Skip to main content Skip to navigation

Integrating Personalized Parsons Problems with Multi-Level Textual Explanations to Scaffold Code Writing

Project Overview

The document explores the innovative application of generative AI in education, specifically through the implementation of a system called CodeTailor, which focuses on enhancing the coding skills of novice programmers. By leveraging large language models (LLMs), CodeTailor personalizes the learning experience by offering tailored Parsons problems accompanied by multi-level textual explanations. This approach aims to improve students' understanding and engagement in coding by adapting to their individual learning needs. The findings indicate that such personalized learning tools not only foster better comprehension of programming concepts but also promote active participation in the learning process. Overall, the integration of generative AI in educational contexts exemplifies a significant advancement in creating adaptive learning environments that cater to diverse student requirements, ultimately leading to improved educational outcomes in coding and programming proficiency.

Key Applications

CodeTailor

Context: Introductory programming courses for novice programmers, specifically those learning Python.

Implementation: CodeTailor provides real-time, on-demand personalized Parsons problems based on the student's current code state, offering varying levels of textual explanations to aid understanding.

Outcomes: The initial study showed that CodeTailor is engaging and beneficial for learning, helping students better understand programming concepts.

Challenges: Some students reported challenges in understanding the meaning of certain code blocks, which could limit the effectiveness of the scaffolding provided.

Implementation Barriers

Understanding Barrier

Students may put code blocks in the correct order without fully understanding the rationale behind the correct solution, which compromises learning.

Proposed Solutions: Incorporate multiple levels of textual explanations to enhance comprehension of the code blocks and their purposes.

Project Team

Xinying Hou

Researcher

Barbara J. Ericson

Researcher

Xu Wang

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Xinying Hou, Barbara J. Ericson, Xu Wang

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies