Updated June 16, 1998
CRS Report for Congress
Received through the CRS Web
National Tests: Administration Initiative
Specialist in Education Finance
Education and Public Welfare Division
The Clinton Administration is encouraging states and local educational agencies
(LEAs) to administer new national tests to 4th grade pupils in reading and 8th grade
pupils in mathematics each year beginning in 2000. Participation in the tests would be
voluntary and would not affect a state or LEA’s eligibility for federal aid programs.
These tests would be based on existing assessments that were developed with federal
financial support. The federal government would oversee development of the tests,
paying the costs for this as well as technical assistance, and the initial round of test
administration, using funds under the Fund for the Improvement of Education (FIE).
According to the Administration, the testing program is authorized under current
statutes, and results from these tests are necessary to inform pupils and their parents
about their achievement in comparison to national and international norms. Opponents
of the national test initiative argue that: it is not authorized; it should be scrutinized by
the Congress before implementation; the new tests are unnecessary; and the initiative
may indirectly lead to inappropriate federal influence on state and local curricula.
FY1998 appropriations legislation prohibits field testing or other administration of the
tests in FY1998, but test development may continue. In addition, test oversight
responsibility was given to an independent, bipartisan board; and the National Academy
of Sciences was directed to undertake studies on the national tests, linking of scores on
existing tests, and appropriate uses of test scores. H.R. 2846, passed by the House on
February 5, 1998, would prohibit national test-related activities beyond those provided
for in the FY1998 appropriations act. Similar provisions in the Senate version of H.R.
2646 are not included in the conference version of that bill.
In his 1997 State of the Union address, President Clinton announced an initiative
under which states and individual local educational agencies (LEAs) are being encouraged
to administer specific tests to 4th grade pupils in reading and 8th grade pupils in
mathematics. Steps have already been taken toward development of the tests, and if the
initiative is implemented, the tests would be administered annually, beginning in the
Congressional Research Service ˜ The Library of Congress
spring of 2000. These tests would be based on two existing tests developed by a nongovernmental entity with federal financial support — the National Assessment of
Educational Progress (NAEP) 4th grade reading test and 8th grade mathematics test.
Scores on the new national tests could be linked to aggregate scores on not only these
NAEP tests but also, in the case of mathematics, scores on the Third International
Mathematics and Science Study (TIMSS).1 The federal government would pay the costs
of necessary modifications to the tests; technical assistance; and in at least the initial year,
for test administration and scoring. This report provides an overview of the initiative,
issues that have arisen with respect to it, and the congressional response.
According to the Administration, “[C]omparing the results of these voluntary tests
to NAEP and TIMSS will give parents and teachers a very clear ‘benchmark’ to measure
how their children and students are performing in relation to those of other states. And,
in the case of mathematics, they will also be able to know where they stand in terms of
the international average.”2 These would be “national tests” in the sense that the same
tests would be administered to all, or nearly all, pupils3 in the relevant grades in
participating states or LEAs throughout the Nation. No current test fits this description.
While one or more tests are administered to virtually all pupils of selected grade levels
in almost all states, these tests vary from state to state. Some states develop and
administer their own tests, others are members of multi-state consortia that develop
assessments (e.g., the New Standards Project), while still others administer tests
developed by commercial publishers (e.g., the Stanford Achievement Tests). Additional
current tests are taken by only a relatively small sample of pupils (e.g., NAEP), or are
taken only by pupils intending to apply to postsecondary educational institutions (e.g., the
Scholastic Assessment Test).
Participation in the testing program would be voluntary for states, or for LEAs that
individually choose to participate. Participation would presumably not be voluntary for
LEAs in states that choose to participate, schools, or students. Thus far, 7 states4 and a
number of individual LEAs have agreed to administer the proposed tests. The
Administration states that participation in or results from these tests would not affect a
state or LEA’s eligibility for assistance under federal aid programs.5 Test results for
individual pupils would be reported to them and their families, as well as schools and
The TIMSS tests were developed and administered under the auspices of the International
Association for the Evaluation of Educational Achievement (IEA), members of which include
education research organizations and government agencies in a variety of nations.
Testimony by Secretary of Education Richard Riley before the House Subcommittee on Early
Childhood, Youth and Families, April 29, 1997 (hereafter cited as “Riley testimony”).
The Administration’s stated intention is that the tests be administered to as many pupils as
possible, including those with disabilities or limited proficiency in English. There has been
debate over Administration statements that the reading test would be offered only in English.
Alaska, Kentucky, Maryland, Massachusetts, Michigan, North Carolina, and West Virginia.
In California, the state superintendent of public instruction has agreed to participate, but the
Governor has not, and it is unclear how this dispute will be resolved.
Unless otherwise noted, information on the national test initiative discussed in this report is
based on material found on the “Voluntary National Tests” site on the Internet’s World Wide
LEAs, but other dissemination of scores would be limited to aggregate data, and “no data
from individual students will come to the U.S. Department of Education.”6
FY1997 and FY1998 appropriations for Title X, Part A, of the Elementary and
Secondary Education Act (ESEA), the Fund for Innovation in Education (FIE), are being
used for development of the national tests. Administration officials have stated that they
believe the FIE provides authority for this initiative, and they have not proposed any
additional authorizing legislation. Approximately $10 million from the FY1997 FIE
appropriation, and $12 million from the amount requested for FY1998, are being used for
national test development and field testing.7 ED has estimated that an additional $15
million from FY1999 FIE appropriations will be used for further test development and
related studies. Administration statements have varied regarding whether the federal
government, or states and LEAs, would be responsible for paying the costs of test
administration after the initial year.
Use of NAEP Frameworks for the National Tests. The national tests would be
based on the frameworks of the existing NAEP tests for 4th grade reading and 8th grade
mathematics. NAEP tests are developed and administered through a cooperative
agreement between the National Center for Education Statistics8 (NCES) and the
Educational Testing Service, a non-governmental organization. Test frameworks are
established through collaboration with teachers, local and state education officials, subject
area and testing specialists. The process is overseen by an independent, bipartisan
National Assessment Governing Board (NAGB).
Major differences between the proposed tests and the NAEP assessments include:
NAEP assessments are administered to a representative sample of students in the
nation, regions, and states, as required by the authorizing statute.9 In contrast, states
and LEAs would be encouraged to administer the new national tests to all students;
No individual pupil takes an entire NAEP test in any subject area or grade level. In
contrast, participating pupils would take the proposed national tests in their entirety;
The national tests would be designed to yield scores for individual pupils, as well as
average scores for schools, LEAs, and states; the NAEP tests are not so designed.
While the NAEP legislation does authorize “state assessments,”10 numerous
limitations are placed on such tests in the NAEP statute (e.g., a limited authorization
period and use of sampling techniques);
All national test questions would be released to the public after each administration
of the proposed tests (test questions would be different each year). Currently, only
selected items are released after each administration of NAEP tests.
Section 411-413 of the National Education Statistics Act (NESA), Title IV of P.L. 103-382
Section 411(b)(1) of the NESA.
States may voluntarily participate in selected NAEP tests. In these cases, the size of the sample
group of pupils tested is increased, so that their scores may reliably predict the performance of
all pupils in the state. Most states have chosen to participate in “state NAEP” tests.
Authorization for the Initiative. ED officials have stated that they have authority
to conduct the testing initiative under the FIE.11 The FIE statute provides broad authority
for discretionary activities intended to promote educational reform. While testing
activities precisely resembling those of the Administration initiative are not specifically
mentioned in the FIE legislation, general categories of activity that are similar in nature
are included in the authorizing language. For example, “...the Secretary is authorized to
support nationally significant programs and projects to improve the quality of education,
assist all students to meet challenging State content standards and challenging State
student performance standards, and contribute to achievement of the National Education
Goals” (Section 10101(a)). More specifically, “[F]unds...may be used for...the
development and evaluation of model strategies for...assessment of student learning”
(Section 10101(b)(1)(A)(ii), emphasis added).
Specific authority for federal support of the development of some types of tests is
contained in certain other legislation; however, the types of tests referred to in legislation
other than the FIE are significantly different from those under the initiative.12 With
respect to possible federal legislative restrictions against the test initiative, the current
statutory prohibitions against federal control of education13 focus on areas such as
curriculum and staff; they contain no explicit or specific references to “tests” or
Selected Issues Related to the Administration’s Testing Initiative
Possible Need for More Specific Legislative Authority. Some have questioned
whether the broad, ambiguous language in the FIE statute is sufficient authority for ED
to develop and administer the national tests, and whether the Administration should
propose legislation to the Congress to specifically authorize the testing initiative. Further,
some in Congress oppose the implementation of the test initiative without debate and
consideration of limitations to the program or possible alternatives; a compromise on
FY1998 ED appropriations legislation does not directly address this issue (see below).
Purposes of the Tests and Possible Alternatives to them. The Administration has
argued that the proposed tests should be developed and administered because they would
serve a unique and valuable role. The tests would provide to individual students and their
parents information on their achievement which could be compared to that of other pupils
not only in their school, LEA, and state, but also the nation at large, and for 8th grade
mathematics, a variety of other nations. In addition, the tests might provide a national
benchmark against which the wide variety of state and local standards for curriculum
content and pupil performance might be measured. The Administration has argued that
Hearings by the House Committee on Appropriations, Subcommittee on the Departments of
Labor, Health and Human Services, Education, and Related Agencies, 1998, Part 5. p. 411.
For example, Section 220 of the Goals 2000: Educate America Act (P.L. 103-227),
Assessment Development and Evaluation Grants, authorizes the Secretary of Education to make
grants to states, LEAs, or consortia for developing, testing, and evaluating state assessments.
Section 103 of the Department of Education Organization Act (P.L. 96-88), Section 438 of the
General Education Provisions Act, and Section 318-319 of the Goals 2000: Educate America
parents will be sufficiently motivated to support higher academic standards only when the
standing of their own child in comparison to national and international averages is made
known to them. No current testing program would provide such information.
However, it may be questioned whether new individual pupil tests are necessary for
this purpose. Existing tests such as NAEP and TIMSS already make it possible to
compare aggregate national, and in most cases state, scores with those of the Nation as
a whole and many foreign nations. Further, some argue that pupil achievement scores
could be compared to national or even international averages without pupils having to
take the same test. For example, it might be possible to statistically link scores on varying
state and other tests currently being administered to NAEP and TIMSS scores. FY1998
ED appropriations legislation provides for a study of such linking possibilities by the
National Academy of Sciences (NAS). An interim report, Equivalency and Linkage of
Educational Tests, was released by the NAS on June 12. It concludes that it is “not
feasible” to compare “the full array of currently administered commercial and state
achievement tests to one another”, due to wide variation in the nature and purposes of
these tests. The NAS is still considering the feasibility of linking scores on “specified
subsets” of current commercial and state tests, to be discussed in their final report.
Possible Evolution of the Nature and Role of the Tests. While participation in the
proposed tests would be voluntary for states, or LEAs that participate individually, and
would not have any effect on federal aid programs, some have expressed concern over the
possible evolution of the testing initiative. At some future time, might participation be
required, or might test results affect federal aid eligibility? Some have speculated that the
current or a different Administration may also want to make state participation in the
testing program mandatory in the future, or condition the receipt of federal aid on progress
in raising state average scores on the tests, especially since the tests are being developed
under broad authority without direct statutory limits on their use.
The President has been quoted as saying that he would like to see voluntary tests
developed in additional subject areas and grade levels in the future,14 but there have been
no official statements from the Administration about long-term plans for this initiative.
It is unclear whether new authorizing legislation would be required for expansion of the
range of tests included in the initiative, but presumably any effort to make the tests
mandatory or to link eligibility for federal financial aid to test participation or scores
would require specific statutory authorization. Nevertheless, at a more subtle level, it is
possible that widespread, voluntary adoption of the tests might indirectly influence
curriculum content in states and LEAs. If, as some argue, school systems tend to “teach
what they test”, then expanded administration of tests based on NAEP frameworks,
especially if substantial consequences are attached to test results, might significantly
influence choices about what pupils are taught.
There are prohibitions against any sort of direct federal “control” of school curricula,
and states or LEAs that participate in the national tests will presumably continue to
administer, and devote primary attention to, their own assessments. Further, the new tests
will be based on the NAEP frameworks, which have been developed by independent
Clinton Hopes Test Proposal Would Be Only the Beginning. Education Week. May 29, 1997.
committees of state and local education officials, subject area specialists, and others, not
ED officials, and under the oversight of NAGB.
Some have questioned whether the process of developing the tests is sufficiently
insulated from influence by ED officials. In response, the Administration and some in
Congress sponsored legislation to give NAGB responsibility for overseeing the national
tests, and this proposal has been adopted (see below). Finally, some policymakers argue
that the national tests would be unfair to many disadvantaged pupils who attend schools
with limited resources. Advocates for recent immigrants and other limited-English
proficient (LEP) pupils in particular have argued that the reading test should be offered
in Spanish and other non-English languages. FY1998 appropriations legislation requires
that the needs of LEP pupils be considered in test development (see below).
P.L. 105-78, FY1998 appropriations legislation for ED, provides that no FY1998
funds may be used to field or pilot test, administer, distribute, or implement any national
tests. However, the development of the national tests in 4th grade reading and 8th grade
mathematics may continue. Oversight responsibility for the national tests has been given
to NAGB, which is responsible for assuring that the tests reflect NAEP content and
performance standards; are free from racial, cultural, and gender bias; are in a form likely
to yield accurate information on achievement; and provide for equal participation of all
children, including LEP and disabled pupils. The NAS is to conduct 3 studies, due to be
completed by September 1, 1998: one on the feasibility of linking pupil scores on
different commercial standardized achievement tests and state assessments with each
other and with NAEP test scores (see discussion of the interim report above); a second on
the technical quality of test items developed for the 2 national tests, including their
reliability, validity, and freedom from racial, cultural, or gender bias; and a third on
safeguards and appropriate uses of pupil scores on the national and other tests. Finally,
P.L. 105-78 provides that no state, LEA, or school may be required to participate in any
pilot or field test of the national tests currently being developed; no student may be
required by the federal government to participate in any national test (including pilot or
field tests); nor may any private, parochial, or home-schooled student be required by any
government agency to participate in such national tests without written parental consent.
More recently, on February 5, 1998, the House passed H.R. 2846, which would
prohibit the use of ED funds for development, testing, or administration of any federally
sponsored national test without specific and explicit statutory authorization. Thus, it
would allow the test development activities in FY1998 to continue, but would not allow
any further test development after FY1998, and would extend the prohibition against
assessment testing or administration beyond the end of FY1998. No Senate action has yet
occurred on H.R. 2846. Provisions similar to those of H.R. 2846 in the Senate version
of H.R. 2646, the Education Savings Act for Public and Private Schools, are not included
in the conference version of the latter bill.