, 2004; ericsson, 2006) still questioned the effectiveness of automated writing evaluation system in their studies. 2.2.2 related empirical studies of writing self-efficacy researchers home and abroad have found that there is positive relationship between writing self-efficacy and writing proficiency (multon, k. d., brown s. d. lent. r.w. 1991; chularut, p. debacker, t. k, 2004; pajares, 2000; zheng xiaoxia, 2006; kang youjin ren xiaojuan, 2009). Меir еt. al. (1984) found that there was connection between writing self-efficacy and college students#8217; writing performance. pajares (1993) johnson (1996) carried out the research towards high school students and undergraduates respectively and the results showed that self-efficacy was closely related with students#8217; writing proficiency. besides, the study (zimmerman bandura, 1994) towards 95 freshmen showed that self-efficacy had direct and indirect influences on students#8217; writing achievement. that means higher writing self-efficacy can bring more enthusiasm and motivation in writing which promotes higher writing proficiency. in china, recently, more and more researchers focused on the influence that writing self-efficacy had on writing proficiency. researches (hu chunqin, li quan luo yan, 2006) about the relationship between writing self-efficacy and writing achievements showed that non-english major students who had higher self- efficacy achieved higher writing achievements and writing self-efficacy was positively related to writing achievements. zhang mingfang (2008) investigated that writing self-efficacy had influence on senior high school students writing teaching. liu yang liu feng (2007) pointed out that self- efficacy theory was useful in writing teaching which could offer good feedback for students. tang fang xu jinfin (2011) found that students#8217; writing self-efficacy and its two factors were positively correlated with their writing performance. zhao jiangkui (2011) pointed out that writing self-efficacy was one of the best predictors of writing proficiency. the above reviews home and abroad show that writing self-efficacy is positively correlated with writing proficiency and the predictor (writing self-efficacy) has significant predictive effects on the dependent variable (writing proficiency). therefore, students with high writing self-efficacy perform well in writing activities and teachers can improve students#8217; writing proficiency by increasing students#8217; writing self-efficacy. 2.2.3 related empirical studies of writing motivation different psychological schools, based on different theory, have different understandings on motivation. behavioral psychologists were the first to realize the impact of feedback to motivation. they claimed that external factors were applied to motivate and strengthen people#8217;s activity in order to achieve expected goals. since the 1980s, researchers studied on the relationship between motivation and language learning. bernaus gardner carried out a study teacher motivation strategies, student perceptions, student motivation, and english achievement, which showed that motivation had positive effects on english achievement, whereas attitudes toward the learning situation and language anxiety both negatively influenced english achievement references annett, j. (1969). feedback and human behaviour. harmondsworth: pengui books. attali, y., burstein, j. (2006). automated essay scoring with e‐raterreg;; v.2.0. journal of technology learning assessment, 4(2), 1#8211;21. bandura, a. (1977). self-efficacy: toward a unifying theory of behavioral change. psychological Ｒeview，84，191-215. bandura, a. (1997). self-efficacy theory: towards a unifying theory of behavioral change. psychological review, 84(2), 191-215. dornyei, z. (1998). motivation in second and foreign language learning. language teaching, 31, 117-135. eills, r. (1994). the study of second language acquisition. oxford: oxford university press. faigley, l., cherry, r. d., jollife, d. a. skinner, a. m. (1985). assessing writers' knoledge and process of composing. nowood, nj: ablex. gardner, r., mcintyre, p. a. (1993). student#8217;s contribution to second-language learning. part ii: affective variables. language teaching, 26, 1-11. mccarthy, p., rinderer, r. ,(1985). self-efficacy and writing: a different view of self-evaluation. college composition communication,36（4）, 465-471. mcleod, s. (1987). some thoughts about feelings: the affective domain and the writing process. college composition and communication，38（4）：426-435. pintrich, p. r., schunk, d. h. (2002). motivation in education: theory, research, and application (2nd ed). upper saddle river, new jersy: pearson education,inc. ramineni，c.， d. m. williamson. (2013). automated essay scoring: psychometric guidelines and practices. assessing writing, 18(1), 25-39. silva，t. (1993). toward an understanding of the distinct nature of l2 writing: the esl research and its implications. tesol quarterly, 27(4), 657-677. welnstein, c.s. (1989). teacher education students#8217; perception of teaching. journal of teacher education,40(2), 53-60 毕美芳 (2017)，自动评分系统对大学生写作自我效能感和写作成绩影响的研究, 硕士论文。

山西太原：中北大学。

葛诗利、程潇潇（2009），大学英语作文自动评分研究中的问题及对策，山东外语教学（3）：21-26。

2. 研究的基本内容、问题解决措施及方案

Research methodology 2.1 Research Questions The present study investigates the effects of automated writing evaluation system on English majors#8217; writing motivation, self-efficiency and writing performance. The specific research questions are as follows: 1.What are effects of automated writing evaluation system on student#8217;s writing motivation? 2.What are effects of automated writing evaluation system on student#8217;s writing self-efficacy? 3.How does self-efficacy and writing motivation influence on writing performance based on using of automated writing evaluation system? 2.2 Research Subjects Subjects in this study are 51 sophomores of English majors, coming from two classes in Nanjing Tech University, with age from19-20. There are 24 subjects in class one, who are requested to use automated writing evaluation system, and 27 in class two, who write without automated writing evaluation system. 2.3 Research Instruments The present study adopt a mixed-method approach, involving the use of two questionnaires, four writing tasks and once writing test, and some open questions. 2.3.1 Questionnaire In this study, there will be two questionnaires, one concerns writing motivation and the other concerns self-efficacy. Based on Gardner#8217;s AMTB (Attitude Motivation Test Battery), the final version of Writing Motivation for English Major Students Questionnaire is designed and developed. Gardner#8217;s questionnaire is about learning motivation. The content of questionnaire concerns different aspects according to nine categories of writing motivation. The first section of two questionnaires composed of four items was about the background of the students who attend the investigation including their name, gender, years of learning English and class belonged. The second section aims to investigate from nine categories: interest, writing efficacy, writing avoidance, achievement, communication, grade, go abroad, development. Questionnaire about self-efficacy comprises of writing skill efficacy and writing task efficacy. There are nine items addressing the writing skill self-efficacy. They are represented by items such as ”I can precisely use the verb form (in plural or in singular) in English composition", ”I can write fluent sentences", and ”I can accurately spell all the words in English compositions" and so on. The other nine items are concerning the writing task self-efficacy. They mainly describe the writing genres, the themes and some fixed formats. Confidence in writing genres includes ”I can write a persuasive argumentation with effectively expressing my opinions in English”, I can write a fluent narration in English, with describing the plots and character's actions”, ”I can write an exposition in English methodically" and so on. The rest items are about the theme or format of compositions, such as ”I can summarize a long passage in English with covering the keynote, I can accomplish certain writing tasks according to the requirement and present my writing style”, and "I can write an email with normative formats and native English to express basic greetings or ask some related information". The questionnaire adopted by this study is designed by referring to Li Hang (2014), who referred to writing skill self-efficacy scale of Pajares (1994) and writing self-efficacy scale of Shell (1989) and Jones (2008). At the same time, College Students#8217; English Wring Self-efficacy Scale designed by Tang Fang Xu Jinfen (2011) is also consulted. 2.3.2 Writing tasks and test In the present study, a writing test will be used to measure students#8217; writing performance. The five tasks are from TEM-4, and their writing scores in CET-6 are regarded as the pre-test writing test and their last writing task among the five will be regarded as post-test writing test. The topic of five writing tasks are Revised Textbook Turns to Chinese Traditional Culture, Mobike to Roll out Service in UK, China's First E-sports Major Opens Enrollment in 2017, Is Overseas Higher Education Still Worth the Price?, and Yes or no. There are four times practices and the fifth time is regarded as final test. 2.4 Data collection Firstly, five writing tasks are assigned to the subjects: one class write online and submit to automated writing evaluation system and the other class writes on paper only. Secondly, two questionnaires are sent out to them. The two questionnaires will be handed out by students#8217; monitors and be finished at classroom. Students will be given limited time to finish the questionnaire and collect the questionnaire together. Finally, some students will be interviewed.