Buy the textbook. Attend lectures. Take tests. Get graded on a curve. This is the way most people remember their college or university learning experience. Learners, however, are a diverse group, with a wide range of personal histories, different approaches to learning, and varying degrees of technological savvy.
Engaging today’s students with yesterday’s methods creates tension among many learner populations, where many traditional approaches to learning prove unappealing. Combine modern economic and social issues, and traditional learning looks out of sync with reality. Rather than slog through lectures, dry reading assignments and long-form essays that don’t seem relevant to their life experience, many students simply leave school.
The students who don’t complete college, but want to continue to learn, potentially join the ranks of non-traditional students: working adults seeking to learn new skills, part-time students with a desire to complete degrees later in life, many of whom hold down jobs and raise children while taking classes — even those without a high-school diploma.
High-achieving students also have an issue: The pace can be too slow, selections too narrow, and learning experiences too mundane or seemingly irrelevant to what they need to know.
How, then, does education meet these students where they are? Not where they want them to be, or where educational systems is comfortable delivering learning experiences. One answer: personalized learning.
At last week’s Bill & Melinda Gates Foundation U.S. Education Learning Forum, I facilitated the personalized learning breakout, an event to reflect on the status of improving learning during the Gates Foundation’s 15th year. I was joined by Chris Bustamante, President of Rio Salado College; Nivine Megahed, President of National Louis University; and Jean Floten, Chancellor of Western Governors University (WGU), Washington. All the panelists served primarily non-traditional students with mostly non-traditional faculty employment models.
The role of data
As with many topics these days, the conversation about personalized learning quickly turns to technology. With technology, learning can be consumed at the pace of the learner, content can be delivered employing the most appealing media, auxiliary material and refreshers on concepts are just a click away, and assessments integrate with the flow. Sophisticated technology supports competency, not just adequacy—and fellow students, mentors and faculty can all be reached by e-mail at minimum, and often through more sophisticated collaboration technology.
Personalized learning, however, is much more a vision than a reality. Algorithms process student performance data into information, providing learners and educators alike with more visibility and better decision-making tools. However, the result is often procedural, more focused on interventions that keep learners on the path than on making the path more interesting. Much of the design of personalized learning feels industrial, like a factory engineered for efficiency. But education is early in the use of data, and much remains to be discovered about how to use data to co-create learning experiences.
The first section of the panel discussed how data drives personalized learning. Floten shared how WGU creates more personalized experiences not using rules that pace students through coursework, but an innovative delivery model that disaggregates traditional teaching duties among content acquisition experts, faculty who teach and tailor content to meet the competency-based mission of WGU, along with mentors and assessors—an ecosystem of support and access available when the learner needs it. Floten describes their system as “anytime, anyplace and at any pace.” WGU’s competency-based model requires students to pass assessments when they are ready. A pre-assessment shows students what they already know so they may focus on what they need to know, before taking the final assessment that covers all of the dimensions required to demonstrate competency. Tuition is a flat rate that permits students to move through the curriculum at their own pace, often accelerated; capturing any number of qualified credits during a quarter. The approach to learning delivery and business model are synchronized to deliver assistance where needed, and to rapidly recognize success.
But WGU’s data goes much deeper. Each student’s learning profile is captured in the equivalent of a baseball card and dashboard, sharing key statistics and insights with administrators, mentors, educators and the students themselves. “Whatever data we have,” Floten said, “the students have it, too.” Information collected about course content and assessments is used to improve the curriculum and inform data-driven experiences at the learning level, ensuring that course materials are available to meet the various learning styles represented by students in a course. Everyone involved has access to student data showing learning strengths and weaknesses, as well as specific performance indicators that track the investment of time and the quality of the work. The data-driven insights lead to quickly triaging issues before they grow into problems. Unlike many institutions that hand off graduates to an alumni association for fundraising after graduation, WGU keeps track of graduate success, including job retention, salaries, and employer satisfaction, and correlates that information back into their continuous improvement processes.
Bustamante explained that Rio Salado meets personalized experiences at the calendar, allowing students to start many courses anytime, no need to wait for quarter or semester breaks, just for the start of a new week. Their rich data systems allow them to manage a much more flexible schedule than would be possible within information technology.
Megahed, who describes herself as a data geek, has slowly transformed National Louis into a data-driven institution that simultaneously addresses cost, quality and completion. The school has reengineered its general education experiences with an emphasis on first-generation and underserved students. She encourages her team to adopt a data mindset, using the increasing wealth of new information as a key input to decision making. Megahed taught the first course on their new adaptive learning platform herself, convincing colleagues she was a kinesthetic learner. She found using the system an incredible personal learning experience. While teaching she flipped the classroom, putting lectures into technology for anytime, anywhere consumption. Class time concentrated on the application and exploration of ideas learned during asynchronous learning.
The National Louis system monitors student activity throughout the process and provides a framework for learner engagement. Megahed said she would write students on Sunday night and mention that she could see they haven’t started an assignment. On a following Wednesday in class, one student shared that she told her husband, “Hey, I just received a note from my teacher that said I better get to work so, I told him he needed to get to work finishing up the dinner.”
Rapid technological change
The second section of the panel focused rapid pace of technological change, the disruption it causes to plans, and the assumptions it often undermines. I shared that I used to ask how many people in an audience owned a tablet, but now that most do, that question doesn’t drive home the realities of technology disruption. So I brought out a Samsung Gear VR headset and asked how many owned a virtual reality headset. Only two people across the dozen of attendees in two sessions raised their hand. They were owners of Google Cardboard, headset swag from another conference.
Virtual reality wasn’t on any panelist’s agenda just yet, but National Louis will likely look at VR technology like all other technology: adopted as the result of successful experimentation. Her team has created a skunkworks of sorts that evaluates the efficacy of technology, testing it to sees how it might serve learning and operational goals, including reducing costs. They are currently testing, for instance, personalized push notifications that include delivering both kudos and concerns to students. This is intended to see if this type of technology will help keep students motivated. Megahed sees performance and insight data delivered from information systems as central to building scalable models. Coaches, for instance, will be able to handle more students as they better understand how to allocate their energies to those who most need it.
As personalized learning advances, William Gibson’s quote about the uneven distribution of the future comes to mind. While panelists shared their experiences, many institutions represented at the conference don’t yet employ even the most nascent technology-based personalized learning experiences. As the Foundation points out, sessions like this, and the programs they fund, are designed to increase awareness and share lessons learned.
The challenge remains to create real personal learning experiences that leverage technology, positively engage learners and assist students in obtaining the competencies they seek. The data required to drive personalized learning is being defined. How that data helps curate and deliver specifically personal, ad hoc learning experiences from the wealth of content available remains a project for Artificial Intelligence and machine learning researchers.
But perhaps a better place to begin is with definitions. Many views of personalized learning are conceptually vague. Entrepreneurs, perhaps better than educators, segment markets looking for place to deliver solutions. Dror Ben-Naim, CEO of Learning Forum exhibitor Smart Sparrow, sells a rule-based design system for learning experiences, and sees a need for a deeper conversation about what we mean by personalized learning. Ben-Naim sees a hierarchy of personalization that starts at the recruitment and admissions process and ends with adaptive feedback about student misconceptions (see his model at edsurge).
In these discussions, it is important that policy continue to evolve with technology. Future personalized learning systems must avoid narrow, mechanistic approaches that constrain degrees of freedom so that learners can still develop highly sought-after skills, like innovative thinking, that drive so much of the global economy—and that their experiences with learning provide not only feedback to them, but feedback to instructors about the success of their instructional design.
While the panel did not lead to a consensus, the panelists did share eight lessons they apply to keep themselves mentally nimble, to challenge their own assumptions, and to help guide their institutions forward. Personalized learning today requires an open mind perhaps more than an open architecture.
8 Lessons from the Personalized Learning Panel
- Remember learning is very personal! Adaptive learning platforms are still very one-dimensional. The intricacies of learning are very multi-dimensional. Supporting student learning must include non-cognitive, socio-emotional variables to ensure student success.
- Rapid prototyping and iteration must become a way of life – and leaders must keep reminding themselves, their faculty, staff and students that failure is one step closer to success.
- Keep in mind that technology and analytics are just tools. Successful implementation is not the result!
- Pioneer change. If you don’t have the political will, perhaps leading change isn’t your thing. Change is not for the faint of heart or anyone seeking reelection. Leaders must develop the institutional will to make change happen.
- Create and deliver personalized, blended, competency based learning experiences. Remember that what you do is more important than what you call it.
- Set Goals. If you don’t have goals, you can’t measure progress. Don’t be afraid to mix, match, and adapt. Just because you decided you wanted to go in a direction should not mean you keep going that direction after you discover the initial decision was incorrect.
- Engage students. Never stop finding better ways to reach them.
- Understand the systems and the science. Collaborate and learn from the work of others. Take the initiative to co-create the future, even as you navigate the uncertainties that lie ahead.
Although personalized learning, and the evidence for its efficacy, hasn’t developed sufficiently for anyone to claim a good recipe for success, the demand for personalized learning certainly exists. Floten shared that she entered a classroom recently and saw this challenge sprawled on the smartboard: “Give it to me just in time, just enough, and just what is right for me!“ Perhaps researchers and academics should stop refining their definitions and use that simple proclamation as their litmus test.
