作为21世纪核心技能，批判性思维受到国内外教育界广泛关注。近年来，批判性思维培养和评价在外语教育领域引起高度重视。外语教育界开展的批判性思维评价研究主要通过分析外语写作文本，推理批判性思维能力。迄今为止，相关研究主要关注大学外语学习者，针对高中生的批判性思维能力评价研究相对匮乏，而且以往研究构建的批判性思维能力评价框架主要以理论构建和经验总结为主，缺乏通过实证研究进行效度验证。发展心理学和脑科学研究表明高中生已具备一定的批判性思维能力，《普通高中英语课程标准（2017年版）》对发展批判性思维能力也提出了明确要求。鉴于此，本研究以批判性思维三元结构模型为理论框架，采用质性研究和量化研究相结合的混合研究方法，构建高中英语写作中的批判性思维能力评价框架并对其进行效度验证，主要回答以下两个研究问题： 研究问题1：如何构建高中英语写作中的批判性思维能力评价框架？ 研究问题2：高中英语写作中的批判性思维能力评价框架效度如何？ 针对第一个研究问题，基于批判性思维三元结构模型，首先将思维元素与批判性思维内涵和外语写作相关研究对接，初步筛选外语写作中的批判性思维能力评价指标。然后通过对6名专家型高中英语教师和4组高中学生进行访谈，结合高中英语写作实际和高中生心智发展状况初步修订评价指标。为使评价指标更具科学性和可行性，进一步通过对4位英语学科专家深度访谈修订评价指标，并通过调查问卷征求更多学科专家和一线教师的意见。7位学科专家和35位高中英语教师在线填答了问卷，根据问卷调查结果，并结合批判性思维三元结构模型中的思维标准和评价等级，进一步细化评价指标在高中英语写作中的表现。最终构建的评价指标包括提出问题、表达观点、提供证据、推理论证、得出结论、阐释评价，每个评价指标按照清晰性、准确性、精准性、相关性、完整性、逻辑性、重要性、公正性等思维标准分为0、1、2、3、4五个水平。 针对第二个研究问题，数据来自3个省市6所高中共913名高中生，其中男生394人（43.2%），女生519人（56.8%），包括高一、高二和高三3个年级。首先基于批判性思维能力评价指标确定写作测评框架、写作任务类型、写作话题等，并在此基础上编制双向细目表，编写试题，然后通过专家外审、预测试、学生访谈等步骤进一步修订题目，最后形成难度、区分度、文本易读度等指标符合测量学规范的英语写作任务。根据基于论证的效度验证模式，本研究通过英语写作测试、被试答题过程分析、评分员有声思维、评分员访谈以及沃森-格来泽批判性思维测评收集数据。根据基于论证的效度验证框架和数据特征，对量化数据进行描述性统计、相关分析、多面Rasch模型分析、差异分析，对质性数据进行转写、编码和语义分析。研究结果表明，评分推论环节的效度证据包括评价指标构建基于理论基础和相关文献、指标构建过程科学、指标具有适切性、评分过程科学、评分员能按照评分标准评分、评分员一致性良好等。概化推论环节的效度证据包括观测样本具有代表性、不同写作任务评分具有一致性、被试能力范围分布合理、写作任务难度合理等。外推推论环节的效度证据包括完成英语写作任务所需的批判性思维能力与理论预期一致、英语写作中的批判性思维能力评分与其他测量相关。解释推论环节的效度证据包括英语写作中的批判性思维子技能相关合理、评分标准区分度良好、不同评分代表不同的批判性思维能力。 本研究构建高中英语写作中的批判性思维能力评价指标并对其进行效度验证，弥补了国内外相关研究的不足，在理论、实践和方法层面上均具有创新性。从理论层面讲，本研究构建的批判性思维能力评价框架基于丰富的理论基础并经过实证研究检验，不仅是一个教育测评工具，也是批判性思维的理论模型，是对既往批判性思维评价模型的有益补充。从实践层面讲，本研究构建的评价框架和效度验证过程既是特定学科领域开展批判性思维评价研究的有益尝试，也是核心素养评价研究的有益尝试。从方法层面讲，本研究采用混合研究设计，综合使用多种数据收集和分析方法，使研究数据和分析结果之间可以实现三角验证，从研究方法角度保证了研究结果的信度和效度，可以为后续研究提供方法论方面的借鉴和参考。本研究符合批判性思维测评研究的发展趋势，既对批判性思维测评研究具有一定的启示意义，又对外语教学中的批判性思维培养具有启示意义。 但是，本研究也存在一定的局限性。比如写作文本的多样性和样本的代表性问题还需要进一步完善，评分员对批判性思维能力评分标准的理解和把握可能受重语言轻思维的惯性因素影响，研究者的主观性也可能会影响质性数据分析结果的精准性。本研究虽以批判性思维三元结构模型为理论框架，但研究者只关注了其中的二元，即思维元素和思维标准，并没有考查思维倾向。尽管本研究提供了很多效度证据，但是对评价框架的效度验证是一个长期过程，需要持续不断地从不同方面收集证据，证明分数解释和分数使用的有效程度。

As a core skill of 21st century, critical thinking has been widely concerned by educational circles at home and abroad. In recent years, the cultivation and evaluation of critical thinking has attached great attention in foreign language education. The evaluation of critical thinking in foreign language education mainly focuses on inferring critical thinking ability through written text analysis. So far, relevant researches mainly focus on foreign language learners in universities. There is not enough research about high school students. Moreover, the evaluation framework of critical thinking ability constructed in previous studies are mainly based on theoretical construction and experience summarization, which lacks validation through empirical research. Research in developmental psychology and brain science showed that senior high school students have been equipped with certain critical thinking ability. And the English Curriculum Standards in Ordinary Senior High School (2017 edition) sets clear requirements for critical thinking development. Therefore, this study takes three-dimension structural model of critical thinking as the theoretical framework and adopts the mixed research method to build the evaluation framework of critical thinking in senior high school English writing and verify its validity, mainly answering the following two questions: Research question1: How do we construct the evaluation framework of critical thinking ability in senior high school English writing? Research question 2: How about the validation of the evaluation framework of critical thinking ability in senior high school English writing? For the first research question, the evaluation indicators of critical thinking ability in foreign language writing were firstly selected through the connection of thinking elements with the connotation of critical thinking and the related research on foreign language writing. Then the evaluation indicators were initially revised according to senior high school English writing practice and their mental development status based on interviews with 6 senior high school English teachers and 4 groups of senior high school students. 4 discipline experts were further interviewed in order to make the evaluation indicators more scientific and feasible. Moreover, 7 discipline experts and 35 senior high school English teachers were surveyed online for more suggestion for evaluation indicators. Finally, according to the survey, thinking standards and evaluation levels in three-dimension structural model of critical thinking, the evaluation indicators of critical thinking ability in senior high school English writing was further refined. The final evaluation indicators include questioning, expressing opinions, providing evidence, reasoning and argumentation, drawing conclusion, and evaluation. According to thinking standards such as clarity, accuracy, precision, relevance, depth, width, fairness, integrity, logicality, importance, each evaluation indicator is divided into 5 levleds, including 0, 1, 2, 3, and 4. For the second research question, data were collected from 913 senior high school students in 3 provinces, including 394 male students (43.2%) and 519 female students (56.8%), covering grade 1, grade 2 and grade 3. First, writing assessment framework, the types of writing task, writing topics, etc. were determined based on the evaluation framework, and the bidirectional checklist and then items were compiled. Then further revision was made through the expert examination, pilot test and interview with students in order to form English writing items with the difficulty, degree of differentiation and readability in line with measurement standards. In accordance with argument-based approach to validation, the data were collected through English writing test, writinging process analysis, thinking aloud of raters, interview with raters and the Watson-Glaser critical thinking test. According to the validation framework and data characteristics, descriptive statistics, correlation analysis, many-facet Rasch model analysis and variance analysis were conducted for quantitative data, and qualitative data was transcribed, coded and analyzed as well. The results showed that validity evidence for the scoring inference includes the construction of evaluation indicators based on theories and related literature, the scientific process of indicator construction and scoring, raters scoring based on scoring rubric, and good consistency of raters; the validity evidence for the generalization includes the representativeness of the samples, the consistency of writing tasks, the reasonable distribution of the subjects; the validity evidence for the extrapolation includes the consistency between critical thinking skills needed to complete an English writing task and theoretical expectations, and the positive correlation between critical thinking skills in English writing and other measures; the validity evidence for the explantion includes the reasonable correlation of critical thinking sub-skills in English writing, the scientific differentiation of scoring rubric, and different scores representing different critical thinking abilities. The evaluation framework of critical thinking ability in senior high school English writing were constructed and validated in the present research, which may fill in the deficiency of relevant research at home and abroad. And it is innovative in theory, practice and research method. From the perspective of theory, the evaluation framework of critical thinking ability constructed in the present study is based on abundant theories and tested by empirical research. It is not only an evaluation tool, but also a theoretical model of critical thinking, which may be a great supplement to the previous critical thinking evaluation model. From the perspective of practice, the evaluation framework and validation process in the present study is not only an attempt to evaluate critical thinking in specific subject, but also an attempt to evaluate core literacy. From the perspective of methodology, this study adopts mixed research design and uses a variety of data collection and analysis methods in order to enable the triangulation among the research data and the results. The reliability and validity of the research results are guaranteed from the perspective of research methods, which can provide enlightenment in methodology for future research. The present research was in line with the development tendency of critical thinking assessment, which has implications for critical thinking assessment and critical thinking cultivation in foreign language teaching. However, there were some limitations. For example, the diversity of writing tasks and the representiveness of samples need to be further improved, the raters’ understanding of scoring rubrics for critical thinking ability in English language writing may be influenced by the traditional habits that emphasize language over thinking, and the subjectivity of the researcher may also affect the accuracy of qualitative data analysis. Although this study takes three-dimension structural model of critical thinking as the theoretical framework, the researcher only focused on two of them, that is, thinking elements and thinking standards, and the thinking disposition was not examined. Although this study provides a lot of validity evidence, validation of evaluation framework is a long-term work, which requires continuous evidence collection from different aspects to prove the effectiveness of score interpretation and score utilization.