Curmudgucation: Can Personalized Learning Deliver
A new report published by the National Education Policy Center looks at the current state of K-12 personalized learning and finds that there are many reasons for school districts to think twice about embracing this hot new trend. “Personalized Learning and the Digital Privatization of Curriculum and Teaching” was written by Faith Boninger, Alex Molnar and Christopher M. Saldana of the University of Colorado Boulder, and it lays out several areas of concern. As the model spreads—and concerns spread with it—this report provides a clear view of the objections to modern personalized learning.
A History Lesson
Imagine a technology “which gives tests and scores—and teaches.” Or a call for a revolution in which science and technology would “combine to modernize the grossly inefficient and clumsy procedures of conventional education” as well as saving teachers time by freeing them from administering and scoring tests. All of that comes from Sidney L. Pressey, the inventor of the first real teaching machine, patented in 1928. Neither personalized learning nor the problems that come with it are new. In fact, the idea of personalizing learning through some sort of mass customization is almost 100 years old; ironically, one of the common pitches for current techno-privatized education is as an antidote to classrooms that supposedly have not changed in 100 years.
B. F. Skinner emerged in the 1950s with an alternative approach to Pressey’s, but as the authors note, the two both claimed that their approach would provide students immediate feedback, allow them to work at their own pace, and provide them more personal attention from teachers. “Both Pressey and Skinner also assumed that a student’s ability to provide the required response to a question demonstrated competency/mastery—and therefore ‘learning.’”
Modern personalized learning has not left any of this behind.
The Modern Version
You would assume that personalized learning meant something like “a humane school and classroom environment and open, flexible teaching strategies,” or “increasing students’ agency over their own learning” or “addressing needs of the whole child,” but a different sort of model has been spreading.
Much of the push for this wave, as with several previous education waves, comes from the Bill and Melinda Gates Foundation, which in 2014 funded a group of organizations to develop a “working definition” of personalized learning. The hard and fast definition still eludes experts and marketers, but the Gates definition (which arguably also informed the definition of Chan-Zuckerberg, now emerging as major players in the field) “offers a tech-friendly vision of an individualized, data-heavy, mastery-based education system.” It is based not on research about teaching and learning, but presents itself as “common sense,’ which, the authors points out, obscures several problematic assumptions.
It assumes that children are, and should be, focused primarily on their own goals, their own objectives—themselves--and not on community connections or goals. It’s about a series of tasks, and not relationships. It assumes that learning involves moving along a linear track ( first learn A, then B, etc) and that complex learning can be broken down and measured as small bits and pieces. First pass a punctuation unit, then a sentence completion unit, then a simple reading unit, and be declared able to write an explicative paper for a work of literature., because computer software can assess the first three, but not the last.
The modern Gatesian model implies constant assessment, feedback and record-keeping. But that means the “learning” must be doled out in small bites and bits that lend themselves to the kind of assessment and record-keeping that a software can handle.
The authors argue that personalized learning has been essentially taken over by a privatized corporate approach, because personalized learning smells like money. Lots of money. But because these programs value data most of all, “they reflect a restricted, hyper-rational approach to curriculum and pedagogy that limits students’ agency, narrows what they can learn in school, and limits schools’ ability to respond effectively to a diverse student body.”
While personalized learning talks a big game about student agency, in fact most models are top-down instruction; the student may choose a speed or even a topic for an exercise, but it’s the software writers who set the major goals, determine the sequence of units, and decides what will prove mastery. Needs and gaps are determined by someone other than the student, who becomes an object to be acted upon by software that is trying to elicit the desired response and behavior from her. Students may be able to move through their list of modules faster, but there’s no support for the notion that learning has anything to do with learning better.
In fact, personalized learning has a limited idea of what an education actually is. Modern personalized learning envisions a series of discrete skills and scraps of knowledge, acquired in a particular sequence. This ignores everything we know about integrating learning into prior knowledge and the world at large. It stifles creativity and critical thinking; rather than forge paths and develop a personal relationship with a body of knowledge, personalized learning calls on students to just move down a path that has already been laid out with pavement, guardrails, and penalties for daring to wander.
Nor is that concrete path supported by research. The research base for modern personalized learning is weak, with little clear support for the idea that this approach can work any better than traditional methods. Too often the pitch is simply magical thinking tied to computers (It’s on a computer, so it will be awesome).
Computer technology often comes with an presumption of unbiased objectivity. But software is written by humans, and it reflects their biases and assumptions, the culture that they breathe, and the culture of the tech world is overwhelmingly white and male. Nor do the venture capitalists who are doing much of the funding of these programs free from cultural biases of their own.
The authors note one of the central ironies of modern personalized learning; though it claims to be about tailoring instruction for the individual student, it actually requires all students to get their education from a single one-size-fits-all delivery system. That system is not centered on a human teacher; they are reduced to the role of “coach” or “mentor,” with little control over the education process. Where more control is given, the teacher spends more time on the computer, modifying, writing, adding, and otherwise maintaining the program.
The digitized approach to personalized learning involves collecting vast amounts of data. Even if the company honestly has no intention of ever putting that data to other uses, the fact that such a data bank exists means that it can be stolen. And since most of these programs come from businesses with investors and owners to keep happy, the pressure for monetizing must be huge. One need only look at Summit Learning, one of the most prominent personalized learning platforms: its software was developed with assistance from Facebook engineers, and the Summit Learning Program has been split off into a separate company with a four-person board of directors that includes Priscilla Chan and the CFO of the Chan-Zuckerberg Initiative. Summit is free to schools that want to use it, but Mark Zuckerberg, the head of Facebook, knows a thing or two about how to get money out of a digital platform people use for free. He is not, however, known for his careful handling of user privacy. If data is the new oil, then digitized learning would bring in a data gusher of epic proportions. Summit wants us to know that it has privacy policies in place, has signed the Student Privacy Pledge, and maintains a Privacy Center on their website that details all the ways they are protecting student privacy.
Personalized learning as currently pitched is not really new, and there are reasons to believe that it cannot deliver on most of its promises. This report lays out in considerable detail why school district leaders should think long and hard before making their students part of this new digitized version of an old revolution that has, for 100 years, failed to launch.
This blog post has been shared by permission from the author.
Readers wishing to comment on the content are encouraged to do so via the link to the original post.
Find the original post here:
The views expressed by the blogger are not necessarily those of NEPC.