In any society, education plays a critical role because it determines the perspectives of individuals’ life. Academic performance is highly correlated with the individuals’ future career and occupational choices. To assess academic performance over time, there are studies such as Programme for International Student Assessment (PISA) organized by the Organization Economic for Co-Operation and Development (OECD), Trends in International Mathematics and Science Study (TIMSS) and Progress in International Reading Literacy Study (PIRLS) organized by International Association for the Evaluation of Educational Achievement (IEA). More specifically, science achievement is about understanding and applying the fundamental knowledge of science, drawing conclusions based on data and evidence and developing the significance of science and technology in daily lives (OECD, 2017). In the literature, empirical research have been made to specify the underlying factors of students’ achievement (Contini, Di Tommaso, & Mendolia, 2017; Sheldrake, Mujtaba, & Reiss, 2017; Kılıç Depren, Aşkın, & Öz, 2017; Kılıç Depren, 2018). Zhang, Khan, and Tahirsylaj (2015) have used regression methods to measure the factors affecting students’ performance using the dataset of PISA 2009 participating countries. The research of Delprato and Chudgar (2018) has found out the systemic school factors on the privatepublic performance gap using the Teaching and Learning International Survey (TALIS)-PISA dataset in three countries. Based on a multilevel model, Giambona and Porcu (2018) have identified the most important factor on students’ achievement was school size. In this research, it was assumed that the students’ science achievement scores were reflecting a mixture distribution that is a sample of students representing various backgrounds, with some or all backgrounds associated with different distributions and mean scores. Thus, Quantile Regression Mixture Model (QRMIX), which is a unique approach in the education literature, was used to determine the factors affecting students’ science achievement in Turkey and Singapore. Abstract. In the last decade, the usage of advanced statistical models is growing rapidly in many different disciplines. However, the Quantile Regression Mixture Model (QRMIX), which is a developed approach of the Finite Mixture Model (FMM), is an applicable new method in the educational literature. The aim of the proposed study was to determine factors affecting students’ science achievement using the QRMIX approach. To reach this aim, data of the Programme for International Student Assessment (PISA) survey, which has been conducted by the Organization Economic for Co-Operation and Development (OECD) every 3 years, was used. Dataset used in the research is composed of 6,115 students from Singapore, which is the top-performer country among the participant countries, and 5,895 students from Turkey. The results showed that the factors affecting students’ science achievement and its importance on the achievement differentiated according to the achievement levels of the students. In conclusion, it was revealed that Turkish students with the lowest science achievement level should be supported with home possessions, perceived feedback, and environmental awareness and Singaporean students with the lowest achievement level should be supported with perceived feedback, enjoyment of science, and epistemological beliefs.

中文翻译：

---

确定影响土耳其和新加坡学生科学成就水平的因素：分位数回归混合模型的应用

在任何社会中，教育都起着至关重要的作用，因为它决定了个人的生活前景。学习成绩与个人未来的职业和职业选择高度相关。为了评估一段时间内的学业成绩，有一些研究，例如合作与发展经济组织 (OECD) 组织的国际学生评估计划 (PISA)、国际数学和科学研究趋势 (TIMSS) 和国际阅读进展国际教育成就评估协会 (IEA) 组织的扫盲研究 (PIRLS)。更具体地说，科学成就是关于理解和应用科学的基础知识，基于数据和证据得出结论并发展科学技术在日常生活中的重要性（经合组织，2017 年）。在文献中，已经进行了实证研究以明确学生成绩的潜在因素（Contini、Di Tommaso 和 Mendolia，2017 年；Sheldrake、Mujtaba 和 Reiss，2017 年；Kılıç Depren、Aşkın 和 Öz，2017 年；Kılıç Depren , 2018)。Zhang、Khan 和 Tahirsylaj (2015) 使用回归方法使用 PISA 2009 参与国的数据集来衡量影响学生表现的因素。Delprato 和 Chudgar (2018) 的研究使用三个国家的教学和学习国际调查 (TALIS)-PISA 数据集，发现了影响私人公共绩效差距的系统性学校因素。基于多层次模型，Giambona 和 Porcu (2018) 确定了影响学生成绩的最重要因素是学校规模。在这项研究中，假设学生的科学成就分数反映了混合分布，即代表不同背景的学生样本，其中一些或所有背景与不同的分布和平均分数相关。因此，使用教育文献中独特的分位数回归混合模型 (QRMIX) 来确定影响土耳其和新加坡学生科学成就的因素。抽象的。在过去十年中，高级统计模型的使用在许多不同学科中迅速增长。然而，分位数回归混合模型 (QRMIX) 是有限混合模型 (FMM) 的一种开发方法，是教育文献中一种适用的新方法。拟议研究的目的是使用 QRMIX 方法确定影响学生科学成绩的因素。为了实现这一目标，使用了经济合作与发展组织 (OECD) 每 3 年进行一次的国际学生评估计划 (PISA) 调查的数据。研究中使用的数据集由来自参与国家中表现最好的新加坡的 6,115 名学生和来自土耳其的 5,895 名学生组成。结果表明，影响学生科学成绩的因素及其对成绩的重要性因学生的成绩水平而异。综上所述，