楚州中学“教师个人专题研究”的研究方案

Perception of English Lexical Stress by Chinese EFL Learners in Senior High School: An Experimental Study

高中生英语单词重音感知实验调查

詹腊梅

Abstract

Stress is an important suprasegmental feature in English; it is the core of intonation, the focus of information, and the carrier of tone and emotion， thus plays an important role in acquisition of English prosody as a second language. English is a stress-timed language, whereas Chinese is a syllable-timed language. Due to this significant prosodic difference between English and Chinese, Chinese EFL learners often have difficulties in acquiring English prosody. So far, lots of studies have been done in the field of L2 English intonation produced by Chinese EFL learners from tertiary levels, however, a parse number of the research at home has ever focused on the perception of lexical stress acquired by junior high school EFL learners (hereafter JHS-EFL learners), who are supposed to be close to the Critical Period in L2 acquisition. Based on the above reasons, the present study intends to address the following research questions:

1) To what extents can Chinese JHS-EFL learners perceive lexical stress correctly?

2) Does syllable structure affect JHS-EFL learners’ perception of lexical stress?

3) Does vowel type affect JHS-EFL learners’ perception of lexical stress?

4) How does Chinese JHS-EFL learners’ perception of lexical stress develop across grades?

5) Does JHS-EFL learners’ perception of lexical stress correlate with their English proficiency?

The subjects of the present study consist of 274 Chinese JHS-EFL learners across three grades (grade7, grade 8 and grade 9). The material of the lexical stress perception experiment includes 152 two-syllable words, 76 three-syllable words, 28 homonyms, 26 non-sense words, and 21 words with the same root. The material of the perception experiment was recorded by a world-renowned English phonetician at the Laboratory of Phonetics in the University of Cambridge, UK. Then the recorded data were segmented and manipulated via recording software Cool Edit Pro2.1 so as to get ready for the treatment for the experiment. The perception experiment was conducted in Hongze Middle School among three grades of JHS-EFL learners during 3 periods of time (all were at the same extra curriculum teaching time in each afternoon). The data were collected in class and analyzed by statistical software SPSS 12.0.

The results of data analysis reveal the following findings:

1. Chinese JHS-EFL learners’ overall correct perception rate (CPR) of all target English lexical stress is 58%. Categorically speaking, the CPRs of lexical stress in non-sense words, two-syllable words, three-syllable words, heteronyms, and words with the same root are 57%, 64%, 50%, 60% and 41% respectively. In the subcategories, for the two-syllable nonsense words, the CPRs of its lexical stress on the first, second are 71% and 61% respectively; for the three-syllable nonsense words, the CPRs of its lexical stress on the first, second, third is 53%, 54% and 49% respectively; for the 2-syllable words, the CPRs of its lexical stress on the first, second syllable are 64 % and 61 % respectively; for the 3-syllable words, the CPRs of its lexical stress on the on the first and second syllable are 51 % and 49% respectively; for the heteronomy, the CPRs of its lexical stress on the first, second syllable are 55% and 65% respectively; for the words with the same root, the CPRs of its lexical stress with the root of photo, able, possible and simple are 39%, 45%, 38% and 42%.

 2. Syllable structure has an effect on the perception of lexical stress by Chinese JHS-EFL learners. For the words with the syllable structure of CV.CV, the correct rate of them is 67%; for the words with the syllable structure of CV.CVC, the correct rate of them is 62%; for the words with the syllable structure of CVC.CVC, the correct rate of them is 65%; And there is a significant difference between them, Sig. =.000, demonstrating the significant influence of syllable structure on Chinese JHS-EFL learners’ perception of target L2 English lexical stress.

3. Vowel types in the stressed syllables have an effect on the perception of lexical stress by Chinese JHS-EFL learners. For the words with the long vowels in the stressed syllables, the correct rate is 60%, that of short vowels is 56%, and there is a significant difference between them (Sig.=.000); For the words with the diphthongs in the stressed syllables, the correct rate is 61%, that of pure vowels is 58%, and there is a significant difference between them (Sig.=.000); For the words with the front, central and back pure vowels in the stressed syllables, the correct rates are 54%, 59% and 63%, and there is a significant difference between them ( Sig.=.000); For the words with the front, central and back diphthongs in the stressed syllables, the correct rates are 61%, 60% and 59%, and there is a significant difference between them (Sig.=.000). These reveal the indispensable role of vowels, such as vowel nature and vowel length, in Chinese JHS-EFL learners’ perception of L2 lexical stress.  

4. The Chinese JHS-EFL learners’ correct perception rate of lexical stress increases with the increase of the levels of grades. The overall correct perception rates of lexical stress across the junior three grades are 53%, 57% and 62% respectively; The correct perception rate of non-sense words by subjects from three grades are respectively 50%, 57% and 63%; The correct perception rate of two-syllable words by subjects from three grades are respectively 60%, 64% and 66%; The correct perception rate of three-syllable words by subjects from three grades are respectively 45%, 48% and 58%; The correct perception rate of heteronyms by subjects from three grades are respectively 56%, 61% and 62%; The correct perception rate of words with the same root by subjects from three grades are respectively 37%, 39% and 48%.

5. Pearson correlation analyses show no significant correlation between Chinese JHS-EFL learners’ correct perception rate of English lexical stress and their English proficiency (r=.042, Sig. =.484>.05), indicating the lag behind nature of Chinese JHS-EFL learners’ L2 lexical stress competence.

The unfavorable part of the results above displays some problems in the study of L2 phonetic acquisition. Firstly, with the increased number of syllables within words, Chinese JHS-EFL learners’ correct perception rate of lexical stress decreased; and the more complex the lexical structure, the lower correct perception rate of the lexical stress. Secondly, both syllable structure and vowel types have a significant effect on Chinese JHS-EFL learner's perception of lexical stress. Thirdly, with the increase of grades and exposure to L2 English, the correct perception rate of lexical stress developed significantly and profoundly, which well demonstrated the substantial role of explicit L2 classroom teaching in junior high school and the learnability of L2 lexical stress. No doubt, these findings shed light into L2 phonetic teaching and learning in classroom contexts in China.  

Key words: lexical stress; Chinese JHS-EFL learners; perception

摘要

重音是英语语言的一个重要超音段特征, 语调的核心因素,是信息焦点、语气、情感的直接承载, 在英语韵律学习中扮演很重要的角色。英语是以重音计时的语言,而汉语是以音节计时的语言。由于英汉韵律之间巨大的差异，英语学习者在英语韵律习得方面存在问题。到目前为止，许多关于此方面的研究主要涉及大学生层面的英语语调的产出，然而，很少有研究涉及初中生英语单词重音感知（此后JHS-EFL 学习者），且他们处于二语习得的关键期。基于以上原因，本研究旨在解决以下研究问题：

1、初中生单词重音感知的情况如何？ 

2、音节结构对初中生单词重音感知是否有影响？

3、重音节中的元音类型对初中生单词重音感知是否有影响？

4、初中生单词重音感知跨年级发展的情况如何？

5、初中生单词重音感知能力和英语水平是否相关？

本研究涉及三个年级（七、八、九年级）共274个初中生对单词重音的感知实验。研究语料囊括152个双音节单词，76个三音节单词，28个同形异音词，26个无意义词以及21个同根词。实验感知材料由世界著名的英国语音学家在英国剑桥大学语音学实验室录制而成。然后用录音软件Cool Edit Pro2.1对语料进行切分和处理，以为实验做好准备。感知实验在洪泽实验中学三个年级的初中英语学习者中进行，分为三个时间段（都是在每个下午的课余时间）。数据收集是在班级进行的，并用SPSS 12.0 统计软件对其进行分析。

数据分析的结果带来以下发现：

1、初中生单词重音感知情况较差，总体感知正确率为58%，分类而言，无意词，双音节词，三音节词，同形异音词以及同根词的感知正确率分别为57%，64%，50%，60%和41%。其中，对于双音节的无意义词而言，重音在第一个和第二个音节的感知正确率分别为71%和61%； 三音节的无意义词，重音在第一个，第二个以及第三个的感知正确率分别为53%，54%和49%； 对于双音节词而言，重音在第一以及第二的分别为64%和61%； 对于三音节的词而言，重音在第一以及第二的分别为51%和49%; 对于同形异音词而言，重音在第一以及第二的分别为55% 和65%；对于同根词以photo, able, possible和 simple为词根得感知正确率分别为39%，45%，38%和42%。

2、音节结构对于初中生单词重音感知存在影响。对于CV.CV结构的单词感知正确率为67%，对于CV.CVC结构的单词感知正确率为62%，对于CVC.CVC结构

的单词感知正确率为65%，三者之间的Sig.=.000, 表明音节结构对初中生英语单词重音感知存在较大影响。

3、重音节中的元音类型对于初中生单词重音感知存在影响。对于重音节中元音类型是长元音的单词感知正确率为60%，短元音的为56%， 且二者存在显著差异，p=.000；双元音的为61%，单元音的为58%, 二者存在显著差异，Sig.=.000；其中前单元音的为54%，中单元音为63%，后单元音为59%，三者存在显著差异，Sig.=.000；前双元音为61%，中双元音为60%，后双元音为59%,三者存在显著差异，Sig.=.000. 这些都证明了元音在初中生英语词重音感知中扮演举足轻重的角色，比如元音的性质和音长。

4、初中生单词重音的感知正确率随着年级的增长而增长。三个年级总体单词重音感知正确率分别为，53%，57% 和62%；三个年级的无意义词感知正确率为，50%，57% 和63%；三个年级的双音节感知正确率为，60%，64% 和66%三个年级的三音节感知正确率为，45%，48% 和58%；三个年级的同形异音词感知正确率为，56%，61% 和62%；三个年级的同根词感知正确率为，37%，39% 和48%；

5、初中生单词重音感知与学生的英语成绩没有显著相关性。其中单词重音感知率与三个年级英语成绩的Z-score的相关性结果为，r=.042，Sig.=.484>.05，无相关性。表明汉语初中生在英语单词重音感知方面存在滞后性。

本研究发现阐明了二语语音习得方面存在的一些问题：第一、随着单词音节数量的增加学生的感知正确率呈现下降趋势，而且单词的结构越复杂学生的感知率越低；第二、音节结构和重音节中元音的类型都对学生的单词重音感知存在影响。第三、随着年级水平的提高以及学生英语输入量的提高，学生的感知正确率也有显著和较为深刻的变化，说明显性初中二语课堂教学的重要性以及二语词汇重音的可学习性。毫无疑问，本研究对中国英语单词重音的教学有着深远意义。

关键词：单词重音;初中生; 感知

The overall correct perception rate of English lexical stress is 58%, and that of categorical words, non-sense words, two-syllable words, three-syllable words, heteronyms and words with the same root is respectively, 57%, 64%, 50%, 60% and 41%. The correct perception rates of two-syllable words with the stress on the first and second syllable are 64% and 61%; and that of three-syllable words with the first and second syllable stress are 51% and 49%; and that of the heteronyms with the stress on the first and second syllable are 55% and 65%. In general, the subjects’ of lexical stress is relatively poor, for the overall correct rate is 58%. Besides, the number of syllables has effects on the perception, because the correct rate of two-syllable words is higher than that of three-syllable words. What’s more, the stress patterns of also plays an important role in the perception of lexical stress, on the whole, the subjects’ perception of words with the stress on the first syllable is better than that of words with the stress on the second syllable, except for the heteronyms. The reason for the subjects’ perception of heteronyms with the stress on the second syllable is better than that of ones with the stress on the first syllable is that all the heteronyms included in this thesis have the weakened vowels in the first syllables when their second syllables are stressed, so it is more easy to perceive the heteronyms with the second-syllable stress. What’s more, the morphological structures of words also affect the perception of lexical stress; on the whole, with morphological structures becoming more and more complex, the subjects’ perception becomes poorer and poorer.

The correct perception rates of words with the syllable structures of CV.CV, CV.CVC and CVC.CVC are respectively 67%, 62% and 65%. And the differences between the correct perception rate of words with these three syllable structures are all significant, F (53, 220) = 11.936, Sig. =.000; F (53, 220) = 21.312, Sig. =.000; F (53, 220) = 17.801, Sig. =.000; To the words with the syllable structure of CV.CV, the correct perception rate of them by three grades, grade7, grade 8 and grade 9, are respectively 63%, 67% and 70%. To the words with the syllable structure of CV.CVC, the correct perception rate of them by, are respectively 58%, 62% and 66%. To the words with the syllable structure of CVC.CVC, the correct perception rate of them by three grades, are respectively 61%, 64% and 69%. In a word, subjects performs best in perceiving words with the syllable structure of CV.CV, then in the words with the syllable structure of CVC.CVC, while relatively poor in the words with the syllable structure of CV.CVC. Possibly, it may because that CV is the universal, unmarked and the basic syllable structure in all languages, and it is considered as the core syllable of language and CVC is the complete structure of syllable. And form the results of the correct perception rate of words with each three syllable structure by three grades subjects, a conclusion can be made there is a growing trend of subjects’ perception as the increase of levels of grades. Besides, no matter through the statistics drawn from One-way ANOVAs of overall correct perception rate of words with these three syllable structures or that of each grade subjects’ correct rate of perception of words with these three syllable structures, it could be summarized that the syllable structures play an important role in the perception of lexical stress.

The overall correct perception rates of words with long and short vowels are 60% and 56% and the difference between them is significant, t=8.468, Sig.=.000; that of words with the pure vowels and diphthongs in the stressed syllables are 58% and 61%. To the words with the pure vowels in stressed syllables, the correct perception rates of words with the front, central and back pure vowels are 54%, 63% and 59%, and the differences between them are all significant, F (64, 209) =18.369, Sig. =.000; F (64, 209) =8.596, Sig. =.000; F (64, 209) =28.590, Sig. =.000. While to the words with the diphthongs in stressed syllables, the correct perception rates of words with the front, central and back diphthongs are 61%, 60% and 59%, and the differences between them are all significant, F (57,216) =17.031, Sig. =.000; F (57,216) =15.390, Sig. =.000; F (57,216) =11.402, Sig.=.000. Generally speaking, subjects do better in the perception of words with the long vowels in the stressed syllables than that of words with the short vowels; and they are better in perceiving words with the diphthongs in the stressed syllables than words with the pure vowels in the stressed syllables. It may because the duration of long vowels and diphthongs is longer than that of short vowels and pure vowels, so subjects can perceive them more easily. To the words with the pure vowels in the stressed syllables, subjects are best in perceiving words with the central pure vowels in the stressed syllables, while to the words with the diphthongs in the stressed syllables, subjects are best in perceiving words with the front diphthongs in the stressed syllables. And form the results of the correct perception rate of words with each type of vowels by three grades subjects, a conclusion can be made there is a growing trend of subjects’ perception as the increase of levels of grades. Besides, by the statistics of the Paired t-tests of correct perception rate of words with the long and short vowels and that of words the pure vowels and diphthongs, from that of One-way ANOVAs of overall correct perception rate of words with each type of pure vowels and diphthongs or that of each grade subjects’ correct rate of perception of words with each kind of pure vowels and diphthongs, it could be summarized that the vowel types play a significant role in the perception of lexical stress.

Overall correct perception rate of lexical stress by subjects from the grade 7, grade 8 and grade 9 are respectively 53%, 57% and 62%, and the differences between the them are all significant, F (53,249) =39.657, Sig. =.000; F (53,249) =47.379, Sig. =.000; F (53,249) =22.408, Sig. =.000. To the non-sense words, correct perception rate of them by subjects from the grade 7, grade 8 and grade 9 are respectively 50%, 57% and 63%; to the two-syllable words, correct perception rate of them by subjects from the grade 7, grade 8 and grade 9 are respectively 60%, 64% and 66%; to the three-syllable words, correct perception rate of them by subjects from the grade 7, grade 8 and grade 9 are respectively 45%, 48% and 58%; to the heteronyms, correct perception rate of them by subjects from the grade 7, grade 8 and grade 9 are respectively 56%, 61% and 62%; to the words with the same root, correct perception rate of them by subjects from the grade 7, grade 8 and grade 9 are respectively 37%, 39% and 48%. No matter from the results of overall correct perception rate of lexical stress by subjects from three grades or that of correct perception rate of each kind of words by subjects from three grades, they all present that there is a growing trend of subjects’ perception of lexical stress as the increase of levels of grades. Besides, to the three-syllable words, the gap of correct perception of them by subjects from grade 9 and the other two grades is relatively bigger; and the same situation also takes place in the words with the same root, which may because these two kinds of words have more syllables than others, thus perception of them may be a bit difficult to the lower grades students.

No matter from the results of the correlation between the overall correct perception rate of English lexical stress and Z-score of English proficiency, or that of the correct perception rate of each kind words and Z-score of English proficiency, or that of the overall correct perception rate of English lexical stress of each three grades and their Z-score of English proficiency, they all show that there is no significant difference between correct perception rate of English lexical stress and the English proficiency. Possibly it is for the reason that the final examination of English of each grade focuses more on reading and writing skills of students.

Two theoretical implications will be presented by the study. On is the L1 transfer, which is greatly shown through the overall perception by Chinese JHS-EFL learners in the English lexical stress. Two aspects can be used to support it. One is that generally the subjects’ perception of words with the stress on the first syllable is better than that of words with the stress on the second syllable. It may be affected by their mother tongue, in which words are mainly stress on the first one. Thus, it is obvious to see that the negative L1 transfer gives rise to the subjects’ better perception in the words with the first-syllable stress. The other is that the syllable structure plays an important role in the Chinese JHS-EFL learners’ perception of English lexical stress. It may because that the syllable structure of Chinese is simple and the main factors affecting the stress of Chinese are tones and semantic features instead of the syllable structure, therefore it might bring barriers to Chinese EFL learners in the acquisition of English lexical stress.

The other is the Input Hypothesis. It could be supported by the findings of the cross-grade development of Chinese JHS-EFL learners’ perception of lexical stress, it shows that the increase of the subjects’ perception abilities with the increase of the levels of grades. For with the increase of learners’ levels of grades, they receive more input of English, so they can do better in perception than before.

The pedagogical implications made by this study are presented as follows: firstly, the overall correct perception rate of Chinese JHS-EFL learners is not very high, which indicates they have problems in the acquisition of the suprasegmental of English and it also suggests that they may have difficulties in the communication, for the stress is the core of intonation and it is also the focus of information and the carriers of tone and emotion. Besides, for the perception and production are reciprocal and the worse perception may bring about the errors in the production. So the teachers should attach high importance to the perception of stress in process of teaching and at the same time try to raise the students’ awareness of stress. Secondly, subjects are relatively poor in the perception of polysyllable words, thus teachers should put more efforts in teaching the stress of polysyllable words. Thirdly, the syllable structures have effects on subjects’ perception of lexical stress, so teachers could introduce the differences between the syllable structures of Chinese and English into students to improve their levels of perception of lexical stress. Fourthly, the vowels also play an important in the subjects’ perception of lexical stress, especially the duration of them; therefore, teachers should strengthen the training of students’ abilities to perceive the duration of vowels in stressed syllables. Finally, the subjects from high grades perform better than those from low ones in the perception of lexical stress, which suggests that the more input may bring better perception, so the teachers should increase the input of English.