Constructing Collaborative Individualized Education Plans Using an Artificial Intelligence Tool-CoIEP

Framework

This research is situated within two complementary theoretical frameworks that guide our investigation of pre-service teachers' perceptions of CoIEP. First, we employ a Human-in-the-Loop (HITL) AI Framework (Russell & Norvig, 2020), which maintains the essential balance between artificial intelligence support and human professional judgment in IEP development. The HITL approach emphasizes collaborative decision-making processes while positioning AI as a supportive tool rather than a replacement for educator expertise, particularly crucial in special education where human judgment and stakeholder relationships are paramount (Billingsley & Bettini, 2019). Second, we utilize a socio-technical-pedagogical lens to examine usability, which allows us to analyze how pre-service teachers interact with CoIEP across three critical dimensions: social (collaborative interactions with stakeholders), technical (interface and functionality), and pedagogical (instructional decision-making and professional learning). This dual theoretical foundation enables us to comprehensively examine how AI-supported tools can enhance pre-service teacher preparation while ensuring meaningful human oversight in the IEP development process (Bettini & Gilmour, 2024).

Methods

This qualitative study employed a socio-technical-pedagogical framework (Pham et al., 2023) to investigate how pre-service teachers interact with CoIEP, an AI-powered IEP development system. Nine pre-service special education teachers participated in semi-structured interviews designed to explore their experiences across three key dimensions: social (collaborative features and stakeholder interactions), technical (system usability and interface design), and pedagogical (professional learning support and decision-making processes) (Pham et al., 2023). The interview protocol specifically addressed participants' experiences with CoIEP's various components, including the PLAAFP Writer, IEP Goal Creator, and SDI Generator, examining how these tools supported their professional learning and IEP development process (Russell & Norvig, 2020). Data analysis followed a qualitative deductive approach (Bingham & Witkowsky, 2021), utilizing the socio-technical-pedagogical framework as our coding scheme to systematically analyze interview transcripts. This analytical approach allowed us to identify patterns in participants' experiences and perceptions while maintaining alignment with our human-in-the-loop theoretical foundation (Zeichner & Liston, 2013), particularly focusing on how pre-service teachers balanced AI support with their professional judgment in IEP development (Billingsley & Bettini, 2019).

Results

Our analysis of pre-service teachers' interactions with CoIEP, though not completed yet, is revealing insights across three key dimensions of the socio-technical-pedagogical framework. In the technical dimension, participants have expressed appreciation of the system's structured interface and navigation but have also identified opportunities for improvement in revision capabilities and prompt understanding. Several participants, so far, have expressed a desire for more focused revision options, noting that the system's tendency to regenerate entire sections when only minor changes were needed could be inefficient. Analysis results across the social dimension is demonstrative of evolving trust relationships with the system, particularly enhanced by the presence of evidence-based citations and research links. Participants consistently express value for CoIEP's potential for collaborative IEP development, especially in facilitating team-based decisions and parent communication. Across the pedagogical dimension, participants are highlighting CoIEP's effectiveness as a learning tool, particularly appreciating its evaluation features and research based suggestions. Participants have noted that the system helped them better understand IEP requirements and provided valuable guidance in developing compliant documentation. A key finding at this junction in analyzing the data, was the participants' evolution in system usage across sessions, demonstrating increased sophistication in prompt formulation and greater appreciation for the system's evidence-based foundations. These initial results suggest that CoIEP does, in fact, show promise as a professional learning tool for pre-service teachers while maintaining the critical human-in-the-loop approach to IEP development.

Importance

This study addresses critical needs in special education teacher preparation and IEP development through innovative AI integration. The research is particularly valuable to conference audiences for several reasons. First, it responds to the persistent challenge of special education staffing shortages (Bettini & Gilmour, 2024) by examining how AI-supported tools can enhance pre-service teacher preparation in IEP development. Second, our investigation of the human-in-the-loop approach demonstrates how AI can serve as a professional learning tool while maintaining the essential collaborative nature of IEP development required by IDEA. The study's socio-technical-pedagogical framework provides a structured lens for understanding how pre-service teachers interact with AI tools across social, technical, and pedagogical dimensions, offering valuable insights for teacher preparation programs and educational technology development.

References

Bettini, E., & Gilmour, A. (2024). Addressing special education staffing shortages: Strategies for schools. EdResearch for Action. https://edresearchforaction.org/research-briefs/addressingspecial-education-staffing-shortages-strategies-for-schools/

Billingsley, B., & Bettini, E. (2019). Special education teacher attrition and retention: A review of the literature. Review of Educational Research, 89(5), 697-744. https://doi.org/10.3102/0034654319862495

Bingham, A. J., & Witkowsky, P. (2021). Deductive and inductive approaches to qualitative data analysis. Analyzing and interpreting qualitative data: After the interview, 1, 133-146.

Pham, M., Singh, K., & Jahnke, I. (2023). Socio-technical-pedagogical usability of online courses for older adult learners. Interactive Learning Environments, 31(5), 2855-2871.

Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach (4th ed.). Pearson.

Zeichner, K.M., & Liston, D.P. (2013). Reflective Teaching: An Introduction (2nd ed.). Routledge.