Setting
The study was conducted in 32 elementary, intermediate, and charter schools in Delaware, New Jersey, and Pennsylvania in the 2007–08 school year. To be eligible for the study, schools had to have at least two fourth-grade classrooms, not track students based on academic performance, use the same math curricula within fourth-grade classrooms, have more than one fourth-grade teacher, have adequate technology to implement the intervention, and have no evidence of use of Odyssey Math within the past 2 years.
Study sample
The analytic sample included 32 schools. All fourth-grade teachers within each school participated in the study; the analytic sample comprised 60 teachers in the intervention group and 62 teachers in the comparison group. Data from 1,223 intervention students and 1,233 comparison students contributed to the classroom-level means used in the analytic sample. The baseline student sample included 1,403 intervention students and 1,451 comparison students. Approximately 7% of the baseline student sample were English learners, about 25% belonged to a racial or ethnic minority group, and about 18% of the students were eligible for free or reduced-price lunch.
Intervention Group
Students in the intervention group used Odyssey Math for 60 minutes per week between October 2007 and April 2008 as a partial replacement of their schools’ core math curriculum. Teachers used Odyssey Math’s online assessments and tools to improve students’ math skills.
Comparison Group
Students in the comparison group continued to use their school’s core math curriculum. The core math curricula used by these schools included Everyday Math, Scott Foresman, Harcourt Brace, and Saxon Math. Participating teachers were encouraged to spend a similar amount of daily and weekly instruction time on mathematics. Five students in the comparison group switched groups mid-year and received the Odyssey Math curriculum. These students were analyzed as part of the comparison group, their originally assigned condition, thereby maintaining the integrity of the randomization.
Support for implementation
Two professional development training workshops taught by Compass Learning were offered to intervention teachers and to interested school administrators. The first workshop was offered in four regional locations in August 2007 and provided 5–6 hours of training. In total, 37 of the 60 teachers attended the August workshop, and 23 intervention group teachers attended a makeup day. The second workshop was offered on a single day in January 2008, and all 60 intervention group teachers attended. Both of these workshops provided training on the Odyssey Math software. The workshops were followed by one-on-one coaching sessions with the intervention teachers in their classrooms and involved 1–2 hours of coaching per teacher. These workshops and coaching sessions were offered in addition to the standard professional development opportunities available to teachers.