This chapter reports how the writing assessment of an English course in a Hong Kong university was re-designed to empower students to use AI and GenAI tools. The re-designed writing assessment helps and encourages students to use GenAI tools (e.g. ChatGPT) and other AI tools (e.g. Grammarly) responsibly in one of their main assignments, an essay writing assignment. After completing the essay, students were asked to document their use of GenAI and other tools in their reflection. The research team obtained informed consent from all 81 students taking this course in one semester to analyze their reflection. The research team adopted a machine learning approach (including topic modeling and sentiment analysis) to examine the reflection to find out how students described their GenAI-empowered assessment experience. Results suggest that students were in general positive about the use of GenAI tools in their writing assessment. In their reflection, they indicated that they mainly use it for fixing grammatical mistakes and improve their writing quality in general. With students’ positive response, this chapter describes opportunities for (1) using machine learning approach to students’ reflection; (2) empowering students with GenAI tools in writing assessment; (3) re-designing writing assessments, but this chapter also suggest concerns with the equity of GenAI tools provision.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Implementing GenAI-Empowered Assessment: An Exploration with Machine Learning Techniques

  • Dennis Foung,
  • Linda Lin

摘要

This chapter reports how the writing assessment of an English course in a Hong Kong university was re-designed to empower students to use AI and GenAI tools. The re-designed writing assessment helps and encourages students to use GenAI tools (e.g. ChatGPT) and other AI tools (e.g. Grammarly) responsibly in one of their main assignments, an essay writing assignment. After completing the essay, students were asked to document their use of GenAI and other tools in their reflection. The research team obtained informed consent from all 81 students taking this course in one semester to analyze their reflection. The research team adopted a machine learning approach (including topic modeling and sentiment analysis) to examine the reflection to find out how students described their GenAI-empowered assessment experience. Results suggest that students were in general positive about the use of GenAI tools in their writing assessment. In their reflection, they indicated that they mainly use it for fixing grammatical mistakes and improve their writing quality in general. With students’ positive response, this chapter describes opportunities for (1) using machine learning approach to students’ reflection; (2) empowering students with GenAI tools in writing assessment; (3) re-designing writing assessments, but this chapter also suggest concerns with the equity of GenAI tools provision.