Shifting School
Assessing Educational Leaders Who Are Trained to Practice a Supportive and Inclusive School Culture : Integrating AI Learning in Our Leadership Preparation Program
Paper Presentation in a Themed Session
Douglas Hermond,
L.S. Spencer, Jr.
The mission of our Educational Leadership department is to train educational leaders to meet the challenges of our schools, particularly those that struggle to meet standard academic expectations. Our investigation determined whether our leadership graduates are capable of optimizing the success of every student, as specified by the National Educational Leadership Program standards. Such organizational resilience occurs along three leadership skill dimensions: a. Leaders can use data to advocate for a supportive and inclusive school culture, b. They can evaluate educational resources, technologies, and opportunities that support the educational well-being of each student, and, c. They can advocate for equitable, inclusive, and culturally responsive instruction and behavior support practices. Additionally, AI has become ubiquitous in the educational landscape, with educational leaders struggling to integrate AI into school curricula and instruction. To address this nascent challenge, we interjected the following queries: d. Did Leaders received competent guidance on leading schools to integrate AI into the learning and teaching process so each student can be successful? e. Can they apply practical solutions to educators so that they can balance the value and challenges of AI? Our survey of recent graduates indicate that our prospective leaders are adept at advocating for a supportive and inclusive school culture, and are competent at monitoring, cultivating, and advocating for culturally responsive instruction and behavior support practices among teachers and staff. The data also indicates that we are in the very early stages of preparing our leaders for the influence of AI in our urban schools.
How to Optimize Gen AI Tools in K-12 Mathematics Classes: An Investigation of K-12 Teachers Embodying ChatGPT 4.0 as a Student's Learning Partner in the U.S.A. View Digital Media
Paper Presentation in a Themed Session
Hsuehi Lo
The study investigates K-12 teachers’ levels of AI literacy and to identify the challenges and opportunities of implementing ChatGPT 4.0 in their math classes. ChatGPT 4.0 played the role as students’ partner to create simultaneous prompting questions in their mathematics classes in the U.S.A. Research (Saclarides & Harbour, 2023) shows one-on-one interactive learning process vis-à-vis simultaneous prompting questions (Morse, 2023; Sönmez & Alptekin, 2020) produced high effective K-12 math learning outcomes. The study examines how Gen AI tools can support simultaneous prompting questions in one-on-one learning process in K-12 math classes. Fifty-three K-12 in-service teachers took an AI literacy survey as a baseline to prepare for their ChatGPT 4.0 training. Twenty K-12 teachers implemented inquiry-based model so students have self-learning time to work with ChatGPT 4.0’s simultaneous prompting questions. After one month of ChatGPT 4.0 intertwined in math learning process, the twenty K-12 math teachers participated in the follow-up interviews to identify the challenges and opportunities of using ChatGPT 4.0 in their math classes. The findings show that significant challenges of using ChatGPT 4.0 are the limitation of understanding students’ questions. K-12 math teachers agreed that ChatGPT significantly save their time so they can spend time in focused students. ChatGPT 4.0 provided more opportunities for K-12 teachers to improve math methods with AI. How K-12 teachers optimize AI tools in their mathematics methods and how AI algorithm can be trained by a neural network in different math classes is discussed.
Exploring the Limitations and Potential of Large Language Models in Junior High School History Education
Paper Presentation in a Themed Session
Yun Jung Hsu
This study utilizes ChatGPT to evaluate the probability and specific topics of artificial intelligence hallucinations (AI hallucinations) in junior high school history subject using contemporary large language model technology. Furthermore, it explores whether large language models can become a reliable learning tool that students can trust at this stage. Because large language models possess multiple abilities that learners must achieve in the cognitive field, they have been widely used in teaching to improve students' learning efficiency. However, when large language models generate text, they may produce answers that are inconsistent with the facts due to variations in training datasets across languages. This situation is called "AI hallucination". History is a subject that emphasizes local knowledge and culture. The traditional culture or proper nouns involved may never have appeared in the language model's training datasets, thereby increasing the probability of AI hallucinations and potentially affecting students' learning. Therefore, this study uses the item bank of the National Academy for Educational Research to test and quantify the probability of AI hallucinations occurring when answering questions in ChatGPT. The results show that the probability of AI hallucinations in each unit of Taiwanese history ranges from 1.54% to 10.77%.Therefore, in the humanities and social science subjects, as the amount of digital text in artificial intelligence training increases, the probability of hallucinations can be significantly reduced, thereby improving the quality of generated content for educational use.
