On the quality of research in education

From my inbox:

The What Works Clearinghouse (WWC) has released one new report this week examining research on strategies that aim to improve instruction for children with disabilities.

The Lovaas Model of Applied Behavior Analysis uses one-on-one instruction for young children, during which a teacher cues a behavior, prompts the appropriate response, and provides reinforcement to the child to reduce behaviors that interfere with learning. As the child progresses, the instruction moves into group settings and focuses on social and pre-academic skills in preparation for preschool.

Fifty-eight studies reviewed by the WWC investigated the effects of the Lovaas Model on children with disabilities. One study meets WWC evidence standards and a second study meets standards with reservations. The two studies included 51 children in two locations ages 18 to 42 months with autism or pervasive developmental disorder. Based on these two studies, the WWC found the Lovaas Model to have potentially positive effects for cognitive development for children with disabilities and no discernible effects for communication/language competencies, social-emotional development/behavior, and functional abilities. Read the full WWC report now at http://ies.ed.gov/ncee/wwc/reports/ece_cd/lovaas_model/index.asp

The thing that struck me was the first two sentence of the third paragraph. Out of 58 studies, exactly two were not tossed out of the WWC on quality grounds. That is pretty sad indeed, and testament to the value of the WWC at setting standards.