ERIC Identifier: ED312773
Publication Date: 1989-00-00
Author: Bowers, Bruce C.
Source: ERIC Clearinghouse on
Educational Management Eugene OR.
Alternatives to Standardized Educational Assessment. ERIC
Digest Series Number EA 40.
An American educator who was examining the British educational system once
asked a headmaster why so little standardized testing took place in British
schools. "My dear fellow," came the reply, "In Britain we are of the belief
that, when a child is hungry, he should be fed, not weighed." This anecdote
suggests the complementary question: "Why is it that we do so much standardized
testing in the United States?"
WHAT ARE THE MAIN USES OF STANDARDIZED TESTING IN AMERICAN PUBLIC SCHOOLS?
Advocates of standardized testing assert that it simply
achieves more efficiently and fairly many of the purposes for which grading and
other traditional assessment procedures were designed. Even critics of
standardized testing acknowledge that it has filled a vacuum. As Grant Wiggins
(1989a) puts it, "Mass assessment resulted from legitimate concern about the
failure of schools to set clear, justifiable, and consistent standards to which
it would hold its graduates and teachers accountable."
Standardized testing is currently used to fulfill (1) the administrative
function of providing comparative scores for individual students so that
placement decisions can be made; (2) the guidance function of indicating a
student's strengths or weaknesses so that he or she may make appropriate
decisions regarding a future course of study; and, more recently, (3) the
accountability function of using student scores to assess the effectiveness of
teachers, schools, and even entire districts (Robinson and Craver 1989).
WHAT PROBLEMS HAVE ARISEN AS A RESULT OF WIDESPREAD USE OF STANDARDIZED TESTING?
The phrase "test-driven curriculum" (Livingston,
Castle, and Nations 1989) captures the essence of the major controversy
surrounding standardized testing. When test scores are used on a comparative
basis not only to determine the educational fate of individual students, but
also to assess the relative "quality" of teachers, schools, and school
districts, it is no wonder that "teaching to the test" is becoming a common
practice in our nation's schools. This would not necessarily be a problem if
standardized tests provided a comprehensive, indepth assessment of the knowledge
and skills that indicate mastery of a given subject matter. However, the main
purpose of standardized testing is to sort large numbers of students in as
efficient a manner as possible. This limited goal, quite naturally, gives rise
to short-answer, multiple-choice questions. When tests are constructed in this
manner, active skills, such as writing, speaking, acting, drawing, constructing,
repairing, or any of a number of other skills that can and should be taught in
schools are automatically relegated to a second-class status.
WHAT ALTERNATIVES TO STANDARDIZED TESTING HAVE BEEN
It is reasonable to assume that the demand for test results that
can be compared across student populations will remain strong. The critical
question is whether such results can be obtained from tests that attempt a more
comprehensive assessment of student abilities than the present standardized
tests are capable of providing. An ancillary, but equally critical, question is
whether such tests are too costly to be widely administered.
Suggested alternatives are based on the concept of a "performance-based"
assessment. Depending on the subject matter being tested, the performance may
consist of demonstrating any of the active skills mentioned above. For example,
in the area of writing, drawing, or any of the "artistic expression" skills, it
has been suggested that a "portfolio assessment," involving the ongoing
evaluation of a cumulative collection of creative works, is the best approach
(Wolf 1989). For subjects that require the organization of facts and theories
into an integrated and persuasive whole (for example, sciences and social
sciences), an assessment modelled after the oral defense required of doctoral
candidates has been suggested (Wiggins 1989a).
A third approach, which might be termed the "problem solving model," can be
adapted to almost any knowledge-based discipline. It involves the presentation
of a problematic scenario that can be resolved only through the application of
certain major principles (theories, formulae) that are central to the discipline
under examination (Archbald and Newmann 1988).
CAN PERFORMANCE-BASED ASSESSMENTS BE USED TO COMPARE STUDENTS ACROSS DIFFERENT SETTINGS?
Performance-based assessment is more easily
scored using a criterion-referenced, rather than a norm-referenced approach.
Instead of placing a student's score along a normal distribution of scores from
students all taking the same test, a criterion-referenced approach focuses on
whether a student's performance meets a criterion level, normally reflecting
mastery of the skills being tested.
How can such an assessment be reliably compared to similar assessments made
by other teachers in other settings? It has been suggested that American
educators adopt the "exemplary system" being called for in Great Britain. In
this system, teachers involved in scoring meet regularly "to compare and balance
results on their own and national tests" (Wiggins 1989b), thus increasing
reliability across settings. Clearly, however, such an approach (similar to the
approach currently in use for the scoring of Advanced Placement essay exams)
could be prohibitively expensive if carried out on a large scale. A key question
is whether the costs associated with this labor intensive scoring system would
be offset by the presumed instructional gains obtained from an assessment model
that rewarded a more thorough and holistic approach to instruction.
HAVE THERE BEEN ANY STATEWIDE EFFORTS TO PROVIDE ALTERNATIVES TO STANDARDIZED TESTING?
California has probably made the greatest effort
in this direction, beginning in 1987 with its statewide writing test and
continuing with its current development of performance-based assessment in
science and history (Massey 1989). The Connecticut Assessment of Educational
Progress Program uses a variety of performance tasks in its assessment of
science, foreign languages, and business education (Baron 1989). (However, this
assessment includes only a sample of students at any given grade level, and, in
addition, every year there is change in the subjects for which performance tasks
are required.) Vermont education officials are currently seeking legislative
approval for funds to pursue a portfolio assessment approach in addition to the
current standardized testing (Massey 1989).
WHAT IS THE PROGNOSIS FOR A GENERAL SHIFT AWAY FROM STANDARDIZED TESTING AND TOWARD PERFORMANCE-BASED TESTING?
psychometric terms, the tradeoff in such a shift is to sacrifice reliability for
validity. That is, performance-based tests do not lend themselves to a cost- and
time-efficient method of scoring that, in addition, provides reliable results.
On the other hand, they actually test what the educational system is presumably
responsible for teaching, namely, the skills prerequisite for performing in the
real world. The additional costs involved in producing reliable results across
different settings for performance-based tests are unknown.
The question is whether a majority of educators will echo the sentiments of
George Madaus, director of the Center for the Study of Testing, Evaluation, and
Educational Policy, who believes that performance-based testing "is not
efficient; it's expensive; it doesn't lend itself to mass testing with quick
turnaround time--but it's the way to go" (Brandt 1989).
Archbald, Doug A., and Fred M. Newmann. "Beyond
Standardized Testing: Assessing Authentic Academic Achievement in the Secondary
School." Reston, VA: National Association of Secondary School Principals, 1988.
65 pages. ED 301 587.
Baron, Joan B. "Performance Testing in Connecticut." EDUCATIONAL LEADERSHIP
46, 7 (April 1989): 8. EJ 387 136.
Brandt, Ron. "On Misuse of Testing: A Conversation with George Madaus."
EDUCATIONAL LEADERSHIP 46, 7 (April 1989): 26-29. EJ 387 140.
Livingston, Carol, Sharon Castle, and Jimmy Nations. "Testing and Curriculum
Reform: One School's Experience." EDUCATIONAL LEADERSHIP 46, 7 (April 1989):
23-25. EJ 387 139.
Massey, Mary. "States Move to Improve Assessment Picture." ACSD UPDATE 31, 2
(March 1989): 7.
Ralph, John, and M. Christine Dwyer. "Making the Case: Evidence of Program
Effectiveness in Schools and Classrooms." Washington, D.C.: U.S. Department of
Education, Office of Educational Research and Improvement, November 1988. 54
Robinson, Glen E., and James M. Craver. "Assessing and Grading Student
Achievement." Arlington, VA: Educational Research Service, 1989. 198 pages.
Wiggins, Grant. "A True Test: Toward More Authentic and Equitable
Assessment." PHI DELTA KAPPAN 70, 9 (May 1989a):703-13. EJ 388 723.
Wiggins, Grant. "Teaching to the (Authentic) Test." EDUCATIONAL LEADERSHIP
46, 7 (April 1989b): 41-47.
Wolf, Dennie P. "Portfolio Assessment: Sampling Student Work." EDUCATIONAL
LEADERSHIP 46, 7 (April 1989): 35-39. EJ 387 143.