Introduction

On arrival in Thailand in November 2003 as a teacher at Dhurakijpundit University I was allocated, among other subjects, two classes of "Beginner English Conversation Skills". At the outset I explained to my students that 'conversation' meant 'talking', and that meant them talking with each other and with me. However, it proved to be very difficult, and in some cases impossible, to get students to participate in any meaningful way. Almost no student would voluntarily answer a question generally addressed to the class. Few would attempt to answer a question directed to them. A common response would be for the student to turn to friends and talk in Thai before offering a monosyllabic 'yes' or 'no' in reply to the question. Paired-students would come to speaking tests armed with a prepared written conversation, which they would then read. In short, spontaneous, 'meaningful' conversation was non-existent.

I learned from Thai fellow teachers that many of the students were the product of traditional teacher-fronted schooling where the teachers did the talking and the students listened and they were simply ill-prepared for the active dialogue required of a university English class.

The following semester I was allocated an intermediate conversation skills class, which included a large number of my previous students from the beginner class. Faced with the prospect of another four months frustration, I radically changed the syllabus evaluation system in order to experiment with an instant reward and recognition system to encourage classroom participation.

Literature Review
The literature abounds with reports of how widespread this problem is in Asian ESL situations.

The underlying problem of getting Thai English learners to speak English was highlighted in a recent Suan Dusit Poll of 1,024 residents of Bangkok and adjacent areas (The Nation, 2004). Of those polled, 95 per cent said English was crucial to their careers and a further 5 per cent said English was rather important 'for basic communication skills', a total of 100 per cent. Despite this, 43 per cent said that in communicating with English speaking foreigners they were 'afraid to communicate or make mistakes' and a further 42.33 per cent admitted to a 'lack of daily use' as a major problem. Fear of speaking or making mistakes (31.58 per cent) and lack of personal expertise (18.52 per cent) were the major factors cited as contributing to the problem of learning English.

Hadley (1997) and Jeffrey (2003) discuss the situation in Japan and report on a participation points system to encourage classroom communication that they trialed and implemented at several Japanese universities and colleges. They hand out tokens, poker chips or marbles to instantly recognize a student's participation in class. These are 'cashed I' at the end of each class for a recorded credit towards their final assessment. The merits of this simple system are that the students get positive and instantaneous feedback, they overcome their passivity and participation becomes motivational (Jeffrey, p.1). He reports that it is one of the simplest, yet most effective, techniques that he has learned as an ESL teacher.

The following case study was based on the models implemented by Hadley and Jeffrey. The objective was to determine if the results they reported could be replicated in the Thai cultural setting, and more particularly at a Thai private university.

Methodology
The experiment was run across two 'Intermediate English Conversation Skills' classes, Group 1 by a highly fluent English-speaking Thai teacher, Group 2 by me. Class numbers were 31 and 30 respectively.

A distinctive yellow 'Credit Certificate' was given to each student at the start of the 30-class semester on which they could receive a 'credit stamp' worth half a mark for class participation. Up to three credit stamps a class were accredited to the students' assessment, to a total of 40 per cent of the final mark. Other assessment factors were attendance (10 per cent), self-access learning in computer laboratories (10 per cent), and mid-term and final listening and speaking tests (10 per cent each). This represented a significant shift from the previous system where 80 per cent of assessment was through a series of listening and speaking tests which, based on my observations and experiences the previous semester, were highly artificial and did not represent 'real-world' listening or speaking skills.

The experiment was conducted in close co-operation and consultation with the Thai teacher of the other class; monthly monitoring of the students' accumulated scores; maintenance of diary notes; an end-of-semester student survey; and analysis of the final assessment scores.

Implementation
The new system of assessment was carefully explained to students at the first class and they were given their 'Credit Certificates', which they retained to give them 'ownership' of their own progress during the course.

Initially I was liberal with credit stamps, both to demonstrate how the system would be put into practice and to encourage students to start speaking up. Some students were quick off the mark and started accumulating points; others were slower. In week three I explained that credit stamps were now going to be harder to earn and that students had to engage in more meaningful and complex interaction to earn points. The primary driver of conversation was the course book, Let's Talk 3 (Jones, L. 2002). Each class we would work through a unit, playing the voice tapes and doing the individual and pair-work exercises. Simple answers comprised summarizing what they had heard on the voice tape. More complex responses were required when the students did pair-work exercises on such things as their own character trait or preferences. In such cases I would require each team member to describe the character of the partner. If students spoke in simple sentences I would elicit further responses by asking 'why?' or 'because?'. In week five I introduced another mechanism to stimulate discussion and to add variety to the course book material. I asked students to come to classes with an English language newspaper article that they had read beforehand and gave them all, one by one, the opportunity to tell the class what the article was about, why it interested them, and their opinion of it. This proved to be another equalizer between the confident and the less confident students.

By this time most of the students were actively competing to enter discussions and answer questions and it became clear that the more competent students were effectively silencing the less competent or more reticent. To counter this and to give the slower and less confident students the chance to speak up and earn credit points I introduced a system of a show of hands on who wanted to speak. In this way, I was able to wait a few seconds while slower students gathered their thoughts, and if they put up their hands I gave them their chance.

At the end of each month I collected students' credit certificates to check their progress and to record their accumulated scores in case they lost their certificates, which represented the proof of their progress. This also provided me with a useful tool to identify the high achievers, the middle stream, and the students who were trailing the class. In this way I was able to keep an eye on the under-performers and would give them preference to speak up if they indicated they wanted to. Credit certificates were returned to students so they could continue to monitor their progress.

Observations
An initial problem was the architecture of the allocated classroom. It was fitted with immovable desks set in formal rows. Apart from the difficulty of stimulating authentic discussion in such an environment, it proved very difficult to move around the classroom to award credit stamps. I managed to find an alternative classroom where we re-arranged the chairs, fitted with desk flaps, into an open-ended circle. This meant that all students were facing each other and I was able to move around freely, asking questions, encouraging discussion, and awarding credit stamps.

Another problem, which was insoluble under present circumstances, was the numbers of students in each class. While 31 and 30 students respectively in the two classes is certainly not high by many Asian standards, it is still too high in a situation where conversation and interactivity between teacher and students and students and students is the goal.

The system of instant reward of credit stamps requires a higher level of teacher concentration, class management and physical movement around the classroom than teacher-fronted systems. The teacher has to be constantly stimulating responses, keeping an eye on slower students, and moving around awarding credits. The offset to this is far less time spent out of class assessing the results of numerous listening and speaking tests and compiling and maintaining the score records. In the previous semester this comprised four listening tests, four speaking tests, and attendance and self-learning records.

An interesting result of this system and the way in which it was implemented was the compression of students' final results into a narrower bandwidth than that produced by the previous method of assessment. There were two causes for this. The first was my monthly monitoring of the students' progress, which identified lower-performing students, in turn giving them the chance to catch up to the more confident and capable students. It could be argued that this was to the detriment of the more capable students and therefore was unfair. My counter argument to this is that as a teacher I had the responsibility to get the best I could out of all students. The second reason was that some of the best students 'over-performed', scoring their three half-mark credit stamps for most of the 30 classes (which could have resulted in a final score of 45). As the syllabus allocated a maximum of 40 marks for credit stamps, a few students had to be pared back to this limit. This was a problem I had not foreseen when devising the scoring system and will need to be addressed in future applications.

It must be acknowledged that the credit stamp element of the students' final assessment was largely quantitative in that it was a measurement of how many times and to what length they had 'talked' during the semester. The emphasis was to get the students speaking in English, not on using perfect English. While I progressively encouraged, and in fact required, the students to give more complex answers to earn credit points, I did not withhold credit stamps for the use of incorrect grammar. In many ways, the whole experiment was an exercise in building students' confidence.

The Final Results
All students in both classes passed the course. In terms of the 40 per cent credit scores, my class scored from 19 to 40, with one student scoring the below par 19 and four scoring the maximum 40. The spread was one student 19, four students in the 21-25 band; four students in the 26-30 band; 9 students in the 31-35 band; and 12 students in the 36-40 band. The Thai teacher's class performed similarly, with a score range from 24 to 40. Three students achieved the maximum score. One student was in the 20-25 band: 12 were in the 26-30 band; nine in the 31-35 band; and nine in the 36-40 band. Figure 1 shows a comparison of the final participation credits awarded to students in Group 1 and Group 2.

Comparison between Groups 1 & 2 - Final Participation Credits

Figure 1. Vertical axis represents numbers of students. Horizontal axis represents scores out of 40

Figure 2. The four coloured segments represent A, B, C & D grade passes. The numbers represent total number of students in each grade.

The final results are not meaningfully comparable with the results of the conversation skills classes that I taught during my first semester because the classes which were the subject of this case study comprised a different student mix and were at a higher level. It is also acknowledged that some of my intermediate students started in my beginners' class and they had come to know me better, and hopefully were more relaxed and comfortable with my teaching. Nevertheless, my own observations and experiences in running this experiment, and the final overall results, convince me that the credit stamp award system positively affected students' willingness to participate. As reported above, I had to introduce a 'show of hands' mechanism to control over-eager students who were 'drowning out' the more reticent and slower to respond. This active participation helped them achieve higher grades than in my previous classes. Certainly no student failed this course, unlike the previous classes when five students out of 64 failed.

The level of animated participation that was sometimes achieved can be illustrated with some anecdotes. One female student brought a newspaper article to class about the visit to Bangkok of the Stuttgart Ballet. She talked about her love of dance, kicked off her shoes and demonstrated Pointe, Glissades and Arabesques to the class, which while not talking was certainly letting down her hair and participating. On another occasion, a student talked about her favorite singers before proceeding to give a rendition of "I did it my way." On yet another occasion when the topic was interpersonal relationships, a male student openly talked about his dawning realization that he was gay and how he hoped to find a steady boyfriend. The final, and for me most poignant incident, was on the final day that students could earn credit points. One male student, very much the lowest performer, had obviously collected his thoughts for this day. He managed to score the maximum three credit stamps for the day, the first time he had accomplished this, and received a round of cheers from the rest of the students.

Student Reactions
In the last week of the course, the 61 students in the two classes were surveyed on their opinions of the instant reward and recognition evaluation system. 91.8 per cent of students said it was a good method of evaluation. 71.4 per cent said it had made them feel less stressful when speaking English. Anomalously, 93.8 per cent said it had made classes and learning more fun.

Many students offered general comments in the questionnaire. Some students expressed concern that there might be inequity between how the Thai teacher in her class and I in mine determined what sort of utterance would merit a credit stamp. While the end results in the two classes were about the same quantitatively, this concern is valid, since the system requires numerous subjective and instantaneous judgments on the part of the teacher during the class. Close co-operation and consultation between participating teachers, both in planning and implementing the system, and during the course of the semester, should help minimize this concern. Also balancing this concern, there was strict and demonstrable equity and fairness in other areas of the overall assessment. Listening tests were co-written and strictly administered to agreed principals, e.g. each conversation or voice tape was played twice only in each class. The speaking tests were co-adjudicated and marked by both teachers. An analysis of the final results showed that there was an aggregate variance between the two groups across all components of the assessment of between 0.17 and 0.6 percentage points.

Other students expressed concern that the system favored the more capable over less confident peers. I have addressed this issue above. The system allows for early and clear identification of slower performing students and the teacher can then address this problem. I believe that far from disadvantaging the less able, it gives them better opportunities than in the previous system of assessment over many listening and speaking tests.

Another interesting observation from some students was that they were afraid of missing classes because they would lose the opportunity to earn credit points. This helped explain the low absenteeism that I had noticed during the semester.

Conclusion
This case study is based on an adaptation of the work of Hadley (1997) and Jeffrey (2003) in Japan. The aim of the experiment was to see if their results could be replicated in the Thai cultural setting. The experiment was run on two classes of intermediate English conversation skills students at Dhurakijpundit University, Bangkok, from May to September 2004. One class was taught by a Thai teacher, the other by me.

As a scheme to enhance class participation and build students' confidence, the exercise was an outstanding success compared to my previous experiences in 'conversation' classes. While aspects of the final results are open to further or other interpretation, I am confident that this instant reward and recognition system motivated, or compelled, students to open their mouths and speak. In the end, I remain convinced that that is the ultimate objective of a 'conversation' class.

Acknowledgements: Sincere thanks to Ajarn Soisithorn Isarakura, English Instructor, Language Institute, Dhurakijpundit University, for her willing co-operation in trialing this system with me, for making many useful suggestions on implementation, and for giving me constant feedback from her own experiences and from her students. Also to Dr. Harald Kraus, Head of the English Department, Dhurakijpundit University, for his advice on the conduct of this case study and for editing the final draft.

References:
Hadley, G. (1997). Encouraging Oral Communication in the EFL Classroom. Paper presented at the Niigata University General Education and Language Research Group. Niigata City, Japan. Accessed 30/04/04 at www.nuis.ac.jp/~hadley/publications/partpoints/participation.htm

Jeffrey, D. M. (2003). Participation Points System to Encourage Classroom Communication. The Internet TESL Journal Vol. IX, No 8, August 2003. Accessed 30/04/04 at http://iteslj.org/Techniques/Jeffrey-PointsSystem.html

Jones, L. (2002). Let's Talk 3. Cambridge: The Press Syndicate of the University of Cambridge.

The Nation 2004. Learning English: Suan Dusit Poll. 12/09/04, 4A. Bangkok