Adelman, S., Gilligan, D. and Lehrer, K. (2008) How effective are food for education programs? A critical assessment of the evidence from developing countries (Vol. 9). Intl Food Policy Research Inst.Link to Source
The authors include 16 randomised controlled trials and quasi-experimental studies from eight countries in Latin America, the Caribbean, Africa and Asia. They report the following findings:
- School participation: Of five studies, none provide estimates of the causal impact of the evaluated FFE programmes on school participation for all school-aged children within the school’s service area. Even though there is evidence that FFE programmes have a positive impact on school attendance and enrolment, methodological shortcomings undermine the conclusiveness of these results.
- Age at entry: There are no studies evaluating whether FFE programmes affect the age of entry into primary school. This lack of research is possibly due to the costs and difficulties involved in collecting such data.
- Drop-out rates: Three studies provide inconclusive evidence of the impact of FFE programmes on grade repetition and drop-out rates. Even though these studies indicate that FFE programmes prevent school drop-out, the studies suffer from statistical deficiencies and problems in the approach used to identify causal impacts.
- Learning achievement: Two of four studies demonstrate that school meals are related to significant improvements in achievement-test scores, operating through improvements in school attendance and better learning efficiency. However, there is no research available on the relative contribution of these effects separately. Overall, the study findings indicate that FFE programmes support learning achievements. However, econometric problems undermine the validity of these results.
- Cognitive development: Four studies provide evidence that FFE programmes have a positive impact on cognitive development; but these studies differ greatly in terms of the size and nature of such impacts, the micronutrient content of the food, and the measure of cognitive development used.
- Change in calorie consumption: Three of five studies demonstrate that the increase in children’s daily calorie intake is almost equivalent to the size of the calorie transfer received in school. Overall, there is some limited evidence that parents reduce their children’s at-home calorie consumption when they have been fed at school.
- Anthropometry: Five studies measured anthropometric outcomes. The available evidence suggests that FFE programmes can increase children’s body size and muscle mass, but it remains difficult to establish a direct causal chain between intervention and outcome.
- Micronutrient status: Three studies measured micronutrient status. One study provided evidence of reduced micronutrient deficiencies following a FFE intervention. Children with low initial micronutrient levels or higher micronutrient demands appear to derive particular benefit from FFE programmes.
The authors conclude that relatively few high-quality studies exist. Thus, despite a vast literature in this field, only very few studies provide causal-impact estimates of FFE programmes in terms of schooling and nutritional outcomes. In addition, nearly all the available evidence investigates the impact of school meals on these outcomes, while the impact of take-home rations appears to be under-researched.
In recent years, food for education (FFE) programmes have received renewed attention as a policy tool to achieve the Millennium Development Goals. Food for education programmes provide meals and/or take-home food rations in order to attract children to school, boost their learning and cognitive development, and give parents an incentive to encourage their children’s school attendance. However, the debate about FFEs among governments and donors continues, because FFE programmes tend to be more expensive than alternative schooling interventions, and nutritional benefits appear to be small, compared with programmes targeting pre-school children. In addition, many FFE evaluation studies suffer from methodological shortcomings. Thus, this systematic review summarises the most rigorous empirical literature in order to provide evidence showing whether or not FFE programmes deliver the assumed nutritional and schooling benefits to children in developing countries.
This systematic review aims to summarise and assess the evidence from rigorous research on the cost-effectiveness of programmes that provide subsidised school meals or take-home rations, and their impact on nutritional and schooling outcomes for children in developing countries.
The authors included experimental and quasi-experimental studies (including randomised controlled trials and field trials) assessing the cost-effectiveness of free or subsidised FFE programmes or their impacts on educational and nutritional outcomes in primary-school-aged children in developing countries.
The authors conducted a systematic search in published literature, including Medline and Econlit databases, and contacted colleagues in the field to identify unpublished studies.
The authors do not report assessing the quality of the included studies, but they discuss the advantages of specific research designs for establishing evidence of the causal impact of interventions on outcomes. The authors analyse the findings of the included studies in a narrative format and discuss possible factors which may mediate or moderate the intervention–outcome relationships.
The review uses appropriate methods to reduce biases by using clear inclusion criteria and discuss reasons for heterogeneity in the findings of the included studies. However, the review has some major limitations. It remains unclear how comprehensively the search was conducted - the review is restricted to English-language literature and there the reporting on the search is limited. The authors do not report comprehensively on data handling: there is no information on how many reviewers screened articles for their eligibility for inclusion in the review, and the authors do not provide information on data extraction. Moreover, they do not conduct a systematic assessment of the risk of bias in the studies included in the review, and therefore results are not analysed or reported separately on the basis of study quality. But, the authors discuss the methodological quality of included studies narratively when discussing their findings. They also emphasise the importance of more high-quality research in the field and make moderate policy recommendations, and this partially mitigates some of the weaknesses of the review.