ERIC ED452219: High-Stakes Testing: Too Much? Too Soon?

23 Pages·2000·0.74 MB·English

by  ERIC

DOCUMENT RESUME
TM 032 504
ED 452 219
Walker, Sherry Freeland, Ed.
AUTHOR
High-Stakes Testing: Too Much? Too Soon?
TITLE
Education Commission of the States, Denver, CO.
INSTITUTION
ISSN-0736-7511
ISSN
2000-00-00
PUB DATE
22p.; Theme Issue. Published three times a year. ECS
NOTE
Distribution Center, Education Commission of the States, 707
17th Street, Suite 2700, Denver, CO 80202-3427 (Annual
subscription, $20). Tel: [phone]; For full text:
http;//www.ecs.org.
Serials (022)
Collected Works
PUB TYPE
State Education Leader; v18 n1 Win 2000
JOURNAL CIT
MF01/PC01 Plus Postage.
EDRS PRICE
*Accountability; *Educational Testing; Elementary Secondary
DESCRIPTORS
Education; *High Stakes Tests; Scores; *State Programs; Test
Results; *Test Use; Testing Programs
ABSTRACT
This theme issue focuses on the use and consequences of high
stakes tests. The lead article, "High-Stakes Testing: Too Much? Too Soon?" by
Sherry Freeland Walker, introduces the topic and related issues, outlining
the pros and cons of high stakes testing by the states. The problem, some
experts say, is that states have tried to do too much too soon without the
proper preparation and support for everyone involved. "The History of
Testing," by Sherry Freeland Walker, traces the growth of high stakes testing
through the last half century and in the present context of the standards
movement. "High-Stakes Assessments Bring Out the Critics," by Jennifer
Dounay, discusses a number of criticisms of high stakes testing and some
responses from the public. "Why Is 'Teaching the Test' a Bad Thing?" by
Lorrie Shepard, explores the issues of test score inflation, curriculum
distortion, and safeguards against political pressures in testing. "How
States Are Responding to Low-Performing Schools," by Katy Anthes, Susie
Saavedra, Judie Mathers, and Jane Armstrong, describes the interventions
states with high stakes accountability systems are using with low performing
schools. The effects of high stakes tests on teacher education are outlined
in "High-Stakes Testing Pressures Teacher Education" by Michael Allen. Other
"Maryland Moves toward Intervention" (Mary
articles in this issue are:
(1)
"Why
"Texas Test Withstanding Court Scrutiny" (Jill Weitz);
(3)
Fulton);
(2)
"Poor Test
Do We Need High-Stakes Assessments?" (Michael Sentance);
(4)
"Performance Management, Not Just
Results Lead to Math Consortium"; and (5)
Accountability" (Peter Robertson).
(SLD)
Reproductions supplied by EDRS are the best that can be made
from the original document.
State Education
teadeir/
Cr
N
N
tr)
Winter 2000
Number 1
Volume 18
High-Stakes Testing:
Too Much?
Too Soon?
Zr
0
LC)
EDUCATION
U.S. DEPARTMENT OF
Improvement
Office of Educational Research and
INFORMATION
0
EDUCATIONAL RESOURCES
PERMISSION TO REPRODUCE AND
CENTER (ERIC)
DISSEMINATE THIS MATERIAL HAS
Et/this document has been reproduced as
BEEN GRANTED BY
organization
received from the person or
originating it.
5
f!_v_Ja-tiL:e
Minor changes have been made to
improve reproduction quality.
in this
Points of view or opinions stated
document do not necessarily represent
TO THE EDUCATIONAL RESOURCES
INFORMATION CENTER (ERIC)
official OERI position or policy.
1
Education Commission of the States
2
EDUCATION
E EDUCATION
ST
COMMISSION
OF THE STATES
CS
WI&
WINTER 2000
NUMBER
VOLUME 18
HIGH
The mission of the
Education Commission of
TESTING:
the States (ECS) is to help
state leaders identify,
TOO MUCH?
develop and implement
public policy for education
TOO SOON?
that addresses current and
future needs of a learning
society.
by Sherry Freeland
Walker
schools are doing. Policymakers see them as a
ick up the newspaper and you're
way to raise standards and achievement and
likely to see an article about state
hold students and educators accountable. But as
assessments that carry big conse-
support grows on one hand, so does opposition
quences. Parents and students in one state
In this issue
on the other. Are high-stakes tests worthwhile?
protest the test, while policymakers in another
laud or bemoan the outcomes of their state's
Or is the controversy around them likely to
High-stakes assessments
latest assessment. Across the nation, state and derail the standards movement?
bring out the critics. Page 4
district leaders are putting more emphasis on
Is "teaching the test" a bad
Lagging skills
testing and using test results to make more
thing? Pages 7-8
No one disputes that too many American
decisions about students and schools. Will this
How states are responding
students are not gaining the knowledge and
student be promoted to the next grade? Will
to low-performing schools.
skills they need to succeed in college and the
that one graduate from high school? Should
Pages 13-14
workforce. Only about one-third are proficient
this school be reconstituted?
in reading and fewer still in math, according to
High-stakes testing pres-
Using assessment tests for such "high-
sures teacher education.
National Assessment of Educational Progress
stakes" purposes is gaining public support as a
Page 16
way to determine how good a job public
Continued on next page
.
0
0
Continued from previous page
with the new standards," said Richard F.
Elmore, Harvard School of Education profes-
scores. Even the most advanced U.S. students
e
sor, at a recent Washington, D.C., conference.
lag behind their peers in other countries on the
While some policymakers are rethinking
Third International Mathematics and Science
assessments, others say the low scores are just
Study. And public opinion shows Americans
'
-
an indication of the work that needs to be done.
.
increasingly critical of public schools overall.
SO
"When we fired this missile," Todd Bankofier
-.0-
.
es
.
-
Almost every state has set content stan-
-
of the Arizona Board of Education said, "we
dards for what students should know and mea-
knew we had to guide it. It's going to take
suring whether students are meeting those
some left turns and some right turns, but it
-
standards is a natural outgrowth. To date, 17
would be wrong to turn it completely back."
states, the District of Columbia and Puerto
.
.
"Doing away with the tests or the conse-
Rico have policies that base promotion or
quences is the easy way out," Robert Schwartz
.
retention on a student's score on a state and/or
and Matthew Gandal wrote in the January 19,
'11
district assessment (see page 11). Twenty-seven
.
2000, issue of Education Week. "It allows us to
-e.
states have high school exit exams (though not
avoid the hard work of improving instruction
all are tied to graduation or test beyond 9th-
.
o-
and restructuring the use of time and resources
.
.
'
grade skills). Polls consistently show public
-
so that all students are given the time and sup-
support for standardized testing.
-
port needed to meet standards."
Pros and cons
100
Confronting the dilemma
Proponents of high-stakes testing argue
-
Jay P. Heubert and Robert M. Hauser of
that it leads to achievement and other gains:
the National Research Council's Committee on
Students know what is expected and that the
Appropriate Test Use recommend in High-
test really counts, so they work harder.
Stakes Testing for Tracking, Promotion and
-
Graduation that policymakers keep the follow-
Schools identify and can address student
0..00
-
.0
0
ing principles of appropriate test use in mind:
. .
weaknesses early.
-
Use the right test. Tests are valid only when
Similarly, schools discover areas of overall
.
used for the specific purpose for which they
weakness, prompting them to refocus
000
-
were designed.
resources where they are most needed.
Remember tests are not perfect. Questions
.
Education across the state is more consistent,
are but a sample of possible questions that
I
11.
eliminating situations where schools in some
could be asked in a given area.
districts are superior to others.
.
Don't use a test as the sole determinant of
OS
-
The public sees gains from year to year and
.
-
. .
.
a major decision. Promotion and graduation
regains confidence in public schools.
decisions should be based on many factors.
Critics say the tests sometimes are too
. - .
.
0
II
-
hard, lead teachers to teach to the test, take
Don't justify bad decisions with a test
.11
00
0 0
time away from instruction, and are expensive.
score or any other kind of information.
Teachers say they're unprepared to teach to the
Tests will not lead to better outcomes if dis-
-
standards, and students claim they're being
tricts and schools lack the services to help
tested unfairly, on content they haven't yet had.
students who don't come up to standard.
.
I.
Some parents and students are calling for an
The answer to who's right
the critics or
end to high-stakes testing, and some policy-
the supporters
seems to be both. If the right
makers are reexamining plans to tie tests to key
test is used in the right way, in conjunction
decisions such as graduation or to make high-
with other measurements, it can be an effective
stakes tests the central part of an accountability
way to assess student learning. Without atten-
system (see pages 4-6).
tion to factors such as discrimination, curricu-
lum and accuracy, however, it can be
.
Too much, too soon?
detrimental to both students and schools alike.
00
-
The problem, some experts say, is that
This issue of State Education Leader looks
states have tried to do much too soon without
at the controversy around high-stakes testing.
the proper preparation and support for everyone
Freeland Walker is ECS publications director
involved.
Education Commission of the States
"Teachers and principals simply do not
STATE EDUCATION LEADER
4 0000
know how to do what they are expected to do
VOL. 18
WINTER 2000
NO. 1
2
1=1
D
j
H
by Sherry Freeland
I
Walker
I
[normal test-
f
ing has become the
kudzu of modern
American society, a
requirements. Most took the form of multiple-
esting is big news these days, and
healthy vigorous
71
choice items that students either passed or
the stakes are getting higher and
grower penetrating all
failed and primarily pinpointed gains at the low
higher. As business and the public put
end of the spectrum. The tests did little if any-
more pressure on public schools and students to
available space. 00
thing to measure how much students were
achieve at higher levels, the use of testing is
Gary Natriello, Columbia
learning or how advanced their skills were.
expanding rapidly.
University's Teachers
Throughout the last century, the uses of
College, and Aaron M.
Standards movement
standardized testing, and the reasons for using
Pallas, Michigan State
Growing criticism of public schools led
it, have grown considerably. As Gary Natriello
University
policymakers and educators to turn toward test-
of Columbia University's Teachers College and
ing to measure higher skills and to gain support
Aaron M. Pallas of Michigan State University,
for raising standards. The late 1980s saw the
say, "formal testing has become the kudzu of
rise of assessment tied to accountability for stu-
modern American society, a healthy vigorous
dent and school performance, although states
grower penetrating all available space."
were relying heavily on nationally published
Half a century
standardized tests, rather than assessments
Standardized testing has been a feature of geared to individual state standards.
public schools for half a century, initially serv-
The early days of test results tied to
ing largely to compare schools and students
accountability, however, were criticized as
against a standard set by testing companies.
showing an inflated pattern of scores. Because
the tests suddenly had high stakes, teachers
Another use was to "sort" students, such as
were teaching to the test, critics said. They
identifying those considered fit for higher
based their reasoning largely on the fact that
education versus those who would be better
gains on the National Assessment of
suited to vocational school.
Educational Progress tests were not as high as
The 1970s saw an eruption of interest in
scores on other assessments.
"minimum competency testing." Then, as now,
While the current wave of education
say Robert Linn and Joan Herman of the
National Center for Research on Evaluation,
reform continues to emphasize accountability, it
is more tied to the setting and implementing of
Standards and Student Testing, "reformers
state standards, both content (what students
sought to improve education by holding educa-
should know) and performance (how well they
tors and students accountable for achieving stan-
are able to do it). States are aligning assess-
dards of performance, using tests for high school
graduation and or grade-to-grade promotion."
ments to their standards and demanding much
By the early 1980s, nearly three-fourths of
more from students than they have previously.
Education Commission of the States
STATE EDUCATION LEADER
the states had minimum competency testing
Freeland Walker is ECS publications director El
VOL. 18
WINTER 2000
NO. 1
3
5
11;r1r%ic04,
of
W.
ASSESSMENTS BRING OUT THE CRITICS
by Jennifer Dounay
or many people, both in and outside
Kaplan, known for its SAT and ACT prepa-
the education policymaking field, the
ration books, has released books to help stu-
concept of assessing students on their
dents and parents of young children prepare for
knowledge and skills seems a perfectly innocu-
standardized tests in Florida, New York, Texas
ous proposition. After all, why shouldn't pupils
and Massachusetts.
WAs the assessment
be held accountable for learning what they have
"Dumbing down" of the curriculum
been taught during a given school year or by a
stakes have increased
Another criticism is that the curriculum
certain milestone in their school careers?
may be "dumbed down" as a result of state-
for both students and
This proposition, however, is not as simple
mandated testing. Some people fear rote mem-
as it may appear. As the assessment stakes have
schools, various 'stress
orization may be stressed rather than
increased for both students and schools, various
problem-solving skills and that teachers will
"stress points" in the system are causing some
points' in the system
focus on subject areas or facts most likely to
students, parents and others to question the
are causing some stu-
appear on assessments, rather than more com-
validity of assessment and accountability
plex skills, such as critical thinking.
systems.
dents, parents and
There also is widespread concern that sub-
Too much pressure
jects not tested (for instance, fine arts or physi-
others to question the
Parents in some states are asserting that
cal education) will be accorded less class time
validity of assessment
some high-stakes tests place undue pressure on
or set aside altogether, as some elementary
young children. Stories of increasing numbers
schools have done with recess, to spend more
and accountability
of children suffering from sleep disorders and
time on academics.
systems.
other stress-related maladies have appeared in
Critics also argue that too much time is
the press in the past few years.
taken away from instruction when students are
Districts across the nation have offered
coached on testing techniques and then spend
Saturday and summer tutorial classes to give
hours taking the tests.
children extra time to work on skills that may
Score discrepancies
be tested. The Hartford, Connecticut, schools
Parents, as well as the general public, also
offered classes during the 1999 spring break to
doubt the integrity of a state assessment when
help 3rd, 5th and 7th graders prepare for the
scores do not match their children's grades or
Connecticut Mastery Test scheduled for the
achievement measured by other tests. Numer-
fall. (To the district's credit, scores did improve
Education Commission of the States
STATE EDUCATION LEADER
ous media articles have profiled students with
significantly.)
VOL. 18
WINTER 2000
NO. 1
6
Individual students likewise will feel the
"A" or "B" averages who attain low scores on
effect of the MCAS. The class of 2003 will be
state assessments or fail to pass high school
the first whose high school graduation will
exit examinations. Parents and students wonder
depend upon students' scoring at the proficient
whether grades are inflated or if the bar on the
or advanced level on all of the 10th-grade tests.
state assessments has been set unreasonably
Like Virginia, scores so far have been low. In
high. Parents in affluent areas of New York
1999, only 34% of students reached those lev-
such as Rye, Great Neck and Mamaroneck
els in English language arts, 24% in mathemat-
were shocked, for example, when, according to
ics, and 24% in science and technology.
a November 1999 New York Times article, one
in five of their children failed the state's new
Minority discrimination
8th-grade math assessments.
Some test critics point out that students
State issues
from predominantly white and middle- to
upper-class districts score the highest on high-
Discrepancies between indicators of stu-
stakes and other assessments. An analysis of
dent achievement have shown up at the state
the 1998 MCAS tests, conducted by the Gaston
and district levels as well. For example,
Institute for Latino Community Development at
Virginia began in 1998 to assess 3rd, 5th and
Students from
the University of Massachusetts-Boston, found
8th graders as well as high schoolers on the
predominantly white
that cities with the highest proportions of
state's Standards of Learning (SOLs) in
Hispanic test takers fared worst on the 10th-
English, history/social sciences, mathematics
and middle- to upper-
grade math assessments, with failure rates
and science. Starting in the 2006-07 academic
class districts score
nearly as high for African-American students.
year, only schools whose pass rates meet or
While the statewide average failure rate for stu-
exceed 70% in the four subject areas will be
the highest on high-
dents of all races on this assessment was 52%,
eligible for accreditation, with the exception of
it was 83% for Hispanic students and 80% for
3rd-grade science and history, whose minimum
stakes and other
African-American students.
pass rate for accreditation will be 50%.
assessments. 00
Testing programs in other states have
Results of the spring 1999 tests reveal
turned up similar gaps in minority achievement,
only 6.5% of Virginia
much work to be done
although Texas' system
the Texas
schools met the pass-rate standard in all four of
Assessment of Academic Skills (TAAS)
the subjects. In Fairfax County, where students
recently survived a legal challenge that claimed
posted an average SAT score of 1095 in 1998
the high school exit exam discriminates against
(versus a national average of 1005) and where
Hispanics and blacks (see page 10 for more).
91% of students continue to postsecondary edu-
While recognizing the differences in passage
cation, only 54% passed the SOLs in 1998.
rates among blacks (60%), Hispanics (64%)
Because of these discrepancies, Virginia
and whites (86%) in the spring 1999 adminis-
has taken measures to evaluate the fairness of
tration, U.S. District Judge Ed Prado wrote:
the SOLs assessments. In February 1999, test-
ing experts from three universities declared the
"The evidence suggests that the State of
SOLs valid and reliable. And a new SOLs Test
Texas was aware of probable disparities
Technical Advisory Committee has been com-
and that it designed the TAAS account-
missioned to report annually on the assess-
ability system to reflect an insistence on
ments' validity and reliability and propose
standards and educational policies that are
suggestions and recommendations for future
uniform from school to school."
changes.
Mistakes and cheating
Massachusetts' assessment results likewise
have raised eyebrows in that state. The
High-visibility examples of security
Massachusetts Comprehensive Assessment
breaches, teacher and administrator cheating,
System (MCAS) tests 4th, 8th and 10th graders
and mistakes made by testing companies also
in English language arts, math and science and
have shaken the public's confidence in assess-
technology. In September 1999, the State Board
ment systems.
of Education voted to rate schools in two-year
Essay questions for Ohio's 4th- and 8th-
cycles based on their students' performance.
grade writing assessments had to be rewritten
Schools that do poorly must submit improve-
after a paper quoted students discussing the
ment goals to the state which, if unmet in two
essay questions before some schools in the
years, will open the schools to state takeover.
Education Commission of the States
state had administered them. Rhode Island
STATE EDUCATION LEADER
VOL. 18
WINTER 2000
Continued on next page
NO. 1
5
r7
I
I
Continued from previous page
Isolated instances of civil disobedience as
well as organized resistance to high-stakes
postponed administering mathematics and
assessments have appeared in several states.
English assessments for 4th, 8th and 10th
Students in some Massachusetts cities sat out
graders last year after widespread security
the spring 1999 administration of the MCAS. A
breaches were discovered.
teacher in Harwich refused to give his students
Test-tampering cases in Houston, Austin
the 8th-grade history test after noticing that
and eight other Texas districts may have been
some questions dealt with the Civil War, which
the impetus for the September 1999 creation of
students had not yet studied. Groups such as
the state's Public Education Integrity Task
the Coalition for Authentic Reform in
Force. In New York City, 52 teachers and
Education and Cambridge Parents Against the
QQ [Policymakers]
administrators were named in a December 1999
MCAS have been established. Parents in sev-
report for helping students improve their test
must remember that,
eral cities, including Boston, "have encouraged
scores by a variety of means.
their children to boycott the test or have taken
Mistakes in scoring also have occurred.
while scores may
them out of the public schools," according to an
Writing assessments for 4th, 7th and 10th
October 31, 1999, Boston Globe article.
reflect improvements
graders in Washington State were rescored by
Likewise, the Christian Science Monitor
hand and subsequently released two months
in schools or the tests
reported that "in certain Detroit suburbs
par-
behind schedule after scoring mistakes were
ticularly Birmingham, Troy and Farmington
themselves, the final
discovered in summer 1999. In September
protesting parents have refused to allow their
1999, testing company CTB/McGraw Hill
goal of states' stan-
children to take the state test. In some towns,
informed officials in Indiana, North Carolina,
fewer than 15% of students participated in state
South Carolina, Wisconsin and New York City
dards and assessment
testing
a number so small as to render any
that their tests may have been scored incor-
results meaningless." The same article notes
systems is not neces-
rectly. Ramifications of the blunder were espe-
that students intentionally have failed tests or
cially strong in New York City, where more
sarily the race for
refused to take them in California, Wisconsin
than 8,600 students were erroneously placed in
and Illinois as well.
ever-higher scores but
summer school as a result of "low" test scores.
What's next?
the race for students'
Backlash
What's a policymaker to do? After all, test-
Such cases of confusion, potential unfair-
solid preparation for
ing experts themselves caution that when
ness and frustration have led to public outcry
higher standards and new assessments are
the workplace or
against tests in some locales and responses
implemented, scores will reflect the greater
from decisionmakers. The results of the math
postsecondary
challenges placed upon students and the teach-
portion of Arizona's new assessment instru-
ers who must prepare them.
education. PP
ment, which members of the class of 2002 must
There are no simple solutions. Policy-
pass to graduate, revealed that 0% of the
makers, however, must be cautious to avoid
44,245 students who took the test exceeded the
alienating their constituencies or dismissing
standard in math and only 11% met the
parents' concerns. Above all, they must remem-
standard.
ber that, while scores may reflect improve-
In response to cries from parents, students
ments in schools or the tests themselves, the
and educators across the state that the test is
final goal of states' standards and assessment
too difficult, the state board agreed to reexam-
systems is not necessarily the race for ever-
ine the scoring levels. Likewise, the Virginia
higher scores but the race for students' solid
state board has indicated it is open to discus-
preparation for the workplace or post-
sion of changing the history portion of the
secondary education.
SOLs, on which significantly fewer students
Dounay is an ECS research associate.EI
attain the proficient level than in other subjects
that the state tests.
Education Commission of the States
STATE EDUCATION LEADER
VOL. 18
WINTER 2000
NO. 1
6
by Lorrie Shepard
[)"i] test-driven
curriculum encourages
teaching of skills in
"generalize" to the intended curriculum con-
ccording to a recent survey reported
isolation, which may
tent. In fact, controlled studies have shown that
by Education Week, testing is the
students may not be able to answer the same
number one accountability tool,
deny students the very
questions if asked even in slightly different
adopted in 48 of 50 states. Test results are
ways.
intended to focus attention on raising student
activities that might
In one classic experimental study, all stu-
achievement. Yet, critics complain that the
have made the prob-
dents in a study were taught to translate from
emphasis on testing leads to problems of
Roman to Arabic numerals. The group tested in
"teaching the test." What is meant by that, and
lems understandable
the same order did well, but when the other
why is it a bad thing?
and useful.
from
group was asked to translate in reverse
Typically, teaching the test means devoting
the drop-off in
Arabic to Roman numerals
extended time to subject areas that are tested,
performance was startling. Students lost from 35
such as reading and math, to the exclusion of
to 50 percentile points, showing they never
other subjects. Test format becomes a template
understood how the-number system really works.
for how tested subjects are taught. Worksheets
and practice assessments mirror the anticipated
Curriculum distortion
accountability tests as much as possible. A
The negative effects of teaching the test on
recent study in Texas, for example, found that
student learning are the flip side of test-score
teachers in urban schools were required to use
inflation. In a nationwide survey for the
test-prep materials from September through
National Science Foundation, the majority of
March, when the Texas Assessment of
teachers acknowledged shifting instructional
Academic Skills test was given.
emphasis from nontested to tested topics and,
Test-score inflation
at the same time, reported negative impacts of
mandated testing on curriculum and learning.
When tests are developed initially, they are
Although critics originally feared that testing
designed to reflect curriculum frameworks or
would take instructional time away from
content standards. Particular test questions are
"frills," such as art and citizenship, research
intended only to be samples of the full curricu-
shows that untested subjects such as social
lum. How students do on the test is supposed to
studies and science have been relegated to
show how well they have mastered that curricu-
Education Commission of the States
Friday afternoons or even eliminated.
lum. But if students practice only questions that
STATE EDUCATION LEADER
imitate the test, test performance may no longer
VOL. 18
WINTER 2000
NO. 1
Continued on next page
7
Safeguards
Continued from previous page
Developing new forms of the test each year
Even in tested subjects, instruction is
is one limited safeguard that prevents practicing
focused only on skills covered by the test. In a
on specific test items. In addition, the move-
study by Mary Lee Smith, elementary teachers
ment toward performance assessments is aimed
had given up reading real books, writing and
at correcting the distorting effects of multiple-
long-term projects and were focusing on word
choice test formats. The more that extended
QQ The movement
recognition, recognition of spelling errors,
tasks on tests reflect the actual kinds of written
language usage, punctuation and arithmetic
toward performance
expression, problem solving and applications of
operations.
knowledge that are intended in the curriculum,
assessments is aimed
Unfortunately, a test-driven curriculum
the less likely it is that teaching to the test will
encourages teaching of skills in isolation,
distort either learning or test-score gains.
at correcting the
which may deny students the very activities
The content of a test alone, however, can-
that might have made the problems understand-
distorting effects of
not be sufficient safeguard against political
able and useful. Practicing only test-like for-
pressures. Ultimately, the best remedies are
multiple-choice test
mats also elicits different cognitive processes
(1) to put less weight on a single indicator
than working with more extended and challeng-
formats.
when judging the quality of schools and (2)
ing curricular materials. For example, students
acknowledge accurately that the responsibility
are asked to read artificially short passages and
for student achievement is shared among stu-
search for answers to formulaic questions. They
dents, parents, teachers, school administrators,
practice finding mistakes rather than doing sig-
community leaders and policymakers.
nificant writing on their own, and they learn to
Shepard is professor of education, University of
guess by eliminating wrong answers.
Colorado at Boulder.
xccal"Er3
ST ,yE EDUC
ION
r
Published three times a year. Annual
subscriptions are $20. Subscribe now
Name
Address
City, State, ZIP
Telephone (
FAX and e-mail
Check enclosed Payable to Education Commission of the States
Payment Method:
MasterCard
Credit card: American Express
Visa
Credit card account #
Expiration date
Cardholder's signature
Mail check/order form to: ECS Distribution Center, Education Commission of the States,
707 17th Street, Suite 2700, Denver, CO 80202-3427
Education Commission of the States
FAX credit card orders to: 303-296-8332, e-mail [email protected], or call 303-299-3692.
STATE EDUCATION LEADER
VOL. 18
WINTER 2000
NO. 1
L
8
I