Since 2018, the Friday Institute for Educational Innovation and the NC State College of Education have provided funding for more than 25 research projects through Catalyst Grants.

Catalyst Grants support collaborative research projects between faculty and staff of the two groups and aid research that could lead to further inquiry and support. This past year’s Catalyst Grants were investigating how to use computational thinking to empower young girls; the levels of burnout, secondary stress and compassion for pre-service teachers; and the ways instructors can use artificial intelligence as an instructional aid in online courses. 

“It’s really clear that there is something actually catalyzing about the work,” said Krista Glazewski, Friday Institute executive director and associate dean for translational research in the College of Education. “Research is an active pull, and when we engage in a way that reminds us of something that we are able to be a part of, and even speaks to potential, or it speaks to change, or it speaks to a possibility, I’m always reminded about just how hopeful and optimistic the world can actually be. So we are moving forward knowledge in these tiny little increments, but we are also contributing to an act of hope.”

On Oct. 2, three Catalyst Grant project teams presented their work from the last year, since their grants were initially awarded. Learn more about the three projects’ yearlong work below.

Empowering Rural Young Girls: Enhancing Computational Thinking Through a STEAM Club

Research has shown that exposure to computing and computational thinking (CT) at an early age can foster an interest in science, technology, engineering and mathematics (STEM) fields. Also, hands-on learning and arts integration can deepen the understanding of CT, physical computing and general STEM concepts.

“We are hoping to use this approach to help students connect those abstract computational thinking concepts to tangible, hands-on experiences,” said Joey Huang, assistant professor of learning, design and technology in the College of Education. 

Partnering with West Middle School’s Creative Computing Club in Mount Gilead, North Carolina, Huang, along with Rebekah Davis, a research scholar at the Friday Institute; and former computer science research scholar at the Friday Institute Carrie Robledo, led a five-session workshop where 15 students explored both digital and physical computing. Sessions focused on their proprietary ArtAlgo platform to connect digital art to physical interactions. The team introduced CT and physical computing concepts and skills to students through making and crafting activities and fostered positive engagement, motivation and empowerment toward computing and STEM. They also incorporated professional development for teachers and club facilitators, equipping them with computer science (CS) instructional training and other CS integration strategies, ensuring that the teachers have the necessary knowledge and skills to effectively support and enhance CT learning in the club.

The insights and data gathered through this work have been instrumental in shaping research questions, design goals and a long-term vision for supporting creative computing and AI-assisted learning in informal settings. Researchers have seen how youth treat AI as a tool, peer and partner, and their research raises questions of student agency, trust and AI design.

The Impact of SEL Professional Development & Wellness Skills on Teacher Retention

Teacher attrition is highest for new teachers in North Carolina, with an attrition rate between 11.7% and 17.5% for teachers with less than five years of experience.  One common factor to predict teacher retention is teacher perceptions of health-related quality of life. 

Initially, the purpose of this project was to determine the effectiveness of providing the College of Education’s elementary and middle school-level pre-service teachers with strategies and skills for resilience and self-care, as well as an understanding of SEL and how to implement it in their classrooms. But the project team, including Alicia Fischer, research scholar at the Friday Institute; Gail Jones, Distinguished University Professor of Science Education in the College of Education and senior faculty fellow at the Friday Institute; and Krista Holland, former professional learning scholar at the Friday Institute; shifted to focus on the levels of burnout, secondary stress and compassion for pre-service teachers between the fall and spring semesters. 

The team found a significant difference in levels of secondary stress, burnout and compassion between the semesters, with participants reporting higher levels in the fall than spring. In the future, the team hopes to discover why those differences exist.

My Teaching Copilot: Prototyping an AI Assistant to Facilitate Cognitive Presence in Collaborative-Inquiry Online Learning

As more learners enroll in online courses, many instructors are not prepared to facilitate collaborative inquiry in virtual environments, resulting in a perceived lower quality of instruction. This project aims to remedy this gap by pioneering the application of large-language models through the OpenAI platform to create an assistant to support online instructors in discussion board facilitation in real time. 

“Most online instructors have huge work,” said Ela Castellanos-Reyes, an assistant professor of learning, design and technology in the College of Education. “We don’t have enough hours in the day. We do care about our students. We want to do better.”

The proposed project builds on Castellanos-Reyes’ ongoing work of automated classification of online learners’ cognitive presence using GPT models and the production of an AI-adapted cognitive presence codebook. Cigdem Meral, research assistant at the Friday Institute, worked with Reyes on this project. 

This project team conducted a mixed-methods study collecting data from nine instructors using My Teaching Copilot, an OpenAI-powered assistant pre-loaded with an AI-adapted cognitive presence codebook and instructional advice that aimed to support online instructors’ understanding of cognitive presence in students’ text interactions. My Teaching Copilot classified and then proposed questions that the instructor could use to support students to move from one specific phase of the inquiry to another. The researchers investigated to what extent the large language model compared to instructors’ and researchers’ manual examination of online learners’ text-based interactions in terms of reliability and efficiency. 

The team found that a human-in-the-loop approach is necessary to support the data and questions the AI assistant provides, but there is a high need for help facilitating collaborative knowledge construction through a tool like this that offers more efficiency for online instructors.