Service learning has been a part of America's K-12 education landscape for more than 30 years. Yet today, service learning is found in less than 30% of K-12 schools in the United States, even though it's achieved a substantial footing in American institutions of higher education and the primary, secondary, and higher education systems of many other countries (Spring, Grimm, and Dietz 2008). Skepticism over service learning's educational merit continues despite published research reviews showing a consistent set of positive outcomes for students. Indeed, reviews of K-12 service learning research include close to 70 studies, most of which have found positive impacts on participating students' academic, civic, personal, social, ethical, and vocational development (Conrad and Hedin 1991; Furco 1994; Andersen 1998; Billig 2000; Shumer 2005).
For an innovation to gain traction in today's educational environment, strong and compelling evidence of its effectiveness must be secured. According to the U.S. Department of Education, evidence is secured when the effects of an educational intervention are tested under certain research conditions. Of the 68 studies cited in the K-12 service learning literature, only 25% have been tested under these conditions. If service learning is to be embraced by more educators and schools, then future investigations must incorporate the kinds of research design that can raise the status of service learning as an evidence-based practice. But the K-12 service learning studies that have been able to produce "possible evidence" and "strong evidence," as defined by the U.S. Department of Education, show a consistent set of outcomes for students across all six of the aforementioned educational domains. Four outcome areas within these six domains are especially noteworthy.
IMPROVED ACADEMIC ACHIEVEMENT
Because of the important role that standardized subject-matter exams play in schools, researchers have explored the relationship between service learning and performance on academic-content examinations. For example, Weiler, LaGoy, Crane, and Rovner (1998) assessed differences in reading and language arts performance between primary and secondary school students (n = 775) enrolled in 12 classrooms that offered service learning and students (n = 310) from eight comparable classrooms that did not. To assess the effect of academic service learning, this study concentrated on 15 classrooms in which service learning was well-designed and well-implemented, based on a set of established quality indicators. The researchers identified eight classrooms with characteristics (grade level, nature of student body, etc.) similar to the service learning classrooms to serve as comparison sites that did not do service learning. They collected scores on students' subject-matter achievement tests, student surveys assessing their attitudes toward school and community service, and observations of classroom practice. Findings revealed statistically significant differences between the two groups, with service learning students outperforming the other students in the reading and language arts portions of the California Test of Basic Skills. In addition, the students engaged in service learning reported that they had learned more in their service learning classes than in nonservice learning classes.
Positive but limited effects in subject-matter achievement from participating in service learning were noted in a recent large-scale study using student panel data from 1988-2000 National Educational Longitudinal Study, or NELS (n = 15,340) to assess the relationship between high school students' participation in community service and performance in mathematics, reading, history, and science (Davila and Mora 2007). By analyzing NELS 1992 data on community service work, Davila and Mora concluded that students' engagement in community service was related to positive but small gains in scholastic achievement in mathematics, science, and history. …