|摘要: ||本研究主要探討學生因素（學生正向影響、學生自信心、學生社經地位、學生性別）與教師因素（教師年資、教師期待、教師準備度、教師性別）對參與2007年國際數學與科學教育成就趨勢調查（TIMSS 2007）之台灣、香港、新加坡三個華人社會國家/地區的四年級學生科學成績之影響。本研究主要回答兩個研究問題：在所選擇的國家/地區中哪些學生層因素與學生科學成績相關以及如何相關？在所選擇的國家/地區中哪些教師層因素與學生科學成績相關以及如何相關？因此本研究樣本選擇為亞洲國家/地區參與TIMSS 2007四年級學生與他們的科學教師，包括台灣150間學校中4131位學生與174位科學教師；香港126間學校中3791位學生與150位科學教師；新加坡177間學校中5041位學生與360位科學教師。資料來源取自TIMSS 2007 四年級學生科學問卷與教師科學問卷。而本研究使用次級資料分析，並使用階層線性模式6.08版（HLM 6.08）分析數據，以小於0.05作為顯著的標准。|
This study examined the effects of selected student factors (students’ positive affect toward science, students’ self-confidence in learning science, family socio-economic status, student gender) and teacher factors (the number of years teacher have taught, teachers’ expectations, teachers feel “very well” prepared to teach the science, teacher gender) on science achievement at the fourth grades in Taiwan, Hong Kong and Singapore of Chinese societies who participated in the Trends in International Mathematics and Science Study (TIMSS) 2007. This study aimed to answer the following research questions: Which factors are associated with student science achievement at student level, and how are they associated in each selected country? Which factors are associated with student science achievement at teacher level, and how are they associated in each selected country? This study analyzed the responses of fourth-grade students and their science teachers from the participating Asia countries include the Taiwan TIMSS 2007 sample consisted of 4131 students and their 174 science teachers in 150 schools, the Hong Kong TIMSS 2007 sample consisted of 3791 students and their 150 science teachers in 126 schools and the Singapore TIMSS 2007 sample consisted of 5041 students and their 360 science teachers in 177 schools. The data of this study were collected by TIMSS 2007 Student Questionnaire and TIMSS 2007 Teacher Questionnaire in science. Thus, the secondary analysis was used in this study. A hierarchical linear model (HLM 6.08) was used to analyze the data; a significance level of 0.05 was used.
The results of this study showed that ICC suggests that multilevel analysis can analyze the data of Taiwan, Hong Kong and Singapore. The results showed that PATS had a significant positive effect on student science achievement in Taiwan, Hong Kong and Singapore. SCS had a significant positive effect on student science achievement in Taiwan, Hong Kong and Singapore. SES had a significant positive effect on student science achievement only in Taiwan and Singapore. TG had a significant effect on student science achievement only in Taiwan. TCHES had a significant positive effect on student science achievement only in Hong Kong. However, SG, TY, TP had a non-significant effect on student science achievement in Taiwan, Hong Kong and Singapore. Besides, four variables of student level and four variables of teacher level interaction effects had no significant effect on student science achievement in Taiwan, Hong Kong and Singapore.