CHILDREN NOT READY TO LEARN? THE INVALIDITY OF SCHOOL READINESS TESTING 1 lorrie a. shepard University of Colorado at Boulder School readiness testing is in disrepute. For example, the Goal 1 Technical Planning Subgroup, asked to advise the National Education Goals Panel about how to measure readiness for school, de- clared instead that they would assiduously avoid the use of the term “readiness” throughout their en- tire report, “because current practice contains so much evidence of the potential misuse of readiness tests and of the misapplication of developmental screening tests” (1991, p. 8). Although it is a nat- ural part of good teaching to assess what children know and what they are ready to learn next, the use of readiness tests as gatekeepers to deny school entry is problematic for several reasons. First, it keeps low performing children out of school who most need the learning opportunities provided in school. In turn, removal of less ready children contributes to the escalation of curricular demands making kindergarten expectations inconsistent with the development of normal 5-year-olds. Height- ened attention to readiness concerns, in the media and in neighborhood talk, also fuels the “red- shirting” trend whereby affluent parents whose children do well on readiness measures nonetheless hold their 5-year-olds out of kindergarten for an extra year. Last, but not least, readiness tests lack the technical rigor to make accurate predictions or placements. The purpose of this article is to review both the technical and policy aspects of readiness test- ing in local school districts. The first section is a summary of the policy context. Why did so many school districts implement readiness testing during the 1980s? What were the intended and unin- tended consequences of accompanying policies such as kindergarten retention and redshirting? What have professional associations done to redress the effects of perceived negative practices and what, if any, have been the effects of those efforts? In the second section, the types of readiness tests and developmental screening measures misused as readiness tests, are reviewed along with evidence of their substantive and technical validity. In the concluding section, policy and assessment alternatives are addressed. The Policy Context of Readiness Testing Prevalence of Readiness Testing There are no national data available on the extent of readiness testing or the number of children affected. In 1988, at a time when readiness testing was at its peak, Gnezda and Bolig conducted a survey of early childhood specialists and testing directors in the 50 states. According to their report, testing specifically to determine readiness for school before kindergarten was mandated in four states and occurred locally in an additional 26 states; readiness testing before first grade was mandated in six states and occurred locally in 37 states. Considering the two readiness practices jointly, there were only three states that did not report the occurrence of either pre-kindergarten or pre-first-grade readi- ness testing. In the 23 states with either mandated developmental screening or readiness testing, most of the mandates were enacted in the mid-to-late 1980s and were attributed by the respondents to “ed- ucational reform efforts, concern over the accountability of schools and of teachers, greater interest in identifying and serving children “at risk” of school failure, and a downward extension of acade- Psychology in the Schools, Vol. 34(2), 1997 © 1997 John Wiley & Sons, Inc. CCC 0033-3085/97/020085-13 85 1 Portions of this article have been adapted with permission from Shepard, L. A. (1991). “Readiness testing in local school districts: An analysis of backdoor policies.” In S. H. Fuhrman & B. Malen (Eds.), The Politics of Curriculum and Testing. London: Falmer Press.