【作者】 韦斯林；

【导师】 王祖浩； 柳秀峰；

【作者基本信息】 华东师范大学 ， 学科教育， 2011， 博士

【摘要】 科学素养是当今国际科学教育共同追求的目标。科学素养的核心与关键在于理解科学。近40年的科学教育研究显示,幼儿到大学毕业生普遍持有大量的错误科学概念,其中相当一部分错误概念十分顽固、旷日持久,它们严重制约着学生对科学知识的充分理解与把握,造成大量学习困难、障碍。化学教育中,学生化学学习的一个瓶颈是难以在不同水平化学表征(即宏观、微观和符号表征)之间建立联系和灵活转换,从而催生大量的错误概念,造成学习障碍。学生实现多水平化学表征之间的灵活转换,需要充分理解化学模型和建模。为帮助学生顺利突破这一瓶颈,近年来,计算机模型广泛运用于科学教学中,它们强化宏观现象、微观运动与符号之间的相互联系,有效促进学生模型建构以及建立化学表征与转换,增强学生的概念理解、推理思维、科学探究技能以及问题解决能力。计算机模型作为学习工具运用于科学教学与学习中,促进学生对科学概念、科学模型的理解,这已成为当今科学教育中的一个重要趋势。既然学生逐渐广泛运用计算机模型发展自己对科学的理解,那么建立基于计算机模型的学生科学理解的测量与评价自然就显得十分必要和重要,然而这方面研究目前仍十分滞后。本论文研究创造性地将Rasch模型、计算机建模、物质结构概念理解、测量与评价4种元素有机整合,以Rasch模型以及测量建构“四基石”方法为指导,以物质结构为切入点,以计算机建模为主要学习工具和评价内容,以测量工具的开发、检验和运用为主线,从理论和实证研究相结合的层面上,深入、微观探索了计算机建模环境下中学生物质结构理解测量与评价的建构。理论研究部分,主要对与本研究密切相关的概念、理论进行综合分析与讨论,包括化学多水平表征、科学模型、计算机模型、学习进程、物质概念理解、物质结构理解、Rasch模型、测量建构“四基石”方法等。这些方面为本论文研究提供了重要的理论基础和方法论指导。在此基础上进一步建构了中学生物质结构学习进程,作为理论框架指导测量工具的开发与修订、数据的分析与解释。实证研究部分,包括三个方面研究：3个测量工具(即化学反应、溶液、酸和碱)的开发与质量检验、测量工具之间的等值化研究、学生物质结构理解的初步研究。研究对象包括中国初三、高一以及美国10&11年级约1600名学生。每个测量工具均经过两轮试测,第一轮测试结果通过Rasch模型的建模与分析,在此基础上对测量工具进行修订,并运用于第二轮测试中；第二轮测试数据再次运用Rasch模型进行分析和质量检验。同时,运用“锚—测试”设计,对3个测量工具进行等值化研究,形成原始分→0—100Rasch刻度分的分数转换表,并提出物质结构学习进程的各水平难度。基于中国初三、高一学生在化学反应、溶液、酸和碱3份测试中的测试结果,我们进一步研究中国两个年级学生基于计算机模型的物质结构理解的特征、规律与发展趋势,在定量和定性分析基础上提出了计算机建模环境下中学生物质结构学习及发展的“生态化”模型。研究发现,所开发的计算机建模环境下中学生物质结构概念理解的测量工具具有良好的信度、效度；“锚—测试”等值化设计合理、有效,充分将计算机建模环境下中学生物质结构概念理解的不同测量相互建立联系,从而更好地描述学生的学习与进步；所建构的计算机建模环境下中学生物质结构学习进程及其认知水平与序列模型充分反映了学生学习与发展的特征；高一年级学生计算机建模环境下物质结构的理解明显高于初三,高一约处于物质结构学习进程水平3,初三约处于水平2；学生的理解随着年级增长呈现出一定的规律性；学生对物质结构的理解包含物质结构模型的理解以及物质结构知识的运用,是以物质结构为中心、4种成分(物质组成与结构、物理性质与变化、化学性质与变化、守恒)相互联系、相互作用的“生态”发展过程。本研究的突出贡献在于,建构了基于Rasch模型的计算机建模环境下学生科学概念理解的测量与评价的范式,建构中学生物质结构的学习进程,开发并实证了3份计算机建模环境下中学生物质结构理解的测量工具,探索了测量工具之间的等值化方法,初步揭示了初三、高一学生计算机建模环境下中学生物质结构理解的特征与规律,具有重要的方法论、理论与实践意义。在本论文的最后,我们进一步讨论了研究的启示、局限以及未来的研究方向。更多还原

【Abstract】 Currently, science literacy is a common goal of science education around the world. To achieve the science literacy, students should develop their essentially understanding of science conceptions, what is the core component of science literacy for k-12 students. One concensus emerged in science education in last four decades had been shown that a wide range of misconceptions, alternative conceptions about science are held by students from kindegarden to senior high school and even to university sutudents, many of which are robust and static. Those misconceptions/alternative conceptions constrain students from fully and successfully understanding science conceptions. In chemistry education, one commonly agreeed primary source of students’ misconceptions is their inabilities in making connections among different levels of representations (macroscopic, submicroscopic and symbolic representation). The inability to connect different levels and translate chemical representations has been shown to limit students’development of a strong conceptual understanding of chemical phenomena and concepts. Research further suggests that transforming different representations seamlessly requires adequate understanding of models and modeling in chemistry. To facilitate students to overcome those difficulties, numerous research has focused on how to enhance students’visualization abilities in chemistry and help students make connection among different chemistry representations. Recently, chemical educators have begun to use computer models to help students make connections among three levels of representations. Research has found that, computer modeling can make the submicroscopic representations dynamic, visual, and interactive, which enable students understand submicroscopic world behaviors, make connections among different levels of reprensentation, construct more science conceptions, and improve their reasoning, scientific inquiry skill and problem-solving ability.Inevitably, there is an obvious trend in science education of using computer modeling as learning tools to improve students’ understanding of science concptions and science models. While chemistry teaching and learning based on computer models and modeling has shown promise to improve student understanding of chemistry concepts, measurement and assessment of students’ conceptual understanding basing on computer modeling is pretty important and necessary. However, such area has lagged behind, because there are no computer modeling based instruments currently available to assess students’conceptual understanding of chemistry concepts. The present study creatively used Rasch modeling to develop computer modeling-based instruments for measuring middle school students’ conceptual understanding of structure of matter, with Wiloson’ (2005) "Four Buidling Blooks" as methodology framework. Conducting both theoretic and emperical studies, we further explored how to develop measurements and assessments on students’ understanding of structure of matter basing on computer models.In theoreticl study, some key conceptions and theories, which are closely related to the present study, were discussed and summarized, e.g., multiple representations in chemistry education, science modeling, computer modeling, learning progressions, understanding of matter, understanding of structure of matter, Rasch modeling, and "four building blocks" of measurement constructing. Those theories are rationale for the study. Basing on numerous literatures analysis and synthesis, the present study proposed the learning progressions of structure of matter, which was used as theoreticl framework to guide the development, revision of measurement instruments and analysis and interpretation of data.In empirical study, there were three parts:(1) development of measurement instruements of high school students’ understanding of structure of matter basing on computer modeling, (2) equating of these measurement instruments, and (3) primary investigation on how students understand structure of matter basing on computer modeling. Around 1,600 students of grade 9 and 10 in China and grade 10&11 in the United States took these tests in two rounds of Pilot-tests. The measurement instruments used in Pilot-test I were modified basing on the analysis of data through Rasch modeling, and then used in Pilot-testⅡ. Similarly, the data generated from Pilot-testⅡwas modeled and analysized by Rasch modeling to investigate the quality of measurement instruments. Anchor-test design was used to explore the equating of three measurement instruements, and a table of conversion between raw scores and 0-100 Rasch scale scores was produced, which is more convenient for the users and makes them avoid conducting Rasch analysis every time the instruments are used. The simultaneous estimations of items in three test forms also can be used to calculate the difficulties of each understanding level in a test form. Consequently, the validated measurement instruements was used to assess high school students’ understanding of structure of matter basing on computer modeling. In term of quantitative and qualitative study of students’ response, the study proposed a dynamically ecological learning model of structure of matter.The courrent study has shown that, the measurement instruments we developed and validated in this study have good reliability and validaty; anchor-test design of equating of measurement instruments which measure the same contruct is reasonable and effective, and it is useful to compare different tests with same contruct and draw students’ progress of understanding over time; the learning progressions and dynamically ecological learning model of structure of matter appropriately describe students’learning characteristics and developmental model; the resuls of students’ examination of three tests shown that grade 10 students’ understandings of structure of matter basing on computer modeling are significantly higher than grade 9 students in China; grade 9 students appropriately stand in level 2 and grade 10 appropriately stand in level 3 of learning progression of structure of matter; students’understanding structure of matter involves understanding structure models of matter and using structure of matter to describe, interpret or predict properties and change of matter, which essentially integrate four components of understanding, i.e. compositions and structure, physical properties and change, chemical properties and change and conservation; the four components are connected and interactive and form a dynamically ecological learning model with the structure of matter at the center.In this study, we had explored an innovative method to use Rasch modeling to develop computer modeling-based measurement instruments on students’ understanding of structure of matter. It provides an example for developing computer modeling-based measurement instruments focusing on some science concepts. We developed the learning progressions of structure of matter, and three measurement instruments. The measurement instruments were validated and equated. We manifested the understanding of high school students about structure of matter basing on computer model. Thus, the current study has both profoundly theoretic and practical meanings. At the end of this dissertation, the implications, limitations and further steps were also discussed.更多还原