1. 研究目的与意义（文献综述包含参考文献）

1. introduction1.1 research backgroundwriting and speech are two cardinal skills in language learning. since the modern linguistics emerged as a momentous field, writing has been deemed prescribed and modified (dai, 2013: 5-6). consequently researchers lay more emphasis on speech which is primary in the development of language. however, several dimensions must be factored in for researchers to weigh writing against speech in pedagogical improvement. first, writing lays the groundwork, either semantical or grammatical, for the standard use of target language, especially to the second language learners. second, writing is a high demanding job but also rewarding. the second language learners confront a myriad of exacting requirements to promote their writing such as lexical richness, sentence structure, layout design, cohesion and transition and so on. undoubtedly, writing deserves investigation. with the implementation and constant amendments of english curriculum standard for senior high school (2020), more emphasis has been laid on the cultivation of students pragmatic competence. one the one hand, english writing is overweight in communication competence. senior students are supposed to be capable of description of experiences and truth, delivery of messages, views demonstration and emotional expression. on the other hand, the documentary also sets three levels of linguistic competence. from low proficiency to high one, it demands that students are able to state issues, convey views and exhibit value orientations on the basis of expert accumulation and flexible application of linguistic knowledge and skills. obviously, it puts forward higher requirements to english writing proficiency of senior students in high school.1.2 purposes of the researchit is universal acknowledged that english writing is the soft underbelly of chinese english learners. due to the lack of advanced expression and comprehensive organization, there exists a discrepancy between writers profound thoughts and awful writing proficiency. consequently, the purpose of this research includes two levels. first and foremost, it is important to figure out the relationship between lexical richness and writing proficiency from lexical variation, lexical sophistication, and lexical density these three dimensions. second is to use evidences concluded in this study to provide some meaningful suggestions to the second language learners.1.3 organization of the researchthis thesis consists of six chapters. following the introduction, chapter 2 reviews publications and researches on this field. it also considers, from a variety of perspectives, how those researches conducted first time, the main definition of lexical richness, and reviews some major findings of them. chapter 3 depicts the methodological approach adopted in the study. lexical richness will be measured from three dimensions (lv, ls, and ld) with the help of r and spss program. all the samples are timed composition in examination. key findings from an analysis of the research data are presented in chapter 4. the results are based on the quantitative research techniques but some qualitative analysis can be included. then chapter 5 includes a detailed account and interpretation of the findings of the study, in comparison with relevant researches and with reference to the current question in research. chapter 6 summarizes the findings and provides some pedagogical implications to english teaching.2. literature reviewthis segment will be divided into three parts which include the definition of lexical richness, theoretical basis and preceding empirical studies on lexical richness. 2.1 lexical richnesslexical richness was first used to compare spoken and written language, and later as an indicator of composition quality. an outstanding composition, among other factors, capitalizes well on vocabulary (grobe, 1981:75-85). generally speaking, it is a loose umbrella term which is comprised various dimensions of lexical proficiency, such as lexical originality, lexical density, lexical sophistication, lexical individuality, and lexical variation (linnarud, 1986; laufer, 1991; read, 2000). at first, linnarud (1986) supposes that lexical richness is measured by lexical variation, lexical density, lexical sophistication, and lexical individuality. later on, lexical originality was substituted for lexical individuality by laufer in 1991. however, read (2000) thinks that lexical originality is not applicable to measure peculiarities of one participants writing and evaluate the development of lexical proficiency. in other studies, a frequency-focused definition of lexical richness is used (nation 2001; nation and waring, 1997). it indicates that low-frequency words manifest themselves in sophisticated knowledge of vocabulary. it is obvious that lexical richness can be elaborated on from at least three dimensions. they are lexical variation, lexical density and lexical sophistication. moreover, wan (2011) also agrees that lexical richness is composed of these three. consequently, this study will employ this definition as general construct. more narrowly, the definitions of these three dimensions need to be figure out. first, lexical density suggests the percentage of lexical words in the text. second, the percentage of unconventional words in the test is measured as lexical sophistication. third, lexical variation, also called lexical diversity, is the ratio between different words in the text and the entire number of running words. considering the samples are all writings in examinations, which means the same amount of time and the same topic, this study can perfectly circumvent these shortcomings of ttr (type-token ratio) in lexical variation. 2.2 theoretical basisthis segment will introduce two hypotheses to the study. they are input hypothesis (krashen, 1985) and comprehensible output hypothesis (swain, 1993). input hypothesis promoted by krashen (1985) defined the current level of learner as i and the upper stage of english proficiency was described as i 1. 1 embodied the gap between the current level and target level. there is only one necessary sufficient condition for second language acquisition, and that is the presence of comprehensible input. krashen (1985) believes that an ideal input of language is guaranteed by a host of memorization of linguistic materials. moreover, input should neither be so far beyond their reach that they are overwhelmed nor so close to their current stage that they are not challenged at all. the compositions of a class are collected at three times as samples in this research, respectively in the beginning, mid-term, and end of the semester, so that data can be accurately gathered as they make progress in english learning.comprehensible output hypothesis (swain, 1985) deems that comprehensible output is also significant in second language acquisition. guided by this theory, noticing function plays an important role in learners cognition that helps them recognize the gap between themselves and native speakers by expressing themselves out. it may bring their attention to something they need to discover about their second language. metalinguistic function helps learners check their language as they convey something both semantically and grammatically. whats more, this function has its counterpart in the input hypothesis and that is monitor hypothesis. while the acquired system is able to produce spontaneous speech, the learned system is used to check what is being spoken. compositions in examination bring these two, metalinguistic function and monitor hypothesis, into full play. students have enough time to choose and employ syntactic rules. they pay attention to the accuracy of language forms to make them stand out in exams. 2.3 preceding empirical studies2.3.1 relevant research on lexical richness abroadsince 20th century, several foreign researches have been conducted in the field, which aroused great interest from foreign researchers. totally, there are three periods of development of lexical richness. the first period was in the 1980s when lexical richness was measured, as one factor, to distinguish the quality of l1 writing (grobe, 1981). later, this dimension was applied to the l2 study on account of the significant role vocabulary play in the construction of meaningful texts in the second language. linnarud (1986) conducted a comparative study to analyze and synthesize the differences in english writing by swedish students and english students. in the study, english writing by english students outperformed the ones by swedish students. swedish students were more likely to use high-frequency words in english writing. as for the four dimensions, with reference to lexical variation, lexical novelty, lexical collocation and lexical individuality, english group was always superior to swedish group. in addition, english students also had effective use of adjectives and adverbs. this study suggests that l2 learners are supposed to pay more attention to authentic expression and confer with native speakers so that they can correct improper uses caused by the interference of their first language in their interlanguage.mcclures study in 1991 brought researches in the field to the second period. she studied the lexical density in the writing of native english students in a bilingual environment and concluded that monolingual environment was more conducive to the improvement of lexical diversity. besides, lexical proficiency in her study was also increased as aging. laufer (1991) launched an investigation on english majors writing in israel and he found that lexical sophistication could be improved in an academic year while no obvious strides existed in lexical variation. a new way to measure lexical richness called lfp (lexical frequency profile) was verified to be reliable by laufer and nation in 1995. it tests lexical richness in free production and provides robust results for two pieces of writing by one student. the third period started in 21st century. researches in this period are more comparative and contrastive. jarvis (2002) claimed that lexical diversity was a vantage point to study the relationship l2 instruction, l2 proficiency, age, vocabulary knowledge, and first language interference. in his study, two of the curve-fitting formulae were verified as accurate models of the type-token curves of more than 90% of texts of participants. in conclusion, these foreign researches have made a great difference to influence of lexica richness in l2 acquisition. they are instrumental in preliminary definition and dimensions of lexical richness. however, there is much for exploration.2.3.2 domestic research on lexical richness researches on lexical richness in china were conducted in the late 1990s. according to lu (2010), many studies have bear out the fact that indicators like lexical accuracy, fluency, and grammatical ability can be used as effective and trustworthy indicators of learners esl language proficiency. bao (2008) found that lexical variation was related to text length in his multidimensional study of l2 written lexical richness completed by 150 college students. and the lexical sophistication turned out to be the best dimension to distinguish different groups in vocabulary use, which always maintained a linear development pattern. in 2010, bao investigated the changing patterns of language complexity in the spoken and written compositions by efl learners covering different linguistic proficiency and registers. it indicated that lexical complexity was entwined with language proficiency and registers. bao (2011) tested the relationship between lexical richness and the quality of timed writing. again, lexical sophistication was the best measure in distinguishing the writing competence of the participants.another empirical research conducted by zhao guanghui (2017) to examine the influence of lexical richness training on college english writing turns out that lexical richness training can contribute to improving writing proficiency.all these studies elaborate on the lexical richness and verify the reliability of each dimension. it is widespread acknowledged that lexical richness is one of the significant factors that affected writing quality.referencesgrobe, c. (1981). syntactic maturity, mechanics, and vocabulary as predictors of ratings. research in the teaching of english, 15, 75-85.jarvis, s. (2002). short texts, best -fitting curves and new measures of lexical diversity. language testing, 19(1), 57-84.krashen, s. d. (1985). the input hypothesis: issues and implications. london: longman.laufer, b. (1991). the development of l2 lexis in the expression of advanced learner. modern language journal, 75(4), 441-448.laufer, b. nation, p. (1995). vocabulary size and use: lexical richness in l2 written production. applied linguistics, 16(3), 307-322.linnarud, m. (1986). lexis in composition: a performence analysis of swedish learners written english. malmp: liber forlag malmo.mcclure, e. (1991). a comparison of lexical strategies in l1 and l2 written english narratives. pragmatics and language learning, 2, 141-154.nation, i. s. p. (2001). learning vocabulary in another language. cambridge: cambridge university press.nation, i. s. p. r. waring. (1997). vocabulary size, text coverage and word lists. in schmitt and mccarthy (ed.), vocabulary: description, acquisition and pedagogy. cambridge: cambridge university press.read, j. (2000). assessing vocabulary. oxford: oxford university press.swain, m. (1985). communicative competence: some roles of comprehensible input and comprehensible output in its development. new york: new bury house publishers.鲍贵（2008），二语习得作文词汇丰富性多维度研究，《外语电化教学》，（5）：38-44。

鲍贵（2010），英语学习者语言复杂性变化对比研究，《现代外语》，（2）：166-176。

鲍贵（2011），英语学习者词汇复杂度的最新测量及验证，《山东外语教学》，（6）：44-52。

2. 研究的基本内容、问题解决措施及方案

3. Method3.1 ParticipantsThe students of one class in the third grade in Xiancheng senior school consists of the participants of this study. Totally, the intact class has 36 students, with 27 male and 9 female. Tests are conducted without telling the students that this is an experiment for research. Instead it is carried out in the periodical examinations though their academic life. Consequently, students will play their parts to the full, concentrating on their expressions to exhibit their language aptitude. As a result of this, the outcoming will be more reliable and authoritative. 3.2 Materials3.2.1 Compositions108 compositions are collected for this study, one piece of work each time and three times in a semester. Compositions are derived from the examination held in the beginning, middle, and end of one semester. The time intervals of two corresponding points are identical so that the teaching schedule, students assignments, and their basic improvements in language learning can be controlled at an equivalent level. Additionally, timed writing also controls the length of text so that LV will not be impacted.3.2.2 Vocabulary of CET-4 and CET-6The study will employ the vocabulary of CET-4 and CET-6 as the list of advanced words which may be used in students writing. Nowadays, CET-4 has become inelastic demand for graduation in university from a utilitarian perspective. It can also become a standard-bearer for students in senior school to improve their vocabulary knowledge. By using this two list compositions will be scanned to show how many advanced words they used in their writing, which is also one of dimensions of lexical richness.3.2.3 Vocabulary of English Curriculum StandardThis word list is also used to assess the lexical sophistication of compositions. The average level of participants will manifest itself in this procedure. 3.2.4 Text Analysis ToolsConsidering the three dimensions of lexical richness, the packages KoRups, Quanteda, and Text Mining are required in R studio. KoRups is used to study the lexical variation and quanteda is applied to conduct researches in lexical sophistication. The last one aims at the data of lexical density.3.3 Procedures and AnalysisBefore the experiment, researcher will be informed of the topic of writing and then evaluate the difficulty of them. During the experiment, participants will be required to concentrate on their writings and finish it in fixed time but without telling them it is an experiment. All the writing papers will be collected as research materials. In the meantime, the grade of writing and their assessed language aptitude will also be gathered. Researcher will collect such materials for three times.All the compositions will be printed in computer and pretreated before they are analyzed. In order to tokenize each composition and establish corpus, numbers, punctuations, symbols, and hyphens need to be removed. Each word will be programmed into lower case and changed to its word stem. Additionally, Stopwords, which are used frequently but usually do not have significant meanings, are also removed to simplify the computation so that each parameter of lexical richness will be more distinct.Following the data collection, the texts were analyzed for lexical richness calculating the following measures:Lexical variation: TTR; mathematical transformations of TTR.Lexical sophistication: comparison with vocabulary list of CET-4 and CET-6; comparison with vocabulary list of English Curriculum Standard.Lexical density: Content words/ total words.After calculating the lexical richness indices, the data will be statistically analyzed using correlation analysis and repeated ANOVA.