TESTIMONY OF
BERYL LEVINGER
SENIOR DIRECTOR, EDUCATION DEVELOPMENT CENTER
AND
DISTINGUISHED PROFESSOR, MONTEREY INSTITUTE OF INTERNATIONAL STUDIES
BEFORE THE
U.S. SENATE COMMITTEE ON AGRICULTURE, NUTRITION AND FORESTY
JULY 27, 2000
Good morning, esteemed members of the Senate Agriculture Committee. Thank you
for inviting me to share my views with you this morning on the World School
Feeding Initiative that has been proposed by His Excellency George McGovern and
The Honorable Robert Dole. I will focus my comments on how this initiative
might be modified to optimally contribute to the healthy growth and development
of girls and boys living in Africa, Asia, Latin America and the Caribbean.
Before I begin my testimony this morning, permit me to introduce myself. My name is Beryl Levinger, and I am a Senior Director at the Education Development Center, a nonprofit organization headquartered in Newton, Massachusetts. I have worked in the field of international education and poverty-alleviation for more than 30 years. Over the course of my career, I have provided short- or long-term technical assistance to over 70 countries striving to meet the health, nutritional and educational needs of their citizenries.
In the last 15 years, I’ve authored several major studies on health, nutrition and learning including three books that are especially relevant to my testimony today: a comprehensive review of international school feeding programs published by USAID; an examination of the influence that health and nutritional factors exert on schooling outcomes, published by the United Nations Development Program; and, a review of the factors that contribute to the development of human capacity, also published by UNDP.
Allow me to summarize what I have learned in the course of this research and fieldwork on the impact of school meals on attendance, enrollment, and learning in settings of extreme poverty.
Let me begin with the question of learning outcomes. There is a substantial body of research to
support the following assertion: the level of a student's cognitive performance
is, in part, a function of the adequacy of his or her diet. The importance of
this research is that it establishes a theoretical and empirical framework for
a major claim made by advocates of the proposed school feeding initiative,
namely that when such programs provide undernourished participants with an
adequate diet, cognitive development outcomes can be reasonably anticipated. Unfortunately, this assertion is only partly
correct. The following caveat is required: meaningful cognitive improvement will only occur when a facilitative
learning environment is present to complement the food that a child receives.
I’d like to shed some additional light on the critical relationships underlying this caveat. Protein-caloric malnutrition often leads to substantial impairment of physical growth, including altered brain development, particularly if the nutritional deficits are early, severe, and long lasting. It also is the case that children who have experienced protein-caloric malnutrition tend to show reduced levels of intellectual development and school performance.
Research findings suggest that the interactions of malnourished children with their environments make them less likely to seek out, utilize, and respond to available opportunities for learning and social interactions. Although in the late sixties and early seventies it was assumed by many researchers that the brain changes produced by malnutrition led directly to an impairment of learning, which was often irreversible, more recent studies have led most investigators to abandon this position.
Currently, the most widely accepted hypothesis is that
malnutrition exerts its major influence on behavioral competencies through
dysfunctional changes in attention, responsiveness, motivation, and
emotionality, rather than through a more direct impairment of basic ability to
learn. This situation implies hopeful prospects for the reversibility of these
effects when the child's learning
environment is intellectually facilitative. In other words, teachers need to
engage children as active partners in learning. They must provide frequent
feedback and encouragement while engaging children in stimulating learning
tasks. In most developing countries this entails investments in teaching
training, texts, and other learning materials.
The truth today is that most schooling in the developing world is far from this mark. Children sit in severely overcrowded classrooms—or outdoors—with poorly trained teachers and spend countless hours repeating meaningless phrases in a language they often do not understand. They have no books, no blackboard, and frequently no desks or chairs. We are all too familiar with the results of this environment: millions of children never enroll in school throughout Africa and Asia and millions more drop out before completing even the first four grades of primary.
For those children who do attend, little learning takes place. In one recent study in Ghana that used criterion-referenced testing, less than 3 percent of all sixth graders had achieved basic language arts and math skills stipulated by the curriculum. The test used was designed so that the average score should have been 90 percent. Similar results have been noted throughout Africa where countries have measured student mastery of curriculum objectives. In an environment of such extreme educational impoverishment, school feeding may get more children to come to school—although as I’ll show in a moment, this assumption is questionable—but it is doubtful that feeding will get them to learn more. Why? Because the educational environment itself allows very little learning to take place.
Ratoosh, in his analysis of research related to nutrition and psychological development cites abundant empirical evidence to demonstrate that improvement of a child's diet alone can lead only to small changes in cognitive and social development. Meaningful change in this area only occurs when dietary change is accompanied by enrichment of the child's psychological and social environment.
Here’s an example of one study that illustrates the need for changes in the learning environment to accompany food distribution. Researchers at INCAP, a renowned Central American nutrition research institute located in Guatemala conducted a seven-year longitudinal effort concerned with the effects of protein-calorie supplementation on children's physical and mental development. Over 600 children aged five to seven were included in the study.
The investigators found that differences in food intake (as opposed to nutritional status measures) over a two-year period from age five to seven could not be used to predict changes in psychological test performance on any cognitive measure between the ages of five and seven. It did not matter whether differences in food intake were defined in terms of home nutrition, food supplements ingested, membership in experimental feeding groups, or attendance at supplementation centers. In other words, given information about a child's test performance at age five, one could not predict differences in improvement on that test over the next two years on the basis of information about protein-calorie intake over the two-year period. This is probably because the dietary supplementation was not accompanied by a cognitively oriented treatment program.
In contrast, consider the well-known Cali Preschool Study, which examined the effects of a combined program of nutritional supplementation, cognitive stimulation, and health care on the cognitive development of lower class preschool children in Cali, Colombia. The investigation involved 240 3-year-old subjects who were assigned to either a nutrition plus stimulation plus health care condition or to a nutrition plus health care only treatment. Results obtained at the end of the study's second year showed that subjects experiencing two years of the comprehensive intervention improved in verbal reasoning and general knowledge, whereas children in the nutrition plus health care only groups did not show comparable improvements. Furthermore, the performance of the nutrition plus health care only groups on the cognitive measures was not substantially different from that of low socioeconomic status subjects in the control group.
In summary, then, the proposed initiative needs to include provisions for a portion of the commodities to be monetized, preferably over a three-year period. Funds obtained through monetization should be used by Private Voluntary Organizations to engage parents as partners in the educational enterprise; to train teachers in active learning methods; to create motivational textbooks and other learning materials that are cognitively stimulating; to improve sanitation so that schools are not major disease vectors; and to create classroom learning environments that are conducive to learning.
I’d like to now turn to the question of school feeding in relation to attendance and enrollment. Many studies explore this relationship. Interestingly, the most positive relationships generally are found in the least rigorous studies. When control groups have been used and attendance records consulted (in contrast to soliciting teacher impressions), the findings are more ambiguous. I should also note, that US PVOs have generally been the sponsors of some of the most rigorous evaluations of school feeding programs.
In general, where parental perceptions of school quality are very low and poverty is extreme, feeding cannot overcome the factors that lead parents to keep their children—particularly their daughters—at home. However, if families live at the border of the terrain that separates extreme poverty from marginal self-sufficiency, and if the quality of schooling is at least sufficient so as not to dampen demand, then feeding can bring children, especially girls, to school. Once again, though, the quality of the schooling is critical in terms of school feeding impact.
Three types of studies have been conducted to assess the impact of school feeding on attendance and enrollment: retrospective analyses, comparative studies, non-comparative studies, and studies examining the determinants of school attendance and enrollment.
To date, retrospective analyses—before and after studies—have not yielded results in which policy-makers can have confidence. Most fail to use enrollment ratios based on solid demographic data, lack data on contextual variables that might influence school attendance, and do not report longitudinal changes. Because of the inherent weaknesses in this type of study and the inconclusive nature of their findings, they do not lend support for the hypothesized relationships among SFPs, attendance, and enrollment.
I have also reviewed six studies that examine the impact of feeding programs (SFPs) by comparing data on attendance and enrollment between SFP and non SFP schools. Most were inconclusive. The evidence suggests, however, that SFPs may be most effective in meeting their attendance-related objective in settings where attendance is not already high and where children come from rural, relatively low socioeconomic backgrounds.
Several of the studies also point to the need for program regularity (that is, a program where meals are provided to children everyday that school is in session) to achieve an impact on children's school-going habits. In most cases in the developing world where imported commodities were used, program regularity was too low to act as a magnet for attendance and enrollment.
Non-comparative studies are generally very favorable in terms of the impact of school feeding on attendance and enrollment. However, such studies rely on impressionistic data, generally furnished by teachers that cannot be relied upon in matters of policy-formulation.
In summary, the impact of SFPs on attendance and enrollment is intimately linked to the quality of the schooling that is being offered. Once again, there is an important role for PVOs in improving educational quality through partial monetization of commodities.
Permit me to comment briefly on how SFPs influence nutritional status and hunger the third area of expected program benefits. There is little evidence to support nutritional status change as a result of school feeding. There are many reasons for this. Parents often provide one less meal at home so that the food received in school is not additive in terms of a child’s dietary intake. The programs are too irregular to have any meaningful impact on nutritional status. Even when regularity is achieved, there are too few days a year when school is in session to make a significant improvement in the nutritional status of children who must eat well 365 days a year to overcome severe protein-energy malnutrition.
This is not to say that feeding cannot make a contribution to the alleviation of hunger and malnutrition. Breakfasts may be preferable to lunches, because the “substitution phenomenon” (i.e., withholding food at home to compensate for what the child has received in school) is less likely occur since most parents don’t typically offer their offspring a nutritious breakfast that can be replaced by a school meal. Breakfasts can also offset the hunger and related learning difficulties linked to short-term fasting. Such difficulties include short-term attention deficits, irritability and reluctance to engage in cognitively difficult tasks.
Finally, I’d like to note one of the complexities of SFPs—the need to adjust ration size and timing to fit a program’s specific objectives. An SFP designed to serve as an incentive to enroll children in school must be primarily viewed as an income-transfer program that offsets the opportunity costs of school attendance. As such, meal substitution is desirable and take-home rations are an effective vehicle for this. However, take-home rations have a limited impact on alleviating a child’s hunger and malnutrition since the ration is divided among all family members and cannot be targeted to a particular child in the household.
In contrast, SFPs designed to alleviate short-term hunger are best designed around breakfast. However, since few families in developing countries provide their children with breakfast, the income-transfer impact of breakfast is minimal, and, consequently, there may be little effect on attendance or enrollment if opportunity costs of schooling constitute a barrier to enrollment for bypassed children.
Along these same lines, programs designed to improve nutritional status must be targeted to children who are indeed malnourished. This means that the meal must be additive (i.e., the “substitution effect” must be eliminated or significantly reduced). Of course, when the substitution effect is eliminated, the program’s income-transfer effect is also eliminated.
From the foregoing discussion, it should be clear that a single SFP can seldom accomplish all its purported benefits because of contradictory design requirements. It is almost impossible to design a school feeding initiative that can simultaneously bolster attendance, improve learning outcomes, reduce malnutrition, and raise enrollment levels of under-represented populations.
In conclusion, I’d like to offer a few additional observations relevant to the proposed initiative:
· Host governments are expected to significantly contribute to the cost of the program over time. Is there a hidden trade-off among adequately paid teachers, quality textbooks, sufficient classrooms, parental outreach and the costs of a feeding program? I believe there is, and it is not one I would be willing to make. I do not believe that food alone can lead to improvements in learning, attendance and enrollment in countries where poverty is rampant and school is nothing more than meaningless repetition of phrases in chaotic conditions. School lunch programs worked in the US precisely because the quality of education in the schools where lunch was served was already quite high.
· US PVOs must play a major role in implementing the proposed initiative. Such organizations as CRS, CARE and Save the Children already have major education initiatives underway that are designed to introduce the qualitative elements so necessary if parents are to enroll their children in school. Make no mistake about it. In study after study, we see that parental perceptions about school quality are often the key factor in the decision about whether and for how long a child goes to school.
· Monetization, with at least a three-year window for spending monetized funds, is necessary in order to introduce the education quality elements that are required to transform a school feeding program into a potent intervention.
· We must not mistake Food for Education with Food for Learning. Food for Education entails getting children into schools regardless of whether their presence in the schoolhouse truly contributes to overall development goals. Food for Learning must be our vision. To enact it, we must build strong, productive linkages between the consumption of a meal and everything else that occurs during a typical school day.
To
derive the full educational benefit of a School Meal Program, investments must
be made to improve the quality of teaching and learning. Malnourished,
vulnerable children require a school environment that is particularly
facilitative for them in order to overcome their nutrition-induced learning
handicaps. Teacher training, textbooks and instructional materials to promote
active learning and individualized instruction will multiply investments in
school meal programs. Furthermore, if parents perceive that the quality of
education is poor, the school meal will not be a sufficient incentive for them
to enroll their children.
PVOs have an important role to play in the transformation of SFPs into Food for Learning. I hope that the proposed initiative entails specific provision for their participation as well as for their monetization of commodities so that the needed investments in quality can be made. Only then will feeding lead to meaningful societal transformation.
Mr. Chairman, thank you for this opportunity to testify. I will be glad to answer any questions you might have.
NOTE: For those interested in the details of the studies I’ve cited today, kindly refer to the three books I mentioned at the beginning of my testimony. All three are available at my website:
http://www.miis.edu/gsips/faculty/blevinger/beryl.htm
School Feeding Programs in Developing Countries: Myth and Potential http://www.edc.org/INT/CapDev/sfp.txt
http://www.edc.org/INT/NHEA/index.html