Under the No Child Left Behind act, districts receiving Title I funds are required to offer free tutoring and other supplemental educational services to students from low-income families who attend a Title I school that has not achieved Adequate Yearly Progress for at least three years. District personnel may be asked to help parents select a provider from a state-approved list. They may also be asked what the research says about tutoring as an intervention strategy. Although high-quality research on this topic is limited, available studies provide useful insights and caveats.
Tutoring under NCLB
In the summer of 2007, RAND released the first federally funded evaluation of school choice and supplemental educational services under NCLB. After examining data from nine large urban districts, RAND researchers concluded that tutoring had "a positive influence on reading and math scores in five of the seven districts where there were enough students to examine effects." Although academic gains were small during the first year, they increased after students received a second year of tutoring, indicating a positive cumulative effect. Comparisons of commercial tutoring services versus district-operated tutoring programs fielded mixed results as to which was better at producing academic gains. Meanwhile, students who elected transfer to better performing schools over tutoring showed no significant improvement in test scores, although researchers cautioned that these findings should not be viewed as nationally representative.
Only weeks before RAND released its report, Patricia Burch, a policy analyst at the University of Wisconsin-Madison, described the evidence base for tutoring and other supplemental education service provisions as "quite limited." She cited two important district-level studies of how these NCLB services affected student performance-one in Minneapolis and another in Chicago-but deemed them "methodologically inadequate" for the purpose of making broad policy recommendations in districts.
Two other districts (Los Angeles Unified School District and Pittsburgh Public Schools) as well as three states (Georgia, New Mexico and Tennessee) have also examined the relationship between tutoring and student achievement. The verdict: In general, tutoring doesn't seem to hurt, and sometimes it can help, with some students posting small gains on state reading and math exams.
Inadequate state and district resources for monitoring providers, implementation and results may help explain why the research base on NCLB tutoring remains small (Sunderman, 2007).
Findings from Other Studies
Tutoring can be delivered in a variety of ways (e.g., in small groups or one-on one) and circumstances (at the school or in other settings). Several studies have shown that tutoring can help low-achieving students improve their academic performance (e.g., Elbaum et al., 2000). In 2004, researchers at McREL examined 20 years of research on the effectiveness of out-of-school-time strategies in assisting low-achieving students in reading and mathematics. After limiting their analysis to 56 studies that used comparison/control groups, they performed a metaanalysis to estimate the impact of various strategies on student achievement. The team found that "overall, the largest average positive effect size (.50 representing a gain of 19 percentile points) occurred for the reading strategies that used one-on-one tutoring."
One-on-one tutoring is also integral to Reading Recovery, a research-based program that targets the lowest-achieving readers in the first grade. In 2007, the What Works Clearinghouse gave the program a rare thumbs-up: Based on five studies that met its standards for rigor, the clearinghouse noted positive effects on general reading achievement and alphabetic skills and potentially positive effects on fluency and comprehension.
However, researchers and analysts caution against adopting a one-size-fits-all approach. What works for one student may not work for another. And what's true in one content area may not hold true in another. A case in point is the 2004 McREL analysis, which found that for low achievers in math, tutoring alone did not improve mathematics performance. The results in math were slightly better, however, for interventions that combined social and academic strategies. The study also found that students in early elementary grades are more likely than older students to benefit from tutoring in reading, while older students may benefit more from extra help in math.
Carla Thomas McClure is a staff writer at Edvantia, a nonprofi t education research and development organization (www.edvantia.org). For references used in this article, go to www.districtadministration.com.