本研究主要目的在探討都市原住民族學生在父母教育期望、學生學習動機及學生學業成就的現況，並探討不同背景變項間的差異情形，分析都市原住民族學生在父母教育期望、學生學習動機及學業成就的相關情形，最後並探究父母教育期望及學生學習動機對學生學業成就之回歸預測分析。 本研究採問卷調查法，以臺北縣六所國民中學原住民族重點學校的學生為研究對象，共計595人，以描述性統計、t考驗、單因子變異數分析、Scheffé事後比較、皮爾森積差相關及多元迴歸分析等方法進行統計分析。 本研究得到以下的主要發現： 一、 都市國中原住民族學生在父母教育期望、學生學習動機及學生學業成就均為中上程度。 二、 都市國中原住民族學生在父母教育期望、學生學習動機及學生學業成就，因背景變項不同而有顯著差異。 三、 都市國中原住民族學生在父母教育期望、學生學習動機與學生學業成就有顯著的低度正相關。 四、 都市國中原住民族學生在父母教育期望和學生學習動機有顯著的中度正相關。 五、 都市國中原住民族學生在父母教育期望、學生學習動機對學生學業成就具有顯著的預測力。 / The main purpose of this study was to understand the relationships among parent education expectation, learning motivation, and academic achievement of urban aboriginal junior high school students. Survey method is used in this research. By purposeful sampling, 595 valid samples were acquired. The instruments used in the research are included Parent Education Expectation Scale and Learning Motivation Scale. The statistical methods used to analyze the data were descriptive statistics, t-test, one-way anova, Scheffé method, pearson product-moment correlation coefficient analysis, and multiple regression analysis . The conclusions of major were as follows: 1. Parent education expectation, learning motivation, and academic achievement of urban aboriginal Junior high school students were in medium-high range. 2. There were significant differences of parent education expectation, learning motivation, and academic achievement of urabn aboriginal Junior high school students. 3. The correlations among parent education expectation, learning motivation, and academic performance were positively significant. 4. The correlations between parent education expectation and learning motivation were positively significant. 5. Urban aboriginal junior high school students’ parent education expectation and learning motivation, could significantly predict their academic performance.

影響學生學業成就因素，是值得關注的議題。本研究欲瞭解教育指標和家庭資本對學生學業成就之間的影響關係，以臺灣、日本、韓國、美國和義大利五個國家已公佈的教育指標資料和TIMSS 2007資料庫問卷調查結果的24383份追蹤樣本，進行現況分析、變異數分析、相關分析、多元逐步回歸分析、結構方程式分析，結果發現： 一、教育指標中，在學率和數學及科學學業成就為正相關。中等教育師生比、識字率、教育經費占國民所得比與數學及科學學業成就為負相關。 二、教育指標對數學學業成就的解釋變異量為25%，對科學學業成就的解釋變異量為9.7%。 三、家中藏書量和學習資源對數學及科學學業成就的變異數分析中可得知擁有越多藏書和資源的組別大於擁有較少藏書和資源的組別。 四、父母親學歷對數學及科學學業成就的變異數分析中可得知美國在此分析中未達顯著，臺灣、日本、韓國和義大利都是學歷越高的組別大於學歷低的組別。 五、自我期望對數學及科學的學業成就的變異數分析中可得知自我期望越高學歷的組別大於自我期望低學歷的組別，但美國學生在數學及科學學業成就及韓國學生在數學學業成就的自我期望皆以「不知道」的組別最高。 六、家長支持對學生數學及科學學業成就的變異數分析可得知家長支持度高的組別普遍大於支持度低的組別，但是義大利的家長支持在數學及科學學業成就和臺灣的家長支持在科學學業成就上都是支持度最高的組別小於支持度最低的組別。 七、家長參與對學生數學及科學學業成就的變異數分析可得知家長參與度高的組別普遍大於參與度低的組別，但是臺灣的家長參與在科學學業成就上是參與度最高的組別小於參與度最低的組別。 八、家庭資本和學業成就之間的相關分析可得到母親學歷和學業成就無關，家庭資本變項中的自我期望與數學學業成就也無關，父親學歷與學業成就為負相關，其它變項與學業成就為正相關。 九、家庭資本對學業成就的解釋變異量以台灣最高，家庭資本對數學學業成就的解釋變異量為27.5%，對科學學業成就的解釋變異量為28.8%。 十、由SEM結果得知家庭資本中的家長支持和家長參與對數學及科學學業成就未達顯著差異。教育指標和家庭資本對學業成就具有一定的解釋力。 依據上述研究結果將提出相關結論與建議以供參考。 / The issue of student achievement has been concerned. In this research, will focus on the relationships between educational indicators and family capital on student academic achievement, the participants include students in Taiwan, Japan, South Korea, the United States and Italy. The published educational indicators and TIMSS 2007 survey results of 24,383 samples for the analysis of variance, Pearson correlation, stepwise multiple regression, structural equation modeling found that: 1. The relationship between educational indicators, student academic achievement in mathematics and science as a positive. Secondary education teacher/ student ratio, literacy, education expenditure/ GDP ratio of and academic achievement in mathematics and science as a negative. 2. The educational indicators of academic achievement in mathematics accounts for 25% variance, in science accounts for 9.7% variance. 3. In analysis of variance, the collection of books and learning resources on academic achievement in mathematics and science indicate that a group has more is higher than the group has fewer. 4. In analysis of variance, parental education on academic achievement in mathematics and science find that in United States is not significant in this analysis. In Taiwan, Japan, South Korea and Italy, groups of higher education is the highest. 5. In analysis of variance, self expectation on academic achievement in mathematics and science indicate that a group with high self expectation is highest, but U.S. students in math and science academic achievement and Korea students in mathematics can find out that the students who "do not know" is the highest group. 6. In analysis of variance, parental support in student academic achievement in mathematics and science that groups with high parental support is higher than those with low parental support, but parental support on student academic achievement in mathematics and science in Italy, and parental support on academic achievement in science in Taiwan indicate that the highest group is less than the lowest one. 7. In analysis of variance, parental involvement in student academic achievement in mathematics and science that groups with high parental involvement is higher than those with low parental involvement, but in Taiwan, parental involvement on academic achievement in science indicate that the highest group is less than the lowest one. 8. The correlations between family capital and student academic achievement indicate that mother’s education is in no relation with academic achievement in mathematics and science, self-expectation is in no relation with mathematics academic achievement, the relationship between father’s education and academic achievement is negative, other variables and academic achievements is positive. 9. The variance of family capital in academic achievement in Taiwan is the highest: 27.5% in mathematics and 28.8% in science. 10. SEM results show that parental support and parental involvement is in no relationship with academic achievement in math and science. Educational indicators and family capital on academic achievement can be explained. Based on the results of the study, conclusions and recommendations will be presented for reference. Key words: Educational indicator, Family capital, Student academic achievement, TIMSS 2007.