DC is in the forefront of a movement to make education a more personalized experience, relying in part on technology to tailor learning to each student’s needs and interests. The approach promises to ensure that advanced students are challenged and struggling ones engaged, even if they share the same classroom.
In any given classroom, some kids grasp the material easily while others struggle. Under the prevailing model, teachers have generally taught to the middle, with the inevitable result that some kids are bored and others are lost. The personalized learning movement aims to engage and challenge all students, wherever they may be.
Personalized learning models vary, but most are premised on the idea that students should progress at their own pace, moving on only when they’ve demonstrated mastery—sometimes of content they’ve chosen for themselves. Most personalized learning models are also examples of “blended learning,” so called because they rely on a blend of traditional teacher-led and technology-based instruction.
Some states have encouraged this mastery-based approach by changing the criteria for high school graduation. Instead of only counting credit hours, they allow students to demonstrate competency in a variety of ways. The DC State Board of Education is currently considering whether to allow that personalized alternative here.
But DC is already a hotbed of personalized learning, thanks to recent initiatives in both DC Public Schools and the philanthropic sector. DCPS has introduced school-wide blended learning programs in 17 schools, and many other schools in the system are using the approach in at least some of their classrooms.
And DC’s CityBridge Foundation has launched two initiatives designed to promote personalized learning in both DCPS and charter schools. One is a yearlong fellowship that has so far introduced over 50 local teachers to methods of using technology to personalize learning.
The other initiative, called Breakthrough Schools: DC, is a competition for new or existing schools that are devising new personalized learning models. Now in its second year, the program has provided funding and technical support to 13 schools, some of which aren’t yet operating. Two of the schools are part of the DCPS system and the rest are charters.
Ultimately, Breakthrough Schools: DC–modeled on a national competition funded by Next Generation Learning Challenges—could reach as many as 20 DC schools. Each school receives as much as $500,000 over the course of several years to plan and launch its model. Funding for the $6 million program comes in part from CityBridge and in part from other local and national philanthropists, including the Broad and Gates Foundations.
Personalization targets an old problem with new methods
Personalized learning is the latest solution to an old problem. Many schools used to sort students of different ability levels by putting them into different tracks. But that approach has largely fallen out of favor, partly because it tends to reinforce racial and socioeconomic segregation. Nowadays, a single class may include a student who is far above grade level, a student with learning disabilities, and another who is just learning English.
Teachers are told they need to differentiate their instruction to reach each of these students, but that takes a lot of training and talent. Some say it’s simply impossible. Technology can make differentiation easier, both by supplying data on student performance and by providing computer software that meets students at their own levels.
Proponents say personalized learning teaches kids to be self-directed, preparing them for what they’ll encounter in college—and in life. They also argue that students are more likely to be engaged in learning when they’re able to set their own pace and get immediate feedback, and to be less inhibited about making mistakes when the only witness is a computer.
There hasn’t been much evidence that personalized learning actually produces better results. Last year, however, two studies—one by a Stanford University research group and the other commissioned by the Gates Foundation—found that some low-income schools using personalized models had positive outcomes on test scores, college prep course completion, and other measures.
There are reasons to be cautious about personalized and blended learning
Personalized learning embraces so many different kinds of models that it’s hard to evaluate its success overall. And while DCPS has recorded some encouraging results with blended learning, research on its efficacy in general has been inconclusive.
Despite the lack of evidence, schools in DC and elsewhere are rapidly embracing these innovations. But there are reasons to proceed with caution.
For one thing, if kids are allowed to progress at their own pace, many may opt not to challenge themselves. After all, learning can and sometimes should be difficult, and people—especially children—often need to be pushed to work hard.
And some personalized learning models have students choosing not only the pace of learning but also the content. That could be problematic. Students generally don’t have the expertise to decide what they should be learning. That’s why we have teachers.
Beyond that, if the students in a class are all studying different things, their learning can become fragmented. They may lose the opportunity to exchange ideas through group discussion, which can be enriched when students of varying backgrounds have the chance to participate.
The blended aspect of personalized learning also has potential pitfalls. At the elementary level, DCPS and many other schools have groups of children rotating between stations in a single classroom, with each group spending a third of their time working at computers loaded with software geared to their needs.
Rigid adherence to that model may not allow teachers enough time to explain concepts to the class as a whole. Students can also lose valuable instructional time in making the transitions or because of technological glitches. And in the many classrooms that have only one teacher, the students working at computers are largely unsupervised while the teacher works with another group.
From what I’ve observed, those students don’t always stay on task. Even if they do, much of the software currently available has no connection to what students are learning from their teachers. Students may spend hours every week practicing reading comprehension skills rather than acquiring knowledge, an approach that is particularly harmful for low-income students.
Older methods of personalization are worth trying too
Given those possible flaws, we shouldn’t lose sight of old-fashioned, low-tech ways of personalizing learning. One would be having students write about what they’re studying, something schools don’t often do these days. Some students can write a sentence, others a paragraph, and others an entire essay. DCPS has been piloting a writing program that has had encouraging results with students of varying needs and abilities.
Writing assignments are a low-tech way to provide teachers with individualized data about what students are understanding. They also develop students’ analytical skills—and, if teachers give the right kind of feedback, their writing skills, which are often far from adequate.
And then there’s the time-honored version of personalization employed by the wealthy: tutoring. To some extent, computers replicate what tutors do, having students practice skills and monitoring their progress. But, at least in their present configuration, computers can’t provide the emotional connection that research has shown is important in stimulating learning.
The obstacle to bringing tutoring up to scale, especially in high-poverty schools, has been that it’s too expensive. But at least one school has figured out a low-cost model that has boosted achievement dramatically: Hire bright college grads for a year or two, as Teach for America does, and have them provide “high-dosage” tutoring throughout the school day in exchange for housing and a stipend. Other schools are now trying to replicate that model. (Of course, as with computer-based instruction, to be truly effective tutoring needs to be connected to the content students are learning rather than just focused on free-floating comprehension skills.)
Of course, the high-tech and low-tech approaches don’t have to be mutually exclusive. Used thoughtfully, computers can free up teachers’ time to work with students one-on-one or in small groups, building relationships and doing other things only humans can do.
And personalization, if balanced by whole-group activities that create dialogue and a sense of community, is a more realistic approach than assuming that all students are proceeding in lockstep just because they happen to be the same age.
So by all means, let’s experiment, judiciously, with these new approaches to an old problem. But at the same time, let’s try to find ways to use older pathways to personalization that are tried and true.
A condensed version of this post is available at Greater Greater Education.