随着21世纪课程教学改革的开展，项目学习已经成为国内外教育界关注的焦点问题。项目学习的开展已经成为各国数学课程改革的重要内容。构建项目学习测评框架，开发项目学习测试工具，可以有效地指导教育者评价学生的能力和素养的发展水平，检验教学成效和提升教学质量。特别地，八年级学生正处于抽象思维发展的关键时期，在开展数学项目学习探究活动上具有较大潜力。因此，进行八年级学生数学项目学习表现的测评研究具有重要意义。

本研究依托于“2022-2023 初中数学素养测评与诊断改进项目”，以“八年级学生数学项目学习表现的测评研究”为主题展开探索，内容包含“构建数学项目学习的测评框架”、“开发数学项目学习的测试工具”和“八年级学生数学项目学习表现的研究”三个方面，方法上将质性研究和量化研究相结合，一方面，开展学生访谈、预测试及多次试卷修订工作，以保证测试工具的质量；另一方面，利用项目反应理论估计学生的能力值，并结合具体试题分析学生在各个过程维度上的表现。最后，研究者对测试结果进行分析，得到本研究的主要结论如下：

第一，构建的数学项目学习测评框架合理稳定。本研究所构建的数学项目学习测评框架分为试题情境、活动过程、认知水平和数学素养四个维度，每个维度下又分为若干二级维度。基于Rasch模型的大规模测试结果显示，测试工具与该框架的匹配程度较高，说明本研究的测评框架合理稳定。

第二，编制的数学项目学习测评工具科学有效。在经历学生访谈、预测试、专家评定后，形成的数学项目学习测试工具共包含15道试题，施测对象为两所学校的八年级学生。测试结果显示，测试工具的整体信度为0.76，每道试题的区分度均在0.4及以上，拟合度都介于0.7到1.3之间。

第三，八年级学生数学项目学习测试表现具有明显特征，具体表现为：（1）八年级学生数学项目学习测试的总体表现存在差异；（2）不同水平学生在数学项目学习测试中各维度的表现存在差异；（3）不同学校学生在数学项目学习测试中各维度的表现存在差异.

本研究构建的测评框架契合数学项目学习的理论研究需要，所开发的测试工具适用于开展八年级学生数学项目学习大规模测评，研究结论能为当代数学教育研究者和一线数学教师提供一定的参考依据。

With the development of curriculum and teaching reform in the 21st century, project-based learning has become a focus of attention in the education industry at home and abroad. The implementation of project-based learning has become an important part of mathematical curriculum reform in various countries. Building a project-based learning assessment framework and developing project-based learning testing tools can effectively guide educators in evaluating the development level of students' abilities and literacy, testing teaching effectiveness, and improving teaching quality. Especially, eighth grade students are in a critical period of developing abstract thinking and have great potential in carrying out mathematical project-based learning and exploration activities. Therefore, conducting research on evaluating the performance of eighth grade students in mathematical project-based learning is of great importance.

This study is based on the "2022-2023 Junior High School Mathematics Literacy Assessment and Diagnosis Improvement Project" and explores the theme of "Evaluation Research on Mathematical project-based learning Performance of Eighth Grade Students". The content includes three aspects: "Building a Evaluation Framework for Mathematical project-based learning", "Developing Testing Tools for Mathematical project-based learning", and "Research on Mathematical project-based learning Performance of Eighth Grade Students". The method combines qualitative and quantitative research. On the one hand, student interviews, pre-tests, and multiple paper revisions are carried out to ensure the quality of testing tools; On the other hand, using project response theory to estimate students' ability values and analyzing their performance in various process dimensions in conjunction with specific test questions. Finally, the researchers analyzed the test results and concluded that the main conclusions of this study are as follows:

Firstly, the constructed mathematical project-based learning evaluation framework is reasonable and stable. The mathematical project-based learning assessment framework constructed by this research institute is divided into four dimensions: test scenario, activity process, cognitive level, and mathematical literacy. Each dimension is further divided into several secondary dimensions. The large-scale testing results based on the Rasch model show a high degree of matching between the testing tool and the framework, indicating that the evaluation framework of this study is reasonable and stable.

Secondly, the developed mathematical project-based learning assessment tool is scientifically effective. After undergoing student interviews, pre-tests, and expert evaluations, the mathematical project-based learning test tool consists of 15 questions, and the test subjects are eighth grade students from two schools. The test results show that the overall reliability of the testing tool is 0.76, and the discrimination of each test question is 0.4 or above, with a fitting degree between 0.7 and 1.3.

Thirdly, the performance of eighth grade students in mathematical project-based learning tests has obvious characteristics, specifically as follows: (1) there are differences in the overall performance of eighth grade students in mathematical project-based learning tests; (2) There are differences in the performance of students of different levels in various dimensions of mathematical project-based learning tests; (3) There are differences in the performance of students from different schools in various dimensions of mathematical project-based learning tests

The evaluation framework constructed in this study is in line with the theoretical research needs of project-based learning in mathematics. The developed testing tool is suitable for conducting large-scale evaluations of project-based learning in mathematics for eighth grade students. The research conclusions can provide a certain reference basis for contemporary mathematical education researchers and frontline mathematics teachers.