An Experimental Study of the Effects of Listening on Speaking for College Students

As China enters WTO, more college graduates with higher oral English proficiency are required. However, we learned that even students in some distinguished universities are lack of this ability. Based on her teaching experiences and the theory proposed by Krashen and some other well-know foreign languages teaching researchers, the author of this thesis formulated two hypotheses: 1) Students’ listening ability and their oral English production ability are correlated. 2) Teachers who bring listening and audio-visual materials into oral English class are likely to have better teaching results.


Background
As China enters WTO, as international relationships become closer, as unions and partnerships across nations become more widespread, an increasing number of jobs are likely to require a person not only to be good at his specialty, but also to acquire higher proficiency in oral English.
What affects an employee's ability to communicate effectively the most in a multinational company environment is oral English ability.For about 90 percent of the job openings we see in China, oral English (at a business level) is a must, with strong reading and writing abilities preferred.These positions usually require candidates to interface often with international managers as well as communicate updates and information to China-based and home office senior management.Beyond just speaking, candidates must be able to express themselves accurately and clearly.
In spite of the fact that English language courses are required in colleges and universities, the students' oral English ability is far from satisfying.The possible reason for such embarrassing cases might be that students who have good command of English in reading and writing might not be good enough at speaking in English.Students may be qualified in reading or writing some English materials, but might fail in oral English communication.For college English teachers, there is a must to at least partly solve this problem.Luckily, nowadays, most of them have realized this problem and some of them tried possible ways to improve students' oral English.These tentative reforms in teaching have undoubtedly positive effects on improving students oral production ability, however, in the writer's view, those methods in oral English teaching could not be the panacea for all students at college levels, and they may not work for students whose English proficiency level is either too high or too low.Such being the case, some students who have lower or higher proficiency level might suffer from the courses rather than improve their oral English.

Significance of the study
The significance of this study is two-fold.From the research perspective, although listening has been widely used in oral English classes, it just has been used for imitation, but not been seen as a way to make students' oral English more authentic.This study might provide some insight into the research in this field; from the teaching perspective, the findings of this study might promote the effectiveness of listening on oral English teaching and learning, hence promoting the college English reform in our country.

Research questions and hypotheses
The central purpose of this study is to verify the truth that listening some appropriate English materials has some effects on improving students' oral English.This study draws on insight from listening and speaking classes of two parallel classes, and it is hoped that this tentative study night shed some light on further research in this field.
Research shows that listening has proved to be a very effective tool to improve students' oral English in oral English teaching.A tentative proposal is put forward to compare the students' results of the final test in two parallel classes (one is experimental class, the other is control class) and investigate whether listening has some effects to make students' oral English more authentic.The following hypotheses are therefore formulated: Hypothesis 1: Students' listening comprehension ability and their oral English ability are correlated.
Hypothesis 2: Since listening has some positive effects on improving students' oral English ability, teachers who bring listening and audio-visual materials into their oral English class are likely to have better teaching results (1981a;1982) andLong (1983b;1983c) have argued strongly that SLA is dependent on the availability of comprehensive input before the learners' internal processing mechanism can work.

Krashen
Krashen presents the case for comprehensible input in the form of the input hypothesis.He argues that for SLA to take place, the learner needs input that contains exemplars of the language forms which according to the natural order are due to be acquired next.Input must consist of 'i+1'.Krashen (1982:21) writes: … a necessary (but not sufficient) condition to move from stage 'i' to stage 'i+1', where 'understand' means that the acquirer is focused on the meaning and not on the form of the message.
Thus acquisition takes place when the learner understands language containing 'i +1'.This will automatically occur when communication is successful.Krashen emphasizes that input does not need to be 'finely tuned' in the sense that it is linguistically adjusted to contain 'i+1'.It requires only rough tuning, which is automatic if the focus is on successful communication.Krashen talks of the input 'casting a net' in order to make certain that it is of an optimal size, providing a build-in review of language forms already acquired that the focus is on meaning and not form.
Long (1983 c) considers in some detail how input is made comprehensible.One way is by the use of structures and vocabulary which the learner already knows.However, this type of input cannot foster development, because it supplies no new linguistic material.Another way is by a 'here-and-now' orientation, which enables the learner to make use of the linguistic and extra linguistic contexts and his general knowledge to interpret language which he does not actually know.A third way is through the modification of the interactional structure of conversation.Long considers interactional adjustments to be the important ones for SLA and points out that these occur even when there are no formal modifications.A 'here-and-now' orientation, together with interactional adjustments, are the main source of comprehensible input.They ensure that communication proceeds, which exposing the learner to new linguistic material.
Krashen bring listening-based methods together through the notion of 'comprehensible input'.He claims that 'acquisition' can take place only when people understand messages in the 'target language' (Krashen and Terrell,1983).Listening is motivated by the need to get messages out of what is heard.Foreign language learner acquires a new language by hearing in contexts where the meaning is made plain to them.Ideally the speech they hear has enough 'old' language that the student already knows and makes enough sense in the context for the new language to be understood and absorbed.How the teacher gets the message across is not particularly important.
Krashen claims that all teaching methods that work utilize the 'fundamental pedagogical principle' of providing comprehensible input: "if X is shown to be 'good' for acquiring a second language, X helps to provide CI (Comprehensible Input), either directly or indirectly.(Krashen, 1981b) Krashen's code breaking approach to listening became a strong influence on language teachers in the 1980s.It is saying essentially that L2 acquisition depends on listening: decoding is code breaking.It did not, however, lead to a generation of published listening -based main course books.

2.2Present oral English teaching situations in colleges
As more and more emphasis has been given to students' communicative competence, therefore textbooks have designed to cater to this need.Take New Horizon College English and New college English for example, there is an additional textbook on listening and speaking for New Horizon College English which, according to the editors, should spent at least two hours for each unit.In classrooms, students usually take task-based assignment, for example, group discussion on certain topics, role-play or pair work as to fulfilling some designed tasks, individual presentation to deliver a public speech, and so on.In New College English, students will meet listening and speaking section before they begin intensive reading part, and at the end of each unit, there are group discussions and pair works concerning the central topic of a particular unit.As recommended by the editors, one third of the in-class time should be allocated to listening and speaking, and speaking should be the larger part.Because of the limitation of class time, students are often required to continue their unfinished tasks as after-class assignment.However, in practice, many students fail to do this mainly due to their boredom to the stiff assignments.
Most of our college English teachers practice the instructions according to the syllabus of the textbooks being used.Two to three hours of oral practices mainly focus on the materials offered in the textbook.Little contextual alteration might have been made, because the teachers rely too much on the textbooks.They assume textbook is the best, at least is better than their own mind.
On the contrary, textbook materials are not necessarily suitable for all students at various levels of different majors.What is easy for computer science majors might appear tough for art students.Furthermore, some tasks seem too outdated to keep pace with the fast-developing society.Also in this kind of speaking class, the teacher is the only model-source, but who would ensure that he/she is qualified in speaking English if he/she is non-native English speaker.Owing to the above reasons and more, it's no doubt some problems may arise.Students show less interests and motivation in these materials, students are able to speak English, but actually Chinese-English in terms of pronunciation, intonation, cultural knowledge, which is far from authentic English, and might lead to communicative failure in some cases.
During the five-year teaching life, the author observed and examined many oral English classes given by teachers of different age groups.She found that they all have something in common: the teachers' main goal was to have students talk and interact, in the hope that through interaction and minimal interference form the teachers the students would practice the linguistic forms they had learned and entered into meaningful dialogue with one another.However, we have seen the important role that context played in the construction of meaning.By failing to take advantage of the full range of contextual possibilities, the teachers often unwittingly constrained classroom discourse to superficial, linguistic exchange, thus only partly achieving the goal they had set for themselves.Doing justice to the full context of the foreign language classroom raises interesting issues that require a new type of pedagogy.

Listen to perfect students' pronunciation and intonation
Views on teaching pronunciation have changed dramatically over the last half-century of language teaching.In the heyday of audio-lingualism and its various behaveristic methodological variants, the pronunciation component of a course in our program was a mainstay.Language was viewed as a hierarchy of related structures and at the base of this hierarch was the articulation of phonemes and their contrasts within English and between English and native language.Oral English classes consisted of imitation drills, memorization of patterns, minimal pair exercises.
In the 1970s', as the language teaching profession began to experience a revolution of sorts, explicit pedagogical focus on anything that smacked of linguistic nuts and bolts was under siege by proponents of the various non-directive, "let-it-just-happen" approaches to language teaching.As we became more concerned with authenticity, real-world tasks, naturalness, non-directive teaching, and process, we became less concerned with the product: language itself.Pronunciation instruction became somewhat incidental to a course of study.It was not ignored entirely, but in the interest of promoting fluency-based instruction, accuracy-based focus on English phonology became, for many, an afterthought.By the mid 1980s', the cutting edge of the profession turned in a different direction.With greater attention to grammatical structures as important elements in discourse, to a balance between fluency and accuracy, and to the explicit specification of pedagogical tasks that a learner should accomplish, it became clear that pronunciation was a key to gaining full communicative competence.(Brown,1994) But the current approach to pronunciation starkly contrasts with the early approaches.Rather than attempting only to build a learner's articulatory competence from the bottom up, a top-down approach is taken in which the most relevant features of pronunciation-stress, rhythm, and intonation-are given high priority.Instead of teaching only the role of articulation within words, or at best, phrases, we teach its role in a whole stream of discourse.Rita Wong (1987:21) reminds us that: …contemporary views (of language) hold that the sounds of language are less crucial for understanding that the way they are organized.The rhythm and intonation of English are two major organizing structures that native speakers rely on to process speech….Because of their major roles in communication, rhythm and intonation merit greater priority in the teaching program than attention to individual sounds.
Wong's comments reflect an approach that puts all aspects of English pronunciation into the perspective of communicative, interactive, whole language view of human speech.
The most common problem among our English learners is that no matter how accurate the international phonetic alphabet may be, we cannot produce accurate pronunciation owning to the influence of "accent" of our native language.Because some English phones don't exist in our Chinese language at all, so it's quite natural for our students to find some similar sounds to replace the real one.It's quite common as we see in English consonants [r] [z] and so on.
Sometimes, this kind of Chinese 'accent' is understandable for beginner learners but for students of college levels, such errors need to be rectified in order to make their oral English more native-like.Since no Chinese phonetic equivalent exists for students to imitate, they should draw on authentic English listening materials.Since all the materials are read in native-English voice, by focusing their attention on the "right sound', students can distinguish the nuance between 'real' pronunciation and Chinese 'substitute'.Only by doing this as a first step, can students follow the second step of imitation.
Besides improving their pronunciation of words that are difficult to produce, perfecting the relevant features of pronunciation, i.e. stress, rhythm, and intonation is also a very important task for college students who aim at speaking more authentic native-like English.We know that stress, rhythm and intonation in the English language are quite different from those of Chinese, and we have been influenced by Chinese from the day we speak our first language.All the facts prove that it's tough for Chinese learners of English to get rid of the influence of our mother tongue.In addition, different local accents might have various negative effects on students' English stress, rhythm and intonation.Fortunately, college students have laid foundation of English study in middle school and have their own self-teaching ability, they can seek help from appropriate listening and audio-visual materials.
Owing to students' age and their various innate phonetic abilities, imitation is not an easy task for some students at all.However, college students have their own advantages to capitalize on some benefits.For instance, they have clear and higher goals to keep motivating themselves through years of English learning; competitive English learning environment on campus make them pay more attention to their communicative ability; also with in and out of classroom helps from professional teachers, they can get more instructions and guide concerning their pronunciation in oral English; last but not least, college students can access to more authentic listening materials for imitation.Practice makes perfect.The more they listen, and the more they follow the native speakers, the greater progress they'll make in improving their oral English.

Listen to develop cultural competence.
We have been concerned up to now with the ways in which speakers give meaning to utterances by shaping the context in which these utterances are produced and received.We have gone from the premise that meaning is not in the spoken text, but in the dialogue between the learner and the text.In both cases, social and personal voices intersect to create what Nostrand (1989:51) calls 'the central code' of a culture: The central code consists not only of customs and proprieties; it involves above all the culture's 'ground of meaning': its system of major values, habitual patterns of thought, and certain prevalent assumptions about human nature and society which the foreigner should be prepared to encounter.
The term 'authentic' has been used as a reaction against the prefabrication artificial language of textbooks and instructional dialogues; it refers to the way language is used in non-pedagogic, natural communication.As Little and Singleton (1988:21) point out 'an authentic text is a text that was created to fulfill some social purpose in the language community in which it was produced.'As spoken exchanges, authentic texts require participants to respond with behaviors that are socially appropriate to the setting, the status of the interlocutors, the purpose, key, genre, and instrumentalities of the exchange, and the norms of interaction agreed upon by native speakers.
Since Widdowson examined the concept of authentic text in 1979, it has been become a commonplace to say that authenticity does not lie in the text but in the uses speakers and readers make of it.As Widdowson wrote in 1979: 'It is probably better to consider authenticity not as a quality residing in instances of language but as a quality which is bestowed upon them, created by the response of the receiver.Authenticity in this view is a function of the interaction between the reader/ hearer and the text which incorporates the intentions of the writer/speaker… Authenticity has to do with appropriate response.'(Widdowson 1979:166) According to the Oxford English Dictionary (1989 2 nd edn.), the term 'authentic' has at least four meanings: 1) in accordance with a socially established usage or tradition (= from a duly authorized source); 2) entitled to acceptance or belief, as being in accordance with fact (= real, trustworthy); 3) the result of a recognizable communicative intention (= sincere, not supecious); 4) compatible with an identifiable, undisputed source or origin (= original, genuine).
Perhaps one of the main authentic activities within a language classroom is communication about how best to learn to communicate.Perhaps the most authentic language learning tasks are those which require the learner to undertake communication and metacommunication.Communicative approaches to language teaching, whether they be of the functional-notional type of the seventies or one of the proficiency orientations of the eighties, expose learners as much as possible to spoken or written texts that have not been fabricated for pedagogic purposes.It is hoped that, by making communication more authentic, learners will be able to better understand the speaking customs and ways of life of the target country, and thus behave more appropriately in native-speaker environments.( Kramsch,1993) What does the foreign language mean for the foreign language learner?Many things.For example, the obligation to adapt, to repeat the conventionally sanctioned phrases, to play a role, to identify.But it also means being able to compare one's own world of language with that of others, to broaden one's experience with language and language use, to insert some uncertainty into ways of speaking one had hitherto taken for granted; it means border crossing, blockade, disturbance-in sum, to use Humboldt's words, it means 'acquiring a new way of viewing the world.' (Hunfeld 1990:15)

Overview of the design of the study
The studies at Qingdao University of Science and Technology show that there's a correlation between students' listening and speaking ability, from which the author drew the hypothesis that listening might have some effects on improving students' oral English.The use of TSE (Test of spoken English TOEFL) and Oral Proficiency Interview Scale are to assist in the decision of students' oral English ability.Yet till now little research in China has been conducted to test the effects of listening to improve students' oral English.In light of experiments of Qingdao University of Science and Technology, the author assumes that teachers who introduce listening to their oral English classes are likely to have better teaching results.After careful investigating the nature of TSE, we know that it can be used as the criterion in deciding on students' oral English ability.Thus, an empirical study has been conducted to test the truth of the hypothesis.This chapter intends to show the readers the methodology of the study, including the profile of the participants, the instruments, and the scoring procedures.

Subjects
The study was carried out in Qingdao University of Science and Technology.The participants were all second-year students (n=50) of Computer Science majors who were grouped into two different groups.Group 1 is the control class, class 2 is the experimental class.
The majorities of class 2 students were interested in the 'listening' method and practiced it in their one-year English learning process.However, in the real practice, it was a little difficult to ensure that all of the students in class 2 practice the method seriously and persistently.Group 1 students still stick to the traditional methods in oral English class without the help of 'listening', which aims at making their oral English more authentic.

3.3.1
The TSE TSE (Test of Spoken English) designed by Educational Testing Service (ETS) was used in this study.The purpose of this test was to identify the oral English ability of the participants.The test is composed of 12 questions which fall into seven broad catogories; namely, giving directions, recommending places, describing pictures, describing charts and graphs, presenting schedule changes, performing language functions and talking about topics.The participants are required to finish each task within limited time.

The CET (Listening comprehension section)
CET(Band 4) Listening Comprehension Section, designed by the National College English Teaching Committee (the CET committee for short) is also used in this study, aiming at knowing students listening comprehension ability.The CET Listening section includes short conversations, passages and sometimes compound dictations.

Design
At the beginning of the school year, the TSE and CET were conducted on different days within a one-week period and the CET came first.The participants' performance on the two tests was studied by means of descriptive analysis.The relationship between the participants' listening and speaking ability was also discussed by the correlation analysis.

Scoring of the CET
All the objective test items went to the machine scoring; the subject items such as spot/compound dictation were graded by competent markers.

Scoring of the TSE
The score record will consist of one score of communicative language ability, which is reported on a scale of 20-60.Raters evaluate each question and assign score levels using descriptors of communicative effectiveness related to language task/function, coherence and use of cohesive devices, appropriateness of response to audience/situation, and linguistic accuracy.The assigned score levels for each question are average.Because of this averaging, the scores are reported in increments of five (i.e.20,25,30,35,40,45,50,55,60 60.Speaking proficiency is functionally equivalent to that of a highly articulate well-educated native speaker and reflects the cultural standards of the country where the language is natively spoken.
55. Speaking proficiency is regularly superior in all respects, usually equivalent to that of a well-educated highly articulate native speaker.50.Able to use the language fluently and accurately on all levels normally pertinent to professional needs.
45. Often able to use the language to satisfy professional needs in a wide range of sophisticated and demanding tasks.40.Able to speak the language with sufficient structural accuracy and vocabulary to participate effectively in most formal and informal conversations on practical, social, and professional topics.
35.Able to satisfy most work requirements with language usage that is often, but not always, acceptable and effective 30.Able to satisfy routine social demands and limited work requirements.

Data analysis
This chapter will mainly deal with the quantitative analyses of the study.
Since the samples are small (N<30) and the groups independent, the t-test for independent samples is carried out to determine whether the differences between group 1 and group 2 in their mean scores are significant at the 5 percent level.SPSS version 11.5 has been used to compute descriptive statistics and perform Pearson product-moment correlation.Descriptive statistics is conducted in order to examine the participants' performance on each test; Pearson product -moment correlation is conducted to investigate whether there is correlation between students listening and oral English ability.

The collection of raw data
In the correlation experiment between listening and speaking, altogether 20 average students were chosen from Group 1 and Group 2(ten in each group).They were tested on listening and speaking separately and the scores were given by competent markers with the least possible errors.The aim of the tests was to see if there's correlation between students' listening and oral English ability.Therefore, the truth of the hypothesis can be clarified.
Insert Table 1 Here

Data for t-test
The test was given in the beginning of the school year, 50 students (25 students are from Group 1, the other 25 students are from Group 2) participated in the same oral test.And the test scores are shown as the following: Insert Table 2 Here After the period of a school-year, Group 1(control group) and Group 2(experimental group) students were tested on another oral test and the test score for each student is collected.
Insert Table 3 Here

Correlation analysis between listening and speaking ability.
In this section, we will mainly discuss whether there is a systematic relationship between the participants' listening and speaking ability.Pearson product-moment correlation coefficient will be applied, which takes into account the exact magnitude of each score on each variable.The expressions for calculating such coefficient are so devised that a value of +1 is obtained for perfect positive correlation, a value of -1 for perfect negative correlation, and a value of zero for no correlation at all.

Insert Table 4 Here
As we saw from table IV, the value of the correlation coefficient r=0.689, the two variables (listening and speaking) are thus positively correlated.The question which now arises is just how great the correlation coefficient must be in order that we may claim a significant correlation between the variables.We may set up a null hypothesis that there is no correlation between the two variables, that is for the population from which the sample was drawn, the value of 'r' is zero.The question is now, how large must be the coefficient for there to be, say a 5 percent chance or less of obtaining the observed result with a sample of N pairs from a population in which there is actually no correlation?Table I of the Appendix IV gives the critical values of 'r' for various numbers of pairs of observations N.For our language test scores, we have N=20, and the critical value of 'r' at the 5 percent level in a non-directional test (that is, if our alterative hypothesis is simply that the two variables are correlated) is 0.444, while that in a directional test (that is, if our alternative hypothesis is that there is positive correlation) is 0.378.Therefore, whichever alternative hypothesis we have set up, the correlation coefficient is in fact significant at the 5 percent level.That is, there is significant positive correlation between listening and speaking.As regard to the students, those with higher listening ability will, naturally be good at speaking.

Descriptive analysis of students' performance
This section will examine in detail the performance of the participants on the TSE before and at the end of the school year; the performance of the participants on the TSE in the two groups (the experimental group, and the control group) will be discussed and compared.

Pre-experimental data analysis
Insert Table 4 Here When testing the significance of differences between two means, if either or both of N1 and N2 fall below 30, we will use t-test.In the t-test, if the calculated value of 't' is greater than or equal to the critical value as determined from table II Appendix IV, we can reject the null hypothesis.
The null hypothesis: there is no significant difference between means of Group 1 and that of Group 2.
According to the above test, the calculated value t is -1.369.The number of degrees of freedom (df) is (N1+N2-2) or 48.Table II of Appendix IV tells us that a value of 1.684 is needed for significance at the 5 percent level.Since our calculated value (t= -1.369) is much smaller than this, we cannot reject the null hypothesis and we conclude that we have been unable to show a significant difference between the two means, and thus the performance of the groups are, on average, not much different.

Post-experimental data analysis
Insert Table 5 Here The null hypothesis: There are no significant differences between the two means.
The table shows that t= -2.315.The critical value for the 5 percent level and 48df is 1.684.Since the value of 't' exceeds the critical value, we can reject the null hypothesis.As the average score of Group 2 (79.24) is much higher than that of Group 1 (74.20),we can accept the alternative hypothesis: there are significant difference between the two means, and conclude that we have been able to show a significant effect of the experimental condition on the performance of the oral English test on average.We can also observe the result from table III.For most students in Group 2, their scores are much higher than those in Group 1.In a word, Group 2 students perform better than Group 1, which proves that listening to more authentic materials can help to improve students' oral English.

Findings based on correlation analysis
As we all know that listening and speaking are closely interrelated, that's the reason why listening and speaking courses are usually incorporated into one.Does a learner's listening ability correlate with his/her speaking ability?That's the question haunting on the author's mind.If the hypothesis is true, then we can say that listening and speaking have correlations, and thus the later research can be based on this theoretical foundation.
From the data collected for correlation analysis, we found that students' listening and speaking scores are listed in two separate lines.SPSS version 11.5 has been used to perform Pearson Product-moment correlation, which is conducted to investigate the relationship between listening and speaking.The author of this thesis made two opposite hypotheses (the null and alternative hypotheses), which needed to be verified by quantitative analysis.From the analysis in 4.2 we concluded that there is significant relationship between the participants' listening and speaking ability, which shows that if one's listening score is high, her/his speaking is good in general, and vice versa.

Findings based on descriptive analysis
The raw data for t-test in 4.1 offered us participants' scores in Group 1 (control Group) and Group 2 (experimental Group) before (table A) and after (table B) the experiment.5.1.2.1 Pre-experimental comparison between Group 1 and Group 2 According to the analysis result drawn from the first oral English test(pre-experiment test), we see that although the mean score( 75.20) of Group 2 is a little higher than that of Group 1(72.52),t< the critical value.So we accepted the null hypothesis and concluded that the performances of the two groups are not much different.

Post-experimental comparison between Group 1 and Group 2
In order to test the effects of listening on improving students' oral English , another test should be designed to test students' performance in both Group 1 and 2 separately after the one-year experiment period.Compared with the result of the first test, the second oral English test indicated significant differences of students' performance in the two group: students in group 2(experimental group) got significantly higher scores than Group 1 students did.

Summary
From the above test results and data analysis, we can draw a conclusion that listening does have some positive effects on improving students' oral English.The author based her research on the correlation analysis between students' listening and speaking ability and made the hypothesis: since there is a correlation between the learner's listening and speaking ability, the students who score higher in listening might obtain higher score in oral English test.By doing the experiments among two groups of students, the author of the thesis verified the truth of her hypothesis.Therefore, if we insert more listening and audio-visual materials into our oral English class, students can not only improve their listening ,but also as learners they can learn skills and knowledge from the native speakers, correcting their mixed English and getting closer to native-like authentic English.

Implications of the study
The result of this study show that introducing appropriate listening and audio-visual materials into oral English class can bring in better teaching and learning results.And this study is of great importance to universities/colleges at the similar level.If we change our traditional ways of oral English teaching, and combine audio-visual means and oral practice into one, then a better result might be achieved.First of all, more vivid materials might be accessible to our students, which provide students with more authentic linguistic and cultural knowledge to learn.Second, our English teachers can easily find various materials to fit for students at different levels.Most important of all, the atmosphere in our class will be more active, and students' interests in English learning will be greatly stimulated.

5.3Limitations of the study
Owing to the limitations of research time, the experiment period only lasts for a school year, which is too short to prove the truth of the hypotheses from a strict scientific approach.However, it is the longest possible time the author can have.And also the author of this thesis relies her analysis only on one single test to get the result of the research, which is far from enough.Actually, at least two or more tests should be carried out to verify one point, because of the busy schedule of both the teachers' and the students', the author failed to do so.
In addition, the author of this thesis chooses altogether 50 students as samples of the experiments, and they are in two different classes of the same major.The small sample size is not very persuasive to prove the truth of the result, however, it is the biggest effort the author was able to do.If more students participate in the said experiments, the author's research result can be more persuasive and acceptable.

Conclusion
Listening and speaking have been very essential in communication and therefore, very significant in English language teaching.The skill of listening with comprehension is an essential part of communication and basic to foreign language learning.The ability to listen to English effectively is very significant, because good listening is also an important step to good speaking.Listening consists of reciprocal listening which refers to those listening tasks where there is the opportunity for the listener to interact with the speaker, and non-reciprocal listening which refers to tasks where the transfer of information is in one direction, only from the speaker to the listener.As listeners, we do not simply take language in like a tape-recorder, but interpret and get information from what we hear according to our purpose in listening as well as our learned knowledge.
Oral English, as distinguished from written English, consists of short, often fragmentary utterance, in a range of pronunciation.There is often a great deal of repetition and overlap between one speaker and another, and speaker frequently use non-specific references.According to Brown and Yule, there are two basic functions of oral English.
They are the transactional function, which is concerned with the transfer of information, and the interactional function, which has the primary purpose of maintaining social relationships.According to the development stage of speaking skill, we can make a basic distinction between dialogue and monologue.While most English learners at college level can use English at the first stage, few of them can do at the second stage.This is the skill which needs to be learned and practiced for college students who aim at improving their English.
During the early 1980s', there was much talk of listening-based methods of English teaching, and classroom research has confirmed that there are distinct advantages to listening-based methods.One of the major schism in contemporary teaching methodology is between those who prefer students to listen for information without speaking and those who require students to practice communication by both listening and speaking.
Krshen brings listening-based methods together through the notion of 'Comprehensible Input'.He claims that acquisition can take place only when people understand messages in the target language.Listening is motivated by the need to get messages out of what is heard.Foreign language learners acquire a new language by hearing in contexts where the meaning is made plain to them.Ideally, the speech they hear has enough 'old' language, (i.e.i) and makes enough sense in the context for the new language(i.e. 1) to be understood and absorbed.In contrast with listening, speaking is an output process.Krashen's Input Hypothesis shows us the significance of listening to speaking as well as the way of choosing appropriate listening and audio-visual materials for our oral English class.
Appropriate listening and audio-visual materials can make students' pronunciation more native-like, because the materials chosen can offer students perfect native voice instead of fabricated ones from other sources.Also by getting access to authentic, real life listening and audio-visual materials, students can develop their cultural competence which enables them to respond with behaviors that are socially appropriate to the setting, the status of the interlocutors, the purpose, key, genre, and instrumentalities of the exchange, and the norms of interaction agreed upon by native speakers.In a word, learners will be able to better understand the speaking customs and ways of life of the target country, and thus behave more appropriately in native-speaker environments.
One test is designed to verify if there is a correlation between students listening and speaking ability, and the other two tests are to show the effects of listening on students' oral English.By analyzing the data collected from the tests we draw two conclusions: 1) there is a significant correlation between students' listening and speaking ability.2) listening to more appropriate authentic materials can help to improve students' oral English.Therefore, it is advisable for college teachers to bring in more authentic listening materials into their class, and for students to try every possible means to get access to positive listening materials.
The research carried by the author is only a tentative one, so it can not be free from limitations.The author of this thesis calls for more work in this field by foreign language teachers and researchers.
Scale can help to provide more detailed description of the participants' oral production ability.(DuZihua and Jordan Singer,2001)

Table 4 .
Correlation between speaking and listening

Table 5 .
Raw oral test results 1