<< back to Early Education/Cognitive Development

The High Stakes Testing Trap
by Samuel J. Meisels, Ed.D., President of Erikson Institute

We live in a test-obsessed society. In virtually his first official act of office, President George W. Bush decreed that every child in Grades 3-8 should be tested annually in reading and math and that federal aid to schools should be linked to children's scores on these assessments.

State and national policy makers devote the majority of their statements about education to extolling the virtues of high stakes testing. Sometimes they refer to tests as part of the educational process; mostly, they describe them as a kind of multiple-choice expeditionary force designed to seek out and destroy lazy teachers and indolent children.

The fact is that more tests are being given in the US than ever before.

Currently, we know that:

  •  All states are testing their children on statewide exams (usually in grades 4, 8, and 10 or higher-though with the No Child Left Behind Act (NCLB) this will take place annually from grades  3-8);

·   Three-quarters of the states use these tests to create "school report cards" and nearly 40% use the test scores to identify low performing schools;

·    At least 18 states already have the authority to close, take over, or overhaul schools that are identified as failing these tests (with NCLB all states will have this responsibility);

·    One-third of the states sanction poor performing schools;

·    Only seven states provide extra funding for all of their low-performing schools, and just nine states allocate money for remediating failing students.

Testing in America is epidemic--despite complaints by students, teachers, and parents. Actually, in several states grassroots campaigns are encouraging parents to keep their children home on test days. Further, in a recent survey  conducted by the Association for Supervision and Curriculum Development, it became clear that many parents are confused about these tests. Half the parents who responded to the survey said that they were not sure what their children were being tested on or how the test scores would be used. But testing continues to be high on the policy agenda,  despite these misgivings by  parents and others.

Test taking for young children

Focusing now on young children, what is the test taking experience like for them? We know the following:

1.  Children are assessed on isolated skills in settings that are devoid of context, rather than being evaluated on tasks that are part of natural settings where they are asked to use what they know and have had experience with previously.

2.  For whole group testing of young children, the entire experience is dominated by how to fill in the bubbles, move the marker, or be sure that everyone is in the right place. These activities may be related to test taking, but they have little to do with the skills we're interested in evaluating.

3.  The tests are characterized by multi-step directions, visual clutter, and confusing instructions that   advantage children who have had prior experience with complex    information or with similar types of questions and answers.

4.  The tests are focused on decomposition, decontextualization, and objectivity, but it is not clear that by breaking a complex skill like reading into its constituent parts that mastery of the whole can be inferred.

High-stakes testing    

Of great concern are high-stakes tests, which may result in long-term negative consequences for children. High-stakes refer to significant rewards or sanctions such as retention in grade, placement in special classes or extra-year programs, or designation as either a gifted or slow learner. Research has shown that the labeling that accompanies these high-stakes tests can have long term impact on teachers' perceptions of children's ability to learn; stigmatization of children and tracking them into low achieving groups; and also on children's self-perceptions, estimates of their own abilities, and motivation and achievement.

Low-stakes tests refer to teaching and learning--an irony at best.

Along with high-stakes tests comes what the scholar George Madaus calls the "first principle of high-stakes testing"--it is that "if students, teachers, or administrators believe that the results of an examination are important, it doesn't matter whether this is really true or false--the effect is produced simply by the perception." This is sort of the Emperor's New Clothes phenomenon in assessment. In effect, if the test is given, it must be accurate. We tend to believe the tests, rather than the kids. (How many times do we hear: "I know that he knows it, but he can't do it on the test"?)

The fact is that too many policies about assessment are being made without any awareness of their long-term consequences. What are these consequences?

·    A system is established in which indicators of learning overwhelm   attention to learning itself;

·    Attention is directed to what is    measured, overlooking nearly everything else that is not measured;

·    A standardized pedagogy becomes emphasized for use with a non-standard, diverse student population;

·    Few rewards for innovation or risk taking are available on the part of teachers or students;

·    All of this results in a distortion of the motivational climate for teaching and learning.

It is not high standards that we are concerned with; it is high stakes, and it's time for a change.

Assessments need to reflect how children learn

Howard Gardner points out that tests should be designed to highlight--rather than obscure--different cognitive features. Tests are more frequently used to identify weaknesses rather than to designate strengths. Gardner challenges us to create assessments that more accurately depict reality--a reality in which people and especially young children, learn things in different ways, at different times, and in different settings.

Our challenge is to transform assessment information into meaningful instruction. We can do this through assessments that focus on performance, rather than on conformity to criteria drawn from outside of the child's lived experience.

Remember, the best way to assess a child's performance is to study performance, not something else, and the best way to improve a child's performance is to teach the child, not test him.

Fundamentally, we are trying to create instructional assessments--those that are:

·   Primarily focused on individual learning, not group reporting;

·   Not designed to rank and compare students or to be used for high stakes, but to be a tool for the teacher;                      

·   Intended to clarify what students are learning and have begun to master by providing information relevant to  understanding individual students' learning profiles;

·   Designed to guide instructional decision-making and provide instructionally relevant information to teachers so that teaching can be enhanced and learning improved.

In short, we're trying to say that high-stakes testing is inconsistent with meaningful educational outcomes at the outset of school and often beyond. This is a time of dramatic developmental change, a critical period of transition from home to school, and an interval of heightened sensitivity to socialization, openness to exploration,   and trying out of the self in relation to others. It's not a time to highlight failure or to impose narrow views of learning and achievement. High standards are appropriate. High stakes are not.