Dr. Steve Geinitz – Research Opportunities

Canvistant: A digital teaching assistant for the Canvas LMS

Canvistant is an evolving piece of software built on top of the Canvas LMS that provides additional utilities and enhances tried-and-true instructional methods. Longer-term development goals include the integration of machine-learning models that can provide personalized follow-up assessments, predict student performance trajectories, and provide the instructor with detailed insights about how students are progressing through the course content.

Benefits and usage examples

By automating collaborative quizzes, intelligently assigning student pairs, generating tailored follow-up questions, and analyzing quiz outcomes to highlight where students may need support, Canvistant helps reduce administrative overhead and provides richer visibility into class learning patterns.

Skills a student researcher needs to contribute

Useful preparation includes Python programming and basic software-engineering practices. Experience with machine learning and/or data analysis is helpful but not required. More importantly, a curiosity about learning analytics and a willingness to work iteratively on applied software systems is what is needed.

Technology-Enhanced Pedagogical Methods

This line of research investigates and develops scalable, technology-supported extensions of proven teaching strategies such as collaborative learning, peer instruction, and continuous assessment. The goal is to design digital tools and algorithms that help instructors implement these practices more effectively and personalize them to the needs of different learners. Research threads may include modeling student knowledge over time (e.g. knowledge tracing), generating formative assessments, and evaluating how technology influences participation, understanding, and academic outcomes.

Benefits and usage examples

These methods can support dynamic student pairings, adaptive quizzes, data-informed study recommendations, and real-time feedback loops between students and instructors. Benefits include improved engagement, more accurate identification of at-risk learners, and enhanced opportunities for students to learn through structured collaboration.

Skills a student researcher needs to contribute

A background in statistics or machine learning is valuable, as is familiarity with Python or another language for data analysis. Students interested in educational techniques, human-centered computing, and/or experimentation in educational settings will find natural entry points. A comfort with designing small studies or analyzing data is also useful. Above all, most important is a desire to improve student outcomes and a willingness to commit to carrying out a study from start to finish.

Please email [email protected] if you are interested in any of these or for more information.