【摘要】 师范生教学实践能力评价是教师教育质量保障的关键环节，但传统评价体系长期存在评价维度碎片化、功能异化与主体协同不足等问题，难以全面反映师范生的真实教学实践能力。基于此，研究通过系统文献分析，梳理了以师范生教学实践能力为评价对象的历史演变，将其划分为技能本位的量化考核(20世纪末—21世纪初)、标准导向的多元整合(2010年—2020年)、情境化与智能化融合(2020年至今)三个阶段。研究构建了包含数据层、分析层、决策层和展示层的四级评价模型。在此基础上，以Z大学开展的微格教学为例，展示了人机协同评价的结果。研究发现，机器评价和教师评价的结果呈现差异性，这表明二者并非简单的替代关系，而是一种能力互补关系。更多还原

【Abstract】 The evaluation of practical teaching ability of normal university students is a critical component in ensuring the quality of teacher education. However, traditional evaluation systems have long suffered from issues such as fragmented evaluation dimensions, functional alienation, and insufficient collaboration among stakeholders, making it difficult to comprehensively reflect the actual teaching ability of normal university students. Based on this, through systematic literature analysis, the historical evolution of the evaluation of pratical teaching ability for normal university students has been sorted out, which is divided into three stages: skill-based quantitative assessment(late 20th century to early 21st century), standard-oriented multidimensional integration(2010-2020), and the integration of contextualization and intelligence(2020 to present). This paper constructs a four-level evaluation model comprising a data layer, analysis layer, decision-making layer, and presentation layer. On this basis, the article uses micro-teaching conducted at Z university as an example to demonstrate the results of human-machine collaborative evaluation. The study finds that there are differences between machine evaluation and teacher evaluation results, indicating that the two are not in a simple substitution relationship but rather in a complementary state of capability.更多还原