The CGScholar AI Helper effectively supports students' writing development through customized AI feedback, though improvements in feedback length and language complexity are needed.

Objective: To examine the impact of the CGScholar AI Helper on 11th-grade students' writing development in English Language Arts (ELA) and explore how AI-driven feedback can support students' writing improvement.

Methods:

• Qualitative case study involving 6 students and one teacher in a diverse, low-income Midwest US school
• Students completed a 200-word writing assignment comparing two texts about Indigenous values
• Implementation of CGScholar AI Helper providing customized feedback based on teacher's rubric
• Data collection through observations, teacher post-survey, student focus groups, and writing samples
• Analysis of students' initial and revised writings using six evaluation criteria
• Use of reflexive thematic analysis for data interpretation

Key Findings:

• Five out of six students improved in at least one writing criterion
• One student improved in three criteria
• Students showed most improvement in Compare and Contrast, Compose, and Analyze criteria
• AI feedback was perceived as helpful, direct, specific, and actionable by students
• Teacher reported the tool effectively motivated students to revise their work
• Main critique was that AI feedback was too lengthy and language too complex

Implications:

• Demonstrates potential of customized AI feedback in supporting writing development
• Shows value of integrating AI tools with teacher rubrics and materials
• Suggests benefits of AI feedback particularly in large writing classes
• Highlights importance of calibrated AI use in educational contexts

Limitations:

• Small sample size (6 students)
• Incomplete data from some students (missing pre/post-surveys)
• Focus on single implementation case
• Limited to one specific writing task
• All participants were second language English speakers

Future Directions:

• Development of more concise feedback options
• Implementation of chat boxes for feedback clarification
• Enable customization of feedback length
• Conduct broader studies with larger sample sizes
• Strengthen data collection methods
• Further research on AI feedback integration in classrooms

Title and Authors: "The Impact of AI-Driven Tools on Student Writing Development: A Case Study From The CGScholar AI Helper Project" by Raigul Zheldibayeva, Ana Karina de Oliveira Nascimento, Vania Castro, Mary Kalantzis, and Bill Cope

Published On: Not explicitly stated, but references indicate 2024-2025 timeframe

Published By: Not explicitly stated in the provided excerpt