I. Introduction

In recent years, college English curriculum reform is in full swing in colleges and universities all over the country. The direction of reform has shifted from the traditional classroom that emphasizes reading and writing ability to the listening and speaking classroom that emphasizes practical application, especially increasing the proportion of oral English teaching in college English courses. Oral English course is no longer attached to reading and writing courses for non-English majors, but is set as a separate course. Then, how to correctly evaluate oral fluency and how to design oral examination scheme has become an important topic that can not be ignored in college English teaching. Although English test has been developed for hundreds of years, oral test started late, so it is difficult to guarantee its reliability and validity. At the same time, it is also very difficult to design a large-scale oral test task suitable for undergraduate non-English majors. Take the oral English intensive reading course for freshmen in Xiamen University of Technology as an example. Students are very interested in learning spoken English and have made great achievements. However, most students are critical of the final evaluation plan. Because, at present, the final oral exam is to test the students' oral output level by the subject teachers alone, including foreign teachers and Chinese teachers. What are their evaluation criteria? Are the standards consistent? Whether the evaluation results are credible and effective, we have no way to verify. The ultimate goal of this study is to improve the effectiveness of the final oral English exam for non-English majors in our school and their trust in the oral English exam. Scientific and effective test methods and fair test results play a positive role in stimulating students' enthusiasm for learning English and improving their oral English. Based on the previous oral test model and related research results, this study puts forward an oral test system suitable for our school.

Second, the main factors affecting the reliability and validity of oral test

Reliability and validity are two important indexes of large-scale standardized examination, and of course they are also important criteria for the effectiveness of oral examination. However, at present, the oral English test in China is mainly conducted through interviews and other direct oral tests, and the results are only based on the examiner's personal judgment, so the reliability of the oral English test results is poor. The scoring standard of oral test is still based on students' oral fluency and phonetic characteristics. In fact, in the process of oral evaluation, communicative effect and communicative efficiency are equally important. Compared with objective multiple-choice questions, the interactivity and flexibility of oral test cannot guarantee that there will be no individual differences in the scoring process. The existing oral test research pays great attention to the reliability of testers and ensures the consistency of scoring standards among testers.

Another core issue of oral test is its effectiveness, and the design of oral test should be carried out around it. The main factors affecting the validity of oral test are the types of oral test tasks, test methods and scoring standards. According to Koloughlin's research, candidates will feel more formal when speaking into a tape recorder. However, many candidates are obviously more anxious, because they realize that what they say will be recorded, and the recorded oral test has an obvious disadvantage that candidates have no expressions and gestures when communicating. The goal of our oral test design is to show students' oral English to the greatest extent, reflect the real communication situation and fully reflect students' oral communication ability. Then the examiner scores the obtained phonetic templates according to a unified scoring standard, such as the scoring standard of Cambridge Business English BEC, which is a good model.

At present, the main forms of oral examination in China are interview (direct oral examination) and recorded oral examination (semi-direct oral examination). Both forms have their advantages and disadvantages. The advantage of the interview is that the examiner can communicate directly with the subjects face to face, and can judge the oral level of the subjects more intuitively and directly. But the result of the interview has a lot to do with the examiner. For example, the examiner's subjective evaluation criteria, the examiner's education level, the examiner's own language level, and the examiner's mood on that day will all become factors that affect the test results. It is impossible for the same examiner to complete a large number of subjects in a short time. In other words, although the interview oral test is highly valid, it lacks reliability, which is not conducive to the development of large-scale examinations. The recorded oral test is more suitable for large-scale exams. For example, at present, this kind of examination form is adopted in the domestic TEM-4 and TEM-8 examinations. However, the recorded oral test does not reflect the communicative nature of language, and it is more like a person talking to himself, lacking the scene simulation of daily communication. The communication process of language should be that in a certain social environment, the speaker conveys his thoughts, attitudes and emotional words to others. Therefore, in order to achieve the effectiveness and authenticity of this communication, the examiner should participate in the communication process and interact with the subjects. In addition to observing candidates' verbal communication, candidates' body language should also be included in the scoring range.

Third, the design and implementation of the evaluation system.

(A) oral design

What English abilities do non-English majors need? What standards need to be met? In order to solve these problems, the author interviewed the staff of several Sino-foreign joint ventures in Xiamen, and found that what students learned in school can hardly be applied to practical work. When foreign customers visit or communicate abroad, they need to communicate daily in English, such as booking hotels and air tickets and introducing the basic situation of the company. It is difficult for them to speak. They have the following requirements for students' oral English: 1. Accurate and fluent language expression; 2. Express your suggestions clearly; 3. Be able to have simple daily life conversations. These requirements will be added to the scoring standard of our oral English test to ensure that the oral English level of our undergraduate graduates is more in line with the needs of social development.

According to the current situation of the oral examination for non-English majors in Xiamen University of Technology, the oral examination for non-English majors in our school is subject to the following restrictions: 1. There is no unified spoken English question bank, and the test papers are all given by the English teacher of this class; 2. Each teacher completes the interview by direct oral examination, which makes the reliability of the test greatly influenced by subjective factors; 3. The average class size of each teacher is about 150, and the oral examination is required to be completed within one week, with concentrated examination time and heavy workload.

Accordingly, the oral test design of Xiamen University of Technology students (at the end of last semester) is as follows:

① Oral test form: interview-style direct oral test is adopted, with one tester and two students in each group. The tester guides the students to complete the oral test according to the process and records the whole process. After the exam, three English teachers graded the students according to the performance in the video. In principle, there are two teachers from China and one foreign teacher. None of the three teachers is the teacher of the examinee. The average scores of the three teachers are taken as the final oral test scores of the candidates.

② Oral test content: It is divided into two parts. One part is that the examinee completes a 2-minute impromptu speech according to the picture information, and the other part is role-playing. This part of the investigation mainly comes from real life scenes, such as hotel reservation, invitation and rejection, and is limited to 3 minutes, which is completed by both candidates. In addition, this study also added a film dubbing link, and the teacher arranged the film clips in advance and sent them to the students to ensure that the students were fully prepared before the exam.

Compared with traditional reading tasks, film dubbing is closer to life and can stimulate students' interest in learning. In addition, there is a good pronunciation model for students to imitate and learn, and it can also achieve the purpose of testing students' pronunciation. The computer and multimedia projector equipped in the school classroom can basically ensure the smooth progress of this exam, accounting for 20%, while the second exam mainly examines students' critical thinking ability, language organization ability and expression accuracy. In this link, students' personal oral English ability will be fully demonstrated, which requires students' comprehensive level. So this part accounts for 30% of the final overall evaluation results. The last link focuses on students' oral communication ability. Two students in the same group have a dialogue according to the daily life scenes provided by the teacher. Because communicative competence is the focus of our oral test, this link is the focus of the whole oral test, and the score is also the highest, reaching 50%. The daily life scenes chosen by teachers should belong to the teaching scope of this semester, so as to ensure that students have something to say.

(2) Scoring standards and procedures

At present, the scoring methods used in oral English test at home and abroad mainly include overall scoring and analytical scoring. Sub-item scoring is to divide the oral expression of candidates into signs such as pronunciation, intonation and vocabulary, and the total score is the final score of candidates. Most of the oral English tests for non-English majors in Xiamen University of Technology are graded. Its characteristics are that testers have a unified scoring standard, simple operation and high reliability. However, the disadvantage is that the overall oral level of candidates is subjectively divided, so the scoring can not reflect the true oral level of the subjects, and the scoring process takes too long. Overall score, also known as impression score, is also a commonly used scoring method in oral test. Testers make an overall evaluation of the subjects' oral English according to a unified scoring standard. The famous IELTS speaking test abroad adopts the overall scoring model. Therefore, the oral test of non-English major freshmen in our school also adopts the overall scoring model. Referring to the scoring standard of IELTS speaking test, this study divides students' scores into 9 grades according to their overall speaking level.

In this experiment, the teacher keeps the order of the examination room, ensures that the examination is conducted in an orderly manner, and videos the examination process. Then three English teachers who are not in this class score the students' performance in the video as a whole, and take their average scores as the candidates' oral scores. This will not only fail to complete a wide range of exams in a short time, but also make the grading process too hasty and unreliable. Teachers who take part in testing and grading should attend the awareness meeting of unified grading standards to ensure the fairness and justice of unified grading standards and grading.

Four. conclusion

In this experiment, the above-mentioned oral examination system was used for the oral examination of some students (related to mechanical, civil and cultural communication majors) of Grade 20 15 last semester. After the reform, due to the increase of videos and the comprehensive evaluation of many teachers, the reliability of oral English test has been greatly improved, and the enthusiasm of students to learn oral English has been further improved. However, due to the need for three English teachers to correct at the same time, it takes a lot of manpower, and if all non-English major freshmen use this test scheme, they need to use multiple photographic equipment at the same time. Therefore, although this oral test mode is reliable and effective, and it is also suitable for students' oral English level in our school, how to popularize it in the whole school deserves further study.

References:

[1] high. A Review of the Research Hotspots of College Oral English Test [J]. Theory and Practice of Foreign Language Teaching, 2008, (1): 92-96.

[2] Liu Qin and Hu. Investigation on the needs of science and engineering students in oral English teaching [J]. Journal of Xi International Studies University, 20 10, (3): 86-89.

[3] Luo Kaizhou, Han Baocheng. Comment on the automatic scoring system of Order and SpeechRater [J]. Foreign language audio-visual teaching, 20 14, (7): 27-32.

[4] Wang Jie. On the Construction of a Reasonable Mode of College English Oral Test [J]. Journal of Qilu Teachers College, 20 13, (2): 134- 136.

;