Performance Improvement 1995. Evaluating Educational Outcomes of School Health Programs



This study addressed the role of school health programs in improving academic performance. It described the available evaluation literature, analyzed the benefits and limitations of alternative research designs, and offered two major approaches to the design of an evaluation. With this guidance, schools can justify not only the health importance but also the academic importance of their health programs.


This study was undertaken to provide guidance to educators in evaluating the effects of health programs on students' school performance. School performance includes grades and other indicators of educational achievement, attendance and dropout rates, and educational expectations. The three specific goals were (1) to review the state-of-the-art of school health program evaluations; (2) to describe the kinds of measures used and their feasibility for assessing educational achievement following participation in health programs; and (3) to provide two research options for evaluating the relationship between school health programs and educational performance.


There are eight general types of school health programs: health education, health services, efforts to promote a healthy school environment, school food service, physical education and fitness, integrated school and community health promotion, school counseling, and health promotion for school faculty and staff. Both the national education agenda, Goals 2000, and the national health agenda, Healthy People 2000, explicitly recognize the relationship between education and health. Although it is generally accepted that good health is necessary for effective learning, a lack of well-designed evaluations has resulted in little empirical evidence on the link between school health interventions and educational outcomes. The absence of evidence has led to retrenchment and the shifting of resources from health programs to academic programs. To maintain present school health programs and to encourage new programs, strong empirical evidence is needed on how these programs affect students' academic performance. This study offers educators the tools to collect this empirical evidence.


This study, conducted by Mathematica Policy Research, Inc., provides a framework for assessing educational outcomes of school health interventions through the following methods: (1) a literature review of both the current knowledge about the effectiveness of school health interventions and the important design issues for evaluating health and educational outcomes of school health programs; and (2) consultation with an advisory panel of researchers and policymakers on the types of health interventions likely to affect school performance, the relevant educational outcomes, data sources for evaluation, and the types of evaluation designs that are feasible in school settings. "Intervention" was defined as a specific plan, program, program component, or strategy aimed at changing students' health attitudes, behavior, or health status and, subsequently, their school performance.


In review of the evaluations of eight types of school health programs, programs varied in effectiveness for student health outcomes. Gains were moderate to large in students' health knowledge and smaller for changes in attitudes, behaviors, and health status. In addition, programs generally required ongoing intervention to sustain positive behavioral effects. The health interventions that have been most thoroughly evaluated are nutrition programs, physical activity programs, and drug abuse prevention programs.

Fewer studies examined the educational outcomes of health programs. Of these studies, a handful demonstrate improvements in basic academic skills, reductions in tardiness and absenteeism, and reductions in use of drugs.

The study provided a list of the data requirements for measuring student short-term educational performance and descriptive information about students, schools, communities, and school health interventions. Issues and problems about data collection, availability of data from student surveys, and obtaining and using these data are also thoroughly reviewed. Some problems related to data collection are (1) obtaining the release of sensitive data on student behaviors; (2) gaining cooperation of school officials whose schools are not receiving health intervention funds; (3) obtaining informed consent from students and parents; and (4) designing a format to gather data. Gaps in school records regarding health interventions, student demographics, school characteristics, and teachers' evaluation of student behaviors make assessments more difficult.

The study proposes two major alternative designs for evaluating school health interventions: (1) analysis of existing data collected from previous school health program evaluations, including nationally representative samples of students, and (2) large-scale multischool demonstrations using experimental and comparison-group designs and random assignment. The study describes the benefits and limitations of each approach and the most appropriate type of statistical analysis.

Use of Results

This report is expected to provide educators and school health researchers with a blueprint for a well-designed evaluation of health programs and their impact on academic performance. Evaluations of this kind can provide justification for continuing or enhancing school health programs for academic and health reasons.


In press under the title Evaluating Educational Outcomes of School Health Programs.

Agency sponsor:

U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion

Federal contact:

James Harrell
Office of Disease Prevention and Health Promotion
Office of the Assistant Secretary for Health
Switzer Building, Room 2132
330 C Street SW
Washington, DC 20201
(202) 401-6295 Fax: (202) 690-7054

Principal investigator:

Barbara Devaney
Mathematica Policy Research, Inc., Princeton, NJ

View full report


"PerformanceImprovement1995.pdf" (pdf, 941.92Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®