And the 2005 award for "best paper on early childhood intervention" goes to….. MIT’s Michael Anderson for a terrific reanalysis of three early randomised trials of childhood intervention programs – Abecedarian, Perry Preschool, and the Early Training project.* He finds that:
The view that the returns to public educational investments are highest for early childhood interventions stems primarily from several influential randomized trials – Abecedarian, Perry, and the Early Training Project – that point to super-normal returns to preschool interventions. This paper presents a de novo analysis of these experiments, focusing on core issues that have received little attention in previous analyses: treatment effect heterogeneity, over-rejection of the null hypothesis due to multiple inference, and robustness of the findings to attrition and deviations from the experimental protocol. The primary finding of this reanalysis is that girls garnered substantial short- and long-term benefits from the interventions, particularly in the domain of total years of education. However, there were no significant long-term benefits for boys. These conclusions change little when allowance is made for attrition and possible violations of random assignment.
The finding that the programs don’t work well for boys is curious, since job training and mobility programs have also come to similar conclusions for boys and young men. It seems blokes are harder to help.
Of course, we can only do this analysis because far-sighted American researchers in the 1960s decided to set up some carefully designed randomised trials of early intervention programs. Hey, crazy thought – in a season when Aussie politicians are talking about how much they care about kids, perhaps they could play forty-year catchup, and set up some similar randomised trials in Australia?
* Yes Virginia, there are early childhood trials apart from Perry Preschool.