Two Percent Flexibility

CONTEMPORARY ISSUE
IN SPECIAL EDUCATION:
TWO PERCENT
FLEXIBILITY
M. Kalene Meeks
Rebecca Sheffield

Presentation Contents
• What is the “Two Percent Flexibility” issue?
• Five recent studies
• Explanations
• Evaluations of research quality
• Implications of this research for special education
policy and practice.
• March 2011 Update
• Logic model for possible project

Two Percent Flexibility

THE ISSUE

The Issue:
• The Two Percent Rule, effective in 2006, applies to
No Child Left Behind assessment requirements, Sec.
200.6(a)(3), and gives states some leeway in
assessing students with disabilities by allowing states
to develop and administer alternative assessments
based on modified achievement standards (AA-MAS).

• Student scores on AA-MAS may account for up to
two percent of the scores for states’ and districts’
Adequate Yearly Progress determinations.

(Kettler et al., 2011; Two Percent Flexibility, 2011)

The Issue:
• States should determine how their accommodation
policies allow for students’ participation in the
regular assessments before developing an AA-MAS.
• Flexibility: AA-MAS are not mandatory.
• Also, more than two percent of AA-MAS scores may
count towards proficiency if less than one percent of
AA-AAS (alternative assessment based on alternate
achievement standards) scores are counted towards
proficiency. Thus:

AA-AAS + AA-MAS < 3% of students taking tests

The Issue:
Who? Students with disabilities who have difficulty meeting
grade level standards as judged by regular state tests:
• must have an Individualized Education Plan (IEP)
• must have access to grade level curriculum
• disability must be responsible for test difficulties (not sub-par instruction)

Difficult Easy

Regular Test or AA-MAS AA-AAS
Test w/ Accommodations

Note: The AA-MAS is different from the AA-AAS (Alternative Assessment
based on Alternate Achievement Standards). Students taking AA-
AAS tests are not necessarily instructed in the general curriculum.


The Issue:
AA-MAS
• must be aligned with grade level content but may differ in
breadth and depth.
• cannot preclude student from receiving diploma.

•In Texas
• AA-MAS = STAAR-M
• AA-AAS = STAAR-ALT

• Regardless of the tests available, testing decisions must be
made on an individual basis by the student’s IEP team (ARD
committee).

The Issue:
The Two Percent Rule is a good place to start but there
is no consensus yet on its implementation.
• Arguments:
• Two percent permits either too many or not enough
students.
• It is not feasible to develop validated assessments in
a short period of time.
• Many states are trying to modify their existing
assessments rather than develop new ones.

The Issue:

Historical
Flexibility for focus on
Longitudinal
Two Percent states to student
Growth
Rule develop achievement
Model
assessments based
evaluations

The way we see it…

The Two Percent Rule:
• acknowledges that the Department of
Education recognizes the need for
alternative assessments based on
modified academic standards.
• provides a framework in which states
can develop and administer these
assessments.

Lazarus, Cormeier, &
Thurlow (2011)
STATES' ACCOMMODATIONS
POLICIES AND DEVELOPMENT OF
ALTERNATE ASSESSMENTS
BASED ON MODIFIED
ACHIEVEMENT STANDARDS: A
DISCRIMINANT ANALYSIS

Lazarus, Cormier, & Thurlow (2011)
• Restrictive accommodation policies might affect students
ability to participate in regular assessments and might,
therefore, affect states decisions about using AA-MAS.
• States’ policies specifically vary on the amount of
accommodations that can be provided to students on the
regular assessments. These include the following categories;
presentation, equipment and materials, response, scheduling
and timing, and setting.
• These researchers compared AA-MAS use data and attempted
to relate that to the states’ accommodation policies.

Lazarus, Cormier, & Thurlow (2011)
The Questions:
• Do differences in the number of allowed
accommodations on regular assessments
differentiate states that plan to offer AA-MAS and
those that do not?
• Can the number of allowable accommodations
permitted on regular assessments predict:
1. states that plan to use AA-MAS.
2. the likelihood that a state will decide to
develop AA-MAS.

Lazarus, Cormeier, & Thurlow (2011)
Design
• Discriminant analysis
• Correlational design
• Participants: Each of 50 states
• Gathered state data for accommodation policies from
National Center on Educational Outcomes
accommodation policies database (online).
• Survey:
• completed by stated directors of special education.
• collected dichotomous variable (plan to, do not plan to
develop AA-MAS)
• Exposed these variables to discriminant analysis.

Conclusion:
• States that plan to offer AA-MAS allowed
statistically fewer accommodations in four
categories:
• presentation,
• equipment and materials,
• scheduling or timing, and
• setting.
• Presentation accommodations showed an
especially strong direct correlation to states’
decisions about AA-MAS development.

Impact
• Demonstrated that states with
more accommodations on regular
exams have less likelihood of
developing AA-MAS.

•Inversely, states with restrictive
accommodation policies more
likely to offer AA-MAS.

Impact
• Raises questions:
Do 1) accommodation training for IEP teams and 2)
implementation of accommodations with fidelity
(targeted use, rather than just permitting a variety
of accommodations) reduce the likelihood of
students being categorized as eligible for AA-MAS?
• Further study of qualitative aspects of
accommodation application needed, perhaps with
development of an index.

Elliott et al. (2010)

EFFECTS OF USING
MODIFIED ITEMS TO
TEST STUDENTS WITH
PERSISTENT ACADEMIC
DIFFICULTIES

The Questions:
• Do AA MAS eligible students perform better on tests
comprising highly accessible, modified items than on
the original tests?
• If the performances of eligible students improve on
tests comprising modified items, what percentage of
the students are likely to perform at a level deemed
proficient in reading or math?

Design:
• Experimental Research Design
• Participants selected from homogenous group of 8 th graders
with disabilities.
• Students were sorted by those eligible and those ineligible to
take AA-MAS. This was not used for treatment but merely
information for later data analysis.
• From this original group, random assignments were made to 3
possible test sets.

Design:
• Each of these three test sets was given in three parts and
given in varying order (to control for order effect) for a total
of 36 possible unique tests.
• All groups were exposed to both modified and unmodified
forms of the exam; so, they serve as their own control.
• The teachers administering the tests were trained with
common slide set.

Conclusions:
• AA-MAS eligible students did significantly better on
reading and math modified test items

Limitations:
• Only 8th grade students were tested. This design
should be repeated with elementary and high school
students.

Impact:
• Do AA MAS eligible students perform better on tests comprising
highly accessible, modified items than on the original tests?
• The observed positive effect of AA-MAS-type modifications
suggests that this is a viable approach to testing students with
disabilities who have poor test performance histories.

• If the performances of eligible students improve on tests
comprising modified items, what percentage of the students are
likely to perform at a level deemed proficient in reading or math?
• More students eligible for AA-MAS could meet proficiency
with modifications.

Kettler, Rodriguez, Bolt, Elliott,
Beddow, & Kurz (2011)
MODIFIED MULTIPLE-
CHOICE ITEMS FOR
ALTERNATE ASSESSMENTS:
RELIABILITY, DIFFICULTY,
AND THE INTERACTION
PARADIGM

Kettler, Rodriguez, Bolt, Elliott, Beddow, & Kurz
(2011)
Question:

Do Tests composed of modified items
have the same reliability as tests made of
original items?

(2011)
Design
• Experimental design used to test reliability of AA-
MAS
• Three groups of 8th graders defined by AA-MAS
eligible and AA-MAS ineligible chosen from several
states as participant pool (CAAVES project
collection).
• Random participants chosen from pool and took
both original and modified reading and math

(2011)
Conclusion:
• Reliability appears to be consistent on AA-MAS
compared to original tests.
• Analysis revealed shortening the question stem
especially to be an especially effective modification,
and adding graphics might be a poor modification.

Palmer (2009)
STATE PERSPECTIVES ON
IMPLEMENTING, OR CHOOSING
NOT TO IMPLEMENT, AN
ALTERNATE ASSESSMENT
BASED ON MODIFIED
ACADEMIC ACHIEVEMENT
STANDARDS

Palmer (2009)

Federal regulations give
states discretion in
choosing to implement
AA-MAS as part of their
accountability system

Palmer (2009)
• Non-regulatory guidance from the Department of Education advises
that neither grade level assessment (with or without
accommodations) or alternate academic achievement standards
(AA-AAS) are appropriate for the group of students that AA-MAS
should target.
• Because
• Grade level assessment is likely too difficult
• AA-AAS does not reflect the wide range of grade level
content and therefore does not demonstrate what they
know or progress they have made.

Palmer (2009)
Questions

•What are states perspectives on assessments
based on AA-MAS?
•What are the reasons states choose to, or not
to, implement AA-MAS?

Palmer (2009)
Design

• Two surveys were created: one for states that have
chosen to implement AA-MAS and one for those who
have chosen not to.
• Surveys were given to state level directors of
assessment in 24 states and 22 responses were
obtained

Palmer (2009)
Conclusion
Reasons given for not implementing AA-MAS:
• Lack of resources
• Lack of guidance
• Further complicates data comparability
• Violates common expectations for all
• Desire to “call” students “proficient” is not enough of a
reason to develop AA-MAS

Palmer (2009)
Conclusion

Reasons given for choice to implement AA-MAS:
•Improving accessibility
•Improving appropriateness

Palmer (2009)
Conclusion

Concerns of effectiveness expressed
by both sides (“it may not work”).

Palmer (2009)
Impact

•Resources and guidance are needed to
implement AA-MAS.
•Time will bear out
effectiveness or
ineffectiveness of
AA-MAS and its
acceptability among
critics.

Christenson, Decker, Triezenberg,
Ysseldyke, & Reschley (2007)

CONSEQUENCES OF
HIGH-STAKES
ASSESSMENT FOR
STUDENTS WITH AND
WITHOUT DISABILITIES

Christenson, Decker, Triezenberg, Ysseldyke, &
Reschley (2007)

•High stakes testing is required in some states
of both general education students and
students with disabilities.

•The original intent of high stakes testing was
to positively impact educational outcomes.

Reschley (2007)

There is some debate over unintended effects
of high stakes testing, such as
• tests’ tendency to become the objects of
instruction,
•scores that don’t generalize to other
assessments measuring similar academic skills,
and
•negative impact on the motivations of
struggling students.

Reschley (2007)

Unintended and intended effects
have not been well studied in an
across-state project for students
with disabilities.

Reschley (2007)

Question

•What are the intended and unintended effects
of high stakes testing for general and special
education students?

Reschley (2007)
Design:
• Descriptive survey of 249 general education teachers,
special education teachers and school psychologist
from 99 schools in 19 states that have mandatory
high school exit exams.
• Named “Perspectives of Testing and Grade
Promotion Survey”
• Pool of possible participants hosting high stakes
testing was used to randomly select participants
from elementary, middle and high school.

Reschley (2007)
Design:
• Special education teachers instructed to answer
based on their students; general education and
psychologists asked to answer based on students
who did not pass high stakes testing or who
struggled with it.
• Oversampled by sending 3 surveys to each school to
increase return rate. Also offered $500 school supply
lottery for those returned and still only achieved 11.6%
return rate.
• Asked to report on 64 observable events

Reschley (2007)
Analysis
• Data was analyzed for disparities of at least 10%
between special education teachers and general
education teachers answers.
• This disparity analysis may contain a great deal of bias
that is not accounted for in the research design.
• Why? Because of special education and general
education teachers varied in their perspectives of
students with disabilities.

Reschley (2007)
Analysis:
 Responses of general and special education teachers were
ranked then put through Spearmans Correlation for rank
order. Researchers reported that the “high degree of
similarity in the rank order of observable events that were
reported as having increased by the two respondent groups
corroborates the high degree of concordance of the
descriptive statistics reported in Table 2.” (Table 2 was simply
the ranked order of observable events organized in order
from highest to lowest.
This is an invalid conclusion.

Reschley (2007)
Analysis:
“How do I interpret a statistically significant Spearman correlation?
• It is important to realize that statistical significance does not indicate the
strength of the Spearman rank-order correlation. In fact, the statistical
significance testing of the Spearman correlation does not provide you
with any information about the strength of the relationship. Thus,
achieving a value of P = 0.001, for example, does not mean that the
relationship is stronger than if you achieved a value of P = 0.04. This is
because the significance test is investigating whether you can accept or
reject the null hypothesis. If you set α = 0.05 then achieving a statistically
significant Spearman rank-order correlation means that you can be sure
that there is less than a 5% chance that the strength of the relationship
you found (your rho coefficient) happened by chance if the null
hypothesis were true.”
 http://statistics.laerd.com/statistical-guides/spearmans-rank-order-
correlation-statistical-guide-2.php

Reschley (2007)…….Analysis
• Teachers were asked “Are there differences in how grade
advancement decisions are made for students with and
without disabilities? If so, what differences exist?”
• Researchers reported “The majority of both general
education teachers (45%) and special education
teachers (51%) indicated that grade advancement
decisions were either ‘occasionally’ or ‘almost never’
made in the same way for students with and without
disabilities.”
• What’s wrong with that observation?
• 45% is not a majority, 51% is barely a majority, much more
accurately represents a split, and certainly not grounds for
drawing any conclusions.
• Also…

Reschley (2007)…….Analysis…. More Bias
• “Are there differences in how grade
advancement decisions are made for students
with and without disabilities? If so, what
differences exist?” asked of both general and special
Ed teachers.
• There is obvious sampling bias in this question:
• General education teachers have experience with special
education promotions due to inclusivity rules.
• However, the reciprocal experience with special
education teachers and general education students’
promotion decisions is probably non existent.

Reschley (2007)
Impact
• Due to the numerous design problems, we do not
recommend basing impact decisions on this
research.
• However, this might be useful in a more generalized way or as
a starting point for further study.


IMPLICATIONS OF THE
RESEARCH FOR
PRACTICE

Implications of the Research for Practice
• Modified assessments can be developed which
improve our ability to assess the progress of a select
group of students.
• In deciding whether or not to implement a new,
alternate assessment of any type, states should first
look at their accommodation policies. Could the
population of students they are considering be just as
easily tested with the regular state assessment if
accommodation policies were changed?

Implications of the Research for Practice
The existing AA-MAS assessments, when appropriately
designed, should be most beneficial for those students who
meet qualifying criteria (this is called a “differential boost”)
(Elliot et. al, 2010)
• Qualifying criteria need to be determined.
• They are partially set by NLCB.
• Groups like the Consortium of Alternate Assessment Validity
and Experimental Studies (CAAVES, part of Elliot et al. and
Kettler et al.’s studies) have clarified the criteria.

Implications for Practice
Leading to our logic model…
• AA-MAS were designed for a specific population and are not
intended to be used with all students (thus the Two Percent /
Three Percent cap).
• Just as researchers carefully defined a populations for their
studies (Elliot et al., 2010; Kettler et al., 2011), schools must be
careful in how they determine which students should take
modified exams.
• There are no federal mandates; therefore there is a risk for
intentional/unintentional mis-assignment to the AA-MAS.

MARCH 2011 UPDATE

image from thinkchange.org photo from namm.org

March 2011 Update:
• U.S. Secretary of Education, Arne Duncan, announced
that the Department of Education will not continue
to support the implementation of AA-MAS (Kaloi,
2011).
• The CEC has not responded to this change on their
website.

March 2011 Update:
• The NCLD (National Center for Learning Disabilities) was pleased
with the discontinuing of the “Two Percent” rule. They have always
been opposed to the AA-MAS.
• States with AA-MAS in place can continue using these tests until
new assessments are developed (unclear what is coming next).

“Every student with a learning disability should have
every opportunity to achieve graduating from high
school with a regular diploma with their peers”
(Kaloi, 2011).

Update March 2011
The “Proxy” issue:
“The provision of an interim policy —
sometimes called a ‘proxy’ — allowed
…states that were working to develop the
AA-MAS to count some students with
disabilities who failed the general state
assessment as “proficient” for school/district
accountability purpose. While limited to
certain states…this interim policy was in
effect for 5 years and expired in 2009.
During that time, several states used this
proxy but did NOT develop the AA-MAS.
Some states pressured the Department of
Education to extend the interim policy”
(Kaloi, 2011).

March 2011 Update:
NCLD’s additional opposition to AA-MAS:
• “The research basis [not cited] to support the new policy did not include a
statistically reliable number of students with disabilities; however, the
policy targets only these students.
• …Too little was known about how to identify and determine which
students should be taking the modified test. … students who are poor and
from minority groups account for a large portion of failing students. Yet
the 2% rule allows only students with disabilities to take the AA-MAS.
• Many of the AA-MAS developed by states provide accommodations that
are not allowed on the state general assessment. Instead of shifting
students out of the general assessment, very restrictive accommodations
policies should be reviewed and revised to provide the widest range of test
accommodations” (Kaloi, 2011).


LOGIC MODEL

Logic Model
The situation:
• In Texas, which was one of the first states to develop an AA-
MAS, even before the Two Percent Flexibility rule went into
place, the assignment of students to AA-MAS (TAKS-M or STAAR-
M) and AA-AAS has not been done with due consideration and
fidelity. This assignment is not consistent from district to
district and frequently not consistent from school to school
within districts (based on personal observations).

• As long as “M” tests are in place,
consistent and appropriate
assessment determination is
needed.

Logic Model
The situation:
• In a hypothetical district, Star of Texas
Independent School District (STISD), IEP
teams no not have the proper training to
apply consistent, appropriate methods in
determining students’ assignment to the
various state assessment types.
• Students with IEPs are inconsistently
assigned to STAAR-Accommodated, STAAR-
Modified, and STAAR-Alternate.
• Factors such as teacher availability, ease of
implementation, and inadequate
instruction are used in determining
students’ STAAR exam versions.
• Students are inaccurately assessed and
sometimes even insulted by their
assignment to various test administrations.

Logic Model
The situation:
Inaccurate assignment to exams prohibits effective
monitoring of student performance, leading to
reduced effectiveness in students’ educational
planning and implementation.
Students are potentially prevented from achieving
their full potentials.
Schools and districts are not collecting the most
accurate data and may be making formative
decisions based on unjustified measures.

Logic Model
Outputs Outcomes - Impact
Inputs Activities Participation Short Medium Long

Assumptions External Factors

Logic Model
• Standards and rules for who can take the
Inputs AA-MAS
• NCLB Sec.200.6(a)(3) regulations (The
Two Percent Flexibility Rule) (Kettler et
al., 2011)
• State of Texas STAAR-M Participation
Requirements (Texas Education
Agency)
• CAAVES (Consortium of Alternate
Assessment Validity and Experimental
Studies) AA-MAS Participation Decision
Criteria (Elliott et al., 2012)

Logic Model
NCLB Sec.200.6(a)(3)
• “a state may develop a new alternate assessment based on
modified academic achievement standards or adapt its
general assessment… [The AA-MAS] must cover the same
grade-level content as the regular assessment… a State may
employ a variety of strategies to design an [AA-MAS].”
• The AA-MAS is intended for students for whom:
• Standards of regular assessment are too difficult
• Standards of the AA-AAS are too easy
• Disability/ies have prevented them from reaching proficiency
• Disability/ies make it unlikely that they will reach proficiency by the same
standards within the same timeframe as students not eligible.

(Kettler et al., 2011)

Logic Model
Texas STAAR-M Participation Requirements

Key indicators:
• PLAAFP lead ARD committee to conclude “multiple years behind”
• “will not progress at same rate as peers”
• “disability significantly affects academic progress”

Logic Model

Key indicators:
• TEKS-based goals in the IEP indicating modified content
• Modified content specific to area of need
• IEP goals should address, at least generally, how content will be
modified.

Logic Model

Key indicators:
• Require direct and intensive instruction for skill acquisition,
maintenance, and transfer.
• Direct = small group/individualized
• Intensive = continuous and focused

Logic Model

“Modified coursework results in the
student graduating on the Minimum
High School Program (MHSP). Students
who graduate on the MHSP are not
eligible for automatic admission into a
Texas four-year university.”

Logic Model
CAAVES AA-MAS Participation Decision Criteria

1. Student has a current IEP with goals based on
academic content standards for the grade of
enrollment
a. Does the IEP state that the instructional
material/curriculum contains grade level content?
b. Are there statements from IEP members that goals and
instruction align with grade level content standards?


Logic Model
2. Student’s disability precludes him/her from
achieving grade-level proficiency as demonstrated
by performance on assessments that can validly
document academic achievement.
• Previous year’s tests documenting performance at
the lowest proficiency level(s) or equivalent
testing documentation.


Logic Model
3. Student’s progress to date in response to
appropriate instruction is such that, even with
significant growth, he/she will not achieve grade-
level proficiency within the year.
•Written description of research-based instruction and
either:
• Two years of class performance records,
• Three years of state achievement test scores,
• Multiple curriculum-based measurement scores


Logic Model
• People knowledgeable about Two
Inputs Percent Flexibility, students with
disabilities, and the ARD (Admission,
Review, and Dismissal) process
• Rebecca & Kalene
• Professors and Colleagues at
Texas Tech
• Contacts at the Texas Education
Agency (TEA)
• Cooperation from STISD
administration.
• Time and technology (PowerPoint,
email)

Logic Model
Activities:
Outputs • Develop training and checklists for IEP
teams using NLCB, CAAVES, and TEA
Activities Participation
documentation.
• Develop train-the-trainer materials to
enable schools to conduct their own
trainings with additional/new staff
members.
• Present training to STISD administration
for their feedback and approval.
• If permission granted, survey parents and
teachers to determine district’s current test
assignment systems. Attempt to
understand the extent of inappropriate
test assignments.

Logic Model
Activities:
Outputs
• Modify training as necessary.
Activities Participation • Schedule and implement training on
individual campuses with key IEP team
members.
• Collect feedback from training
participants to determine impact.
• Is the process clearer?
• Will rules/policies be adhered to?
• Will administrators/supervisors
monitor for fidelity?

Logic Model
Outputs Participation:
Activities Participation
• Presentation/training team:
• Rebecca
• Kalene
• STISD administration
• STISD school IEP team key members:
• Special education teachers
• Diagnosticians
• Administrators
• Others
• Parents/other teachers (if permission
given for initial survey)

Logic Model
Short-term Outcomes/Impact:
Outcomes - Impact • IEP teams will have increased
Short Medium Long understanding of:
• the intended population for STAAR-M
• moral and legal obligation to
correctly select the test that is best
for each student
• how to use a checklist to improve
assessment assignment
• The IEP teams will have checklists to use
when assigning assessments .
• Administrators will understand and be
able to evaluate whether students are
appropriately assessed.

Logic Model
Medium-Term Outcomes/Impact:
Outcomes - Impact • IEP teams will use checklists.
• IEP teams will base all assessment
Short Medium Long
assignment decisions on the intended
criteria.
• IEP teams will continue the “train the
trainer” model
• Administrators will monitor the fidelity
of assessment assignment and will
follow up when inappropriate
assignments have been made.
• Students will be assigned to STAAR-M
only when it is the appropriate
assessment for evaluating their progress.

Logic Model
Outcomes - Impact Long-Term Outcomes/Impact:
Short Medium Long
 Accurate assignment to exams enables
effective monitoring of student
performance, leading to enhanced
effectiveness in students’ educational
planning and implementation.
 Students are better enabled to achieve
their full potentials.
 Schools and districts collect the most
accurate data and make formative
decisions based on justified measures.

Logic Model

Assumptions:
• Administration will be on board with the goals of the
training.
• Presentation will have a positive influence on IEP team
decision making.
• Presentation will not conflict with existing policies and/or
school will be willing to update current practices and policies
to match with the criteria and indicators suggested.

Logic Model

External Factors
• Limited amount of time will be available for training
• Consideration: Perhaps ultimately an online
training/webinar could be developed
• Current IEPs which mis-assign students may not be
changed before testing this school year
• Parents’ influence on testing decisions
• Some IEP team members may never receive training
• Changing assessment systems in the state of Texas

Logic Model
Outputs Outcomes - Impact
Inputs Activities Participation Short Medium Long

• Survey

Knowledge
• Standards • Check-
•Trainers

Impact!
Actions
lists
• People • Develop
•Admin.
training
• Time • Present
•IEP
training
Teams
• Tech. • Eval.
impact

Reception from district, existing policies Time, existing influences


QUESTIONS?

References
• American Association for Public Opinion Research. (n.d.). Best practices.
Retrieved from AAPOR website http://www.aapor.org
• Christenson, S.L., Decker, D. M., Triezenberg, H. L., Ysseldyke, J. E., & Reschly,
A. (2007). Consequences of high stakes assessment for students with and
without disabilities. Educational Policy, 21, 662-690.
doi:10.1177/0895904806289209
• Council for Exceptional Children. (2011). Two Percent Flexibility. Retrieved
from CEC website http://www.cec.sped.org
• Elliott, S. N., Kettler, R. J., Beodow, P. A., Kurz, A., Compton, E., McGrath, D., .
. . Roach, A. T. (2010). Effects of using modified items to test students with
persistent academic difficulties. Exceptional Child, 76(4), 475-495.
• Gersten, R., Fuchs, L. S., Compton, D., Coyne, M., Greenwood, C. R., Innocenti,
M. S., (2005). Quality indicators for group experimental and quasi-
experimental research in special education. Exceptional Children, 71 (2), 149-
164.
• Kettler, R. J. Rodriguez, M. C., Bolt, D. M., Elliott, S. N., Beodow, P. A., & Kurz,
A. (2011). Modified multiple-choice items for alternate assessments:
Reliability, difficulty, and differential boost. Applied Measurement in
Education, 24, 210-234. doi:10.1080/08957347.2011.580620

References (continued)
• Kaloi, L. (2011). U.S. Department of Education finally backs away from a policy that
masks student performance. Retrieved from NCLD website
http://www.ncld.org/archive/entry/1/149
• Lazarus, S. S., Cormier, D. C., & Thurlow, M. L. (2011). States’ accommodations
policies and development of alternate assessments based on modified
achievement standards: A discriminant analysis. Remedial and Special Education,
32, 301-308. doi:10.1177/0741932510362214
• Palmer, P. W. (2009). State perspectives on implementing, or choosing not to
implement, an alternate assessment based on modified academic achievement
standards. Peabody Journal of Education, 84, 578-584.
doi:10.1080/01619560903241051
• Texas Education Agency (2011). STAAR Modified Participation Requirements.
Retrieved from TEA website
http://www.tea.state.tx.us/student.assessment/special-ed/staarm/partreqs/
• Thompson, B., Diamond, K. E., McWilliam, R., Snyder, P., & Snyder, S. (2005).
Evaluating the quality of evidence from correlational research evidence-based
practice. Exceptional Children, 71 (2), 181-194.

Unless otherwise indicated, all clipart and images are from Microsoft clipart.

Two Percent Flexibility

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Viewers also liked

Viewers also liked (15)

Similar to Two Percent Flexibility

Similar to Two Percent Flexibility (20)

Recently uploaded

Recently uploaded (20)

Two Percent Flexibility

Editor's Notes