The Latest Research

Closing schools in a shrinking district: Do student outcomes depend on which schools are closed? Report from the RAND Corporation (March 2012)

What? This study examines the impact of school closures on student test scores and attendance rates. The study’s sample was a mid-sized urban district that, faced with declining enrollment, chose to make student achievement a major criterion in determining which schools would be closed. Specifically, the district sought to close its low-performing schools and move displaced students to higher-performing schools. The district closed 22 schools at the end of the 2005–2006 school year.

Researchers evaluated the impact of the school closures by comparing student achievement growth before and after the closures. They used student-level data on enrollment, demographics, residential location, and state achievement test performance that had been collected by the district since the late 1990s.

Results of the study suggest that students displaced by school closures can experience adverse effects on test scores and attendance; however, these effects can be minimized when students move to substantially higher-performing schools. As for the schools that receive the transferring students, the study uncovered no adverse effects on students in existing schools due to the influx of students from the closed schools.

The authors note that their analysis does not necessarily support school closures as a means for improving student achievement; rather, the results suggest that students transferring from closed schools will experience improved student achievement only if the students are moved to schools that are dramatically higher-performing than the ones they left.

Where? The report can be found at

Evaluation of the Effectiveness of the Alabama Math, Science, and Technology Initiative (AMSTI). Report from the National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education (February 2012)

What? The purpose of this study was to evaluate the effectiveness of the Alabama Math, Science, and Technology Initiative (AMSTI). First introduced to Alabama schools in 2002, AMSTI is a two-year intervention intended to better align classroom practices with national and state-wide teaching standards, with the ultimate goal of improving student achievement. The program includes professional development, access to the materials and technology necessary to deliver hands-on inquiry-based instruction, and in-school support for teachers.

Schools participating in the study were randomly assigned to either a treatment group or control group (schools in the treatment group participated in the AMSTI intervention, while schools in the control group used their existing programs). A total of 82 schools, with about 780 teachers and 30,000 students in grades 4–8, took part in the evaluation. Data sources included professional development training logs and observations, professional development teacher surveys, and interviews with teachers and principals. Student achievement was measured using year-end scores on the Stanford Achievement Test Tenth Edition (SAT 10).

After one year of AMSTI implementation, students in the treatment schools scored, on average, two percentile points higher on the SAT 10 mathematics problem solving assessment than their control group counterparts, and the difference was statistically significant. Similar results were shown for the SAT 10 reading assessment; however, the effect of AMSTI on student achievement in science, as measured by the SAT 10 science assessment, was not found to be statistically significant after one year.

For year two effects, the researchers emphasize that their analysis should be considered exploratory because the control group started the AMSTI program after year one, so was no longer a “pure” control group in year two. The results of their exploratory analysis suggest that, after two years of AMSTI instruction, students will achieve an average gain of four percentile points in math and five percentile points in science when compared with students who do not receive AMSTI instruction.

Where: The report can be found at

An Evaluation of Number Rockets: A Tier-2 Intervention for Grade 1 Students at Risk for Difficulties in Mathematics. Report from the National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education (February 2012)

What? Number Rockets is a small-group tutoring program for first graders who have been identified as at risk of academic failure. The program has 63 lessons covering 17 topics (e.g., sequencing numbers, skip counting, place value) and includes a highly specified teacher’s manual. The tutoring sessions are 40 minutes each, with a recommended three sessions per week for 16 weeks.

To evaluate the effectiveness of Number Rockets, researchers conducted a randomized study of the program. A total of 76 schools in four urban districts across four states participated. Half of these schools were assigned to implement Number Rockets as a supplement to their regular core mathematics instruction (the treatment group) and half were assigned to conduct “business as usual” (the control group). Nearly 1,000 students were included.

The primary outcome measure for the study was the Test of Early Mathematics Ability–Third Edition (TEMA–3), a broad measure of student proficiency in mathematics. After participation in Number Rockets, findings of the study showed that the treatment group scored significantly higher on the TEMA–3 than the control group.

The researchers conclude that grade 1 students at risk for difficulties in mathematics benefited from participation in the Number Rockets intervention. However, they do note that since the control condition consisted of regular classroom instruction and no added mathematics instruction for at-risk students, it cannot be stated whether the intervention effect was due to additional mathematics instruction time delivered in any manner, or specifically to the design of Number Rockets.

What Works to Prevent or Reduce Internalizing Problems or Social-Emotional Difficulties in Adolescents: Lessons from Experimental Evaluations of Social Interventions. Report from Child Trends (December 2011)

What? Which types of programs can best support adolescents struggling with social-emotional difficulties? To find out, researchers reviewed evaluations of 37 programs designed to prevent or treat internalizing problems (e.g., depressive or anxious mood, negative self-perceptions, emotional distress). This report presents findings from their research.

Programs were identified for the review by searching LINKS (Lifecourse Interventions to Nurture Kids Successfully), Child Trends’ online database of rigorously evaluated social interventions for children and youth. All of the program evaluations used a random assignment design.

Based on the evaluations’ findings, the researchers determined whether the programs worked, had mixed impacts, or did not work at preventing or reducing internalizing problems. Categories were defined as follows: programs in the “found to work” category had positive and statistically significant impacts on a particular child or youth outcome; programs in the “mixed findings” category had varied impacts either on particular outcomes, at different times, or for varied subgroups; and programs in the “not proven to work” category had non-significant or marginally significant impacts on a particular child or youth outcome. Outcome areas included depression/ depressive symptoms, suicidal thoughts or behaviors, and anxiety/anxious symptoms.

Of the 37 programs reviewed, the researchers found that 24 had positive impacts, three had varied impacts, and 10 were not proven to work for preventing or reducing internalizing problems. The approaches that were most effective were therapeutic approaches (e.g., family therapy, counseling) and skills-training approaches that promoted healthy relationships with others and an ability to cope with negative thoughts and feelings. Mixed findings emerged for prevention-focused approaches implemented with a universal population and approaches explicitly designed to increase self-esteem. The approaches that were not proven to work for internalizing problems were those targeted only to females and those that utilized mentoring to improve self-esteem. However, it should be noted that the programs that incorporated mentoring were primarily designed to produce positive impacts on other outcomes outside of preventing or reducing internalizing problems

Where? The report can be found at

This article is available to subscribers only. If you are an existing subscriber, please login. New subscribers may register below.

Existing Users Log In


April 2013