大学英语写作背景下AI素养的测量与应用

Measurement and Application of AI Literacy in the Context of College English Writing

摘要: 本研究旨在开发并验证适用于大学英语写作的AI素养量表，系统揭示学习者在真实写作任务中的AI使用行为及其认知与情感反思特征。研究围绕三个问题：AI素养量表的构建与效度验证；写作任务中AI使用行为的表现及其与素养水平的关联；学习者对AI使用的反思及其所体现的素养特征。研究第一阶段通过215名大学英语学习者的问卷调查数据进行量表开发和验证；第二阶段对37名参与者进行为期12周的纵向追踪研究。研究结果显示：AI素养量表包含情感素养、行为素养、认知素养与伦理素养四个核心维度，具备良好的信度与效度；学习者的AI使用行为可归纳为四类；反思分析表明，学习者普遍经历相似的认知演化过程，高素养者更早表现出对输出内容的批判意识和对学术伦理的关注。

Abstract: This study aims to develop and validate an AI literacy scale applicable to college English writing, systematically revealing learners’ AI usage behavior in real writing tasks as well as their cognitive and emotional reflection characteristics. The research revolves around three questions: the construction and validation of the AI literacy scale; the manifestation of AI usage behavior in writing tasks and its correlation with literacy levels; learners’ reflection on AI usage and the literacy characteristics it reflects. The first phase of the study involves scale development and validation through questionnaire survey data from 215 college English learners; the second phase involves a longitudinal tracking study of 37 participants over a period of 12 weeks. The research results show that the AI literacy scale includes four core dimensions: affective literacy, behavioral literacy, cognitive literacy, and ethical literacy, demonstrating good reliability and validity; learners’ AI usage behavior can be summarized into four categories; reflective analysis indicates that learners generally undergo similar cognitive evolution processes, with high-literacy individuals exhibiting critical awareness of output content and concern for academic ethics earlier.