Recent dates
How might AI help universities connect disparate evidence, identify meaningful patterns, and translate assessment findings into continuous institutional improvement?
What is the problem?
Universities generate substantial evidence about student learning, educational programs, institutional effectiveness, and the student experience.
This information can appear in assessment reports, accreditation materials, program reviews, strategic plans, surveys, institutional research, public dashboards, committee reports, student outcomes, and other sources.
The challenge is often not a lack of information. It is the ability to connect, interpret, and act on information that exists in different formats, was created for different purposes, and may be distributed across organizational units.
As a result, potentially valuable evidence can remain within individual reports or processes rather than contributing to broader institutional learning.
The challenge is to explore how AI could help universities make better use of existing evidence and move from periodic assessment toward continuous learning and improvement.
Who experiences it?
Stakeholders include faculty, academic departments, assessment professionals, institutional researchers, academic administrators, student success professionals, accreditation teams, and university leaders.
Students are ultimately important beneficiaries because institutional decisions about curricula, programs, advising, support services, resource allocation, and academic policies influence their educational experience.
Why does it matter?
Universities invest substantial time and effort collecting information and evaluating programs and outcomes. If the resulting evidence is difficult to synthesize or remains isolated within individual organizational units, institutions may miss opportunities to recognize patterns, identify emerging concerns, share successful practices, or make evidence-informed decisions.
Periodic assessment can also become focused on producing reports rather than generating organizational learning.
More effective use of evidence could help institutions recognize what is working, identify where additional investigation is needed, connect related findings across organizational boundaries, and make improvement a continuous process rather than an episodic exercise.
What is the existing solution and/or what is hard about solving it today?
Current approaches rely heavily on reports, dashboards, surveys, committees, accreditation reviews, program reviews, institutional research, and human analysis.
These approaches provide valuable expertise and judgment, but the evidence involved is often heterogeneous. It may include quantitative data, qualitative findings, narrative reports, strategic objectives, survey responses, recommendations, and historical documentation.
Information may also be produced on different timelines and at different organizational levels.
The challenge is therefore not simply summarization. A useful system would need to identify relationships across disparate evidence, preserve context, distinguish meaningful patterns from superficial similarities, explain the basis for its conclusions, and support rather than replace expert human judgment.
What would a better solution make possible?
A better solution could help universities move from asking, "What does this report tell us?" to asking, "What is our collective evidence telling us?"
AI could potentially help connect findings across reports and organizational units, identify recurring themes, surface emerging issues, recognize areas where different sources of evidence reinforce or contradict one another, identify successful practices that may be transferable, and generate questions that warrant deeper human investigation.
The goal would be a more dynamic institutional learning environment in which evidence collected for one purpose can contribute to broader understanding and continuous improvement.
To explore this challenge, contact:
Lucy Bowen: Director, Administration - Lehigh University
Email: clb222@lehigh.edu