There's a strong case for making social and emotional learning (SEL) skills and competencies a central feature of elementary school. Children who master SEL skills get along better with others, do better in school, and have more successful careers and better mental and physical health as adults. Evidence from the most rigorous studies of elementary-school SEL programs however is ambiguous. Some studies find few or no effects, while others find important and meaningful effects. Or studies find effects for some groups of students but not for others. What causes such variation isn't clear, making it hard to interpret and act on the evidence. What are the sources of variation in the impacts of SEL programs designed for the elementary years? To find out, Stephanie Jones, Sophie Barnes, Rebecca Bailey, and Emily Doolittle examine how the theories of change behind 11 widely used school-based SEL interventions align with the way those interventions measure outcomes. Their central conclusion is that what appears to be variation in impacts may instead stem from imprecise program targets misaligned with too-general measures of outcomes. That is to say, program evaluations often fail to measure whether students have mastered the precise skills the programs seek to impart. The authors make three recommendations for policy makers, practitioners, and researchers. The first is that we should focus more on outcomes at the teacher and classroom level, because teachers' own social-emotional competency and the quality of the classroom environment can have a huge effect on students' SEL. Second, because the elementary years span a great many developmental and environmental transitions, SEL programs should take care to focus on the skills appropriate to each grade and age, rather than taking a one-size-fits-all approach. Third, they write, measurement of SEL skills among children in this age range should grow narrower in focus but broader in context and depth.