论文总字数：37939字

摘 要

词汇复杂度是衡量学生语言水平高低的一个重要因素。学者们对二语学习者的书面和口语产出中的词汇复杂度进行了大量的调查研究。然而，却极少有研究关注不同的任务类型对词汇复杂度的影响。

本文探讨了英语专业学生的两种不同写作类型——摘要写作和评论写作对词汇复杂度是否有影响。本研究以东南大学英语系大二的55名学生为研究对象，他们平均学习英语的年数为9-12年，他们参加高考时在英语考试中均取得了优异的成绩。本研究所涉及的测量词汇复杂度的指标共有三个，即词汇多样性，词汇密度和词汇难度。在电脑软件RANGE 32 and SPSS的帮助下，本研究成功地收集了两个任务类型（摘要写作和评论写作）在词汇多样性，词汇密度和词汇复杂度的不同数据。

研究结果表明，不同的写作任务类型会影响词汇复杂度。在本研究中，数据显示摘要写作在词汇多样性，词汇密度以及词汇难度均相对高于评论写作。因此摘要写作的词汇复杂度要高于评论写作。此结果表明，教师可以将摘要写作作为提高学生词汇能力一种手段。

关键词：任务类型；词汇复杂度；词汇多样性；词汇密度；词汇难度；英语写作

Table of Contents

Acknowledgements i

English Abstract ii

Chinese Abstract iii

Table of Contents iv

List of Tables vi

Chapter One Introduction 1

1.1 Background of the Study 1

1.2 Significance of the Study 1

1.3 Layout of the Thesis 2

Chapter Two Literature Review 3

2.1 Task Types 3

2.2 Lexical Complexity 4

2.2.1 Definition of Lexical Complexity 4

2.1.1.1 Lexical Variation 4

2.1.1.2 Lexical Density 5

2.1.1.3 Lexical Sophistication 5

2.2.2 Some Studies of Lexical Complexity outside China 5

2.2.3 Some Studies of Lexical Complexity inside China 6

Chapter Three Research Methodology 9

3.1 Research Questions 9

3.2 Research Participants 9

3.3 Data Collection 9

3.4 Data Analysis 10

Chapter Four Results and Discussions 13

4.1 The Effect of Task Types on Lexical Variation 13

4.2 The Effect of Task Types on Lexical Density 14

4.3 The Effect of Task Types on Lexical Sophistication 15

4.4 Discussions 15

Chapter Five Conclusion 17

5.1 Major Findings of Present Study 17

5.2 Implications 17

5.3 Limitations and Suggestions for Further Study 17

References 19

List of Tables

Table 1: Descriptive Statistics of Lexical Variation...................................................13

Table 2: Multiple Comparisons of Lexical Variation.................................................13

Table 3: Descriptive Statistics of Lexical Density......................................................14

Table 4: Multiple Comparisons of Lexical Density....................................................14

Table 5: Descriptive Statistics of Lexical Sophistication...........................................15

Table 6: Multiple Comparisons of Lexical Sophistication.........................................15

Chapter One Introduction

1.1 Background of the Study

Task-based Language Teaching starting in 1980s has received many researchers’ attention. Moreover, three aspects of performance, the complexity, accuracy and fluency of language have become hot research topics in the field of Second Language Acquisition.

Many language researchers have acknowledged that the three concepts ought to be the English teaching goals. These three concepts have also been widely used as the indicator of the written and oral evaluation of second language production. However, the study of complexity, particularly lexical complexity has not given much attention as the other two elements, namely, accuracy and fluency of language. Furthermore, lexical complexity plays an important role in indicating the quality of one’s writing task performance. Laufer and Nation (1995) supported this view by claiming that a well written composition makes effective use of vocabulary. It is a tough job for second language learners to express their ideas in writing by using various and complex vocabularies. Even the native speakers find that they often pause because of lacking vocabulary when they have a strong desire to make their ideas come across. Thus, lexical complexity is worthy to be furthered researched in second language learning.

In this thesis, the author aims to find that whether the different writing task types will affect the lexical complexity in second language learners’ writing.

1.2 Significance of the Study

Writing, as a significant part in language teaching and learning, is said to be a difficult task which causes troubles to lots of language learners. Many researchers have devoted efforts to find the factors which influence the language learners’ performance in writing and hoped to find some effective ways to enhance the learners’ writing ability.

Since lexical complexity is an important indicator of writing quality, the importance of it should be valued by the researchers. This study is intended to make a survey of the effect of different writing task types on lexical complexity in English Majors’ writings and hopes to provide some useful suggestions for further second language teaching and learning.

1.3 Layout of the Thesis

This thesis contains five parts. Chapter one is the introduction of the thesis, including the background of study, the significance of study and the layout of the thesis. Chapter two is literature review. The author will first give the definitions of task types, lexical complexity and its three measures------lexical variation, lexical density and lexical sophistication. After that, the author will review the previous studies of lexical complexity both inside and outside China. Finally, the evaluation of the former studies will be given. Chapter three will introduce the research methodology of this study which contains research questions, research participants, the procedures of data collection and data analysis. Chapter four is the core of the thesis. Firstly, the author will report the results of the effect of task types on lexical variation, lexical density and lexical sophistication.Next, the effect of task types on lexical variation, lexical density and lexical sophistication will be discussed. Chapter five is the conclusion of the thesis. The author summarizes the major findings of this study, gives some advice for language teaching and learning, points out the limitations of this study, and offers some suggestions for further study.

Chapter Two Literature Review

2.1 Task Types

After the task-based teaching was introduced in the 1980s, a large number of tasks have been introduced in second language acquisition. The type of writing tasks contains: descriptive, narrative, discursive or the products of demonstrating understanding through the target language (Macaro 2008).

Skehan (1996) pointed out that one can choose the tasks according to the language demands, such as the cognitive demands or the communicative pressure. Therefore, Skehan and Foster focused on three major types------personal tasks, narrative tasks and decision-making tasks. Through investigating the effects of three task types on spoken language measured by accuracy, complexity and fluency, their study in 1996 showed that the personal task had lower complexity than the other two tasks.

There have been other studies on task types. Robinson (1995) investigated twelve adult second language learners’ spoken language production by using two types of task: here-and-now task and there-and-then task. Also Prabhu (1987) divided tasks into reasoning-gap, opinion-gap activity, and activity information-gap activity.

In China, Chen and Wu (1998) investigated the effects of three tasks on writing. The three tasks are: recalling tasks, topic tasks and summary tasks. The results told that the summary tasks had the greatest complexity and the topic tasks had the least complexity, with the recalling task between them. Shao (2003) investigated picture-retelling task, personal task and decision-making task, the results revealed that every task had higher fluency, accuracy and complexity under time-unlimited condition and decision-making task had the highest complexity and personal task was related to the highest fluency.

The present study includes two task types: summary tasks and comment tasks. Because these two tasks are suitable for practicing the writing skills such as the ability of summarizing or restating the reading materials and the ability of expressing their ideas on a given topic, which the national education committee required.

2.2 Lexical Complexity

2.2.1 Definition of Lexical Complexity

Lexical complexity is also called lexical richness, or vocabulary diversity which is widely used by researchers in vocabulary research. Laufer and Nation (1995) defined lexical complexity as the degree of a varied and large vocabulary that a writer uses. In order to measure lexical complexity, they devised a new concept called lexical frequency profile to cope with the proportion of high frequency general and academic words in learners’ writing. Engber (1995) put that lexical complexity is “the diversity of lexical choice and the correctness of lexical form” that is measured by lexical variation, lexical density, error-free lexical variation and lexical error. However, no clear definition for this term has been given except for its different measures including lexical variation, lexical density, lexical sophistication, lexical originality or lexical errors. Therefore, the definition of lexical complexity has usually been ignored by language researchers for they all focus on the various measures of lexical complexity. In this study, three dimensions are used to measure lexical complexity. They are lexical variation, lexical density and lexical sophistication.

2.1.1.1 Lexical Variation

Lexical variation, as an index of lexical complexity, is related to the range of a learner’s vocabulary. Consequently, comparing the number of different words with the total number of text is the most obvious method of measuring lexical variation. According to Laufer (1991), “Lexical Variation was defined as type/token ratio. LV shows how inclined the learner is to repeat the same words in his writing. The higher the LV, the more varied the productive word repertoire”. In this study, the formula of lexical variation is: Lexical Variation=(No.of types/ No. of tokens)*100%.

2.1.1.2 Lexical Density

Lexical density, first used by Ure (1971) refers to the percentage of the number of lexical or content words in the text such as nouns, verbs, adjectives, and adverbs. The lexical or content words are those meaningful words. Some people believed that type/token ratio was in fact the lexical density of a text. However, from previous studies (e.g., Laufer 1991; Engber 1995), we can find that the popular measure of lexical density is the ratio of the number of lexical words to the text’s total number of words. Therefore, the the formula of lexical density in this study is: Lexical Density=(No.of lexical words/ Total No. of text’s words)*100%.

2.1.1.3 Lexical Sophistication

Lexical sophistication, sometimes known as lexical rareness, is the proportion of infrequent or advanced words in a writing. Laufer (1995) pointed out that a writing would have a more prominent lexical sophistication if it had a higher percentage of beyond 2,000 words. Based on Laufer and Nation’s Lexical Profile (Laufer amp; Nation 1995), some researchers used the measure of lexical sophistication as the ratio of the number of the “beyond 2,000” words, to the total number of words of text. Thus, the formula of lexical sophistication in this study is: Lexical Sophistication=(No.of advanced words/ Total No. of text’s words)*100%.

2.2.2 Some Studies of Lexical Complexity outside China

There are a number of studies on lexical complexity outside China. Some of them focus on the comparative studies of native learners and non-native learners (e.g. Hyltenstam 1988; McClure 1991). These studies have generated varied results. Hyltenstam’s (1988) study found no clear difference between native and non-native writers in lexical variation. However, McClure (1991), after comparing the compositions of native English speakers and second language learners in Grades 4 and Grades 9, revealed that native English speakers had more lexical variation than second language learners. And the older students had more lexical variation than younger students.

Ong amp; Zhang (2010) did a research on the effect of different tasks on lexical complexity. Through analyzing the argumentations of EFL students, they found that task complexity had complex effect on fluency and lexical complexity: (1) If the writing time was limited, both fluency and lexical complexity would increase. (2) If the topic sentences were given in writing, there would have a clear increase in lexical complexity but not the fluency. (3) Writing a draft or not did not affect fluency and lexical complexity.

2.2.3 Some Studies of Lexical Complexity inside China

The researches of lexical complexity inside China usually dealt with the different grades, different levels and even the different university types on lexical complexity. Bao (2008) did a research in three groups students’ timed essays in order to discover the developmental mode of vocabulary complexity, which was measured by lexical variation, lexical sophistication, lexical density and lexical originality. The results revealed that the different levels of students showed significant difference in lexical sophistication, while the lexical variation, lexical density and lexical originality could not distinguish the different levels of students. Bao (2011) conducted another study on the effects of different university types and writing complexity on lexical complexity through analyzing the writings of TEM4 and TEM8 tests. The findings showed that: (1) University types and writing levels had no interaction in lexical complexity regardless of the courses levels. (2) University types and writing levels would affect lexical complexity for the difference of different courses but only on the surface lexical complexity.

Some studies concentrated on the relationship between writing types,writing quality and lexical complexity. Bu (2011) did a survey by selecting 120 English major students in a normal university from year one to year four for the purpose of finding the lexical complexity correlations with writing quality. The data, analyzed with software RANGE and SPSS, yielded the following findings: (1) Year Two and Year Four have higher lexical variation than Year One and Year Three, however, there are no prominent differences of lexical variation from Year One to Year Four. (2) Lexical sophistication turns out to be the best indicator of writing ability in four groups. Generally, the number of sophisticated words increases while the number of frequent words decreases in writings as the students further their studies. (3) As for lexical density, the results show that Year One has the highest lexical density while Year Two has the lowest. On the whole, lexical density increases from Year Two to Year Four. (4) There is obvious correlation between lexical sophistication and lexical variation while there exist no significant relations between lexical density and lexical variation and between lexical density and lexical sophistication. (5) As for the relationship between lexical complexity and writing quality, obvious relations are revealed between lexical density and writing quality together with lexical sophistication and writing quality. However, no statistics show a significant correlation between lexical variation and writing quality. Tong (2012) made a comparison between students in independent college and those in universities by investigating the syntactic complexity and lexical complexity in English majors’ writing across three task types--personal descriptive task, expository task and argument task. The results show that: (1) Task types obviously affect students’ syntactic complexity. (2) No clear relations between lexical complexity and task types have been found through the measures of lexical density and lexical variation. (3) University students have higher lexical complexity than independent college students.

2.2.4 The Evaluation of Previous Studies

The literature review shows some merits and limitations of the former studies. First of all, many studies contain three measures--lexical variation, lexical density, and lexical sophistication, which makes us fully understand the situation of lexical complexity in second language learners’ performance. Moreover, most researchers use software to analysis the data which greatly enhances the correctness and effectiveness of the studies. However, some limitations need also be considered in current studies. Many earlier studies deal with lexical complexity with one or two measures. It is impossible for researchers to measure lexical complexity with one measure, thus more measures need to be covered for higher accuracy. Besides, even some researchers investigate the correlations between lexical complexity and writings, few of them have analyzed the relationships between lexical complexity and different writing task types . Therefore, it leaves space for this study to enrich in this field.

Chapter Three Research Methodology

This chapter will describe the methodology of the present study, containing the listed elements: research questions, research participants, data collection and data analysis.

3.1 Research Questions

The thesis intends to answer the following three questions:

1. Do students’ summary and comment show similar lexical variation?
2. Do students’ summary and comment show similar lexical density?
3. Do students’ summary and comment show similar lexical sophistication?

3.2 Research Participants

The participants of the thesis are 55 second year English majors in Southeast University,which comprise 55 students. They were between 20-21 years of age and have learned English for 9-12 years. When they graduated from senior high school, they all did very well in the English test in the University Entrance Examination.

3.3 Data Collection

The raw materials are all compositions collected from Year Two were 55 summaries and 55 comments. The writings were collected with the help of the teacher of these three classes at the end of the academic year of 2014-2015. Then the summaries and comments were keyed into the computer, and the spelling mistakes were rectified but no other changes have made. After that, each of summaries and comments was saved as a separate text file named from summary 1 to summary 55 and from comment 1 to comment 55.

3.4 Data Analysis

The students’ summaries and comments were analyzed through the following steps to get the results of the three measures, lexical variation, lexical density, lexical sophistication with the help of the computer program, RANGE 32 and SPSS.

To examine the lexical density, the author first calculated the number of content words including nouns, verbs, adjectives and adverbs, for the higher percentage of content words indicates the greater lexical density. The author used the mark [CONW] to represent the content words in the text, and marked the content words manually. After that, by using the software Frequency 40 which included in the software RANGE32, the author counted the total number of content words in the text. Finally, lexical density was obtained by using the formula lexical density=(No.of lexical words/ Total No. of text’s words)*100.

Lexical sophistication was worked out by the percentage of the sophisticated words in the text, as mentioned before, the formula for calculation is Lexical Sophistication=(No.of advanced words/ Total No. of text’s words)*100. The data was also analyzed by RANGE 32. The software lists the proportion of words at different word frequency files, 1^st 1000, 2^nd 1000, 3^rd 1000, and “ not on the list” words. The author obtained the lexical sophistication by calculating the “beyond 2000” words.

In order to have a better understanding, summary 20 is taken as an example to explain the detailed calculation analyses.

The author[CONW] uses [CONW]his refugee[CONW] mother’s[CONW] experience[CONW] to introduce[CONW] the British[CONW] who seemed[CONW] to his mother [CONW]self-contained[CONW], self-controlled[CONW], and law-abiding[CONW] yet tolerant[CONW] to others[CONW] and encouraged[CONW] them to laugh[CONW] at themselves[CONW]. They were polite[CONW] and considerate[CONW], the self-confident[CONW] took care[CONW] not to humiliate[CONW] the shy[CONW] or timid[CONW]. Appearances[CONW] in Britain[CONW] could deceive[CONW]. The orderliness[CONW] and restraint[CONW] of political[CONW] life[CONW] in Britain[CONW] also struck[CONW] the author’s[CONW] mother[CONW]. Many[CONW] remarked[CONW] upon the gentleness[CONW] of British[CONW] behavior[CONW] in public[CONW] and their behavior[CONW] when ill[CONW] or injured[CONW] was stoic[CONW].

Then, the author[CONW] said[CONW] the culture[CONW] and character[CONW] of British[CONW] restraint[CONW] have changed [CONW]into the exact[CONW] opposite[CONW]. Many[CONW] young[CONW] British[CONW] people[CONW] believe[CONW] that by drinking[CONW] themselves into oblivion[CONW], they are getting[CONW] rid of inhibitions[CONW] that might otherwise[CONW] do[CONW] them psychological [CONW]and even physical[CONW] harm[CONW]. Lack[CONW] of self-control[CONW] is just as character-forming[CONW] as self-control[CONW]: but it forms[CONW] a different[CONW], and much worse[CONW] and shallower[CONW], character[CONW]. As a result[CONW], the young[CONW] British[CONW] find[CONW] themselves hated[CONW], feared[CONW], and despised[CONW] throughout[CONW] Europe[CONW]. No person[CONW] with the slightest[CONW] apprehension[CONW] of human [CONW]psychology [CONW]will be surprised [CONW]to learn[CONW] that as a consequence[CONW] of this change[CONW] in character[CONW]. Before the English[CONW] and British[CONW] became[CONW] known[CONW] for self-restraint[CONW] and an ironic [CONW]detachment [CONW]from life[CONW], they had[CONW] a reputation[CONW] for high[CONW] emotionalism[CONW] and an inability[CONW] to control[CONW] their passions[CONW]. The moralization[CONW] of the British[CONW] in the nineteenth[CONW] century[CONW] was the product[CONW] of intellectual[CONW] and legislative [CONW]activity[CONW].

In the end[CONW], the author [CONW]would like[CONW] to say[CONW] to Americans[CONW]: excoriate[CONW] sin[CONW], especially in public[CONW].

Lexical complexity:

Lexical variation=(No.of types/ No. of tokens)*100= 155/250*100=62.00

Lexical density=(No.of content words/ Total No. of text’s words)*10 =127/250*100=50.80

Lexical sophistication=(No.of advanced words/ Total No. of text’s words)*100=12/250*100=4.80

In the end, the above calculation of lexical variation, lexical density and lexical sophistication were processed by the SPSS software to investigate the relationship between lexical complexity and different writing task types.

Chapter Four Results and Discussions

The results of the data analysis will be presented in this chapter. After the assistance of SPSS, the statistical analysis is completed. The results will be presented according to the research questions, and the discussions will be made in line with the results.

4.1 The Effect of Task Types on Lexical Variation

Lexical variation, as one measure of lexical complexity, is calculated as the ratio of the number of word types to the word tokens in the text.

In Table 1, the statistical results of lexical variation of the writing task types (summary and comment) are presented.

Table 1 Descriptive Statistics of Lexical Variation

From Table 1, we can see that the average summary lexical variation is 59.2485 per hundred words while the mean score of the comment lexical variation is only 53.6618. Therefore, we can draw a conclusion that summary tends to show a relatively higher lexical variation than comment. Except that, the standard deviation of summary and comment lexical variation are 4.51653 and 6.26406, both lower than 8, which reveals that the levels of lexical variation of the summary and comment are just slightly different.

In order to examine whether the task types affect lexical variation, a Paired Samples T-Test was conducted. The following table reveals the comparison of lexical variation between two the tasks (summary and comment).

Table 2 Multiple Comparisons of Lexical Variation

Note: *The mean difference is significant at the level of .05.

From the above data analysis, we can find that different task types (summary and comment) significantly influences lexical variation (P=.000lt;.05), which means lexical variation varies across task types ( here summary and comment).

4.2 The Effect of Task Types on Lexical Density

The information of lexical density is listed in Table 3 and Table 4.

Table 3 Descriptive Statistics of Lexical Density

From Table 3, it can be found that summary tends to have a relatively higher

ratio of lexical density than comment, for the mean score of summary lexical density is 50.5356, while the comment’s score is 47.8520. The standard deviation of summary and comment lexical density are 4.2215 and 3.1161, both of which are lower than 8. So the subjects’ level of lexical density are not quite different.

In order to investigate whether task types (summary and comment) influence lexical density or not. The results about the comparison in lexical density between two different task types (summary and comment) are reported in the following Table 4.

Table 4 Multiple Comparisons of Lexical Density

Note: *The mean difference is significant at the level of .05.

The Table 4 shows that a statistically significant difference exists in terms of lexical density between two the task types (p=.000lt;.05), which means that task type affects students’ lexical density.

4.3 The Effect of Task Types on Lexical Sophistication

Lexical sophistication is a measurement of lexical complexity, which is used to investigate whether students are good at using the advanced words (beyond 2000 words) in their written or oral productions or not. The results of lexical sophistication in this study is reported in the following Table 5.

Table 5 Descriptive Statistics of Lexical Sophistication

From Table 5, we can see that comment has a relatively lower lexical sophistication than comment. Because the mean score of comment is only 2.8140, while that of summary is 3.3691. As for the standard deviation, both summary and comment’s standard deviation are lower than 8, with summary’s score is 1.58274 and comment’s score is 1.46891, thus, the students’ level of lexical sophistication with summary or comment is quite similar.

Table 6 is the result of a Paired Samples T-Test about the comparison in lexical sophistication between two different task types (summary and comment).

Table 6 Multiple Comparisons of Lexical Sophistication

Note: *The mean difference is significant at the level of .05.

The results in Table 6 reveal that the lexical sophistication is significantly higher in summary than in comment (p=.011lt;.05).

4.4 Discussions

From the above results, we can draw the following conclusions: firstly, different task types will affect lexical complexity in all three dimensions: lexical variation, lexical density and lexical sophistication. Secondly, all the three measurements of lexical complexity are relatively higher in summary than in comment. Therefore, the writing task summary can be used to enhance students’ lexical complexity. The possible reason why subjects tend to have a higher lexical complexity in summary is that when writing a summary, learners have the original text to refer to and they may use some content from the text or some original words, however,when they write a comment, they may tend to use simple words or their English is not proficient enough to achieve the variety in words.

The present study’s results are different from those of the previous studies by Tong (2012) and Tian (2007). Both studies found that different task types had no significant influences on lexical complexity. Reasons for this phenomenon can be listed below: the choice of task types are different between the present study and previous studies. The present study chose summary and comment, while Tong (2012) chose personal descriptive task, expository task and argument task. The measurements of lexical complexity used in the research are different. Tong (2012) investigated lexical variation and lexical density in her study, while Tian (2007) only used lexical variation. So if other measurements of lexical complexity had been used, for example, lexical sophistication, the significant difference of task types in lexical complexity may have been discovered. Last but not least, the time control of the writing task may be a factor that influences the ratio of the lexical complexity. For instance, in Tian’s (2007) study, all the writing tasks are time limited, while data of the present study were collected from students’ homework, thus they have relatively long time to finish.

Chapter Five Conclusion

5.1 Major Findings of Present Study

In the research, we used three measures lexical variation, lexical density and lexical sophistication, to examine the written production under different task types. The results yield that task types significantly influence learners’ lexical complexity on lexical variation, lexical density and lexical sophistication. Summary and comment do not have the same lexical variation, lexical density and lexical sophistication, their lexical complexity are different from each other. Moreover, summary tends to have relatively higher ratio of lexical complexity than comment on all three dimensions: lexical variation, lexical density and lexical sophistication.

5.2 Implications

The implications of the present study can be drawn from both the theoretical and pedagogical aspects.

For another, the present have wide pedagogical implications for the teaching of English writing. To begin with, the results of the data analysis can help the teachers have a better understanding of the students’ attitude towards different writing tasks and difficulties in writing. Therefore, they can choose the appropriate task types for teaching. Besides, the summary task is a good choice for teachers to use as a tool of lexical teaching and lexical enhancing, for summary relates to the higher lexical complexity. To some extent, it also reveals that reading is very important in English learning.

5.3 Limitations and Suggestions for Further Study

Even the present study is carefully designed, some limitations should be acknowledged for the lack of time, energy, experience and other reasons.

One obvious limitations is that students’ writing, both summaries and comments are all students’ homework, they have relatively longer time to finish and they can refer to the dictionaries, which may to some extent affect the final results. Thus, further research should better choose time-limited writings as research materials.

剩余内容已隐藏，请支付后下载全文，论文总字数：37939字