State
Curriculum Standards and Policy on Instructional Materials
Accountability Systems and Assessments
One response to evidence of disappointing achievement by U.S. students
has been the movementaccelerating since the early 1990sto define
and implement higher standards for student learning. The National
Council of Teachers of Mathematics (NCTM) issued and revised mathematics
standards in 1989 and 2000 (NCTM 1989
and 2000), the American Association
for the Advancement of Science (AAAS) published Benchmarks for
Science Literacy (AAAS 1993),
and the National Research Council (NRC) issued the National Science
Education Standards (NSES) (NRC 1996).
These standards documents recommend that schools cover fewer topics
in greater depth, use inquiry-based methods, and focus on understanding
of concepts in addition to basic skills. During the 1990s states
used such guiding documents to develop their own standards and curriculum
frameworks, to create new assessment instruments, and to reform
teacher education.
This section reports on state curriculum standards and testing
and accountability policies.
State Curriculum Standards and Policy on Instructional
Materials
The NCLB Act requires states to immediately set standards in mathematics
and reading/language arts, and to set standards in science by academic
year 2005. In 2002, 49 states and the District of Columbia had content
standards for mathematics (as well as for English/language arts),
and 47 states had standards for science (CCSSO
2002). Many states have recently revised or are in the process
of revising their standards, curriculum frameworks, and instructional
materials. By 2002, exactly half the states had set a regular timeline
for reviewing and modifying their standards (Editorial
Projects in Education 2003).
Standards documents vary greatly in detail, degree of focus, specificity,
clarity, and level of rigor. Evaluations of standards have used
different criteria and methods (Achieve,
Inc. 2002b; AFT n.d.; and Finn and
Petrilli 2000). States also prescribe instructional materials
to varying degrees. In spring 2002, 21 states had no policy prescribing
textbooks and another 4 had a policy of local choice. Of states
that restricted textbook choice, eight produced a list of approved
books and materials for local choice, five selected textbooks, and
nine combined selection and recommendation (CCSSO
2002).
Accountability Systems and Assessments
Assessment Programs in Mathematics and Science
Building on the testing requirements included in the 1994 reauthorization
of the Elementary and Secondary Education Act, the NCLB Act requires
all schools to conduct mathematics and reading assessments during
academic year 2002 in at least one grade of three different grade
spans (grades 35, 69, and 1012). By academic year 2005, states
must test students in grades 38 in these subjects every year and
must test all students once during the grades 1012 span. States
must also conduct science assessments in one grade of the same grade
spans by academic year 2007. The act prescribes rigorous assessments
aligned with state standards but does give states wide latitude
in setting school performance standards. The NCLB Act also requires
states to participate in the NAEP assessments for the subjects in
which the state tests in order to provide policymakers and the public
with common benchmarks for judging the rigor of their own state's
standards, assessments, and performance requirements.
By 2002, many states had already developed and administered tests
based on their curriculum standards and frame-works. For example,
in academic year 2002, 19 states and the District of Columbia required
students to take mathematics and reading tests in the grades identified
by the NCLB Act (Doherty and Skinner
2003).
The NCLB Act requires states to publish achievement data and other
indicators of performance (such as attendance and completion rates)
at the school level, and disaggregated by key demographic characteristics
such as income, race/ethnicity, and English proficiency status.
A total of 29 states and the District of Columbia rated all schools
or identified all low-performing schools in academic year 2002,
but only 12 states relied solely on student test scores for these
evaluations (Editorial Projects in
Education 2003). The other 17 states and the District of Columbia
used test scores along with other information such as attendance
rates, graduation rates, and coursetaking data.
Consequences and Sanctions
Recently implemented state accountability systems differ from previous
waves of reform in that they specify consequences for poor school
and student performance. For students, consequences may include
using test scores to determine grade promotion or retention and
award high school diplomas. For districts and schools, states have
developed a range of rewards, supports, and sanctions based on student
test scores. In academic year 2002, 27 states and the District of
Columbia provided assistance to low-performing schools (for example,
funds for tutoring and additional teacher professional development)
and 17 states financially rewarded schools that meet, or make sufficient
progress toward, high achievement goals (Editorial
Projects in Education 2003). State officials may impose sanctions
on low-performing schools in 22 states and the District of Columbia.
These include reconstitution (18 states and the District of Columbia),
allowing students to transfer to other schools (11 states and the
District of Columbia), and school closure (11 states). However,
only three states permit withholding funds from schools. States
do not necessarily exercise their authority to apply sanctions against
schools and staff; they generally try to raise achievement in a
low-performing school by first providing additional support such
as targeted professional development, new instructional materials,
and tutoring. Of the 30 states that identified low-performing schools
in 2002, 27 provided some form of assistance to these schools (Achieve,
Inc. 2002a).
Implementation Issues in Assessment
The role of standardized testing in accountability systems is controversial.
Proponents of testing say it can improve achievement in at least
two ways. First, it can provide information about how well educational
systems are functioning and insight into where changes may be warranted.
Second, accountability for test results can create incentives for
students, teachers, instructional material developers, and school
administrators to alter their behaviors in ways that facilitate
achievement. Critics worry that, in implementing testing regimes,
school systems will rely on tests that are insufficiently aligned
with their standards and curricula. Such tests would measure school
and student performance poorly, and strong incentives to perform
well on these tests would undermine curricular priorities.
One indicator of alignment is whether tests were customized,
or specifically designed for a state's standards and curricula.
Customization provides opportunities for alignment, although it
does not guarantee it. In the 2002 academic year, 31 states used
only customized tests, 12 used a mix of customized tests and tests
purchased from commercial publishers that develop tests for a national
market, and 7 used tests that were not customized (GAO
2003). Customization will increase over time because the NCLB
Act requires states to either develop tests aligned to their standards
or augment commercial tests with aligned questions.
Critics also doubt that assessments, especially multiple choice
examinations, will effectively measure higher-order thinking and
conceptual understanding, which are key emphases in national mathematics
and science standards. In the 2002 academic year, 12 states used
tests composed solely of multiple-choice questions, while 36 states
used tests that combined multiple-choice items with a limited number
of written-response questions (GAO 2003).
Definitive data on the effects of enhanced accountability measures
do not exist, but the limited studies available suggest that under
some circumstances, these measures may improve student achievement
(Carnoy and Loeb 2002; Raymond
and Hanushek 2003; Roderick, Jacob,
and Bryk 2002).
|