I wrote in my recent post analyzing a study of Teach to One: Math that I welcomed any responses from New Classrooms, the producers of Teach to One: Math. They emailed me with a few thoughts and some additional data, that I’m passing along.
Dear Justin,
We are grateful for your thoughtful analysis of Professor Doug Ready’s report on our first two years implementing Teach to One: Math. You raised some important questions that we are happy to respond to and clarify for others who may also be unclear on the points you raised. Thank you for taking the time to explore our results, and for giving us the opportunity to respond.
Regarding your question about whether the students in year one are included in the year two results, they are. To clarify, all seven schools from year one implemented Teach to One: Math in year two, and are included in Doug Ready’s report. In year two we expanded from serving 3,500 students at seven schools to serving more than 6,000 students at 15 schools.
Between year one and year two, we made numerous improvements to Teach to One, such as:
Refining the algorithms used to create student learning progressions and developing a visualization tool
Creating more than 130 new multi-day learning opportunities called Tasks that help students collaborate, discuss, and debate mathematical problems in real-world contexts
Sourcing and screening high-quality lessons to expand our library to more than 12,000 from 20 different content vendors.
Improved tools for teachers, students, and parents to see student history and maximize out-of-school learning time using the web-based portal
Improving the depth and quality of professional development and instructional support that New Classrooms offers partner schools
To better control for changes in student population, we are providing a side by side comparison of the students’ average growth in schools that we partnered with in both 2012-13 and 2013-14. In five of the seven schools implementing Teach to One in both years, gains increased from year one to year two. In one school gains remained the same, and in one school gains decreased.
This chart includes the year one results (2012-13) as reported by Professor Ready, rounded to the nearest tenth. The year two results are from our own analysis of the data for the same set of schools in the 2013-14 school year, using a ratio of gains to national norms instead of the effect size Professor Ready uses in his report.
We hope that clarifies any confusion about the study. If you or your readers have any other questions, we’d be happy to try to answer them the best that we can. While we are proud of these results, we believe that more research, more development, and more iterative design is needed.
We would also like to clarify for your readers the nature of the Teach to One model. While your term “computer-assisted math instruction” is technically correct because we use computer algorithms to help create unique student schedules every day based on their learning needs, this term suggests students learn solely on a computer, which is not the case. Students learn in a variety of ways, including by teacher-delivered instruction in large groups, teacher-facilitated collaboration with a small group of students, or teacher-supported independent study with computer software or worksheets. Of the eight instructional modalities available to students, just two involve computer-assisted math instruction. Rather than focusing on technology to deliver instruction, Teach to One uses technology to tailor what, when, where, and how students learn each day.
I appreciate New Classrooms sharing this additional data, as well as their efforts to inform their development with research.