Is “personalized” the next generation of learning?

The long-awaited promise of being able to tailor educational environments to meet the unique learning needs of each and every student is upon us. Or, is it? asks David Andrews

In the United States, federal, state, and local educational systems are joining not-for-profit foundations and for-profit vendors in touting “personalization” as the next generation of learning. The Gates Foundation continues to invest in individual schools and entire districts, through its Next Generation Learning Challenge, with the intent of providing anytime, anywhere technology-enhanced educational opportunities with the goal of enhanced learning outcomes for students. Multiple educational systems, with encouragement from the US Department of Education, have followed suit, with the implementation of initiatives encouraging schools to use coordinated access to technology and advanced data systems to enhance customized pathways to academic success.

What we know
● Some of the most extensively tested comprehensive reform models have taken the basic principles of personalized learning and implemented them systematically with high fidelity.
● Advanced technology and rapid feedback allows the delivery of anytime, anywhere instructional support.
● Making personalized learning the cornerstone of 21st century education requires a clear definition of what it is, and is not.

These substantive investments and broad-reaching initiatives are all based upon the belief that recent technological advances can bring mass customization of learning to scale in our educational systems, and that such customization can, and will, maximize academic performance. The belief is spurred by the notion that each student’s maximal learning pathway can be identified and fully supported with advances in big data analytics and an exploding array of easily accessible digital content.

The challenge to fulfilling the promise lies in two areas. First, while there are some common features to a wide array of approaches that have been labeled “personalization” or “next generation learning,” there is no commonly accepted definition of the approach and its core elements. Second, the evidence of effectiveness of this strategy is piecemeal at best.

From common features to a shared definition

Three common features of most approaches touting themselves as “personalized learning” are student-centered, competency-based, and mixed instructional methods. These three features are not exhaustive, but are represented in some form in most descriptions of personalization.

Student-centered refers to both explicitly identifying individual student goals and the use of the most effective customized techniques (instructional strategies) for achieving these goals. One assumption is that student goals are derived from data indicating the next logical progression in their learning pathway, and that students actively contribute to generating and “owning” the goal.

A second assumption is that the effectiveness of instructional strategies will vary depending upon the student and “what works” to maximize their academic gains. Student goals become relatively unique, and the strategies for achieving these goals are diverse.

Competency-based learning suggests that goals are set to achieve specific competencies and that each student can move toward achieving the competencies in a manner and pace that is individualized. Learning for Mastery is a commonly used variation of competency-based learning, and implies that a fixed amount of competency is necessary in order to be deemed competent. Competency-based learning emphasizes what has been learned over how it has been learned, or how much time it took to try to teach it. Competency is the constant and time becomes a variable. Implementation of competency-based learning requires high levels of confidence in the assessments being used.

Mixed instructional methods are used with the prevalence of any one method determined by its past record in contributing to achieving the students’ goal. Blended Instruction is used as a specific reference to mixed instructional methods that combine technology-enhanced delivery of instruction with a variety of more traditional face-to-face methods.

Despite the prevalence of student-centered, competency-based, and mixed instructional methods in emerging definitions of personalized learning, a more integrated and commonly accepted definition of the approach is necessary if we are to test its efficacy. So many approaches are currently calling themselves personalized that it is impossible to speak to the potential integrated impact of the approach.

Evidence supporting personalized learning

The basic tenets of learning and behaviour change are well founded in scientific evidence, and are consistent with the overall strategies touted in personalized learning: set individual goals, incentivize these goals, regularly monitor progress, provide feedback, and make adjustments to strategies toward improved progress. Diagrams with loops and repeat sequences abound, but the core components remain constant. Clinical psychologists provide ample scientific evidence in the literature from the prevention/intervention sciences, most often referring to the model as cognitive behavior therapy. Educators have evaluated parallel models under different names and organizational schemes, the most currently prominent being response-to-intervention (RTI).

In the U.S., Public Law 94-142 (Education of All Handicapped Children Act), as far back as 1975, put federal policy weight behind the overall scheme in requiring Individual Education Plans (IEPs) in special education. While such individual approaches to goal setting, progress monitoring, and structured feedback have had limited public extension beyond special education for the past four decades, many would argue that the approach has always been the foundation of learning and skill development, whether in mathematics, chess, tennis, or sailing.

Some of the most extensively tested comprehensive reform models have taken these basic principles and implemented them systematically with high fidelity. The organizing frame for data-driven decision-making that underlies Success for All (SFA) is built on this behavioral foundation, as is Positive Behavioral and Instructional Support (PBIS). Goal setting, incentivizing, progress monitoring, feedback, and constant adjustments to learning strategies, are stalwart components of SFA and PBIS, two of education’s most evidence-based strategies.

Two techniques are being used to advance personalization in practice. First, evidence-based programs, like SFA, are integrating more comprehensive data systems into their strategies toward better and quicker profiling of students and matching instructional/intervention strategies with student learning trajectories. Second, emerging programs focusing on creating personalized experiences for students are rapidly committing to evaluate their impact. These programs, however, tend to be in their infancy and have little rigorous empirical data of their contributions to student learning. Nonetheless, the relatively few programs that have implemented a visible personalized approach to learning are receiving substantial national attention and have gained public support.

We now have an opportunity to use advanced technology, not only to organize data toward more precise goal setting, monitoring, and feedback, but also to begin mining this more complex data toward a deeper understanding of students and their learning progressions. As importantly, we can share this understanding, in its entirety, with the educators these students will encounter in the future.

With advancing technology and a system that allows more immediate and comprehensive feedback on individual student growth, as well as the strategies most likely to create and sustain that growth, we are able to reach well beyond the traditional classroom and deliver the anytime, anywhere instructional support that will define 21st century learning.

Achieving the promise of personalization as the cornerstone of next generation learning will require a clear definition of what it is, and what it is not. Furthermore, the components of this common definition must be subjected to rigorous and ongoing evaluation to determine their impact on student learning.

About the author

David Andrews is Dean of Johns Hopkins University School of Education. He has dedicated his career to enhancing opportunities for children and youth by strengthening families, communities, and schools.


December 2014