From the National Center for Education Research (NCER)

First-Year Evaluation of Social and Character Development Research Program Completed

The first year of this multi-site evaluation—a study of the efficacy of social emotional learning, character education, and violence prevention interventions designed to promote prosocial behavior, academic achievement, and school climate, and reduce problem behavior—has been successfully completed. Seven school-wide programs were randomly assigned to 88 elementary schools in 6 states. More than 4,000 third-grade students were assessed in the fall of 2004 and spring of 2005 using a comprehensive measurement battery. Data from the 2004-05 school year are currently undergoing analysis by Mathematica Policy Research Inc., the contractor for this study, to determine the impact of the seven different programs after 1 year of implementation in schools. Impact findings are scheduled for dissemination in 2006. Third-graders in this evaluation will be followed through fifth grade.

Highlights From Research Funded Under the Cognition and Student Learning Program

purpose of the Cognition and Student Learning research program is to contribute to the improvement of student learning by bringing recent advances in cognitive science to bear on significant problems in education. The long-term goals of this program are to develop approaches to instruction that are based on principles of learning and information processing gained from cognitive science and to provide evidence of their usefulness in education settings. Here are some observations from studies currently under way:

• Optimizing Resistance to Forgetting. It has long been suspected that learning is enhanced when students repeatedly study the same material in two sessions that are more widely separated in time (as compared to spending the same total amount of time, but doing it within a single session or a single day). What has not been known, however, is whether it matters how much time separates two study periods, for example, one day versus one month. Our results suggest that when the goal is for the student to remember the material for a long time (i.e., years), it is important to study on two or more occasions separated by at least one month. Furthermore, our research actually hints that students would probably retain far more of what they learn in school over the long run (for example, 10 years) if schools offered review sessions separated from the original learning by one year or more. —Harold Pashler, University of California, San Diego

In most mathematics textbooks, each set of homework problems comprises primarily problems based on the most recent lesson. In other words, the homework problems corresponding to a given topic are "massed" into one assignment and not "distributed" across different assignments. The results of our experiments, however, suggest that a distributed-practice format produces much better long-term retention than a massed-practice format.—Douglas Rohrer, University of South Florida and Harold Pashler, University of California, San Diego

R&D Centers Funded

IES recently awarded 5-year $10 million grants to the:

• UCLA Center for Research on Evaluation, Standards, and Student Testing (CRESST) to conduct a randomized study in two states to test a system for assessing middle-school algebra and to develop innovative mathematics classroom assessments in grades 6-8.
• University of Houston to test models of curriculum modifications for English language learners and to test the effect of the Sheltered Instruction Observation Protocol (SIOP), a research-based model of sheltered instruction, on student success in vocabulary and in science.