This research is based on a well designed group experiment in relation to the literature review referenced to “Behavior Modification Techniques, Token System.” Before conducting the experiment, questionaaires are made to the subject (High school teacher) in Macao, to examine about the conditions of the classroom disorder as well as to the misbehavior of the student. The purpose of the experiment is to find out the associated problems and the degree of classroom disorder in respect to students’ misbehavior. The fundamental subject of this questionnaire investigation have divided into three different categories: catholic organized institute; social organized institute and government organized institute. There are teachers participated in questionnaire survey from seven high schools. To further look into the condition of classroom disorder in respect to students’ misbehavior, a class of 34 Form One students are picked from one of the above listed schools, this class of students is used as the subject of the well designed group experiment. The study is to observe the impact of “token system” in response to students’ behaviorism. The result of the experiment is positive; there is significant improvement in classroom order and the causes of the issue are viewed in relative to the various factors affecting the classroom order. By working through the above associated problem, suggestion of solutions is made for other scholars or teachers for further studies. The research is designed to identify the changes after taking the experiment. In the experiment Backward conditioning (ABA) is used. There are three different stages in the experiment design. The first stage is baseline phase; the result is made from the behavior analysis. The second stage is consequence from the independent variable. Causes of classroom disorder are explained by understanding the independent variable. The third stage is reversal phase, by terminating the use of independent variable, working repeatedly on the statistic value in comparison to the test taken before the experiment began, so as to set up the control for the experiment. Data analysis is to examine subject of the experiment and the different time-frame of the investigation. Questionnaires are made for teachers as well as for students. The questionnaires for teachers are given out before the research experiment began. The data collected from the questionnaires is used to find out the condition of classroom disorder in respect to students’ misbehavior and how teachers take care of the problems associated. On the other hand, the questionnaires for students are given out after the research experiment finished; the questionnaire is then used to find out the impact of token system on the condition of classroom disorder in respect to students’ misbehavior. The quantitative data analysis is based on the written report provided by students which reflects to their feelings towards the use of token system and the result of the questionnaires. These two sets of data strengthen the research reliability. In addition, Likert-Type Scale is used to test the reliability of the research. By the comparison in using statistic analysis on the number unit in the token system and the percentage of success in the improvement of classroom order in respect to students’ behavior, are further developed to construct validity. Moreover, in order to further examine the different types of teachers and the various kinds of subject students, in respect to their age, sex, and religion, discovering how the above factors in correlate to the research problems, the results are based directly from the chi-square test. Results of the research are listed as follow: 1. According to the result from the questionnaire, Token system is better than using the traditional way of punishment in the long run as response of the high school teachers 2. By using the token system, the experiment class unit shows a decrease in cases of classroom disorder in respect to students’ misbehavior. 3. Through the duration period of 13 weeks, the experiment class unit shows a gradual decrease in cases of classroom disorder in respect to students’ misbehavior. Zero cases can be predicted with this gradual impact. 4. According to the students’ response in the questionnaire given out after the experiment. Students generally accepted the usage of the Token system. Also they believed the number of the classroom disorder cases has been reduced after using the Token system, and that the impact of the sytem can be maintained for a long time. In addition, the academic result of the students has also improved. Finally, based on the above results, suggestions are made for other scholars or teachers in furthering their studies or practices in token system: 1.The comment made to the students should have an encouraging effect 2.Appraisal should be used appropriately. 3. Pygmalion effect should be use appropriately 4.Caring for the teenages should be enhenced. 5.Student’s diary should be use appropriately 6. Should enhance the strategic studies in eliminating the usage of the punichemnt to the studnt’s emotionally, mentally or psychologically. in the future.

中文摘要 本研究是根據行爲改變術中的代幣制的相關文獻設計的一個群體實驗，研究者在作實驗前，曾對本澳中學教師進行問卷抽樣調查，藉以得知那間學校的學生課堂秩序問題較嚴重，而取樣方式以學校的類別(即天主教、社團及官立)爲基礎，選取七間不同學校類別的中學為調查問卷的取樣範圍。教師問卷是在確立本研究題目前發出的，主要調查教師在其任教學校課堂秩序的情況及教師對違規學生的處理方法。之後，憑教師問卷的數據資料，選取其中一間秩序問題較嚴重的學校，再選取校內正就讀中一的34名常有違規行為的學生為實驗對象，探討接受了代幣制來改變行爲的受試學生，整體的課堂秩序將有明顯改善;並且從實驗的過程中，了解影響實驗效果的因素，針對這些因素提供具體建議，以供實務工作者及研究之用。而學生問卷則是在實驗結束後才發出，旨在調查受試學生對是次實驗的觀感，探視利用代幣制改善中學生上課秩序的可行度。質性研究的資料，就是讓他們用文字表達對上述問題的感想，從而加強問卷及是次實驗結果的信度。為了確保問卷的可靠性和有效性，利用李克特態度四點量表，再作出信度檢驗，信度結果爲 0.7，顯示問卷有高的信度。再探用描述性的統計分析，主要以次數分配、百分比進行比較，來探視填答者對利用代幣制改善課堂秩序適用於中學與否、有效度及持久度。此外，爲了進一步探討不同學校類別的教師，以及受試學生中，不同年齡層、 不同性別和不同宗教，對研究問題看法有否差異，研究者探用了百分比同質性卡 方考驗檢定填答者對主要題目的意見是否有所差異。 本研究的實驗設計，對實驗組進行前測、後測，探倒返設計法(ABA)，第一步先量基準線(以全班爲單位)，以此結果作為了解實驗班級的行為現狀，第二步呈現自變因，看此自變因對實驗班級的行爲發生何種影響。第三階段是倒返階段，停止操弄自變因，再以它的前測分數作為統計分析時之共變數，以控制實驗過程中的無關變項。茲將主要研究結果摘述如下: 一、教師問卷的結果顯示，教師對代幣制在初中班級實施是可行的，而且比會單純用校規懲罰更爲有效及持久。 二、實施代幣制後，受試班級的課堂違規行為的整體記錄次數有所減少。 三、經過實驗處理後十三週，受試班級的課堂違規行爲的整體記錄次數持續減少或等於雾。 四、實驗後發出的學生問卷結果顯示，受試學生普遍對代幣制有高度的接受度，而且他們都認為實施了代幣制後，班級同學的違規行為整體記錄次數減少，而且守規紀錄較單純用校規持久，成績也隨之而提高。 最後，研究員依上述研究結果，為代幣制的應用及後續研究者提出具體建議。 (一)學生品行評定應發揮其激勵作用 (二)善於運用讚揚 (三)善於運用畢馬龍效應(Pygmalion effect ) (四)加強對青少年的關注 (五)善於利用學生手冊 (六)未來需加強消除心罰現象的對策