本研究之主要目的在探討科技大學中特殊身分學生，包括原住民、僑生、外籍生等，可能影響成績表現之因素有哪些？俾提供校方後續在招生政策、教學資源的分配或學習的輔導上能有所參據。為達上述目的，首先藉由相關文獻整理，列出可能影響之關鍵性因素，如性別、男女朋友、經濟狀況、家中排行、家中子女數、同儕成績、住宿等，作為後續探討之方向，接續實施初級資料分析，對同一類別學生、不同二類別學生及三類別學生成績平均數作檢定，另針對目標對象實施問卷調查，其中對僑生全體發出49份問卷，回收有效問卷42份，回收率85.7％，對原住民學生全體發出122份問卷回收有效問卷74份，回收率60.6％，另外籍生部份因尊重其文化差異及業管單位意見，本階段未作進一步調查。
    之後分別以描述統計（Descriptive Statistics）、平均數檢定、單因子變異數分析（ANOVA檢定）、多元迴歸分析、異質變異數檢定等方法進行資料分析，本研究之主要發現如後：
一、就特殊身分學生學業成就相對於該班表現，外籍生明顯優於僑生，而原住民學生相對較差。
二、就成績平均數檢定結果，在顯著水準α=0.05下，上述三類別學生中，各類男生與女生成績平均數皆相等。
三、不同二類別學生成績平均檢定結果，在顯著水準α=0.05下，均顯示成績平均數不相等。
四、三類別學生成績平均數ANOVA檢定顯示，沒有出現完全相同之情形。
五、多元迴歸模型對原住民學生之實證結果發現，各解釋變數多呈現不顯著之情形，雖經White一致性估計法進行估計，結果仍不顯著，代表建立的模型後續修訂仍有相當的改善空間。
六、另多元迴歸模型對僑生之實證結果發現，除住宿因素為一顯著因素外，其餘解釋變數均不顯著，再以White一致性估計法進行估計，結果顯示住宿因素對於僑生的成績影響顯著增加，惟呈現負相關，此情形說明僑生並未因住宿而獲益，或許可解釋為未住宿僑生與家人同住，獲得較佳照顧有關，惟後續校方該如何加強住宿僑生課業輔導，應為當務之急。

The main purpose of this research was to investigate the factors that might affect the academic grade performance of certain minority groups of students in a university of science and technology, such as aboriginal students, overseas Chinese students, and international students. It is hoped that the results could provide a concrete reference for the university in making decisions on student recruitment, allocation of educational resources, and provision of academic guidance. To achieve this purpose, after a review of the relevant literature, a group of possibly influential factors was isolated to be used in the ensuing research, including gender, friends of the opposite sex, financial situation, birth order, number of children in the birth family, peer group grades, and dormitory accommodation. Next, the first stage of data gathering and analysis involved determining the grade averages for each group of students, as well as for any two groups or all three groups together. Following this, questionnaires were distributed to each group, with overseas Chinese students receiving 49 questionnaires, of which 42 valid were returned, for a response rate of 85.7%; aboriginal students receiving 122 questionnaires, of which 74 valid were returned, for a response rate of 60.6%; out of respect for the cultural differences of the international students, on the advice of the concerned administrative unit, they were not included the questionnaire survey. 
 The data obtained were analyzed using descriptive statistics, the determination of averages, one-factor analysis of variance, multiple regression analysis, and heteroskedasticity. The main discoveries of these analyses were as follows:
1.Comparing the academic achievement of the different minority student groups with that of their local classmates, the performance of the international students was clearly superior to that of the overseas Chinese students, while that of the aboriginal students lagged behind.
2.ANOVA analysis of the grade averages of the three groups did not produce results entirely the same as those indicated above.
3.Comparing the grade averages of any two of the three groups, with the level of significance at=0.05, all of the differences were significant.
4.Comparing the grade averages for male and female students within the three groups, with the level of significance at =0.05, there was no significant difference between the rade averages of the male and female students within each group.
5.Multiple regression analysis of the questionnaire results for the aboriginal students showed no significant relationships between the different variables and grade averages. Further analysis through White heteroskedasticity Consistent Covariance martrix estimator also revealed no significant relationships. This indicates that the structure of the model used for the research has considerable room for improvement.
Multiple regression analysis of the questionnaire results for the overseas Chinese students showed a significant result for the relationship between dormitory accommodation and grade averages, but not for any of the other variables. Further analysis through White heteroskedasticity Consistent Covariance martrix estimator showed an increased level of significance for the dormitory accommodation variable, but the influence of this variable was negative, indicating that living in a dormitory is not an advantage for overseas Chinese students. Perhaps those who live off-campus are living with relatives and receiving better care. However, this apparent disadvantage would indicate that the university should look into how to provide academic assistance for overseas Chinese students living in school dormitories.

目       錄
第一章 緒論.........................................................................................1
  第一節 研究動機與目的...................................................................1
  第二節 研究架構..............................................................................2
第二章 文獻回顧..................................................................................4
第三章 成績平均數檢定.....................................................................10
  第一節 資料來源、選取與處理........................................................10
第二節 檢定方法說明.....................................................................14
  第三節 三類別學生成績平均數檢定...............................................19
  第四節 不同二類別學生成績平均數檢定結果.................................21
  第五節 同一類別學生成績平均數檢定結果....................................24
第四章 影響成績表現之因素分析......................................................26
  第一節 計量模型設立.....................................................................26
  第二節 計量模型估計方法說明......................................................29
  第三節 實證結果............................................................................32
第五章 結論.......................................................................................37
參考文獻..........................................................................................40
附錄（問卷暨基本統計資料）..............................................................43


圖    目    錄
圖1-1  研究流程圖.............................................................................3


表    目    錄
表3-1  國立○○科技大學特殊身分學生成績表現統計表.................13
表3-2  國立○○科技大學三類別學生成績平均數檢定.....................20
表3-3  國立○○科技大學不同二類別學生成績平均數檢定..............23
表3-4  國立○○科技大學同一類別學生男女成績平均數檢定..........25
表4-1  原住民學生成績影響—OLS估計結果...................................33
表4-2  White 異質變異數檢定結果…………………………………...34
表4-3  White一致性估計結果…………………………………………34
表4-4  僑生學生成績影響—OLS估計結果………………................ 35
表4-5  White 異質變異數檢定結果…………………………………...35
表4-6  White一致性估計結果…………………………………………36

參考文獻
一、中文部分：
王雅觀(1999)，「高級中學班級組織型態及其班級氣氛與學習狀況之研究」，國立台灣師範大學教育研究所碩士論文。
石培欣(2000)，「國民中學學生家庭環境、同儕關係與學業成就之相關研究」，國立高雄師範大學教育研究所碩士論文。
李玉涵(2005)。「同學、朋友及家庭同儕對學生成績之影響」，東吳大學經濟研究所碩士論文。
李佩怡(1999)。人際關係理論。測驗與輔導。152，3152-3156。
林惠雅(1998)。我們都是好朋友-談幼兒「同儕關係」的發展與建立。成長幼教季刊。9(4)，30-34。
洪瑞瑩(2003)，「人力資本的代間移轉與手足效果-台灣實證」，暨南大學經濟研究所碩士論文。
徐澄清(1998)。社會化過程的重要機轉-同儕互動。健康世界。145，10-14。孫青山與黃毅志（1996）。補習教育、文化資本與教育取得。台灣社會學刊，19，95-139。
孫青山與黃毅志（1996）。補習教育、文化資本與教育取得。台灣社會學刊，19，95-139。
張春興(1994)。教育心理學-三化取向的理論與實踐。台北：東華。
陳永欽(2002)，「家庭背景對子女教育成就之影響」，暨南大學經濟研究所碩士論文。
曾天韻(2000)，「台灣地區出身背景對高等教育入學機會」，台東師範學院教育研究所碩士論文。
劉金花(1999)。兒童發展心理學。台北：五南。
蘇建文(1991)。發展心理學。台北：心理。
蘇秋碧(2000)。國小六年級被同儕忽視學童之研究。台中師範學院國民教育所碩士論文。


二、英文部分：
Coleman, J. S. (1988b), Social capital in the creation of human capital. Supplement to American Journal of Sociology,  94,  95-120. 
Corey,  G. (1991), Theroy and Practice Counseling and Psychotherapy.California：Brooks/Cole Publishing Company Pacific Grove.
Hoxby, Caroline(2000), ”Peer Effects in the Classroom ：Learning from Gender and Race Variation, ”National Bureau of Economic Research Working Paper No. 7867.
Powell, B., & Steelman , L. C. (1990), Beyond Sibship Size：Sibling Density, Sex Composition,andEducational Outcomes. Social Forces 69(1), 181-206.