The hype and reality of personalized learning

The hype and reality of personalized learning: a photograph of children sitting at desks in a classroom. Image attribution: edusurge

K-12 education is spending billions of dollars on personalized learning. 97% of US school districts are investing in this hot educational trend. But what is “personalized learning”? Is this effort worthwhile?

A recent Education Week article, The Case(s) Against Personalized Learning” by Benjamin Herold, includes interesting research findings slotted into a confusing narrative.

What is personalized learning?

Herold begins by describing personalized learning as a high-tech, well-funded push for schools to install computer-based lessons. This software supposedly monitors each student’s responses and “personalizes” subsequent lesson plans. He then goes on to outline arguments that:

  • The hype for this agenda “outweighs the research”;
  • This kind of personalized learning is “bad for teachers and students”; and
  • “Critics are worried that ‘personalized learning’ is cover for an aggressive push by the tech industry to turn K-12 education into a giant data-mining enterprise.”

Given Herold’s initial framing, it’s unsurprising that the article is full of competing perspectives. Juicy journalism perhaps, but the result is that the various authorities quoted talk past each other. That’s because they have different preconceptions of what personalized learning is.

Two extremes

Much of this confusion can eliminated if we think of personalized learning as a spectrum, with two extremes defined by Dan Buckley:

…”the T-route, in which the educational route the learner takes is controlled, decided and evaluated ultimately by the Teacher, and P-route in which the route that the learner takes is controlled, decided and evaluated by Peers (or Pupils if you prefer).”
—The PbyP Approach, Dan Buckley

For example from the article, the author of Schooling Beyond Measure, Alfie Kohn’s view is a dismissal of the T-route:

“…much of what’s marketed as ‘personalized learning’ amounts to little more than breaking knowledge and ideas down into itty-bitty parts, then using extrinsic rewards to march kids through a series of decontextualized skills they had no meaningful role in choosing.”

While Diane Tavenner, the CEO of California’s Summit Public Schools charter network, says:

“…the strongest personalized-learning models offer the best of what both conservatives and progressives want: high-quality standards and content for students, with opportunities to apply that knowledge via self-directed projects, all supplemented by human mentors and technology tools that help students keep track of their own learning.”

is claiming that Summit’s technology platform and instructional model, developed with support from Facebook and the Chan Zuckerberg Initiative, represents a successful mixture of the T- and P- routes.

Adult learning

This brings us to my interest in this topic as a student and facilitator of adult learning. (Which the article doesn’t mention. To be fair, Education Week confines itself to K-12 education — though you wouldn’t guess that from its name.)

The energetic disagreements documented by Herold are fueled by society’s three mutually incompatible ideas about children’s education (i.e., making good citizens, mastering certain bodies of knowledge, and fulfilling each student’s unique potential), as Kieran Egan lays out in his thought-provoking book The Educated Mind. These societal goals, however imperfectly attained, imply the need for a mixture of T- and P-route learning strategies. This is because teachers will always be needed to facilitate the understanding of ideas that have taken centuries for the human race to discover and which society insists are important for young minds to grasp.

For adults, on the other hand, I’d argue that the P-route is by far the dominant successful learning paradigm. We live in a world where the job you’ll have ten years from now probably doesn’t exist yet. (That’s the story of my life since I was in my 20’s.) Self-directed, active, peer-supported, just-in-time learning is now the default mode for most professionals. Every adult learner and meeting attendee needs to create their own environment and structure for life-long personalized learning if they are to be optimally effective in the world of work.

Arguments about how we should educate children will likely (and should) never end. But the case for P-route personalized learning in the adult world of work has never been stronger. And until the normal meeting process reflects this reality, our meetings will be pale shadows of what they could be.

Image attribution: edusurge

Why measurable outcomes aren’t always a good thing

On measurable outcomes: A self-referential comic entitled "Self-Description" with three panels. By XKCD, Comic #688. The contents of any one panel are dependent on the contents of every panel including itself. In the first panel's pie chart, "this image" refers to the entire comic image, the one that can be downloaded from xkcd (and the entire comic as displayed here above). In the second panel the amount of black used in each panel is displayed in a bar chart. This actually makes this panel the one that uses most black. The third panel features a scatter plot labeled "Location of black ink in this image." It is the first quadrant of a cartesian plane with the zeroes marked. The graph is the whole comic scaled proportionally to fit the axes, so the last panel also has to contain an image of itself having an image of itself ad infinitum.

What could be wrong with requiring measurable outcomes?

“Enough of this feel-good stuff! How do we know whether people have learned anything unless we measure it?”
—A little voice, heard once in a while in learning designers’ heads

Ah, the lure of measurement! Yes, it’s important. From a scientific perspective, a better understanding of the world we live in requires doing experiments that involve quantifying properties in a statistically meaningful and repeatable way. Science has no opinion about ghosts, life after death, and astrology, for example, because we can’t reliably measure associated attributes.

The power of scientific thinking became widely evident at the start of the twentieth century. It was probably inevitable that it would be applied to management. The result was the concept of scientific management, developed by Frederick Winslow Taylor. Even though Taylorism is no longer a dominant management paradigm, its Victorian influence on how we view working with others still persists to this day.

But we can’t measure some important things

I’m a proponent of the scientific method, but it has limitations because we can’t measure much of what’s important to us. (Actually, it’s worse than that—often we aren’t even aware of what’s important.) Here’s Peter Block on how preoccupation with measurement prevents meaningful change:

The essence of these classic problem-solving steps is the belief that the way to make a difference in the world is to define problems and needs and then recommend actions to solve those needs. We are all problem solvers, action oriented and results minded. It is illegal in this culture to leave a meeting without a to-do list. We want measurable outcomes and we want them now…

…In fact it is this very mindset, one based on clear definition, prediction, and measurement which prevents anything fundamental from changing.
—Peter Block, Community: The Structure of Belonging

One of my important learning experiences occurred unexpectedly in a workshop. A participant in a small group I was leading got furious after something I had said. He stood up and stepped towards me, shouting and balling his fists. At that moment, to my surprise, I knew that his intense anger was all about him and not about me. Instead of my habitual response—taking anger personally—I was able to effectively help him look at why he had become so enraged.

There was nothing measurable about this interchange, yet it was an amazing learning and empowering moment for me.

The danger of focussing on what can be measured

So, one of the dangers of requiring measurable outcomes is that it restricts us to concentrating on what can be measured, not what’s important. Educator Alfie Kohn supplies this example:

…it is much easier to quantify the number of times a semicolon has been used correctly in an essay than it is to quantify how well the student has explored ideas in that essay.
—Alfie Kohn, Beware of the Standards, Not Just the Tests

Another reason why we fixate on assigning a number to a “measured” outcome is that doing so can make people feel they can show they’ve accomplished something, masking the common painful reality that they have no idea how to honestly measure their effectiveness.

Measured learning outcomes can be relevant if we have a clear, performance-based, target. For example, we can test whether someone has learned and can apply cardiopulmonary resuscitation (CPR) by testing them in a realistic environment. (Even then, less than half of course participants can pass a skills test one year after training.)

This leads to my final danger of requiring measurable outcomes. It turns out that measurements of learning outcomes aren’t reliable anyway!

For nearly 50 years measurement scholars have warned against pursuing the blind alley of value added assessment.  Our research has demonstrated yet again that the reliability of gain scores and residual scores…is negligible.
—Professor Trudy W. Banta, A Warning on Measuring Learning Outcomes, Inside Higher Ed

Given that requiring measurable outcomes often inhibits fundamental change and is of dubious reliability, I believe we should be considerably more reluctant to insist on including them in today’s learning and organizational environments.

[This post is part of the occasional series: How do you facilitate change? where we explore various aspects of facilitating individual and group change.]

Image attribution: xkcd