土生土长的
词汇
课程
词汇发展
计算机科学
人工智能
内容(测量理论)
语言学
内容分析
自然语言处理
教学方法
数学教育
母语教学
语言能力
土著教育
课程开发
心理学
定性研究
语言习得
传统知识
教育学
社会学
作者
Jia Li,Todd Cunningham,Esther Geva,Catherine E. Snow,Andrew Biemiller,Novera Roihan
标识
DOI:10.1080/09588221.2025.2569343
摘要
Given inequitable access to learning resources caused by socioeconomic and historical issues, many Indigenous students have fallen behind grade expectations. This is in particular critical at the high school level, with increased academic English language demands in their learning of content areas. This paper reports on an exploratory study using corpus analysis that examined Indigenous students’ vocabulary use in verbal narratives, a spoken corpus—the representation of adolescent oral language competence at the school. It evaluated the feasibility, in terms of vocabulary coverages, of using both human-created and AI-generated narrative curriculum materials with enriched vocabulary to support culturally responsive vocabulary instruction. The study was conducted in an Indigenous high school in Canada, located in a remote First Nation community where most students spoke Ojibwe as their first language. The results found that Indigenous students’ narratives used a significant percentage of the first 1000 high frequency words and a small percentage of the second 1000 high frequency and academic words. Human-created and AI-generated narratives had a significantly higher percentage of the second 1000 high frequency words than Indigenous students’ narratives. Finally, AI-generated narratives contained a significantly higher percentage of academic words than both Indigenous youth’s and human-created narratives. The implications of the findings indicate that a data-driven corpus-based language pedagogy can be effective developing in innovative vocabulary instructional materials for future CALL interventions to support Indigenous youths. This can be achieved in light of culturally responsive pedagogy by leveraging Indigenous oral literacy traditions of storytelling, the youths’ narrative skills and their interests and aspirations. The present study has shown that some AI tools, along with carefully piloted prompts, have the capacity to efficiently co-develop vocabulary instructional content.
科研通智能强力驱动
Strongly Powered by AbleSci AI