一种汉语自由叙述的多水平编码方法及在心理研究中的应用
A multi-level coding method for Chinese free narration and its application in psychological researcn
梁婉莹 1张雨青2
作者信息
- 1. 中国科学院大学心理系,北京,100049;中国科学院心理研究所青少年人格与健康研究中心,北京,100101
- 2. 中国科学院心理研究所青少年人格与健康研究中心,北京,100101
- 折叠
摘要
质性研究在心理学领域占据重要地位，但当前流行的质性研究方法较难兼顾自然灵动、步骤清晰、标准严谨的特点。相较之下，一种用于汉语自由叙述的多水平编码方法兼备自然与完满的特点，且相对规则化。该法包括划分大块文本以形成大类、提取不同尺寸信息颗粒以形成子类、连接子类三个阶段，与西方扎根理论和中国清代朴学均相似但有本质区别。该法对心理学研究具有支持混合分析与自证（伪）、有分析潜意识的潜力、为发现非预期相关提供可能、提高从定性到定量的精准度等特殊价值；同时具有单独使用存在风险、实施起来时间成本高等局限，有待通过改进现有算法工具、确定各步骤如何分别依托算法工具和研究者实施来突破。
Abstract
Qualitative research is significant in the field of psychology. However, the commonly used qualitative methods face difficulties demonstrating flexibility, clear steps, and rigorous standards simultaneously. In comparison, a multi-level coding method for Chinese free-narration materials shows both naturalness and data-inclusion, and is relatively regularized. It includes three stages: dividing large blocks of text to form categories, extracting information granules in different sizes to form subcategories, and connecting subcategories. It is similar to, while substantially different from, Western grounded theory and Chinese Qing Philology. It has unique value regarding psychological research, by supporting mixed analysis and self-verification/falsification, showing the potential of subconsciousness analysis, creating the possibility of discovering unexpected correlations, and improving the accuracy of qualitative-to-quantitative transformation. Meanwhile, it has limitations such as risks associated with exclusively using it and high time costs, which need to be overcome by improving existing algorithm tools and determining how the implementation of each step relies on algorithm tools or researchers.关键词
自由叙述/质性分析/数据挖掘Key words
Free narration/Qualitative analysis/Data mining引用本文复制引用
梁婉莹,张雨青.一种汉语自由叙述的多水平编码方法及在心理研究中的应用[EB/OL].(2026-04-20)[2026-04-22].https://chinaxiv.org/abs/202604.00231.学科分类
汉语/语言学
评论