Trends Show Improvement in Testing Young Children

Status: 
Archived
Subject: 
K-12 Testing

Practices in testing young children have improved in the past decade, according to results of another survey of the states. "Trends in Early Childhood Assessment Policies and Practices," by Lorrie Shepard, Grace A. Taylor and Sharon L. Kagan, reports that the testing of children through grade 2 has improved between a 1988 survey and their similar 1995-96 survey. The report recommends further changes to reduce the likelihood of harmful consequences and to better enable assessment to support all students.

The authors report that some states have modified laws or regulations that mandated or encouraged various testing practices in order to discourage inappropriate testing, such as readiness testing and multiple-choice testing for accountability. Such testing frequently has harmful consequences for young children, such as retention in grade or placement in low-level school programs. Many of the practices they report on are not state mandates but local requirements and uses of tests that are in varying degrees monitored by state education departments.

Among the findings:

  • "Most states have made an effort to move away from readiness testing and kindergarten retention." Many states have taken active steps to discourage these harmful practices. As the report notes, "Mandated formal testing is problematic because it often leads to misuse of test results." Some states, however, still have mandates to administer these tests.
  • "Almost all state-mandated standardized testing for purposes of school accountability has been eliminated for children below grade 3; some local testing for accountability remains." They found three states testing for accountability in kindergarten, three others in first grade, and seven requiring such tests at grade 2. They also reported 23 states test at grade 3. (The latest survey on state assessment programs by the Council of Chief State School Officers shows a reduction in grade 2 testing to four states, though it appears not all states responded fully to that survey.)
  • "Some states and local districts are moving to new forms of assessment in the early grades that are more supportive of instruction." These include classroom-based assessments that help teachers evaluate each child's learning while accumulating information for public reporting purposes. Mostly this is a local, not a state phenomenon.
  • "Misuse of screening instruments for instructional purposes has apparently decreased since 1988 but continues to some extent." Screening tests should not be used directly for instructional purposes but only to indicate students who might benefit from a more detailed study. The authors recommend that "states eliminate formal, every-pupil cognitive screening requirements and instead establish informal referral mechanisms." They note, "[W]e know of no evidence to suggest that identification of children with special needs is either more accurate in those states with mandated testing or reaches more children earlier."
  • "There is a need for professional training to understand and be able to use new forms of assessment." The authors recommend greater attention to this issue.
  • "Testing of preschool-age children is largely driven by mandates for categorical programs," such as special education. These programs require that children be categorized based on test scores in order to receive program funds. Labeling has been found to lead to over-representation of minority students in categorical programs and to those students staying permanently in such programs.

The study also found that parents, policymakers, and other users of test information needed a great deal of education about the meaning of test results to discourage misinterpretation of the results in making decisions.

The perspective and recommendations of the authors follow the National Association for the Education of Young Children's Developmentally Appropriate Practice in Early Childhood Programs (see Examiner, Winter 1996-97). These guidelines are highly consistent with the Principles and Indicators for Student Assessment Systems of the National Forum on Assessment and thus with the standards FairTest used to evaluate state programs. Some of the changes Shepard and her colleagues found are in line with the more positive results found by FairTest in its study of state testing practices, such as a slow growth in the use of performance assessments.

The report is available on the web at http://govinfo.library.unt.edu/negp/, but has not been published