Navigating Success: How SNHU Uses the Learner Information Framework with AI to Provide Personalized Learning Support
Hear from Chloe Rich, Program Manager of Learner Equity Initiatives and Clay Gendron, Generative AI Data Scientist at Southern New Hampshire University
Three organizations raised their hands to pressure test the Learner Information Framework (LIF) tools and model produced by a collaborative effort of 27 contributors.
We’ve heard from our partners at Western Governors’ University and Opportunity@Work about their projects. Now, we are pleased to pass the pen to Chloe Rich, Program Manager of Learner Equity Initiatives, and Clay Gendron, Generative AI Data Scientist at Southern New Hampshire University (SNHU) to share how they are using the Learner Information Framework at SNHU.
As Program Manager of Learner Equity Initiatives, Chloe leads data-driven projects to promote equitable learning outcomes, aligning with the University Provost’s goals. By fostering collaboration, driving change management, and optimizing processes, the role ensures impactful implementation of learner equity initiatives across online and campus modalities.
Clay is a Data Scientist at Southern New Hampshire University specializing in generative AI projects. He focuses on building personalized learning tools for students and developing internal solutions that help automate tasks, enhance productivity, and improve knowledge sharing at SNHU.
What is SNHU testing with your LIF demonstration project?
Through the innovative combination of data analytics and Generative AI, SNHU is testing how best to support the success of all our learners. We aim to create tools and approaches that help students who have historically been underserved by existing services, identifying and addressing the specific barriers that prevent them from reaching their full potential. This proactive approach ensures timely, personalized support reaches students before they need to seek help, breaking down traditional barriers to accessing institutional resources. To do this, SNHU plans to use LIF tools and the power of Generative AI to provide students with new, personalized academic supports that lead to improved outcomes.
Being part of the LIF community enhances our AI Tutor learning experiences by providing access to more longitudinal data on a student’s educational and career journey—especially if other institutions adopt the framework. This community also offers an opportunity for SNHU to learn from other institutions that build learning experiences or analytical tools using LIF data, provided the elements are standardized.
Being part of the LIF community has given the SNHU team valuable external feedback from individuals who understand the framework concept but are not directly involved in building the tutor. This feedback has not only refined our approach to the tutor but also provided additional considerations that have shaped the AI Tutor’s future development.
Who can benefit from this project?
A wide range of people across the higher education landscape stand to benefit from this effort. This project aims to identify ways to advance learners’ persistence and success, especially for those facing barriers to success. By identifying learners who could benefit from additional support early on and providing tailored resources, the project can help improve engagement, retention, and graduation rates. The AI-driven system empowers students who may not typically seek help, giving them the tools to succeed.
This approach also has the potential to offer institutions a scalable, data-informed system that can transform student services. It provides valuable insights into how institutions can utilize AI to equitably and proactively support students, ensuring more effective and inclusive academic interventions.
How is SNHU approaching this effort?
Our project is structured in three interconnected workstreams.
- Identify Students Who May Need Support: We will start by identifying students who could benefit from additional support to prevent failure and/or drop-out.
- We will pinpoint learning trajectories associated with poor outcomes and student behaviors correlating with these patterns.
- We will train supervised machine learning models to find links between specific behaviors and course failure. These models use academic achievement data from a SNHU-hosted data lake housing LIF-structured data. We could leverage LIF’s extensibility feature to include historical student learning behaviors, such as academic engagement metrics, communication and help-seeking behaviors, learning resource utilization, behavioral and emotional indicators, and external factors like family, work, etc.
- We will establish a baseline using historical data, and then measure model accuracy with previously unseen data.
LIF’s learner-centered data model makes it possible to share this information, with the student’s consent, to other institutions (like any university that the student transferred to or from, or any employment organization) to support a life-long and seamless learning experience.
- Prescribe Personalized Support: Develop Generative AI systems to proactively support students who are at risk of failure and/or drop-out, addressing their unique needs and aspirations. We will use Generative AI models, combined with learner profile data, to identify gaps in students’ study approaches and understanding of course materials.
- Personalized feedback will then be delivered to the student via email and Brightspace (SNHU’s Learning Management System), and connected to student advising teams to ensure consistent messaging.
- Ongoing testing and evaluation will ensure that support does not inadvertently cause harm, such as reinforcing stereotypes or triggering negative feedback responses.
- Support Student Needs with AI Tutor: Create AI-based tutoring support that is responsive to the circumstances and needs of diverse student groups. This system is proactively helping students with course-specific insight, guidance, and feedback that is also personalized to them to make course content and assignments more accessible. Additionally, the system will identify learning behaviors that don’t result in success and respond with recommended behaviors that can improve success.
- This support will be available to students through Brightspace and in a follow-up message of support based on personalized feedback, using learner profile data and records of previous interventions. This is the most critical portion of the project, combining the predictive findings and identified gaps to personalize the learner experience with the AI Tutor.
- With help from learner profiles and the LIF, the tutor will be able to better support a student’s individual needs by calling upon historical data and interactions.
- Testing and evaluation will take place in trial courses, using both quantitative and qualitative measures to assess the AI-based tutoring support’s effectiveness in improving student outcomes and supporting learning.
How is LIF making this effort possible?
LIF components provide critical advantages.
The Translator standardizes student information—including employment and education history—before enrollment, while the Orchestrator enables precise, real-time data access for AI and machine learning systems.
Together, these tools create a centralized learner profile that automatically tracks student behaviors and identifies risk factors for academic challenges. This eliminates manual data entry and creates a single, accessible source of student information that enables real-time support across university departments.
By implementing the LIF tools, we’ve created solutions that address our immediate needs while ensuring flexibility for future innovations and needs. This common approach to organizing learner information provides a strong foundation for continuous improvement, allowing for the seamless integration of new technologies and fostering an easier partner-ecosystem where collaborators can more readily connect with our infrastructure. The LIF data standard also expands our analytical reach by providing access to a wider range of data sources, leading to deeper insights, more informed decision-making, and improved interoperability across our systems.
What do you see as the potential long-term outcomes for this effort?
This initiative has to potential to advance both learning success and institutional effectiveness through three key areas:
- Personalized Interventions – Delivering timely, customized academic support based on individual student needs and learning patterns.
- Predictive Support – Using behavioral and academic patterns to identify students that are expected to need different strategies to succeed based on their circumstances, enabling targeted interventions before challenges escalate.
- Equitable Implementation – Evaluating and mitigating potential risks of AI interventions to ensure equitable benefits across all student populations, with particular attention to marginalized learners.
What’s next? What are your future plans?
SNHU is building a technology framework that personalizes online education through data-driven insights. By leveraging LIF tools and comprehensive analytics, we continuously adapt content, course structures, and support mechanisms to each student’s needs. This dynamic approach optimizes engagement and academic outcomes while creating scalable, individualized learning experiences.
In the next phase of work, our team will test this system with two key groups: SNHU staff/faculty enrolled in online courses and human tutors operating on SNHU’s Campus. We look forward to sharing specific feedback on the AI Tutor’s effectiveness, user experience findings, and preliminary learning outcomes.
