A Position Paper of the Association for Childhood Education International by Vito Perrone
A Position Paper of the Association for Childhood Education International by
Vito Perrone
INDEX (Click on any of the following links to jump to the specified subject)
SOME HARD QUESTIONS ABOUT STANDARDIZED TESTING
CHILDREN AND TESTS
Vito Perrone is a faculty member of the Harvard Graduate School of Education,
where he is also Director of Programs in Teacher Education and Chair of the
Teaching, Curriculum and Learning Environments Program. In addition, he is a
senior fellow at the Carnegie Foundation for the Advancement of Teaching.
In 1976-the middle of a decade when standardized testing of young children
was approaching unprecedented levels and test results were being used for decisions
about kindergarten entry, promotion and retention, and placement in curricular
programs-ACEI issued a position paper calling for a moratorium on standardized
testing in the early years of schooling (ACEI/NAESP/Perrone,1976). Recognizing
that it was making primarily a moral statement, ACEI hoped that its action might
encourage serious discussion about the effects of testing and the active pursuit
of assessment directions that honored developmental traditions and were educationally
sounder than standardized tests .
Pressures to test children continued in the latter years of the 1970s, but there
was also vigorous debate about the negative effects of testing. And support
for more authentic forms of assessment, rooted in close observation and systematic
documentation of children's learning, became more common. After publication
of A Nation at Risk (National Commission on Excellence in Education, 1983),
however, the climate changed dramatically. Testing programs expanded greatly,
especially in kindergarten and the primary grades. The results have been deleterious,
particularly for poor and minority children .
A moratorium is more necessary now than in 1976. It is time for teachers, school
administrators and parents to say more forcefully than ever that testing and
its uses in the primary years must be brought to an end and reduced in frequency
thereafter. This 1991 ACEI position paper is dedicated to that end.
Given the power of standardized testing in society, it comes as a surprise to
many that the history of this form of testing is so short. Produced in 1909,
the Thorndike Handwriting Scale was, for example, the first popular standardized
achievement test used in the public schools.1
A wide variety of achievement and aptitude tests quickly followed.
By the 1930s, a majority of schools in the United States and Canada engaged
in some form of stan-dardized testing, but the scope was exceedingly small by
today's standards. Few people who completed high school before 1950, for example,
took more than three standardized tests in their entire school careers.2
The results were hardly ever discussed, parents didn't receive the scores and
school-wide results were not grist for local newspapers. By contrast to this
earlier period, those who complete high school in 1991 will have taken, on average,
from 18 to 21 standardized tests; many will have taken more, the majority of
them in the K-5 years.3 And test scores
will not only fill newspapers, but also become part of the sales-pitch of real
estate brokers, especially if test scores are high in a particular district.
To understand the overall magnitude of the shift, it should be noted that since
1950 the volume of testing has grown at the annual rate of 10-20 percent (Haney
& Madaus, 1989).
While the tests are problematic at all ages and levels of schooling, they are
particularly questionable for children in the primary grades. These are years
when children's growth is most uneven, in large measure idiosyncratic; the skills
needed for success in school are in their most fluid acquisitional stages. Implications
of failure in these years can be especially devastating.
SOME HARD QUESTIONS ABOUT STANDARDIZED
TESTING
Acknowledging that standardized tests overwhelm much of classroom
practice, Harvard psychologist Sheldon White suggests that we are contending
with "an affair in which magic, science, and myth are intermixed" (1975). He
is offering, of course, an understatement! How many of us actually believe that
an individual's intelligence, achievement and competence can be represented
adequately by any of the standardized tests that fill our schools? Or that one
distribution curve--whatever the metric--is capable of classifying all children?
Or that a particular score on a test can provide a genuinely defensible demarcation
between those who should be promoted to the next grade level and those who should
be retained? Between those who should be provided enrichment and those needing
remediation? Such assumptions defy almost everything we have come to understand
about children's growth, as well as their responses to particular educational
encounters. Teachers and parents know this. When they have a chance to step
back and reflect on their children, few will accept that any test score can
define any child.
Even if one fails to take note of the implicit assumptions of the tests--essentially
that children's knowledge and competence can be measured by the number of correct
answers they supply--an examination of the test items and the composition of
the tests (something those in schools need to do more often) ought to cause
some measure of pause, if not enormous concern.4
Are the questions clear? Do they address the particular educational concerns
of teachers of young children or of parents? Do the tests as a whole provide
useful information about individual children? About a class? Do they help children
in their leaming? Do they support children's intentions as learners? Do they
provide essential information to children's parents? In our experience with
teachers and parents, we have encountered few who can provide an affirmative
response to any of these questions. That teachers and parents can offer so little
positive response surely suggests problems with the tests and the emphasis given
to them.
In contrast, however, almost all teachers responded affirmatively to the following
questions: Do you feel any pressure to teach to the tests? If the tests were
not given or used for the evaluation of individual children, teachers and schools,
would you use fewer skill sheets, workbooks and other simple-response pedagogical
materials? Would you use a broader range of instructional materials, more attention
to integrated learning? Would expectations for all children enlarge? Would you
devote more attention to active, inquiry-oriented programs in mathematics and
science? Would you give more time to the arts? Would the curriculum be more
powerful, more generative? Do you feel that you can assess children's learning
in more appropriate ways than the use of standardized achievement tests?
Used for major educational decisions, as they are in many settings, the various
tests clearly limit educational possibilities for children. We need to understand
this well, for the pressure to use more tests for more purposes continues to
mount.
More on the History of Testing
As noted, testing programs began a substantial upward spiral after
1950; more often than ever before, they became the basis for selection and retention
in numerous educational programs and grade levels. But prior to 1965, the tests
were not used often in the early grades. This is important to understand. A
consensus associated with the traditions of kindergarten, as well as a developmental
perspective, guided the primary grades as a whole: the early years were "special,"
a time for natural growth and development. Where serious testing programs existed,
they generally began in grade 3 or 4.
Testing exploded in the period after 1965, however, especially with regard to
its uses. As evaluation demands grew with the influx of new federal and state
resources for schools, the tests were quickly seen as inexpensive and easy-to-use
measures for meeting the requirements. And with the accountability movement
of the 1970s, the tests became the definers of standards in almost all curricular
areas.
Yearly testing, beginning in grade 3, became more the norm, although in many
school districts accountability demands contributed to the use of annual fall
and spring testing as a means of determining "gains" in achievement. By the
late 1970s, testing started to invade the primary grades. Developmental understandings
began to erode as early years' testing became the big growth area. By the end
of the 1980s, testing of young children had become commonplace. Sixteen states
in the U.S. and districts in 21 other states now require children to take a
standardized test before entering kindergarten; districts in at least 42 states
require students to pass a standardized test before "graduating" from kindergarten.
It is now the exceptional school district that doesn't test K-2 children (National
Commission on Testing and Public Policy, 1990).
While many of the prekindergarten tests are of the paper-pencil
variety, most have a more individual, performance-oriented quality. For measuring
physical development, children are asked to skip or stand on one foot for 20
seconds; on the cognitive level, they are asked, for example, to retell a story
in its proper sequence. With regard to social and environmental experience,
they are asked to count to 10, recognize colors and shapes, manipulate a crayon
or pencil, and follow directions. The results of these "screening" activities
are often the basis for cautioning parents to "wait another year before starting
your child in kindergarten." They are also used as a means of "early identification"
of individuals who, as the preschool screeners say, might be expected to have
difficulty in school and might need special assistance (essentially an early
process of labeling). Although scant evidence exists that such early screening
is beneficial for either children or schools, it has, nonetheless, become almost
universal.
In kindergarten, children typically receive their first paper-pencil test, which
ostensibly gauges "reading readiness." Those who score in the bottom quartile
are encouraged, in some settings required, to spend another year in kindergarten;
or they are placed in a K-1 transitional setting that often leads to later retention.
The underlying rationale is that children benefit from the knowledge teachers
gain from this kind of testing. Yet, teachers gain little if any important knowledge
from such tests. With so little evidence that reading readiness scores correlate
with reading success, their use is unwarranted. It is a scandal to retain children
on the basis of such tests (Shepard, 1987).
Beginning in grade 1 and continuing through the elementary grades, children
in most schools complete at a minimum (and many children take even more) an
annual achievement test battery such as the Metropolitan Reading Test, Metropolitan
Achievement Test, California Test of Basic Skills, Stanford Achievement Test
or Iowa Test of Basic Skills. In a small minority of districts (particularly
those serving middle class, mostly white populations), the tests are rather
benign. They are, as the administrators say, the "staying in-touch with overall
achievement levels." If scores go down significantly, that fact would likely
prompt discussion in these districts, but changes in scores tend not to be dramatic
enough to raise too much concern.
Given the pressures of the past two decades, however, the tests in the majority
of school districts have expanded in their purposes. For example, how well individual
children score determines whether they will be placed in a gifted and talented
program or become eligible for special tutoring. The results of annual achievement
testing also determine eligibility for a variety of enrichment programs, special
classes, foreign language instruction, and the like. The tests also determine
a student's academic level. They become the basis for early tracking and then
ongoing tracking, reflecting the belief that homogeneous achievement groups
facilitate more efficient and effective teaching and learning. That such grouping
on the basis of a test leads mostly to inequity has not been sufficiently considered.
And in recent years, test results have been used increasingly to determine whether
a child should advance from one grade to another. This represents a new dimension
(Meisels, 1989; National Association for the Education of Young Children, 1988).
What Testing Means
for Children
All in all, increased testing results in increased pressure on
teachers and children. In a school guided by developmental concerns, teachers
place much less emphasis on the tests. If, however, tests play a significant
role in grade advancement, or the tests are the primary basis for the school's
so-called accountability, teachers feel compelled to spend considerable time
preparing children to take the tests. In such settings, the tests become the
school curriculum.
Preparation usually begins many weeks before the actual testing. During this
period, two to three hours a day are often devoted to practice tests and exercises,
all alien to the ongoing instruction and the usual student response patterns.
The teachers readily acknowledge that the questions comprising the practice
exercises, similar to those on the real test, are "trivial and unimportant."
Moreover, the possible responses contain words that children likely have never
seen and certainly don't use. The practice time is wasted time, yet some teachers
believe it is important to waste the time: they are preparing students for
the test.
By the time the three days of real testing is completed--after children have
been admonished to "get a lot of sleep," "remember that this test is very important,"
"take all your books off your desks," "leave your calculators at home," "keep
your eyes down on your own papers" and "ask no one for help"--weeks, sometimes
months, will have passed. Time for real books will have been sacrificed for
time spent reading isolated paragraphs and then answering several multiple-choice
questions. Rather than posing problems for which math might be used, in the
process coming to a natural and deeper understanding of math concepts, time
will have been spent on reviewing skills such as addition, subtraction, multiplication,
fractions, division--all in isolation. Little time will have been given to science
and social studies, other than the concentration on factual information that
isn't particularly useful or generative of on-going interest. Time is a valuable
commodity; it should not be wasted in this manner.
When it's over, the frustration of teachers in these schools will be high. They
will feel that their own intentions have been undermined. They will not have
had an opportunity to look carefully at the tests to see what individual children
did with various questions, to inquire why children selected particular responses
as a way of getting closer to their logic, to get some sense of patterns in
various sections of the test, or to determine how closely any of the
questions got to their purposes. In the world of standardized testing, such
issues are not viewed as particularly important. And the scores, when they come
back several months later, will be of little use. Yet, because of their seeming
authenticity, the scores will stand for how well each class and the school as
a whole performed for the year. The scores may also affect the opportunities
afforded individual students. This is all travesty! The substance and integrity
of education are missing.
An educator's principal purpose is to enhance the growth of every child.
When children are labeled "unready" or "slow learners" because of standardized
test results, their educational opportunities generally become narrow, uninteresting
and unchallenging. One-dimensional tasks such as those found in skill sheets,
workbooks and drills figure prominently in their education. Who are the ones
who tend most often to be labeled? A high proportion of children from lower
socioeconomic populations, including large numbers of minorities, are represented
in special education and lower-level tracks.5
This ought to give us serious pause! Our commitment to democratic practice and
equality of educational opportunity forces us to speak out strongly against
any process that consistently produces such results.
Reasons for caution in the use of tests include the possible loss of children's
self-esteem, the distortion of curriculum, teaching and learning, and the lowering
of expectations. Other concerns relate to the tests themselves. For example,
tests used in grades 1 and 2 are different from those used in grades 3-6. The
early tests are picture and vocabulary dependent, while the later ones place
greater stress on content. Consequently, high scores in early testing may not
carry over to later testing. Because tests include diverse subject areas, they
may or may not relate directly to what children have been taught or evoke from
particular children any intrinsic interest. In addition, the multiple-choice
format of standardized tests confuses many children who are not accustomed to
sharing their understandings in that manner. More-over, for a host of reasons
having little to do with their reading ability, children who read very well
may select "wrong" answers from among the limited choices available.
Peculiarities of testing abound. Children who have been routinely encouraged
to be cooperative learners are forbidden to talk while testing. Children who
have been taught to work problems out slowly are told speed is essential. Children
who have come to understand that they must construct answers to problems,
that many answers are possible, are confronted with someone else's answers and
told that only one answer is possible. The message is clear: "Don't take your
time--guess if necessary and forget what you have been learning day in and day
out." Such conditions cause many children undue anxiety, even if the ultimate
consequences of test-taking are not devastating. And we have only touched the
surface.
The Value of Responsive
Educational Environments
Educators of young children have long believed that children
learn in many different ways, demonstrating in the process that they have multiple
patterns of growth and achievement. This belief has given direction to programs
with diversified aims and goals. In these programs, children are respected,
regardless of racial background or socioeconomic class. Their interests become
basic starting points for learning. Such developmental programs tend to support
more formal instruction in reading, for example, only when children are ready
and not simply because they are 6 years of age.
Because teachers in such settings commit themselves to increasing successful
learning experiences and improving children's self esteem, many learning options
are made available. The clock then tends not to determine to such a large degree
when children begin and end teaming activities. Peer interaction and communication
are encouraged. Creative and expressive forms of communication that develop
feeling--the most personal of human possessions--become integral, rather than peripheral,
to a child's life in these classrooms. (Too often a teacher does little with
the creative and expressive arts because they don't relate particularly well
to the normative testing programs. They are not basic enough!)
Static expectations for children, rooted in an array of basal materials and
common curricula, do not reflect the diversity that actually exists in primary
schools. Yet, standardized tests are rooted in standard curricular materials
(basal textbooks, syllabuses, state or provincial guidelines) that have predetermined
expectations all children must meet. To actually develop a responsive, developmental
classroom environment is to risk lower scores on standardized tests. Teachers
and children do not need this kind of external pressure.
In 1976, ACEI called for a moratorium on all standardized testing in the early years of schooling. The Association also affirmed the importance of evaluation in classrooms and schools, acknowledging that careful evaluation was the key "to the qualitative improvement of educational practice and the teaming of children." ACEI's position remains similar now with one exception. We now believe firmly that no standardized testing should occur in the pre-school and K-2 years. Further, we question seriously the need for testing every child in the remainder of the elementary years.
ACEI is not alone in such a position. The National Commission
on Testing and Public Policy recently reached that conclusion after five years
of studying standardized testing intensively. The National Association for the
Education of Young Children, in one of its policy imperatives of 1988, called
for an end to K-2 testing. In its new primary program released in 1990, the
Province of British Columbia outlined steps intended to eliminate all deficit-based
assessment and evaluation in the early grades. And the state of North Carolina,
after a decade of heavy high-stakes testing of K-2 children, ended all such
testing in 1988. What we are seeing is a growing understanding that teaching
to tests increasingly has become the curriculum in many schools especially in
the early years when test are most affected by such a direction.
The fact that test scores are increasing is no longer causing much celebration.
We have evidence that the curriculum is becoming a matter of worksheets, workbooks
and skills; higher order thinking skills and deeper levels of understanding
are being sacrificed, reading for meaning is being set aside; the arts are becoming
nonexistent; exploration of real materials, the science and mathematics of the
world, isn't being "risked"; and time for play, what most teachers and parents
understand to be the work of children, is being seen as a frill.
Such understandings have brought a growing realization that the curriculum in
many schools is not powerful enough, that it doesn't lead to large understandings
or commitments to extened learning. And increasingly, teachers are becoming
more vocal about wanting opportunities to create a more thoughtful and expansive
curriculum. In the process, they are making clear that they know how to address
accountability issues through good documentation of children's actual work rooted
in a solid and generative curriculum. This bodes well for change.
Evaluation
Consonant with Purpose
The need is to engage in assessment that is not only related
to the best practice, but also rooted directly in the instructional process
itself. While many possible entry points to such assessment exist, we share
first the way a group of elementary teachers in New York City responded to a
new city-wide science test for use in grades 3 and 5. We believe the example
is instructive for other assessment areas, as well as other grade levels.
This group of teachers argued that the test (not science assessment itself)
was inappropriate for use in their classrooms. It 1) covered too much ground
too superficially and didn't get close enough to what children actually knew
and understood; 2) didn't honor their slower, more intense, meaning-making,
hands-on, observational and experience-oriented approaches to science; and 3)
was a distraction at a time when serious science inquiry was becoming well-established.
Working with a research psychologist at the Educational Testing Service, the
teachers developed a science assessment that used the district's objectives
and the questions asked on the city-wide test, but made the basic questions
open-ended. They wanted to demonstrate the larger possibilities in an open-ended,
less restricted assessment format.
In the document they prepared as part of their oppositional process, the teachers
wrote that:
... the multiple-choice format... allows no room for pupils to construct or
generate answers based upon their knowledge and thought.... Further, tests which
consist solely of questions for which there is only one correct response constitute
an inappropriate assess mentor model for science education. We are concerned
that testing in this form will undercut science as a process, the investigative,
experimental components of our science program which entail observation, experiment,
and field work. (Chittenden, 1986)
Their critique is worthy of more attention.
The city-wide test asked, "Which of the following trees can be found growing
along the streets of our city? a) Redwood, b) Palm, c) Rubber, d) Maple." While
not suggesting that the question was important, the teachers asked in their
alternative test, "Name some trees that grow along the streets of New York"
The 30 3rd-grade children who took the alternative test named 73 different species
of trees (including the "Central Park tree"). For instructional purposes, teachers
gained entry points they hadn't thought about.
Rather than ask, as the city-wide test did, "Which of the following planets
is the largest? a) Venus, b) Mars, c) Pluto, d) Jupiter," the test prepared
by the teachers asked students to draw a picture of the solar system. The drawings
were enormously revealing. The teachers and the ETS researcher didn't argue
that the questions they asked were wonderful, fully generative, connected to
many of the issues they believed were critical. They did conclude, however,
that their open-ended process provided information more useful to their ongoing
instruction and got closer to children's understandings than the multiple-choice,
city-wide test.
Assessment for purposes that go beyond the school--and that
is what most current accountability efforts are about--need not, of course,
have an individual, every-student basis. More open-ended, perforrnance-oriented
processes that typically take more time and demand more materials, for example,
would likely be seen as more feasible if sampling were to be used. Sampling
could also involve teachers in schools much more directly, making assessment
more than a process "owned and operated" by some distant bureaucracy.
Centrality of
the Teacher in Classroom-Based Assessment
Work in the area of writing represents the most serious break
yet in the power of standardized testing. Those concerned about writing in the
schools argue convincingly that writing cannot be assessed validly outside the
instructional process itself and that writing to a real audience is central.
Further, they assert that writing at its best is situated--in this sense, not
easily standardized in current psychometric or technological terms.
Understanding children's writing cannot begin with one task, a single piece
of work, or with writing that has not been completed within the norms of powerful
classroom practice. Such writing isn't likely to bring forth students' best
and most committed efforts. That understanding alone has changed the assessment
landscape enormously. Teachers who encourage active writing programs make clear
that serious writing takes thought and time, is close to personal experience
or interest and connects to an individuals way of interpreting the world. Children
write what they know and feel about their world--uderstandings
that extend to all curricular areas including social studies, science, math
and the arts.
Teachers recognize that children have much more to talk and write about
in settings where the ongoing school experience of the students is rich: teachers
read a great deal to children, giving emphasis to authorship and personal style;
books are plentiful; active learning is promoted; the world is permitted to
intrude, to blow through the classroom. In this sense, writing is not something
apart; rather, it has a context and that context is important to understanding the
writing. Most writing assessment efforts that have existed, those of NAEP (National
Assessment of Educational Progress), provide little knowledge of contextual
issues.
Experience has also shown that the best person to judge students' writing, who
can monitor their progress as writers, is the teacher closest to them. That
shouldn't surprise anyone. The classroom teacher knows, for example, the questions
a particular child has been raising about various aspects of classroom learning.
When reading a piece of writing, the teacher can refer to previous writing efforts,
a book the child is currently reading, genres of authors the child is most inclined
toward at the moment, a painting completed, a trip recently taken, the new baby
sister, the spring flooding across the community's many glacial lakebeds, the
special meadow colors, the classroom's human mosaic. Thoughtfully responding
to the surrounding context, which is never really separate from the text, the
teacher can better interpret the writing.
It is that teacher, deeply involved with the child as writer, who knows the
next question to raise, when to push and when not to, who can judge the meaning
and quality of a piece of that child's writing. This outlook governs our perspective
about evaluation issues as a whole.
The foregoing becomes clear when reading children's work in the various publications
of the Teachers and Writers Collaborative (see Landnun, 1971; Murphy, 1974)
or the accounts prepared by Don Graves and Lucy Calkins. And we have seen similar creative
and energetic writing in large numbers of elementary schools where active writing
programs have been established.
As we read the wonderful writing, knowing that each of the pieces was completed
over time-not at one sitting, not sitting, not without conversation, not without
several tries, not without some peer response and early teacher response--we
wonder what would have been produced had these writers been forced to write
on April 1 at 10:00 a.m., they had 30 minutes and the readers would be persons
far away. Actually, we don't wonder too much. We have seen the writing and it
isn't the same. And we talk with enough teachers to know that they don't believe
what students produce on those days represents anything approximating their
best work. Many of the students, often the most skilled writers, leave much
of the writing assessment page blank.
Teachers who honor children's work as the genuine product of thought, capable
of evoking thought can certainly describe their students' writing. They are authentic
readers. And they have been convincing in their view that any talk of assessment
is doomed intellectually if it doesn't acknowledge the importance of being close
to the student writer and the surrounding context.
So where does this lead us regarding an assesment program? Having acknowledged
the centrality of the classroom setting, the classroom teacher and work over
time, we are convinced the principal direction is rooted in carefully organized
and considered classroom documentation. Classroom teachers can, for example,
systematically preserve copies of drafts of students' writings as well as finished
pieces. Two to three pieces a month would provide a reasonable collection. Reviewing
them periodically can inform a teacher's ongoing efforts to assist particular
students, an important purpose of documentation. At year's end, the accumulation--organized
chronologically--can be subjected to a careful review, with some of the following
questions serving as a framework: Over time what are the salient features, dominant
motifs? How much invention? What about complexity? Choice of topics? Discourse
frameworks? Connections to ongoing academic and social strengths? Diversity
of word use? Voice? Use of conventions?
This review often provides a perspective missed in the course of addressing
work that stands alone. Such a portfolio is almost always enormously revealing
to parents, bringing the kind of overview, or large picture, that parents often
miss as they interact with their children about the school experience.
Classroom-based review addresses concerns about the ongoing support of individual
students and informs further instructional practice. It also serves as a way
for a teacher to describe children's growth as writers over the course of a
year, as well as inform their subsequent teacher more fully. In addition, students
learn to bring careful self-examination and more solid interpretation to their
own efforts as writers. Such an opportunity should not be missed.
For purposes of a larger school-wide review, randomly selected students from
each classroom in a school might be asked to choose five or six pieces of their
writing to be read by groups of teachers in the school as a whole-providing
the readers with a context of the individual works. At the level of the school,
using such samples as a base-knowing that they were written within the instructional
program itself and not apart from it in a forced, unsituated exercise-provides
readers with more confidence about describing, for example, the writing of 4th-graders
in a particular school. And they should be able to do it with good authority.
Further, as "a community of readers of writing," the teachers involved in this
school-wide review can actually enlarge their understandings of writing, in
the process becoming better teachers and facilitators of writing. If the evaluation
effort doesn't produce these kinds of results, it is quite clearly a failed
and faulty exercise.
A School
District's Assessment Effort
Having argued that the best evaluation is classroom and school-based,
it is still possible to extend the logic of making use of multiple samples,
embedded in best practice, to a school system where a community of readers linked
to shared beliefs can be formed. With each step from the classroom, however,
confidence levels must, of necessity, begin to decline.
With that, we want to share an assessment effort conducted by a school district
that is moving in a more positive direction. The example comes from Grand Forks,
North Dakota, where the Superintendent of Schools agreed to experiment with
a classroom-based, instructionally oriented writing assessment planned by teachers.
Grade 6 was selected as a focus. The Grand Forks teachers began by examining
as a group samples of children's writing. Further, they read collectively some
of the writings of Don Graves, Lucy Calkins and Jerome Harste, among others.
They learned about the diverse ways 6th-grade teachers throughout the system
worked with writing, and they also became deeply involved in a writing-workshop
approach to the teaching of writing. Documenting their own practices, reflecting
together on their experiences, and reading work produced by their children in
the workshop setting, the teachers acquired a healthy outlook on the district-wide
mandate for assessment that would get them close to children's writing and inform
their ongoing practice.
For the assessment studies, they decided to ask students to complete a personal
narrative of their own choosing, within the framework of the ongoing instruction.
In some classrooms, a process approach provided the structure; in other settings,
different processes prevailed. Six hundred fifty narratives were produced. A
holistic process, using as criteria clear message, logical sequence, voice and
mechanics, enabled the teachers to respond descriptively and quantitatively
to the question of "How well do 6th-graders in the Grand Forks public schools
write?"
While not perfect, the process was embedded within classroom practice. Further,
it enlarged school system-wide discussion about writing and writing process,
provided teachers with more experience as readers of written discourse and broadened
insights into the teaching of writing.
A community of writing teachers, persons able to link the teaching of writing
to the classroom context as well as understand ways to make connections between
writing and evaluation is being formed. This is empowerment of a high order.
In addition, because of the way the evaluation process was organized, large
numbeers of students have learned to evaluate their own writing. This contributes
more to writing in the schools than any process that stands apart from teachers
and their ongoing instructional efforts. Although the focus has been on the
connections to other subject fields should be clear.
The foregoing hardly covers all the possibilities in what is increasingly being
defined as "authentic assessment" or "Performance-based assessment." Basic to
such efforts is the close tie of assessment to the instructional process itself.
The interest is not in what students can give back in terms of information but what
they can do, the relationships they are able to make, the understandings they
are able to develop and extend to other learnings.
In addition to the benefits discussed and implied from the systematic record-keeping,
our experience reveals that teachers who document children's learning through
carefully organized records tend also to be more knowledgeable about children
and learning. They become the "students of teaching" that schools need and parents
desire. Teachers able to describe children's learning in great detail are teachers
who are trusted and capable of helping reestablish parental confidence in schools.
This Association for Childhood Education International position paper decries the continuing potency of standardized testing in primary programs. Stressing the inappropriateness of standardized testing, it argues that teachers and parents should oppose using test results to make any important judgment about a child. And it sets forth unequivocally the belief that all testing of young children in preschool and grades K-2 and the practice of testing every child in the later elementary years should cease. To continue such testing in the face of so much evidence of its deleterious effects, its opposition to most of what we know about the developmental needs of young children, is the height of irresponsibility. We know, for example, that testing:
In emphasizing the critical need to seek more active directions for staying close to children's growth, this position paper also presents teachers and schools with a means of entering the assessment arena systematically and beneficially. The classroom setting and the teacher are acknowledged as central to an assessment program that, over time, is rooted in carefully and considered documentation. This kind of systematic, classroom-based review can inform not only a teacher's efforts to help individual children, but also ongoing instructional practice. Additionally, children learn to subject their work to careful self-examination and interaction. Most important, authentic, performance-based assessment guarantees a greater understanding of the growth of individual children, which should reduce the need for any of the testing programs that currently exist.
Association for Childhood Education International/National Association of Elementary School Principals. Perrone, V. (1976). Position Paper. On standardized testing and evaluation. Childhood Education, 53, 9-16.
Chittenden,E. (1986). Alternatives to the New York City science test (p.4).Princeton, NJ: Educational Testing Service.Haney, W., & Madaus, G. (1989). Searching for alternatives to standardized tests: Whys, whats and whithers. PhiDelta Kappan, 70, 383-387
Landrum R. (Ed.). (1971). A day dream I had last night. New York.- Teachers & Writers Collaborative.
Meisels, S. J. (1989). High stakes testing in kindergarten. Educational Leadership, 46(7), 16-22.
Murphy, L (Ed.). (1974). Imaginary worlds: Notes for a new curriculum. NewYork: Teachers & Writers Collaborative.
National Association for the Education of Young Children. (1988). Testing of young children: Concerns and cautions. Washington, DC: Author.
National Commission on Excellence in Education. (1983). A nation at risk: The imperative for educational reform. Washington, DC: Government Printing Office.
National Commission on Testing and Public Policy. (1990). From gatekeepers to gateway: Transforming testing in America. Chestnut Hill, MA: Boston College.
Shepard, L. (1987, January). The assessment of readiness for school: Psychometric and other considerations. Presentaiton at the Natinal Center for Educational Statistics, Washington, DC.
White, S. (1975). Social Implications of IQ. National Elementary Principal, 54(4), 10.
1 Alfred Binet began his work on "mental ability" tests in 1904. This work, which contributed to the development of achievement tests, resulted in the Stanford-Binet test (and the "IQ score") in 1916. While IQ testing became popular in many schools, particularly as a means of separating children for various special education programs, it has been reduced substantially in the past two decades because of court-imposed limitations. We begin this position paper with the premise that all IQ testing should end. Such testing serves no educational purpose. When we speak here of standardized testing, we are referring to the array or readiness and achievement tests that continue to be so dominating in the schools.
2 Students in New York state, who pursued regents' diplomas, were among those who would have taken more. There were also some school districts--generally in urban communities--that made more frequent use of tests for purposes of promotion and graduation. These were, however, exceptions.
3 The difference in numbers is matched by the magnitude of their meaning. The tests were used for many more purposes. They determined a good deal about the educational experiences made available to children and whether they would be promoted or retained; they also became a basis for evaluating classrooms, schools and teachers.
4 A large number of publications have provided thoughtful critiques of sample test items from a variety of popularly used standarized tests. See, among others: D. Meier, Reading Failure and the Tests (New York: Workshop Center for Open Education, 1973); D. Meier, H. Mack & A. Cook, Reading Tests: Do They Hurt Your Child? (New York: Community Resources Institute, 1973); B. Hoffman, The Tyranny of Testing (New York: Collier Books, 1964); National Elementary Principal (Mary./Apr. 1975 and Aug. 1975); Standardized Tests and Our Children (Cambridge: FairTest, 1990).
5 J. Mercer, Labelling the Mentally Retarded (Berkeley: University of California Press, 1972). See also P. Olson, "Power and the National Assessment of Education Progress," National Elementary Principal (July/Aug. 1975) for some of the cultural problems with tests as well as a review of some of the important related court cases. M. E. Leary, "Children Who Are Tested in an Alien Language: Mentally Retarded?" New Republic (May 23, 1970), pp. 17-19, discusses the Diana et al. vs California State Board of Education case regarding the placement of Mexican-American and Black children in special education classes on the basis of test scores. Hobson vs Hanson, Civil Action No. 82-66, U.S. District Court for Washington, DC, 1968, provides an excellent review of standardized tests and teaching. For readers interested in an excellent review of the serious problems of standardized testing and minorities, see R. Williams, Chairperson, "Position Paper of the American Personnel and Guidance Association Committee on Standardized Testing and Evaluation of Potential Among Minority Group Members," 1975; N. Medina & M. Neill, Fallout from the Testing Explosion: How 100 Million Standardized Exams Undermine Equity and Excellence in America's Public Schools (Cambridge: FairTest, 1988); and A. Hilliard, "Testing African-American Students," Negro Educational Review, 1987.
Copyright 1991 by the Association for Childhood Education International.
ISBN 0-97173-1233-1
This page is copyright 
2001 by the Association for Childhood Education International. Please send any
comments to Marilyn Gardner at aceimemb@aol.com.
