科学教育的主要目标是培养学生的科学素养，而教育进展的衡量只能依靠教育测量。自新千年以来，互联网科技迅猛发展，海量信息的传递与获取变得轻而易举，民主进程对公民的科学素养提出了更高的要求。只有在公民科学地分析信息、科学地做出决策、科学地解决问题，民主的重要意义才得以发挥。基于这样的时代背景，各国在科学教育领域开始了不断的探索与实践。科学素养的测评工作必须高瞻远瞩，引领科学教育的发展。通过传统的纸笔教育测评，学生对科学内容的理解这一层次可以得到较好的评价，而学生设计和实施科学探究的能力等则大多依靠文字作答，不便记分，也不够精准。近年来，人机交互测评任务在科学教育测评中展露出了测评优势，尤其是在测查学生设计与实施科学实验、建立与使用科学模型等方面。 本研究主要的目标是测验形式的创新，即开发测量指标良好的交互测验工具。研究基于创新测验形式的视角，搜集、整理、分析了当前国际上的人机交互科学素养测验，开发了一套计算机交互形式的科学素养评价工具。测验的具体创新体现在：交互测验对科学探究与科学思维相关能力的良好测量；在计算机交互的场景上进行创新与拓展；应用基于同一问题解决情境的测验形式。 研究选取了北京某中学共81名学生作为研究样本，采用项目反应理论，对原始数据进行分析，估计得到了测验项目的有关参数以及学生的潜在科学素养水平。并结合因素分析的方法，对测验进行了单维性检验。此外，调查问卷的方式也被用于学生对测验项目、测验设计、考察能力、喜爱程度、各学科的自我效能、科学兴趣方面的调查。为获得测验适用性的有关信息，研究分析了不同性别的学生使用电脑频率是否会影响学生的作答水平。 研究结果表明，测验的项目参数指标良好，大部分项目具有中等以上区分度。测验的一致性系数为0.788。潜在科学素养与科学学科的学业成绩具有中等强度的显著相关，效度良好。根据调查结果，学生对测验的题目描述、操作设计反应良好，考察能力符合预期。学生也更为喜爱交互形式下的科学素养测验。方差分析显示，学生使用电脑的频率不会对学生作答造成显著影响。基于此，科学交互测验具备在更大范围测试的条件。另外，科学素养与学生的自我效能以及科学兴趣有着显著相关。

The main goal of science education is achieving scientific literacy and assessment is the barometer of progress toward this vast and complex goal. Since the internet made it facilitate to gain tons of information, scientific literacy becoming more and more significant to make scientific decisions for citizens in a democratic society. In this case, scientific literacy evaluation must play a role of leader in science education progress. The computer-based test, especially interactive test, has shown its merit nowadays. Compared to paper-based test, it allows students to play simulations so as to show the evidence of the abilities of evaluate and design scientific enquiry as well as interpret data and evidence scientifically and so on. The focus of this study is on developing an evaluation tool of scientific literacy with the application of computer interactive tasks. Concentrated on the innovation of method of assessment, this study did an investigation of overseas interactive assessment and developed an interactive assessment which shown innovation at: Measuring abilities related to scientific enquiry and scientific thinking well; Extending the Item form of interactive tasks; Applying the unitive scenario of problem solving. 81cases in this study were came from a school in Beijing. With the application of Item response theory, the study estimated all items’ psychometrical parameters and students’ potential scientific literacy. Good discrimination shown in the item analysis. The assessment validity was verified by the significant correlation to grade as well as unidimensional analysis that achieved a good fit. While the reliability was support by a decent Cronbach’s α(0.788). To evaluate the adaptation of this assessment, frequency of computer use was analysis as a random factor. Results shown that degree of familiarity to computer has no significant influence to the performance, which means the interactive assessment can be implement in a wider range in China. Moreover, Students expressed positive view to the items’ words and interactions. They also shown more interests toward interactive test compared to paper-based test. Besides how students feel about the interactive test, the questionnaire also contains students confident in science scales and students’ interest in science scales. The correlations between these indicators and potential scientific literacy was significant.