Chapter 5 - Consortium For Longitudinal Studies
Introduction to the Original Evaluation (Excerpt)
The Report from the Consortium of Longitudinal Studies ("the Consortium Study") was a collaborative effort by eleven research groups with longitudinal studies of early childhood interventions that operated between 1962 and 1972. Only eight of the eleven projects were used for most of the analyses described here: the Early Training Project, the Experimental Variation of Head Start Curricula, the Harlem Training Project, the Mother-Child Home Program, the New Haven Follow-Through program, the Parent Education Program, the High/Scope Perry Preschool Project, and the Philadelphia Project. (A ninth study, the Curriculum Comparison Study, was used in the analysis of a limited number of outcomes.) The evaluation sought to determine the long-term effects of selected infant and preschool intervention programs on child participants. The data for the final follow-up were collected through a youth survey, a parent survey, school records, and an IQ test.
Irving Lazar, then of Cornell University, headed the independent analytic team (the "Consortium team") that coordinated the collaboration. The Consortium Study statistically combined findings from selected early childhood education programs that had conducted long- term follow-ups. The studies included in the analysis had relatively rigorous designs, but most had small samples. By assessing the studies as a group, the Consortium team was able to increase statistical power. As a result, they often found statistically significant findings, even when many of the studies individually did not demonstrate such effects.
The Consortium Study has often been cited as evidence that early childhood intervention programs can improve cognitive outcomes and school performance. Its findings were largely consistent with other literature reviews, including the findings that initial IQ gains "fade out" and that early intervention leads to improvements in school performance. Despite the rigor of the study, some methodological issues remain: the small number of studies examined, relatively high rates of attrition, and potential selection bias. Moreover, although this concern was less important when the study was published, the projects themselves operated over forty years ago in a very different societal environment, which may limit their current applicability.
There are currently no comments on this document
HOME - PUBLICATIONS - CONFERENCES - ABOUT US - CONTACT US