By Howard White, CEO, The Campbell Collaboration
Children spend far more time out of school than in school. It is therefore a common-sense part of received wisdom that the home environment matters for child development and child learning outcomes. A logical extension of this view is that interventions to support learning in the home environment will be an effective strategy for tackling the well-known problem of poor learning outcomes in developing countries. Many children leave primary school unable to read or perform simple arithmetic.
I have been prompted to write this blog by a systematic review I am reading which appears in our Campbell-UNICEF Child Welfare Megamap. I will come back to this review below, as it is the third time I have encountered this issue in recent years. And the evidence doesn’t look good.
The first time I came across evidence on this issue was in dealing with a 3ie-funded study of interventions aimed at mothers in India. These interventions had a small, but significant, impact on the learning outcomes of the children of mothers targeted by the intervention. The first draft of the report we received focused on the significance of the effects found. But 3ie requires authors to address cost-effectiveness. In the revised version, the authors compared the cost effectiveness of their intervention with those listed in a review of cost-effectiveness of education interventions. The intervention fared poorly on this comparison: 15 out of 16 interventions listed in the cost-effectiveness review were more cost-effective approaches to improving learning outcomes than the intervention aimed at mothers.
A few years later I was writing the Campbell Plain Language Summary for a review of community- and family-based interventions to improve children’s literacy. There are a wide range of such interventions such as mobile libraries, local-language, literacy instruction outside schools including the teaching of literacy through religious instruction, the distribution e-readers in countries, educational TV and radio, and supporting community members to educate children. I cite this review as an example of why producing a summary of findings table is often necessary to see the wood for the trees. And once I prepared the summary table the evidence was striking. There is no evidence at all for most interventions. Support to parents and peers is largely ineffective in improving literacy - though it has worked in some places. What works is watching television! More precisely, educational TV such as Sesame Street and its many local-language derivatives.
And this morning I am reading Asim et al.’s review of education interventions in South Asia. The forest plot is so striking it moved me to write this blog. The plot is reproduced here, so take a look:
(Source: Asim et al. Figure 2, p. 30).
Interventions aimed at teachers and schools overall improve children’s learning outcomes. But those aimed at household and communities have no effect. And there is not a whole lot of heterogeneity going on. There is pretty much straight down the line zero effect.
Of course, these interventions may have other beneficial effects. As we found in the education review I worked on with Shari Krishtnaratne and Ella Carpenter, you need different interventions to get children into school than you do to help them learn once they are there. Successful demand-side interventions are far more likely to be community and household-oriented. And the evidence suggests that that should be their aim. Get parents to send children to school, not supplement the education at home. If school is broken, fix the school, rather than try to substitute for it at home.
Since joining Campbell, I have struck at how often the evidence sends a clear signal for policy and practice. Here is such a case.