Top Actions to Achieve a Future of Unified Data Models in Higher Ed

• By Rhea Kelly
• 04/02/24

As institutions become more data-rich and data-reliant, their capacity for analytics is not necessarily keeping up. In an Educause QuickPoll of higher education IT leadership focused on data strategy and student success analytics, just 25% of respondents said that the current structure of data functions at their institutions is ideal for meeting data analytics needs — leaving a lot of room for improvement in the use of data on campus.

"This problem is only being exacerbated by the rapid proliferation of AI tools on campuses," according to Educause Senior Researcher Jenay Robert. Calling it an "urgent issue," she noted that the misalignment of data strategy and resources has "implications for data privacy and security, student recruitment and retention, and the holistic student experience."

To help colleges and universities envision a better future of data analytics in higher education and provide practical steps to get there, Educause has released a new "Action Plan" for achieving unified data models across the institution. The report outlines five key goals for the next 10 years:

• Institutional culture, systems, and processes center holistic student success, leveraging ethical, equitable, inclusive, accessible, and human-centered data governance practices to support students' learning experiences;
• Digital literacy training is available for all faculty, staff, and students;
• Data mesh architecture facilitates the unification of disparate data sources;
• Industry-wide interoperability standards for data dictionaries and data management provide the infrastructure necessary to combine data across functional areas and even across institutions; and
• The comprehensive learning record (CLR) has evolved to "CLR 2.0," providing a holistic, longitudinal record of student activity.

The report then provides 23 actions to help meet those goals, organized by individual, department/unit, institution, and multi-institution collaboration. Recommendations for individuals include:

• Seek data literacy training;
• Practice good data stewardship;
• Engage students with data, including their own; and
• Participate and lead in the collection of data.

Departments and units are advised to:

• Redesign workflows and job descriptions to emphasize data responsibilities;
• Prepare data tools for students to use;
• Use a data mesh architecture;
• Ensure that technology infrastructure can scale and adapt to changing data landscape requirements;
• Establish defined data-quality assurance checks;
• Create clear processes and tools for data integration, access, and security;
• Design a data-collection strategy that aligns with your strategic goals and mission; and
• Provide students, faculty, and staff with data literacy training.

Institutions should:

• Establish requirements for institution-level professional development on data literacy;
• Provide financial and cultural support for a data-forward mindset;
• Invest in research and development related to using AI for analytics;
• Gain financial and cultural buy-in from executive leaders to unify and democratize data;
• Establish an institution-wide data community of practice;
• Create institution-wide data governance strategy, guidelines, and policies; and
• Integrate information security in all strategy and processes.

And finally, collaborative actions for institutions include:

• Join a national or international community of practice;
• Create industry-wide standards for unified data models, with flexibility for customization;
• Facilitate shared datasets and tools that are built around industry-wide standards; and
• Advocate for transparency and fair pricing from industry partners.

The full report provides a deeper look at the unified data goals and actions, and provides planning activities and tools institutions can use to build their action plan. The "2024 Educause Horizon Action Plan: Unified Data Models" is available here on the Educause site.