The Curiosity Corner early childhood education program was one of 19 highlighted for its effectiveness by Graham Allen, the MP commissioned by the government to review early intervention the report of which was published in January. But while Allen’s confidence in the effectiveness of the program is backed by a review of the evidence conducted last year by the University of York and John Hopkins University, it isn’t shared by the What Works Clearinghouse, an initiative of the US Department of Education's Institute of Education Sciences.

Its 2009 review of the program concluded that the program “was found to have no discernible effects on oral language, print knowledge, phonological processing, cognition, and math”. By contrast, when the program was included in the systematic review undertaken in 2010 into early childhood education programs by a team of researchers from the University of York and Johns Hopkins University, the review team concluded that Curiosity Corner was one of six early education programs that demonstrated strong evidence of effectiveness.

Developed by the Success for All Foundation, the program works with three- and four-year old children to help prepare them for school, emphasising the attainment of language and literacy skills.

The program comprises two sets of themed units: one for three-year olds and one for four-year olds. The curriculum is delivered over 38 weeks (roughly one academic year) and teachers present daily activities which engage children in problem-solving and literacy-focused tasks. Coaches are provided to assist teachers to implement the program with fidelity.

Why, though, did two reviews appear to come to such different conclusions about the program’s effectiveness, especially when both rely on the same two source studies to provide the evidence for the program’s impact?

One study of the program, conducted in 2008, met the highest standards of evidence. It used a randomized controlled trial design to allocate 18 preschools in three US states to either take part in the program or be part of a control group. The second study, conducted in 2001, was a non-randomized design where children receiving the program were matched to children not receiving the program so as to provide a comparison.

The key difference between the two reviews lies in the periods used to estimate impact. The 2008 study involved children in the intervention group being offered Curiosity Corner at the start of their pre-school year. They were followed up in the spring term and again in the spring of their kindergarten or reception year. The 2001 study followed the progress of three- and four-year old children and measured their language ability at the start and end of the pre-school academic year.

The 2008 study, which used random allocation, found no significant differences between Curiosity Corner children and the controls at the end of preschool. The 2001 non-randomized study did find the program had an impact, with three-year old children scoring significantly higher on expressive language ability. The authors estimated a small to medium effect of the program (+.24) on three- and four-year olds.

From this, it seems straightforward why the What Works Clearinghouse may have concluded no discernible effects. Adjusting for the bias inherent within a non-randomly allocated study, the WWC suggest that Curiosity Corner could be given an average improvement index of between -3 and +7 percentile points. In other words, in a group of 100 children, where the average control group child would be at the 50th percentile position, Curiosity Corner could, at best, raise this child to the 57th percentile position.

But the 2008 study provides additional evidence that is neglected by the What Works Clearinghouse review. The study followed children beyond preschool into their kindergarten or reception year of school, to examine their language and academic skills. At this point in time, there were small but non-significant effects on phonological awareness and math ability and significant differences on children’s literacy skills. This is the evidence used to justify a conclusion of strong effectiveness in the 2010 systematic review.

It is more often the case that studies of early interventions show strong, significant impacts shortly after people have been through a program that may, or may not, dissipate as time goes by. It’s much rarer, however, that non-significant, short-term effects evolve in time to produce long-term effectiveness. However, Curiosity Corner looks to be one such program. It is possible that as a school-readiness program, its greatest effects come to be felt only once the children enter school. It also underlines the importance of including long-term evaluations of programs where possible.

References:
Chambers, B., Cheung, A., Slavin, R.E., Smith, D. and Laurenzano, M. (2010) Effective early childhood education programs: a systematic review. Best Evidence Encyclopedia

Curiosity Corner, WWC Intervention Report: Early Childhood Education, What Works Clearinghouse (January 2009).

Links:
See www.successforall.org