Limitations of Test Scores : What the Experts Say

Don’t put all of the weight on a single test when making important decisions about students and schools (i.e. retention, promotion, probation, rewards). Instead, seek multiple indicators of performance. Include performance assessments and other indicators of success such as attendance, students taking Advanced Placement courses, etc.
Dr. Robert Linn, "Standards-Based Accountability - Ten Suggestions," National Center for Research on Evaluation, Standards, and Student Testing, www.cse.ucla.edu

State policies must not mistakenly assume that the current state tests fully measure achievement of the standards. For standards that are truly important, students should not just get a passing score on a set of miscellaneous test items but should have to show they have mastered each standard. Thoroughly assessing all standards at the state level would be far too expensive and time consuming, so it can only be done by teachers in local schools.
Dr. Ron Brandt (Executive Editor Emeritus of Educational Leadership, published by the Association for Supervision and Curriculum Development), "Changes Needed in Virginia's SOL Program," January 2000
 

"…There are limitations to what any single indicator can tell us. A potential misuse of SOL tests is to give them too much weight in making decisions about students, teachers, administrators, and schools. Whether it is student or teacher competence -- or school accountability -- decisions are best made using multiple factors, and we must balance SOL test data with other information.
Dr. James McMillan (Prof. of Ed. Evaluation, VCU), "SOL Tests Aren't Sole Predictor of Effective Education," Richmond Times-Dispatch (10/15/99)

 

"… even if the SOL tests were beyond reproach, the use of test scores as the ultimate criteria for graduation decisions violates professional standards for test use. Test scores should inform professional opinion, not override it. Similarly, the decision to base school accreditation on SOL test passing rates is based on a simple but indefensible logic: low passing rates mean poor schools; high passing rates mean good schools."
Dr. Lawrence Cross (Prof. of Ed. Research and Evaluation, VA Tech), " Virginia Schools Need Accountability for the Tests as Well as the Tested," VA Forum (July 2000)

" By relying on personal observations alone, a teacher may view a slow learner …as being lazy…and … may be wrong with potentially serious consequences. Tests can help to reduce these types of misjudgments …. This does not mean that tests can serve as a replacement for the personal judgment of professionals. In fact, a test is usually only one source of information among several used to assist in making the most accurate and fair decisions possible. …test results are not perfect. No procedure that assesses people can be. The best decisions come when test results are combined with information from interviews, expert observations, ratings of past work, and so on"
Association of Test Publishers, "Questions about Tests," http://www.testpublishers.org/fquestion.htm, pp. 2, 4

"It is important to keep in mind that test scores give only one picture of how a student is doing in school and that many things can affect a student's test scores. Therefore, it is important to consider other kinds of information as well. The school has more detailed information about how a student is doing."
Stanford 9 score reports, Harcourt Educational Measurement

"We know from decades of research that student achievement on standardized tests… is strongly related to many factors that schools have little or no control over, such as parental education, the number of parents living at home, student mobility, dominant home language, and community culture. …These influences should be considered when interpreting the scores -- not as excuses, but as factors that must be addressed to understand fully the meaning of the test scores. To ignore them is to deny their very real impact on student performance. Schools should be held accountable for what they can influence; communities and families should be held accountable for their contribution to student performance.
McMillan, Ibid.

"…Test scores must be considered, but don’t tell the whole story. Decisions about whether students should get high school credit and whether schools should be accredited must be judgments made by qualified people based on all relevant data..."
Brandt, Ibid.

"Inherently, candidates and elected officials must know that presenting test scores as isolated measures of success or failure is a far too narrow and easy way of assessing education today."
ETS President and CEO Kurt Landgraf , 1/17/01 NY Times

"Inferences from test scores to quality of schooling are problematic and must depend a great deal on contextual information"
Haertel, E. H. (1999). "Validity arguments for high stakes testing: In search of the evidence," Educational Measurement: Issues and Practice, 18, 4, 5-9

"Assessment can provide important information, but … While results can be used for one of many purposes (e.g., accountability, informing instruction, program evaluation), no single method can address all of these purposes well. Most state assessment systems are designed to improve instruction and to inform accountability decisions, but these two purposes are often at odds… Therefore, it is critical to be clear about the purpose of assessment and the limitations of the methods used."
L. Banicky, Ph.D. with H.K. Foss, "Assessing Student Learning," Delaware Education Research & Development Center, University of Delaware (2000).

Position statement on high stakes testing by the Alliance for Childhood

HIGH-STAKES TESTING:
A STATEMENT OF CONCERN AND CALL TO ACTION
April 25, 2001

Political leaders throughout America, including President Bush, are calling for a dramatic increase in standardized testing in public schools. The new tests invariably carry high stakes--that is, the test results are linked to serious consequences for students, teachers, and schools.

We believe that this massive experiment, intended to raise educational achievement, is based on misconceptions about the nature and value of testing and about how children develop a true love of learning. We further believe that this experiment may harm children's health by causing excessive anxiety and stress. Health-care professionals and parents already report that test-related stress is literally making many children sick.

At the same time, we know a great deal about the kinds of schools and assessment practices that will best support children's learning and the development of the skills and capacities most needed in the 21st century. Such schools foster strong bonds between students, families, and good teachers. They help students to frame questions intelligently, to pursue and analyze information, and to think with originality, creativity, and concern for others. Yet these schools and practices--as well as some of our most able teachers--are placed increasingly at risk by the proliferation of standardized tests.   CLICK HERE TO LINK TO THE REST OF THE ALLIANCE FOR CHILDHOOD'S STATEMENT.