Data Fluency as a Strategic Imperative
A Conversation with Ellen Wagner
It's been decades since higher education leaders and analysts first alerted institutions about a "data tsunami" that would challenge both infrastructure and academic programs. How would legacy data centers keep up? And how would curriculum in most disciplines, as well as research and administration change?
We asked Ellen Wagner, predictive analytics reporting (PAR framework) pioneer, seasoned consultant to higher education, and perhaps the ultimate digital learning activist, for a sense of where we've been on our data journey, and for a structured view of the data capabilities higher education institutions can foster for success.
Mary Grush: What was it like to be in the education data trenches "back in the day"?
Ellen Wagner: There were many people, including me, engaged in an extensive conversation about "big data in education" — which was, notably, a national trend fueled substantially by the Bill & Melinda Gates Foundation. The idea was to focus on student success and leverage the growing interest in big data to explore how to address retention, progression, and completion problems. There were maybe 7 crucial years — from about 2009 until 2016 or so — during which predictive analytics matured enough not to be seen as "the devil's work" in education!
There were maybe 7 crucial years — from about 2009 until 2016 or so — during which predictive analytics matured enough not to be seen as "the devil's work" in education!
Grush: Obviously there were many trials and discoveries over the years, but could you pick one lesson learned about the value and role of data — one that was perhaps the most important?
Wagner: Sure. Data are the means of achieving various ends, through analysis and decision making. Being clear about what your ends are — and finding the essential information needed to support the search for answers about how best to achieve those ends — is what's truly important.
Data are the means of achieving various ends, through analysis and decision making. Being clear about what your ends are — and finding the essential information needed to support the search for answers about how best to achieve those ends — is what's truly important.
We obtained insights from exploring data patterns; we evaluated the tenability of our hypotheses. We tested our assertions over and over again to ensure that tenability. But then, we still needed to figure out what we were going to do, actually, with the information we learned, in terms of supporting students better.
So, it ended up not really being 'all about the data' — or having the 'most' data…
Grush: …It's about clarity.
Wagner: Yes.
Grush: And to bring up a more current issue relevant to our data policies and practices, today we often hear a plaintive cry in higher education, especially in instructional circles: "Am I losing my expert status to AI?"
Is this a concern we will one day resolve?
Wagner: I do think that a very important question on people's minds these days revolves around the accuracy and efficacy of student performance data along with the role that AI is and/or isn't going to play in figuring this out for institutions.
A very important question on people's minds these days revolves around the accuracy and efficacy of student performance data along with the role that AI is and/or isn't going to play in figuring this out for institutions.
Grush: So is this unresolved today but something we may eventually look back on as a key lesson learned along with the development of all the other competencies and capabilities surrounding data?
Wagner: Hopefully it will be resolved, along with a myriad of other important questions!
Grush: You've talked about a "three tiered" view of data capabilities — from a data literate enterprise, to data proficient leaders and managers, to data fluent specialists. Could you tell us a bit about each of these tiers?
Wagner: Of course. The ability to leverage data for decision making can be viewed as a spectrum of increasing sophistication and integration.
The ability to leverage data for decision making can be viewed as a spectrum of increasing sophistication and integration.
These levels are still emerging within our contemporary enterprises:
Data literacy encompasses the fundamental skills needed by everyone in a data informed organization. It includes the basic ability to read, work with, analyze, and communicate with data in context. Everyone — students, faculty, staff, top-level administrators — belongs in this tier.
Data proficiency builds upon data literacy but extends further into application, evaluation, and decision making. This tier is particularly crucial for leaders and managers from government, education, and industry who must translate data insights into organizational action without necessarily being data specialists themselves.
Data fluency represents the highest level of data capability and possesses the greatest potential for impact. While data literacy provides foundational understanding and data proficiency enables application, data fluency allows for creation, innovation, and mastery within the data domain itself. Data fluency involves the agency of technical experts and advanced teams working with top-level leadership on issues of strategic importance for the institution.
Data fluency involves the agency of technical experts and advanced teams working with top-level leadership on issues of strategic importance for the institution.
The three tier model of data capabilities has significant implications for how organizations develop their data culture. And the value that data fluency holds for education and training programs has shown the most impact within the context of creating collaborative, strategic plans for driving future success.
Grush: Why is data fluency a must for an institution taking a holistic view of its own data capabilities spectrum?
Wagner: Thinking about data fluency as a strategic imperative is going to provide institutions with a way to cross the "learning efficacy divide". It will make information and data resources more accountable and actionable.
Let's face it, it is going to take more than technological excellence, commitment to innovation, and a vision for the future for any institution to be successful in the coming years. To create a data fluent culture, stakeholders will need to embrace fully the insights that data can spark and the possibilities that data informed decision making enables. This will depend on helping stakeholders imagine what data fluency means for their unique enterprise. It will mean encouraging stakeholders to contribute actively to an institutional culture predicated on actionability. It will mean all of us engaging with data resources and tools and methods to experience, personally, the benefits that data fluency catalyzes.
To create a data fluent culture, stakeholders will need to embrace fully the insights that data can spark and the possibilities that data informed decision making enables… It will mean all of us engaging with data resources and tools and methods to experience, personally, the benefits that data fluency catalyzes.
I think what we're all realizing is that we have reached an inflection point both around our own access to data and technology and the power that we now have to do things more thoroughly and purposefully than we had ever imagined — now, at a time when we're going to have to get smarter about how we step up to new opportunities.
[Editor's note: Image by AI - ChatGPT. For more on data fluency, see also "Navigating the Data Capabilities Spectrum: From Literacy to Proficiency to Fluency", a briefing paper prepared by Ellen Wagner for the 17th annual eLearning Africa Ministerial Roundtable (7 May, 2025). The paper is available on her blog, New Learning Frontier.]