Published by the National
Center on Educational Outcomes
Number 13 / June 2002
Prepared by Rachel Quenemoen and Martha Thurlow
This document has been archived by NCEO because some of the information it contains is out of date.
Any or all portions of this document may be reproduced and distributed without prior permission, provided the source is cited as:
Quenemoen, R., & Thurlow, M., (2002). Including alternate assessment results in accountability decisions (Policy Directions No. 13). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes. Retrieved [today's date], from the World Wide Web: http://education.umn.edu/NCEO/OnlinePubs/Policy13.htm
Alternate assessments provide a
mechanism for students with the most complex disabilities to be included in assessment
systems. Like regular assessments, the purpose of alternate assessments is to provide
valid and reliable assessment data that accurately reflect the states learning
standards, and that indicate how a school, district, or state is doing in terms of overall
student performance.
Background
Two policy issues have emerged with
respect to alternate assessments (1) can alternate assessments be included in
school accountability systems? and (2) what is the best way to include alternate
assessment results in accountability systems? The first question has been debated, with
arguments for including these results focusing on the need to count all students and
arguments against including these results focusing on technical difficulties in
determining how to count alternate assessment results. These arguments are summarized in
Table 1.
Table 1. Pro and Con Arguments for Including Alternate
Assessment Results in Accountability
Pro Arguments |
Con Arguments |
Each students
assessment score is valued. Scores of alternate
assessment can improve. Scores of alternate
assessment are not a negative factor for school accountability index; thus, inclusion is
not discouraged. |
Does not value the
real level of proficiency that must be reached by most students. Could encourage
inappropriate placement of students in the alternate assessment. |
The objective of including alternate
assessment results in school accountability is to ensure that the students who participate
in the alternate assessment are included in the opportunities to learn that are generated
by accountability systems, and ultimately in the improved learning that accompanies them.
Schools can use accountability information based on all students to make policy,
administrative, and instructional decisions that improve schooling practices so that all
students, including students in the alternate assessment, are successful.
Starting Point: Good Alternate Assessments
Alternate assessments are part of the
standards-based reform initiative designed to ensure that all students attain high
standards of learning. Thus, they are one part of the broad effort to:
Define content standards describing
what all students should know and be able to do
Define acceptable levels of
performance
Ensure that all students have
opportunities to learn the content
Develop technically sound
assessments to measure student performance
Develop methods of using the
assessment results to hold schools accountable for students learning
Despite being part of the
standards-based system, alternate assessments involve different assessment approaches from
those used in most general assessment systems. Thus, questions are raised about how
results from alternate assessments can be combined with results from general assessments
for accountability purposes.
To address questions about ways to include alternate assessment results in accountability systems, it is essential that the alternate assessment be well-developed (see Table 2 for some of the characteristics of well-developed alternate assessments). Once this has been accomplished, it is much easier to identify ways in which the results of these assessments can be included in accountability systems.
Table 2. Characteristics of Good Alternate Assessments
There has been careful
stakeholder and policymaker development and definition of desired student outcomes for the
population, reflecting the best understanding of research and practice. Assessment methods
have been carefully developed, tested, and refined. Professionally
accepted standards are used to score evidence (e.g., adequate training, dual scoring,
third party tie breakers, reliability tests and rechecks of scorer competence). An accepted
standards-setting process has been used so that results can be included in reporting and
accountability. The assessment
process is continuously reviewed and improved. |
Accountability Options
Although there are many ways in which
alternate assessment results can be reported (see Bechard in Resources), these approaches
can be merged into two basic options for accountability (see Hill in Resources). The two
options reflect different ways of counting alternate assessment results either as a
score that can cover a range of values and thus can be improved in accountability, or as a
score that is always at the lowest level.
These options have been described in
terms of how the alternate assessment results will be scaled relative to the results of
the general assessment (see Hill in Resources):
Option 1: Scale the results
so that the value awarded for achievement levels on the alternate are the same or similar
to the value awarded for achievement levels on the general assessment.
This approach is based on a belief
that achievement on the alternate assessment is valued just as much as achievement on the
general assessment. The alternate assessment, like the general assessment, must be aligned
to the states content standards (usually by extending those standards). The
alternate assessment also has a defined alternate achievement standard to complement the
desired achievement standard for students in the general assessment.
In this option, improvements in
alternate assessment scores are just as important as are improvements in general
assessment scores. A student in the alternate assessment can achieve proficient status as
defined for this small population of students, just as can a student in the general
assessment. For this approach to work, it is important that the population of students for
whom the alternate assessment is appropriate be clearly defined defined in such a
way that the number of alternate assessment students is stable from one year to the next.
The major objection to Option 1 is
that awarding the same number of points or the same proficiency label for successful
performance on the alternate assessment may devalue proficient performance on the regular
assessment. It also could promote the inclusion of some students in the alternate
assessment who should be in the general assessment. Both of these objections can be
addressed through policy.
Option 2: Scale the results
so that the achievement levels on the alternate assessment are at the lower end of the
scale and achievement levels on the general assessment are at the upper end of the scale.
This approach is based on a belief
that there must be an absolute level of achievement that is the target for all students,
regardless of the significance of their disabilities. A student taking the alternate
assessment with this approach can never achieve proficient status, but instead can only
achieve the lowest achievement level possiblejust because the student participates
in the alternate assessment.
This approach discourages the
inappropriate assignment of students to the alternate assessment because the scores a
student could receive are automatically at the lower end of the achievement scale. On the
other hand, there is nothing that can be done to improve the score of a student in the
alternate assessment because the student will always earn the lowest achievement level
possible regardless of actual improvements in achievement.
The major objection to Option 2 is
that the presence of students who are appropriately assigned to the alternate assessment
is likely to bring down school achievement indices. As a result, there is little incentive
for schools to want these students in their buildings, thereby discouraging inclusive
educational settings for alternate assessment students.
Comparison of Accountability Options
The selection of one accountability
option over another is a policy decision one that should be reached through
stakeholder involvement. The decision also should be based on evidence from states that
have adopted one approach or another, and by consideration of the results of simulation
research.
The effects of adopting Option 1 have
been demonstrated in simulation research conducted by Richard Hill from the Center for the
Improvement of Educational Assessment (see Resources). Three conclusions were supported
when he used state data in simulated formulas for accountability:
The impact of including alternate
assessment scores on school gains is trivial if the number of alternate assessment
participants remains fairly constant from year to year.
Making gains on the alternate
assessment that are comparable to gains on the general assessment introduced little
additional measurement error.
Including alternate assessment
results in accountability appears to lead to better outcomes for the students who
participate in the alternate assessment.
If a school or district is looking
for gains in achievement for all students, the school should be entitled to rewards for
gains in the alternate that are equivalent to rewards for gains in general assessments.
Hill concludes that states should
address several essential questions as they consider options for including alternate
assessment scores in accountability: What is fair? What will encourage the greatest
improvement for every student? What seems reasonable? Thus far, states have answered these
questions in different ways.
Examples of State Approaches to Accountability Option 1
To some extent, the application of
the accountability options in states will reflect the differences in the assessment
systems on which accountability is based. The three approaches described here begin with
the assumption that all students count; the question of whether results can be included
has already been answered: yes they can. The approaches differ in the specific ways that
results are included.
Approach 1: Alternate
In this approach, achievement levels
for the general assessment and alternate assessment are different, as are the descriptors
of the levels. For example, one state refers to the four achievement levels of the general
assessment as novice, partially proficient, proficient, and advanced. In contrast, for the
alternate assessment, there are three achievement levels, with the labels beginning,
partially skilled, and skilled.
In this first approach to including
alternate assessment results in accountability, a set of decision rules is used to
determine when the alternate assessment results are considered. The decision rules reflect
the importance placed on appropriate participation in assessments. Thus, alternate
assessment results are included when there is no evidence of sufficient participation in
the general assessment, and when adequate improvement in general assessment scores
relative to a long-range target is not evident.
In this approach, any school that has
a general assessment participation rate that is less than 95% automatically goes into
school improvement. For those schools without adequate gains in general assessment
performance, other factors are considered as well (for example, participation rates for
the alternate assessment, grade 1 and 2 assessment results, progress of Title I students,
reductions in percentage of students in the novice level).
Schools gain or lose points based on
a combination of participation and progress on the alternate assessment. Specific points
added or subtracted from school scores are as follows:
Add 4 points progress in
average alternate assessment score and/or 99% participation rate in the general assessment
Add 2 points no progress in
alternate assessment score and/or 99% participation rate in the general assessment
0 points decline in average
alternate assessment scores with a 98% participation rate in the general assessment
Subtract 2 points
participation rate in general assessment was 97%
Subtract 4 points
participation rate in general assessment was less than 97%
Obviously, the specifics of the way
in which the alternate assessment factors into the accountability system can be altered,
but the key element of this approach is that the alternate assessment is counted in some
way.
Advantages. An emphasis on
getting all students into the system is likely to result in high participation rates.
Schools can improve their status through full participation and through improved alternate
assessment performance.
Disadvantages. Different
labels for achievement levels and use of alternate assessment results only when general
assessment participation and performance is not considered adequate essentially gives less
weight to the alternate assessment. For schools making adequate gains with their general
assessment students but inadequate gains (or no gains) with their alternate assessment
students, there is no recognition of the need to improve programs for those students with
the most significant disabilities.
Approach 2: Two Alternate
Assessments Counted Differently
In this approach there are two
alternate assessments, one of which is intended for students with significant disabilities
(generally an alternate portfolio), and the other of which is for students who may
not be able to take the general assessment but who are not eligible for the alternate
portfolio (generally some type of lower grade-level assessment or an academic
inventory). Each alternate assessment is included in accountability, but in different
ways.
The academic assessment, which is
most like a general assessment, expands the academic achievement levels of the general
assessment. A common approach is to add levels to each of the existing achievement levels.
For example, there may be Novice 1, 2, 3, and 4; Apprentice 1, 2, 3, and 4; Proficient 1,
2, 3, and 4; and Distinguished 1, 2, 3, and 4. This essentially provides more steps for
demonstrating progress. These scores can be included in accountability in the same way
that general assessment scores are generally one number for each content area
(reading, math, and others).
The alternate portfolio is included
in this accountability approach in a different way from the alternate academic inventory.
While the same labels are given to the alternate portfolio achievement levels (novice,
apprentice, proficient, distinguished) as are given to the general assessment achievement
levels (without the extra levels 1, 2, 3, and 4 given to the academic alternate), each
student receives only one score rather than the multiple scores in the general assessment
and the alternate academic inventory. Thus, the general assessment counts several times,
as does the alternate academic inventory, while the alternate portfolio counts just once.
Advantages. All students
receive scores and all scores count. The use of the same labels conveys a message that
students with the most significant disabilities can become proficient and distinguished,
just as other students can.
Disadvantages. Scores of
students in the alternate portfolio count less than the scores of other students. This
approach essentially averages scores rather than averaging students, and some students
count more than others. Schools may earn awards without any impact on students in the
alternate portfolio. This may lead to less concern about improving programs and
instruction for these students.
Approach 3: Alternate
Assessment with the Same Labels and Same Counting
This approach also uses the same
labels for the achievement levels assigned to the general assessment and the alternate
portfolio assessment novice, apprentice, proficient, and distinguished. The belief
that students who demonstrate proficient performance within the structure of the alternate
assessment should make the same contribution to the school accountability index is carried
into the numbers of the accountability system. The same point values employed for the
general assessment are used for the alternate portfolio assessment. In addition, students
in the alternate assessment are required to demonstrate achievement within the same
multiple content areas as are students in the general assessment, and scores are assigned
accordingly for each content area.
To further support the commitment to
inclusion of all students in the accountability system, this approach generally has a
couple of additional aspects. One is to assign the lowest score possible to any student
who is not in either the general or alternate assessment. To guard against inappropriately
including students in the alternate assessment, an audit point is identified (generally 12%
of all students) to indicate that the level of participation in the alternate assessment
needs to the checked because it might be too high.
Advantages. Scores of
students in the alternate assessment count just the same as scores of students in the
general assessment. This approach essentially averages students, rather than test scores,
so that each student receives equal weight in the school accountability index. With the
scores of all students counting (even those not in either assessment), participation is
encouraged. Further, schools are unlikely to earn awards without having an impact on
students in the alternate portfolio. This may lead to more concern about improving
programs and instruction for these students.
Disadvantages. Without
certain additions, this approach could result in students being inappropriately placed in
the alternate assessment. Additions, such as assigning the lowest score possible to
non-participating students and monitoring participation rates through an audit point, can
result in an approach with no obvious disadvantages.
Summary
The imperative to include all
students in school accountability systems and the decisions that are made based on these
systems is clear. Yet, there are multiple ways in which scores can be included, and the
effects of each approach needs to be carefully considered. The two most basic options (1)
to count achievement on the alternate assessment comparable to achievement on the general
assessment, or (2) to count achievement on the alternate assessment as automatically lower
than achievement on the general assessment, is the starting point for conversations about
which approach to take for accountability.
There are many approaches that can be
taken within each option. In this document we have highlighted those approaches that are
consistent with the first option (where alternate assessment scores have the same or
similar value as general assessment scores), the option that has been supported by
simulation research on the technical, motivational, and policy effects of different
approaches. As implementation of different approaches proceeds, it will be important to
monitor the actual effects of various approaches.
Resources
The Impact of Including Special Education Students in Accountability Systems. Hill, R. (2001). Portsmouth, NH: National Center for the Improvement of Educational Assessment. http://www.nciea.org/cgi-bin/pubspage.cgi?sortby=pub_date
Models for Reporting the
Results of Alternate Assessments within State Accountability Systems
(Synthesis Report 39). Bechard, S. (2001). Minneapolis, MN: University of Minnesota,
National Center on Educational Outcomes.
Use of Alternate
Assessment Results in Reporting and Accountability Systems: Conditions for Use Based on
Research and Practice (Synthesis Report 43). Quenemoen, R., Rigney, S.,
& Thurlow, M. (2002). Minneapolis, MN: University of Minnesota, National Center on
Educational Outcomes.