Research

Designing and evaluating educational tools for novice computing learners.

Research Overview

Central Research Questions

•How can we design educational tools that engage novices while scaling to large classrooms?
•What forms of feedback and scaffolding best support learning in computing?
•How should AI be integrated into programming learning environments to benefit both learners and instructors?

Our Approach

We integrate learning theories from computing education research with human-centered design methods from HCI. We create and evaluate educational tools through qualitative lab studies and large-scale classroom or field studies.

Research Themes

AI-Supported LearningFeedback & AssessmentScaffolding & PracticeParticipatory Design with Educators

Current Projects

Active and recently completed research projects.

Project image placeholder

CodeTailor

Active

2023 – Present

LLM-powered personalized Parsons puzzles that adapt to each learner's needs, providing engaging support while learning programming.

AI-Supported LearningParsons ProblemsLLMsPersonalization

Team: Xinying Hou, Zihan Wu, Xu Wang, Barbara J. Ericson

Collaborators: University of Michigan

Publication: L@S '24 — Best Paper Nomination

Paper PDF →

Project image placeholder

SQL Puzzles

Active

2022 – Present

Evaluating micro Parsons problems with different feedback types as practice for novices learning database querying.

Feedback & AssessmentSQLParsons ProblemsFeedback

Team: Zihan Wu, Barbara J. Ericson

Publication: CHI '24

Paper PDF →

Project image placeholder

AI-Supported Programming Learning Tools

Active

2024 – Present

Understanding learner and instructor needs in AI-supported programming environments to inform design of adaptive features and control mechanisms.

AI in Computing EducationAIProgramming EducationAdaptive Control

Team: Zihan Wu, Yicheng Tang, Barbara J. Ericson

Publication: AIED 2025

Paper PDF →