Luận văn An Action Research on the Use of Continuous Feedback to Improve the First Year Students’ English Pronunciation at the English Department, College of Foreign Languages, Vietnam National University, Hanoi

researchers to gather information about the ways that their particular school operates, how they teach, and how well their students learn. The information is gathered with the goals of gaining insight, developing reflective practice, effecting positive changes in the school environment and on educational practices in generals, and improving student outcomes.” Kemmis and Mc Taggart (1988) identify three characteristics of an action research, which are: “carried out by practitioners”, “collaborative” and “aimed at changing things”. A typical action research needs to go from one step to another. Eileen Ferrance (2000) suggested an action research cycle as follows: Figure 1: Action research cycle According to her cycle, the first step of an action research is to identify the problem. Then, data must be collected through means of interviews, portfolios, questionnaires, etc. Next, the data is analyzed to find out problems. The fourth step is to design a plan of action that will allow the researcher to make a change and to study that change. After the intervention has been carried out, it is the researcher’s task to evaluate its effects in order to justify whether the intervention worked or not. Then, if it has not been very effectively, the researcher would have to consider another method so as to improve the situation. Another researcher that has conducted a thorough investigation into action research, Nunan (1992), defines the framework of a research as consisting of seven steps: Step 1: Initiation (Identify the problem) Step 2: Preliminary investigation (Collect data through a variety of means) Step 3: Hypothesis (Develop research questions) Step 4: Intervention (Devise strategies and innovation to be implemented) Step 5: Evaluation (Collect data again and analyze it to work out the findings) Step 6: Dissemination (Report the result by running workshops or issuing a paper) Step 7: Follow-up (Find alternative methods to solve the same problem) Action research is conducted widely due to its advantages that cannot be denied. When comparing it with other kinds of research, we can see its advantages more clearly. Table 2: Differences between Action Research and Formal Research Topic Formal research Action Research Training needed by researcher Extensive On own or with consultation Goals of research Knowledge that is generalizable Knowledge to apply to the local situation Method of identifying the problem to be studied Review of previous research Problems or goals currently faced Procedure for literature review Extensive, using primary sources More cursory, using secondary sources Sampling approach Representative sampling Students or clients with whom they work. Research design Rigorous control, long time frame Looser procedures, change during study; quick time frame; control through triangulation Measurement procedures Evaluate and pretest measures Convenient measures or standardized tests Data analysis Statistical tests; qualitative techniques Focus on practical, not statistical significance; present raw data Application of results Emphasis on theoretical significance Emphasis on practical significance ( One of the most significant features of an action research is it is very situational, which means it appears to provide specific solutions to specific problems in each context. While other types of researches focus on experimenting or testing a pre-assumed theory or concept, action research aims at “improving the situation”. Therefore, it is very beneficial to the teaching and learning process. When thinking of choosing an appropriate methodology, the researcher, who is also a teacher, found out that action research is the best choice for the purpose of improving the students’ pronunciation. First of all, action research means “act upon research”. It is impossible to assume pronunciation problems that students are having without having adequate amount of observation. An action research enables the teacher to have a thorough investigation on the pronunciation problems of students on a daily basis. Thanks to this, the findings would be up to date and mostly reliable. The teacher can have a thorough look and is able to work out a detailed analysis on the students’ performance during the progress of the research. In short, an action research helps the teacher design a detailed program which is most suitable to the class and therefore benefits students to the most. This program will be consulted for building up a more suitable pronunciation course for students who share the same problems as the subjects of the research. Secondly, in this action research, the data is taken from every student and treated individually. Therefore, the teacher can see students’ progress easily. S/he will be able to get the most up-to-date information and make sensible changes if necessary to the research while applying it to meet the demand of the students. For pronunciation mistakes that persist for a long time, the teacher can spend more time on it by giving more practice time in class, as well as paying more attention when giving feedback to students. Performance of students is collected weekly, and any potential problems can be solved in time. Last but not least, by using an action research on the use of continuous feedback to students individually, the researcher gives the student a strong sense of themselves. This is typically different from other types of research in which the teacher treats all the students in one class the same way. In this case, data is carefully collected and analyzed for each student. When feedback is provided individually, each student can receive a thorough care from the researcher. The students’ awareness as well as their gratefulness for the teacher’s effort will serve as a strong motivation to make progress in their pronunciation. Despite the obvious advantages, the research has some shortcomings compared to other types of research, such as an experimental one. It is sometimes claimed that action research has limited application. This comes from one of its typical characteristic, which is situational. This means the research program may work well in one class but cannot be applied to other classes in different context. If the same research is carried out in another class, the result may not be effective as expected. In other words, action research lacks generalization. However, the reality is that this action research is very applicable in a wide range of classes. This is explainable by two reasons. Firstly, students at the English Department were sorted in different classes randomly. There were no categories to classify them into different classes, such as basing on their hometowns, or the marks acquired at the entrance university exam. Therefore, each class consists of students coming from many parts all over the country. This also means students who learned in the same high schools may be scattered over a number of classes at the department. Therefore, they may have the same pronunciation problems rooted from the same way of teaching method. In addition, due to random sorting, it can be assumed that there is an equal division of students’ level of English proficiency in different classes. Obviously, there may be the possibility that some classes may have more students whose English is better than others. However, this is the common practice in any language classes. Therefore, it should not be treated as a major cause for the impossibility of generalization. 3.3 Background of the study 3.3.1 Participants The participants of the research were selected on the basis of cluster sampling. The researcher, who is also a teacher, was in charge of teaching speaking and pronunciation for first year students in group 07.E9, English Department, CFL, VNU during the first semester. Naturally, they became ideal samples for the research. With the method of cluster sampling, the research ensures the variety of the students’ background, which varies from one to another as the group was chosen by chance. This also enables for a wider range of application to other classes, which, to some extend, shorten the limitation of an action research. However, as the research was conducted during the first semester of the students’ first year, there were some problems relating to the consistency of the participants. After the first two weeks of the first semester, a competition was held at the English Department to select best students for the Fast-track group. Some students in the experimented group attended the competition, passed and then they moved to that group. Thus, they were not able to attend the research program any more. In turn, after some weeks, some new students became members of the experimented group as a result of the redistribution of the students in the department. These students could not be proper participants of the research program due to their lack of participation in the very initial steps of it. In addition, as the program was conducted extensively during the first semester, with a lot of extra-class time, in which students had to come to school to take part in audio recording sessions even when they had no class at school, some students missed some sessions due to their personal businesses. Thus, their performance should not be treated as a trusted source of information. Taken every factor into consideration so as to minimize the risk of invalidity and to ensure that every single participant was treated under the same conditions of the research, the researcher reached the final number of 20 students whose performance would be appropriate data provider for the research. They were the one who participated thoroughly in the research from the beginning to the end. In order not to make other students feel excluded, the teacher still let them participate in the research program. Nevertheless, their performance was not counted on to yield any comments or interpretation of the research. Further observation, investigation and talks with the students revealed more information about them. Most of the students are of the age 18-20 and many are from the countryside. When entering university, they faced many problems in life, not only in getting acquaintance to a new lifestyle and living environment, but also with the new style of studying in university, which requires a great amount of self-study. Therefore, they needed great help from the teachers. As for language competence, some of the students were from gifted English secondary schools, but many were not. The university exam comprised of three subjects, namely mathematics, literature and English, among which the mark for English was doubled. This also meant some students who got a low mark in English, but high marks in mathematics and literature were still eligible to enter the university. In general, the students’ reading skill and grammar were quite good. However, their listening, writing and speaking skill, including their English pronunciation were of a lower level. The good news was the students were willing and determined to improve these weak skills. 3.3.2 The speaking and pronunciation programs A semester at the English Department lasted 15 weeks. Four periods per week (accounting for 180 minutes) was devoted to the teaching of speaking skills and pronunciation. The pronunciation period was often the last period, which lasted for 45 minutes. As the tasks designed for the pronunciation period were closely related to the speaking periods, we would like to present the content of both speaking and pronunciation programs. * Objectives The objectives of the speaking and pronunciation programs were stated clearly stated in the course-outline of Division 1, English Department, CFL, VNU. As for speaking, by the end of the semester, students would be able to: - function in social and personal interactions with considerable confidence - conduct a short discussion on common topics using proper verbal and non-verbal language - have built up basic skills of presentation As for pronunciation, the course aims at building students’ pronunciation in a practical way. It was stated that, “By the end of the first semester, students will be able to have relatively correct pronunciation of vowel sounds, difficult consonant sounds and ending sounds, so that the listeners can understand.” * Materials: The textbook that was used to teach speaking skills for first year students at the English Department was Inside Out (Pre-Intermediate). As for pronunciation, the name of the course-book was Practise Your Pronunciation and Speaking Skills. Practise Your Pronunciation and Speaking Skills was designed by the teachers of Division 1, English Department and had been used for teaching pronunciation for the past three years. The pronunciation exercises in the book were designed in a practical way, in which all the words are taken from the correlative units in Inside Out. Each unit consists of six pronunciation exercises, aiming at helping students familiarize and be able to pronounce the words they have encountered in their speaking lessons. * The pronunciation syllabus All exercises of each unit in the pronunciation course-book must be done beforehand by the students at home. During the pronunciation class, three or four of the exercises, chosen by the teacher, would be corrected in the class. The choice over what exercises to correct in the class depended largely on the teacher, according to what aspects of pronunciation s/he thought the students were weak at. Normally, some students were called to do the exercises. After the teacher gave feedback, the typical activity was reading aloud, either individually or as a whole class after the teacher. The teacher was free to choose any other extra class activity so as to make the lesson more interesting. A typical feature of this syllabus, which made it different from traditional pronunciation course, was that there was no pre-determined focus of each pronunciation lesson for each week. The words that students were to practised were taken from another book based on their appearance. From week 11 to week 14, students were to do pronunciation assignments by groups of 4-5. They were to read a passage chosen by the teacher at home beforehand, record it, and then bring to the class on the day of performance. The students of the group would have to analyze particular features of pronunciation they aimed at when reading the passage (normally suggested by the teacher), and make the whole class aware of those features. In other words, they played the role of the teacher in guiding the whole class through the lesson. * The speaking and pronunciation test Tests of speaking skills and pronunciation were incorporated into one test delivered at the end of the first semester under the name of the Speaking Test. Students took the test in pairs. The test consisted of three parts: 1. Social interaction (2 points) Students introduced themselves to the examiners or introduced themselves to each other. This was aimed at checking students’ ability to make initial interaction. 2. Reading aloud (3 points): Students were required to read aloud a passage of about 70-100 words. This passage was taken from Inside Out (Student book). This part was aimed at checking the students’ pronunciation competence. 3. Topic (4 points): Students were asked to discuss one topic with each other. The topics were within the content of the speaking units in the first semester. Students’ manner was assessed during the whole test and if the students were marked as polite, friendly and natural, they would get the last point to make up the maximum mark of ten. The students’ performance was assessed based on the following criteria: * Accuracy * Fluency * Manner * Timing The marking scale showed that pronunciation was one of the major focuses for the first year students. As we can see, the second part (reading aloud) accounted for 30% of the whole speaking test. In addition, students’ pronunciation also contributed to the other two parts of the test (social interaction and discussion), because it would be assessed against the Accuracy criteria. It was natural that students would not be able to gain high points in other parts of the test if their pronunciation was not good. Therefore, actually, pronunciation accounted for more than 30% of the test. This reflected the concern of teachers about the students’ pronunciation at the very first year at university. 3.4 Instruments In order to ensure the validity of the research, a number of different instruments were implemented, which were largely qualitative. * Observation As the researcher herself was the head teacher of the class whose students participated in the research program and was also in charge of teaching speaking and pronunciation, this provided a perfect opportunity for her to have a thorough observation on the performance of the students, particularly their English pronunciation. As for action research, observation is obviously one of the main instruments because it provides the researcher with a deep insight into how the research program is going. Regarding this study, observation was conducted through the whole research program. The researcher paid great attention to the students’ pronunciation not only in pronunciation class but also in speaking class. This helped her find adequate amount of data to yield any conclusions about the students’ pronunciation problems, as well as how much progress they made during the research. * Questionnaire A questionnaire was conducted at the first week of the research. This aimed at collecting personal information of the students, particularly that related to their studying at secondary schools and their attitude toward learning pronunciation (See appendix). By doing so, the researcher was in the hope of finding possible factors that affect students’ pronunciation. The questionnaire, in cooperation with some interviews, which were taken later, could help the researcher identify possible reasons for students’ weak pronunciation. * Interviews Two interviews were conducted with individual students throughout the research so that the researcher could be able to find explanation for the students’ choices when they answered the questionnaire, and get a deeper understanding of the pronunciation problems that the students made. The first interview was delivered right after the students did the questionnaire. It was aimed at finding evidence for the students’ choices over some questions, such as “My phonetic ability is not good.”, or “My pronunciation is affected by my dialect.” In the interview, the researcher asked the students to explain their choice by providing evidence for that. For example, when a student coming from a central province of Vietnam claimed that her English pronunciation, especially her intonation, was badly affected by her dialect, the researcher checked this by asking her to speak freely about a topic in English. In this way, she could have a closer look at whether it was a real problem that the student had. Another interview was conducted at the end of the research in the hope of finding reasons for the students’ remaining problems. We were fully aware that there might be differences between the pronunciation performance of different participants even though they all participated in the same research program. Therefore, an interview was necessary to justify those differences, which was much beneficial to work out a more effective pronunciation program to apply to other classes. * Mock speaking tests (Audio recording) During the research program, the students were asked to attend two mock speaking tests. The first test was conducted in the second week, and the second test was delivered in week 15. They were designed according to the real speaking test’s model, excluding the first part (social interaction). The two tests were audio recorded and then compared so as to reveal any improvements that students’ made during the progress of the research. There are some reasons for the choice of using mock speaking tests. Firstly, students should be familiarized with the speaking test’s format during the first semester because it was used to assess their speaking and pronunciation. It was the teacher’s task to inform students about the format of the speaking test right at the beginning of the semester so as students were well prepared for it. It might be argued that if the mock speaking test was conducted right in the second week of the research program, the students would not have enough familiarity, as well as preparation, for it. Therefore, their results might be affected badly and it was naturally expected to be worse than the second mock test (conducted at the end of the research program). This might lead to inappropriate conclusions about the effectiveness of the program. However, in fact, this was not problematic, because the researcher only focused on the students’ pronunciation of English consonant sounds. Other aspects of their speaking skills such as manner or fluency were not taken into consideration. Therefore, it would not pose any problems regarding the different time in which the mock speaking tests were distributed. Secondly, the format of the mock speaking tests allowed the researcher to have a thorough look into the students’ pronunciation. The tests consisted of two parts: reading aloud and discussion. For the first part, by providing an appropriate passage for the students to read, the researcher was able to identify almost every pronunciation mistakes regarding consonant sounds that the students made. For the second part, she could check the students’ pronunciation in their real-life speaking. Last but not least, it was widely acknowledged that motivation had a lot effect on the students’ performance. When the mock speaking tests were designed in exactly the same way as they were conducted in the real life, students would be more willing to participate in it. They would also try to have a better preparation for it, which, to some extend, may help reduce the negative effect of fluency over accuracy. This helped us gain a more accurate understanding about the students’ pronunciation of single English sounds. * Passages (Audio recording) Students were asked to read aloud four passages during the research program. This was taken outside the class. This was aimed at finding students’ typical pronunciation mistakes regarding English consonant sounds. The use of reading-aloud exercises has been criticized as unrealistic by a number of researchers. They argue that the spelling of a word may have a negative impact on students’ pronunciation of it. Furthermore, with other aspects of pronunciation such as intonation, rhythm, elision, etc., reading aloud is certainly not a good choice because it makes students “speak in an unnatural way”. Speaking spontaneously obviously reflects more accurately the actual speaking skills of the students. When speaking spontaneously, students are free to choose what they are going to say, and therefore can adopt their rhythm and intonation more properly according to what they want to focus in their speech. This is hardly achieved in reading-aloud exercises where everything is far beyond the control of the students. They are asked to read certain words, phrases and sentences which may not be of their interest; and sometimes it is difficult for them to justify what the main purpose of the writer is. However, when the focus is on the pronunciation of single English sounds, the problem of using reading-aloud exercises seems to be solved perfectly, especially when this kind of exercise is used to enhance students’ speaking performance, not to assess it. As stated in previous chapter, students’ awareness plays an important role in their pronunciation. Therefore, with this type of exercise, students are fully aware that they have to concentrate on the pronunciation of the words and try to pronounce correctly. They do not have to worry about other aspects of connected speech. Therefore, they tend to read more accurately. If this type of exercise is practised at a regular basis and recorded so that students can identify their mistakes, it is expected that the pronunciation of the single sounds will be carved deeply in their minds. Over a long period of time, it is very much hoped that students will be able to pronounce the sounds correctly in their natural speaking, which serves as a basic ground for the promotion of other aspects of connected speech. The passages were taken from authentic audio listening exercises in an IELTS series. This would make it an easier task for the teacher when comparing the accuracy of the students to the target source, with the focus mainly of the pronunciation of single sounds. The passages chosen were t