A new report released last week by the Bill & Melinda Gates Foundation and the RAND Corporation marks one of the first attempts to define and analyze the impacts of personalized learning. “Early Progress: Interim Research on Personalized Learning” is important for schools, organizations, and companies creating educational strategies that tailor instruction to a students’ learning needs because it offers a cautiously optimistic assessment of personalized learning. Two-thirds of the schools in the study saw significant growth in reading and math.
We welcomed these early promising results, and our team of educators, technologists, and designers took special interest in the report’s focus on the need to help teachers with the mountainous task of personalization. In the teacher survey conducted as part of RAND’s research, 60 percent of teachers said “excessive amounts of time I need to spend developing personalized content” is a substantive roadblock to personalized learning (p. 26).
Creating high quality lessons plans that address each student’s unique learning level in a class of 30 students is time-consuming, painstaking work. New Classrooms’ first personalized instruction model, Teach to One: Math, was designed to address that reality and operationalize personalized learning for every student every day. By leveraging data each day to match each student with the lesson that is most likely to be successful—and by suggesting to teachers where in the classroom and with whom that instruction should take place—personalized learning becomes a lot more doable.
In Teach to One: Math, teachers receive daily schedules that provide them with ready-to-use information about the students they’ll be teaching that day and recommendations on lessons especially curated to meet their student’s needs. Teachers share responsibility for students in an entire grade, enabling them to work with groups of students with common needs. Students can encounter the same skill in teacher-led, collaborative, or multi-day learning experiences we call Tasks. Students who can’t easily cluster with other students on the same skill are scheduled for independent learning activities, such as virtual instruction or independent practice.
We use technology to eliminate the vast amount of time teachers need to figure out how to do the personalization, leaving teachers with more time to focus on crafting good lessons and thinking about their pedagogical approach for the lessons that they will teach. Each day, students take short assessments on the skills they studied that day. Using that learning data, our team uses learning algorithms to create customized and dynamic curriculum for each student and provide teachers with recommended lessons to meet their students where they are at. (Our team has reviewed more than 80,000 mathematics lessons from leading digital and print-based publishers and selected roughly 12,000 high quality lessons to offer in the Teach to One model.) Teachers get to focus on teaching. And by turning teachers into a team, teachers are not alone in their personalization efforts. They can collectively utilize their strengths to best meet students at their level.
The report’s authors caution that it is still early to draw conclusions. Importantly, because researchers used a “Virtual Comparison Group” and not an experimental study design to analyze the impact of personalized learning practices, the findings are limited. As the authors note, more research on personalized learning will benefit schools and organizations working on the development and implementation of strategies to tailor instruction to students’ learning levels.
New Classrooms is seeing similarly positive results in schools we partner with to implement Teach to One: Math. In our first full program year, students made growth gains, on average, above the national average despite starting the year well below national norms. You can review our 2012-13 results here. Our 2013-14 school year results will be released later this year.