AmeriCorps is an organization that trains volunteers to serve the community in various civically-minded ways. A recent evaluation examined the effects on students' math achievement of training AmeriCorps volunteers to teach math strategies to struggling math students in grades 4-8. The volunteers taught skills addressing whole and rational numbers, as recommended by the National Mathematics Advisory Council when working with students struggling in math. Specifically, interventionists used scripted protocols to teach three math strategies to struggling students. Each strategy was studied in prior research and shown to have positive effects on achievement: concrete-representational-abstract, which uses concrete objects to teach concepts; cover-copy-compare, which teaches steps for computation and provides practice; and cognitive-strategy instruction, which teaches students to use procedures and reasoning to solve word problems. For each math skill taught, such as multi-digit multiplication, students received instruction using all three strategies and had to pass an assessment before moving on to the next subskill.
AmeriCorps volunteers had to agree to a year-long, full-time commitment and received a four-day training prior to starting the intervention, with additional training one and two months after. Each school received at least one interventionist from AmeriCorps, who was mentored by one school-staff member who was fully trained in the program. Both were supported by a coach for 1-2 hours a month. Training included classroom instruction, using data to inform decisions, and classroom management skills.
Subjects were 489 students in 150 Minnesota schools who were randomly assigned to either receive the intervention at the start of the school year (n=310), or to a control group who would receive the intervention a few months later (n=179). All students had scored below proficient in the prior year's state math assessment, and had autumn scores falling below STAR Math benchmark standards. During the intervention, student pairs with similar math scores were to receive math support 90 minutes a week for a semester. Post-tests using STAR Math were analyzed two ways: the intent-to-treat analysis included all students who received the intervention, and showed significant positive effects as compared to the control group (ES=+0.17); and the optimal dosage analysis that included students who received the targeted 12 weeks of intervention for at least an hour a week (36% of the subjects). Effect sizes for the experimental group increased to ES=+0.24 when students were given the optimal dosage.