A NEW APPROACH TO

TESTI NG AND ASSESSMENT

August 2014
Teachers have a vital perspective
on testing and assessment. As front-
line observers, we experience how
state assessments work with our
specic student populations. As a
result, we have valuable insight on
how to use testing in schools.
SURAJ GOPAL, ninth-grade STEM special
education teacher, Hudson High School of
Learning Technologies
Executive Summary 1
Introduction 5
Design: Improve the accuracy of
standardized assessments 7
Culture: Create and maintain a positive
testing environment in schools 12
Teaching: Use data to improve instruction 16
Accountability: Include data in critical decisions 20
Conclusion 27
Teacher Policy Team Process and Methodology 30
Notes 31
Teacher Policy Team and Acknowledgements 34
1
P
R
E
F
A
C
E
Standardized testing can be deeply benecial to students, teachers, and schools by providing an
important measure of progress, as well as meaningful feedback about areas of success and areas
of growth. As teachers, we know the costs and benets of assessments. This leaves us between two
sides of an often-heated debate, but this is where the evidence leads us. In short, tests have value
so lets take advantage of them. Here is how:
DESIGN: IMPROVE THE ACCURACY
OF STANDARDIZED ASSESSMENTS
A large body of research shows that well-designed
standardized tests can provide valuable information about
students knowledge and teachers performance. In fact,
such tests are often predictive of long-term life outcomes.
It is essential to ensure that all standardized tests are
well-designed and that feedback from teachers is solicited
during all stages of the testing process.
A common concern is that the accuracy of assessments
is undermined by excessive teaching to the test, which
does not contribute to meaningful learning. However,
there is little evidence that test preparation even produces
signifcantly higher test scores when tests are well-designed
and focused on higher-order skills. Teachers and principals
should be strongly discouraged from teaching to the
test because it neither raises test scores nor results in
genuine learning.
Computer-adaptive testing is an important tool for
improving the accuracy of assessments. Such tests do a
better job than traditional assessments of measuring both
high- and low-achieving students, and should be made
widely available for adoption.
Finally, ensuring the quality of state-created tests is an
iterative process. The vast majority of state test items should
be released publicly so that stakeholders, such as teachers
and parents, can ofer feedback on the exams.
CULTURE: CREATE AND
MAINTAIN A POSITIVE TESTING
ENVIRONMENT IN SCHOOLS
In some schools, the negative culture surrounding
standardized testing is pervasive, undermining the value
of assessments and harming teachers morale and students
motivation. A truly pernicious culture can lead to cheating.
As educators, we must work within our schools to create
a positive culture that recognizes the value of testing for
learning and growth. Best practices should be instituted to
deter, detect, and investigate potential instances of cheating.
Policymakers must address the negative impact of excessive
testing by getting an accurate measure of time spent on
assessment and eliminating unnecessary tests. Moreover, the
use of alternate assessments, including holistic, portfolio-
based exams, should be studied to determine whether
they are compatible with data-driven improvement
and accountability.
TEACHING: USE DATA TO
IMPROVE INSTRUCTION
The data from standardized tests can serve as an important
tool for teachers and administrators. Research suggests that
both teachers and schools beneft from thoughtful use of
data. Data-driven instruction can be improved in a variety
of ways, including: ongoing professional development for
teachers; a dedicated data specialist in each school; and data
that is returned to teachers in a timely, disaggregated, and
accessible manner.
2
ACCOUNTABILITY: INCLUDE
DATA IN CRITICAL DECISIONS
Because test scores are important refections of student
learning, assessment data should be a part of consequential
decisions. In fact, there is a large body of literature showing
the benefts of using tests as part of a multiple measure
accountability framework. However, tests should never be
the sole basis for any high-stakes decision. For example, the
current system of denying graduation to any student who
does not pass all Regents exams is misguided and should
be revised to incorporate multiple measures.
Furthermore, when connecting student test scores to
teacher performance, special care must be taken to isolate
the efect of teachers and exclude the multitude of factors
outside teachers control that afect student performance.
Teachers of traditionally non-tested subjects should be
evaluated using growth measures or student learning
objectives on assessments that are designed with signifcant
input from educators.
CONCLUSION
We believe in the value of standardized assessments
when they are used carefully. They can be a critical
tool for teachers and students alike, and we would be
unwise to discard them. At the same time, policymakers,
administrators, and teachers must invest the time, money,
refection, and work necessary to realize the value
of assessments.
Throughout our teams research, a positive culture
of assessments and data-driven instruction was a key recurring
theme for school success. That culture starts with each of us,
in our own classrooms and buildings, and will only happen if
teachers are invested as active participants in the process of
shaping changes to testing and assessment.
Trevor Baisden, founding fth-grade ELA and history lead teacher,
Success Academy Bronx 2 Middle School
E
X
E
C
U
T
I
V
E

S
U
M
M
A
R
Y
CULTURE
CREATE AND MAINTAIN
A POSITIVE TESTING
ENVIRONMENT IN
SCHOOLS
TEACHING
USE DATA TO IMPROVE
INSTRUCTION
DESIGN
IMPROVE THE ACCURACY
OF STANDARDIZED
ASSESSMENTS
ACCOUNTABILITY
INCLUDE DATA IN
CRITICAL DECISIONS
MAKING GOOD USE OF STANDARDIZED TESTS
4
5
Standardized assessments have increasingly become a part of life for schools across the
country. Since No Child Left Behind became law in 2001, there has been a growing attention
to measuring districts, schools, and students progress, with a particular focus on historically
disadvantaged students.
Critics of this trend suggest doing away with standardized
tests entirely, while many proponents argue that we
simply need to stay the course. As a team of 14 teachers,
committed to elevating our profession and ensuring
students succeed, our response is none of the above. We
are unifed in the belief that testing has signifcant value,
with the understanding that the way tests are currently
designed and used must be improved. In this paper, we lay
out a new vision for testing and assessment, beginning with
the design of assessments and ending with the important
decisions that test results should inform.
In New York, testing has dominated the conversation
about the implementation of new teacher evaluation
programs and the Common Core State Standards. We
fnd ourselves frmly in the middle between those who
would do away with testing altogether and those who
do not acknowledge any faws in the current system. But
we are comfortable in the rational middlecomfortable
with the view that as educators we can beneft from the
information these tests provide. We are comfortable with
the idea that our students growth on tests can be one part
of our evaluations, while using that same data to inform
our teaching decisions. Finally, we believe that a standard
measure can be critical in ensuring equality in education.
We believe that disaggregated assessment data shines a
light on populations of students who are not getting the
education they deserve.
We all have a part to play in changing the substance and
the culture of testing. None of the Above has something
for everyone: teachers and principals, state and district
administrators, elected ofcials and policymakers. In
June 2014, in response to concerns about the role of
standardized tests in teacher evaluation, the New York State
legislature passed a so-called safety net that removes the
impact of state assessments on teachers with the lowest
evaluation ratings for two years. Let us say no to the all-
or-nothing approaches, and make the most of this time to
get these tests right.
Teachers see the impact that testing and assessment have on our
practice and our students. Teachers know rsthand what is best for
our students and our practice. Its important for us to have a voice in
the testing and assessment debate because it has a direct impact on
the daily actions of teachers and students.
Christine Montera, social studies teacher, East Bronx Academy for the Future
7
KEY TAKEAWAYS FROM RESEARCH AND EXPERIENCE
At the core of the debate on testing is a critical question: Are standardized assessments reective
of students learning? Test opponents, on one end of the spectrum, claim they are not indicative of
student learning or achievement.
1
At the other extreme are those who argue that a single assessment
on a single day is the only measure that we should use to make high-stakes decisions.
2
As classroom
teachers, we think the truth falls somewhere in between.
There is abundant research showing that standardized
tests are meaningful. Such assessments can predict with
moderate accuracy individuals frst-year college GPA,
3

cumulative college GPA,
4
post-college income,
5
and
success in graduate school.
6
Aggregate international test
scores are also predictive of the economic prosperity
of countries.
7
Additionally, teachers whose students
standardized test scores grow produce an increase in those
students adult incomes and rates of college attendance.
8

This research shows that standardized tests are able to
capture important information about what is happening
in our classrooms.
Standardized tests, however, are not the be-all and end-all;
they do not measure everything that matters. There are
many students who do not test well and end up leading
happy, successful lives. Research indicates that certain
subjective evaluations of teachers are only modestly
correlated with their students test-based success,
9

suggesting what many teachers know: that tests cannot
measure the full value of an educator. Indeed, the teachers
who have the greatest positive efects on students social
and behavioral skills are not always the ones who produce
the highest test score gains.
10
This is why past E4E-
New York papers on teacher evaluation
11
and Common
Core implementation
12
have insisted on multi-measure
evaluation and decision-making for teachers and students.
There are other limitations to standardized tests, which
we will discuss later in this paper, but, in short, tests are
meaningful but dont measure everything.
I MPROVE THE ACCURACY OF STANDARDI ZED ASSESSMENTS
SUMMARY OF RECOMMENDATIONS
When designing tests, follow best practices such as
ensuring alignment to standards, testing higher-
order thinking, and actively soliciting teacher input.

Prioritize higher-order instruction, and eliminate
excessive test preparation that does not contribute to
meaningful learning.
Use computer-adaptive assessments, which improve
tests accuracy by measuring the growth of low- and
high-performing students.
Release the vast majority of state test items publicly
after the assessment window has closed so that all
stakeholders can monitor the quality of the exams.
=STATE =DI STRI CTS =SCHOOLS
8
RECOMMENDATION:
WHEN DESIGNING TESTS, FOLLOW BEST
PRACTICES SUCH AS ENSURING ALIGNMENT TO
STANDARDS, TESTING HIGHER-ORDER THINKING,
AND ACTIVELY SOLICITING TEACHER INPUT.
All tests are not created equal. Anecdotally, as teachers, all
of us have experience with assessments that were poorly
written or were not aligned with the academic standards.
We also all have experience with many well-designed tests
that were fair assessments of our students learning and our
teaching, and that gave us important data that we were
able to use to improve our instruction.
We were heartened to learn about the process that New
York State test questions (technically called items) go
through before they are ever used on an ofcial exam. It
takes a full two years for each item to be approved through
a process that includes extensive feld testing, statistical
validation, and input from a committee of teachers.
13
It is
disconcerting, however, that even after such a thorough
process, there are still concerns from educators about the
quality of these tests.
14
We are glad the New York State Education Department
uses a committee of teachers to validate testing items. The
opportunity to join such a committee should be widely
disseminated so that as many teachers as possible have the
chance to share their voice.
We also believe that there should be a formal system for
soliciting and receiving teacher commentary so that all
educators can share feedback after a test has been given.
We recommend that the State Education Department
send a survey to all teachers who administered tests to
gather feedback on positive and negative aspects of
the assessments.
Improving tests will only be efective with the active participation of teachers
in testing design on a district and state level. Our ability to share insights from
the classroom, as well as the cultural and socioeconomic backgrounds of our
students, will undoubtedly help create high-quality assessments.
Blackfoot U-Ahk, fourth- and fth-grade teacher of students with severe emotional disabilities,
Coy L. Cox School P.369k
DESIGNING QUALITY ASSESSMENTS
When designing all tests, the following practices must be followed:
Classroom teachers need to provide input throughout the process, from the creation of the tests to
feedback after the tests are given. This feedback must be taken into account and meaningfully acted upon.
Tests must be aligned to standards and assess higher-order thinking skills.
15
The diversity of students backgroundsincluding differences in geography, socioeconomic status, racial
identity, disability status, etc.must be considered in test development in order to avoid potential bias.
Test items should be worded to make sure each item measures the specic standard being assessed,
as opposed to students ability to understand a tricky question.
The amount of time given for assessments and the number of assessments given in a single day need
to be age-appropriate.
D
E
S
I
G
N
9
RECOMMENDATION:
PRIORITIZE HIGHER-ORDER INSTRUCTION
AND ELIMINATE EXCESSIVE TEST PREP.
One of the most serious critiques of standardized
assessments is that excessive teaching to the test can
efectively negate the validity of an exam, as students
learn how to score well without learning meaningful
skills or content. Teaching to the test, or drill and kill,
tends to take valuable time away from rich, higher-order
instruction. No teacher gets into the profession for this
kind of mechanized work, and it undermines teachers
and students love of school.
But contrary to the notion that tests can be gamed by
excessive preparation, research suggests that the best way
to prepare for most standardized assessments is through
challenging, authentic work focused on content and
skills.
16
One study that examined students preparation for
the ACT found that improvements from [an ACT pre-
test] to the ACT are smaller the more time teachers spend
on test preparation in their classes and the more they use
test preparation materials. Moreover, the focus on testing
strategies and practice diverts students and teachers eforts
from what really mattersdeep analytic work in academic
classes.
17
In other words, at least for well-designed
assessments, excessive test preparation may actually lead
to worse results.
This aligns with our experience, as well as recent
statements from education leaders. As New York City
Schools Chancellor Carmen Faria said, If we do good
teaching, thats the best test prep.
18
Similarly, New York
State Education Commissioner John King stated, The
best preparation for testing is good teaching.
19
We agree.
Since there is scant evidence that excessive teaching to
the test will lead to higher assessment results, teachers and
principals need to be shown this research. When educators
realize that test prep is counterproductive, more time will
be spent on authentic teaching and learning.
RECOMMENDATION:
USE COMPUTER-ADAPTIVE ASSESSMENTS.
One valid concern about traditional tests is that they
cannot adequately capture the growth of students who
are signifcantly above or below grade level. The good
news is that technology ofers a solution to this problem
computer-adaptive testing adjusts question difculty based
on students demonstrated skill level. This sort of assessment,
which is already relatively widely used, including by the
Graduate Record Examinations (GRE)
20
and the Graduate
Management Admission Test (GMAT),
21
would help
teachers get a better sense of students growth from year
to year.
22
Similarly, computer-adaptive tests give more
accurate information to students and parents. We therefore
strongly support the use of computer-adaptive testing
whenever available, and encourage investment in this
alternative where it does not exist.
Questions have been raised regarding whether computer-
adaptive testing will lead to low expectations for struggling
students.
23
We understand these concerns, but ultimately
disagree: We are not aware of evidence that educators
will lower expectations for their students simply because
tests focus on academic growth. If, for example, data show
that a certain schools students are not making progress,
eforts can be made to help those students and ensure
that teachers are held accountable. In that sense, more
accurate data will help rather than hinder the improvement
and accountability process. Moreover, there is no clear
alternativestudents who are far behind or far ahead
need a meaningful gauge of their progress, and computer-
adaptive tests provide this.
Computer-adaptive testing is
absolutely crucial because many
of my students are far behind and
would benet from a test scaled
to their abilities.
Rachael Beseda, rst-grade special
education teacher, Global Community
Charter School
D
E
S
I
G
N
10
That being said, it is important that computer-adaptive
assessments give all students a fair opportunity to
engage with grade-level content. All tests should begin
with grade-level questions, and only move down once
it becomes clear that students are not at grade level.
Furthermore, such tests should attempt to push all students
to demonstrate higher-order thinking skills. For example,
a student reading below grade level can still be given the
chance to show the same skills as her grade-level peers,
but do so with a less-challenging text.
RECOMMENDATION:
RELEASE THE VAST MAJORITY OF STATE
TEST ITEMS PUBLICLY AFTER THE ASSESSMENT
WINDOW HAS CLOSED.
All tests, especially those used for making high-stakes
decisions, need to undergo careful scrutiny both before
and after administration. We believe there is a healthy
process in place to ensure quality in the creation of New
York State exams. At the same time, it has been frustrating
for many educators that state tests prohibit teachers and
students from discussing the contents of the exam.
24
Right
now, with low public confdence in tests,
25
the state needs
to allocate funds to signifcantly increase the transparency
of state assessments,
26
except for feld test items, which, by
design, cannot be publicly released. These funds will allow
for the printing of additional forms of state assessments
that will give the state the ability to feld test more items,
decreasing the need to reuse (and thus keep hidden
from public view) previous items. This will allow for the
elimination of the widely criticized
27
stand-alone feld tests.
Increased transparency will let educators, parents, and
students give feedback on state tests, which is particularly
important as the Common Core standards are being
implemented. This also ensures that teachers and students
have a better understanding of what to expect on future
exams. We believe that this will not only improve the
assessments themselves by holding test designers and the
New York State Education Department accountable to the
public, but will also help restore public trust in the exams.
Schools and teachers emphasis should always be on high-quality
rigorous instruction. Both research and experience suggest that this is
the best method for preparing for well-designed assessments.
Vivett Hemans, English and language arts teacher, Eagle Academy for Young Men
of Southeast Queens
D
E
S
I
G
N
WHAT IS COMPUTER-ADAPTIVE TESTING?
Computer-adaptive assessments start all students at the same levelin this case, at their grade levels. However,
questions on the test become progressively harder as the test-taker gets more questions right or progressively
easier as the test-taker gets more questions wrong. That does not mean that if a student gets the rst few
questions wrong, the remainder of the test will be below grade level. Instead, the test continuously adapts based
on the students responses. For example, if a student gets the rst few questions wrong, but the next several
questions right, the difculty level will begin increasing as more correct answers are given. This process
allows assessments to meet students where they are at in order to get an accurate measure of their learning
and growth.
ADDITIONAL BENEFITS OF TESTING
Our paper is organized around the two main benets of standardized
assessments: using them for improvement and as a factor in important decisions.
However, we would be remiss if we did not discuss some smaller, but important
additional benets of testing.
Assessments provide evidence of achievement and opportunity gaps. Using
both the NAEP and state tests mandated by No Child Left Behind, policymakers
and concerned citizens have quantitative evidence of the inequities that persist
in our country. Testing not only shows that this is the case, but also helps
quantify the gap and determine whether it is expanding, contracting, or staying
constant. While qualitative evidence is also important in this regard, test scores
can provide the hard data necessary to bring light to the shameful inequities
that persist in our country.
Standardized tests are important to prepare students for success in adult life.
Not only must college-bound students take the SAT or ACT, but all those who
aspire to graduate school must take additional exams. Potential lawyers must
take the LSAT and the bar exam; would-be doctors must do well on the MCAT
and board exams. The list goes on and includes most professions. That is not
to say that the purpose of K12 education should be to prepare students for
assessments, but we would be doing a disservice if we limit students exposure
to the types of high-stakes tests they need to do well on later in life.
There is some evidence that assessments do not simply measure learning,
but actually enhance it. A variety of studies
28
have found that students retain
information better after being tested on it. At this point, it is not clear that this
research applies to standardized tests, but it is a potential value that points to
the necessity of aligning standards what is taught in classto what is tested.
12
C
U
L
T
U
R
E
CREATE AND MAI NTAI N A POSI TI VE
TESTI NG ENVI RONMENT I N SCHOOLS
KEY TAKEAWAYS FROM RESEARCH AND EXPERIENCE
Many of our experiences suggest that in too many instances, the culture of testing and assessment
in New York has turned toxic. No doubt this is not the case in all schools, but for too many of us,
testing has become something to be feared and avoided.
But it does not have to be that way. The negative culture
of testing that permeates some schools must change.
We believe that part of this shift has to come from us
as teachers: We should be focusing on the value that
assessments have to ofer. We cannot be surprised that a
pessimistic culture exists in schools if the adults in those
buildings have counterproductive attitudes about testing.
Teachers cannot solve this problem alone, however. We
need principals to do their part, by setting a positive
building-wide tone about assessments. Moreover, as
discussed earlier, we need principals to communicate
clearly to teachers that excessive test prep will not raise test
scores. Currently, though, it is often principals who
mandate that teachers engage in this counterproductive
practice, feeding a negative cycle that harms
student engagement.
As we will discuss further in a subsequent section, teachers
also need to be given the tools to use test results to
improve instruction. When teachers are supplied with what
we need to make tests valuable, our outlook will change
for the better. Moreover, part of the anxiety that surrounds
testing comes from the feeling that a single test can
determine our students futures. A commitment to using
multiple measures for all high-stakes decisionsanother
topic we will elaborate on in a later sectionwill go a
long way toward eliminating this fear.
SUMMARY OF RECOMMENDATIONS
Measure time spent, by both students and teachers,
on testing and eliminate unnecessary and redundant
exams.
Implement best practices, such as administering
tests in controlled environments and monitoring for
test irregularities, to prevent and detect cheating.

Create or expand pilot programs of schools using
nontraditional tests to determine whether they lead
to positive results for students, and can be used to
evaluate and support teachers and schools.
=STATE =DI STRI CTS =SCHOOLS
13
Finally, accountability must be paired with support
throughout the year. What if teachers and students did
not feel that low test scores would lead to punishments or
poor ratings, but that they would lead to increased support
and resources? To be clear, we do believe in accountability,
but accountability should always go hand-in-hand with
support and resources. Tests should be instructive, as well
as evaluative. It is outside the scope of this paper to address
what such support should look like specifcally, but this
should be a core tenet of any accountability system.
RECOMMENDATION:
MEASURE TIME SPENT, BY BOTH STUDENTS
AND TEACHERS, ON TESTING AND ELIMINATE
UNNECESSARY AND REDUNDANT EXAMS.
One cause of the general frustration directed at
standardized tests is the widespread feeling that there are
simply too many of them. We certainly feel that way. As we
have elaborated, we believe there is value in assessment, but
any such value must be weighed against the time and efort
invested in testing.
The frst and most important step must be to accurately
gauge how much time is being spent on testing. We were
glad that New York State Governor Andrew Cuomos
Common Core Implementation Panel attempted to
address the underlying problem by recommending a 2
percent limit on school time spent on local and state
assessments combined, and a 2 percent limit on test prep.
29

These suggested changes were subsequently implemented
in the State Budget.
30

The goal here is laudable, but we are skeptical of an
arbitrary percentage that does not vary by grade. That is
why we need a genuine fgure for just how much time
and money are spent on testingthis should include
time spent preparing, administering, and grading these
assessments for teachers; money spent developing the
test; time spent by students taking tests (including feld
tests); and instructional time lost on days when tests
are administered.
We think the state took a step in the right direction by
requiring an audit of assessments to make sure districts
are not giving unnecessary assessments based on the
assumption that they are mandated by the state.
31
It is
important that this audit is prioritized so that excessive
testing is reduced as soon as possible.
Once these two audits are complete, districts can make
smart decisions, with the input of teachers, about which
tests are worthwhile and which are not.
14
RECOMMENDATION:
IMPLEMENT BEST PRACTICES TO PREVENT AND DETECT CHEATING.
Though the vast majority of educators regularly administer assessments with honesty and fdelity, an extreme outgrowth
of a counterproductive school culture manifests itself in cheating scandals, which have occurred throughout the country.
32

Some have taken these cheating scandals to mean that standardized tests should be eliminated, but this makes no more
sense than cancelling fnal exams because a handful of students tried to cheat on them. Instead, we should institute best
practicesbased on a U.S. Department of Education symposium on test integrity
33
to ensure that cheating rarely
happens, and how to detect and investigate it if it does.
In order to
PREVENT CHEATING,
34

the state, districts, and schools must:
In order to
DETECT EVIDENCE
OF CHEATING,
35
the state, districts, and schools must:
In order to
INVESTIGATE CHEATING,
36
the state, districts, and schools must:
Develop and disseminate a
standard denition of cheating.
Monitor test results for
irregularities as part of
the testing process.
Establish procedures for
conducting an investigation
if one is necessary.
Train principals and
teachers to administer exams.
Ensure that proctors look for
evidence of irregularities during
assessment administration.
Create standards that will
trigger an investigation.
Keep testing windows short.
Use advanced analytic techniques,
such as erasure analysis, to check
for irregularities.
Provide whistleblower protections.
Administer tests in
controlled environments.
Use trained personnel
to conduct the investigation.
Establish and monitor a chain
of custody for testing materials.
Make the investigation as
transparent as possible.
Store and score test
materials of-site.
Make use of sanctions
when wrongdoing is found.
In sum, these best practices, created by experts in the feld, will help stop cheating in the frst place, while ensuring a fair
process if testing irregularities are found. We emphasize, though, that an ounce of prevention is worth a pound of cure
herea healthy testing culture will go a long way toward eliminating this problem.
RECOMMENDATION:
CREATE OR EXPAND PILOT PROGRAMS OF
SCHOOLS USING NONTRADITIONAL TESTS.
One serious problem with traditional standardized tests,
which often include multiple-choice questions, is that
it can be difcult to continually engage students in such
exams. For students and teachers, so-called bubble tests
have become a chore that must be endured. As discussed
earlier, we believe that schools have an important role in
changing this culture. At the same time, alternatives to
traditional assessments should be explored and tested for
their efectiveness.
The New York Performance Standards Consortium
is a group of 28 schools that have used performance
assessments in place of traditional high-stakes tests.
37
The
Consortium schools boast impressive results, showing
C
U
L
T
U
R
E
15
C
U
L
T
U
R
E
Policymakers should embrace a pilot program for portfolio assessment in
order to see whether this type of assessment can work. I think that project-
based learning and inquiry-based work are things I dont do nearly enough
I rely on more traditional assessments, and teachers need to think of ways to
cater to all students needs and strengths in terms of assessment.
Charlotte Steel, seventh-grade math teacher, Booker T. Washington M.S. 54
their students graduate high school at higher rates than
other demographically similar New York City students.
38

But the fact that these schools produce strong graduation
rates does not mean that performance assessments are
the cause. Moreover, legitimate questions have been
raised regarding the ability to fairly and efciently use
performance assessments to evaluate teachers and assess
student learning.
39

We therefore propose an expanded pilot program that
allows more schools to enter into the Performance
Standards Consortium, while also determining whether
such assessments are compatible with data-driven
improvement and accountability. We recommend opening
up an application for schools interested in joining the
program, and conducting a lottery in order to randomly
accept half of the eligible applicant schools into the pilot.
Under this approach, schools that adopt the performance
assessment model can be evaluated against similar schools
that do not. If this system gets positive results for teachers
and students, it should be expanded to even more
city schools.
16
KEY TAKEAWAYS FROM RESEARCH AND EXPERIENCE
Research is clear that assessment data can be used as a tool for teachers and schools to improve.
40

It has been found, for example, that schools that make thoughtful use of data often produce
signicant gains in student achievement.
41
Research also suggests that access to data can
increase the quantity and quality of conversations that educators have with colleagues, parents,
and students.
42
Data can enhance collaboration among educators
43
and can improve teachers
instruction.
44
There is also evidence that the most successful charter schools make use of data-driven
improvement and instruction.
45
Overall, data can and should be used to help schools and
teachers improve.
46
Unfortunately, this is not always happening. One recent
study found that a new data system introduced in
Cincinnati Public Schools was rarely used by educators and
did not lead to observable student gains.
47
A pilot program
in Pennsylvania produced similar results.
48
The key, then, is
to give teachers the support we need to make good use of
testing data.
RECOMMENDATION:
OFFER HIGH-QUALITY TRAINING THROUGHOUT
THE YEAR FOR TEACHERS ON HOW TO IMPROVE
INSTRUCTION USING ASSESSMENT DATA.
Teachers and administrators need more training on how
to use data efectively. The New York City teachers
contract recently put in place more time for professional
development.
49
Some of that time should be dedicated to
high-quality training on understanding and using student
USE DATA TO I MPROVE I NSTRUCTI ON
SUMMARY OF RECOMMENDATIONS
Ofer high-quality training throughout the year
for teachers on how to improve instruction using
assessment data.
Provide each school with a teacher who serves as a
data specialist.
Ensure that teachers and administrators receive
timely, detailed, and disaggregated data in a
transparent, accessible format.
=STATE =DI STRI CTS =SCHOOLS
T
E
A
C
H
I
N
G
17
data. It is worth noting that while we support school-based
creation of professional development, this may be an area
in which schools need outside support and expertise to
design appropriate programs.
RECOMMENDATION:
PROVIDE EACH SCHOOL WITH A TEACHER WHO
SERVES AS A DATA SPECIALIST.
Teachers need continuous support in using data systems.
We need more than a one-time training. We propose that
at least one teacher in each school receive the designation
of data specialist. This role should come with extensive
training, as well as the responsibility of supporting
and working with staf to use data and integrate this
information into their regular assessment ofand feedback
fortheir students. Additionally, data specialists should
receive compensation for this role that is either monetary
or in the form of a lighter class load. A fnal beneft is that
this position could potentially serve as an additional rung
on a teacher career ladder, a concept that past E4E Teacher
Policy Teams have endorsed.
50
RECOMMENDATION:
ENSURE THAT TEACHERS AND ADMINISTRATORS
RECEIVE TIMELY, DETAILED, AND DISAGGREGATED
DATA IN A TRANSPARENT, ACCESSIBLE FORMAT.
To make full use of assessments, teachers and administrators
need timely, detailed, and disaggregated data in order to
tailor their instruction to address their students needs. The
current system does not supply educators with sufciently
detailed feedback on these exams. Compounding this
problem is the fact that the results do not come back
until the summer, and thus teachers often cannot act
on the data. A high priority must be placed on giving
educators actionable, disaggregated, and timely results
from standardized assessments. Teachers also need access
to a high-quality, easily navigable interface in which we
can access all relevant data. Georgia, in particular, is a state
that has been highlighted for its success in making data
accessible and easy to use for teachers,
51
and New York
should follow suit.
T
E
A
C
H
I
N
G
It is particularly important that teachers receive
thorough and useful training in data-driven instruction.
Unless the results of assessments are used to move
teaching and learning forward, they serve little value.
Michelle Knifn, ninth- to 12th-grade math teacher,
High School of Telecommunication Arts and Technology
COMMON CORE ASSESSMENTS CONSORTIUM
As the Common Core State Standards are being implemented across the country, new testing
consortia are being rolled out that are aligned to the new standards. There are two testing groups:
Smarter Balanced Assessment Consortium (SBAC),
52
which has been adopted at least in part by 20
states,
53
and the Partnership for Assessment of Readiness for College and Careers (PARCC),
54
which
has been adopted by 14 states and the District of Columbia.
55
Field tests took place in the spring of
2014,
56
and the full assessments will be available for use beginning in the 20142015 school year.
New York State has adopted PARCC,
57
but has not yet determined when the new exams will be rolled
out.
58
Below, we discuss aspects of PARCC and how they align with our recommendations:
We are encouraged that PARCC assessments appear to test higher-order thinking skills. Although it
is too early to determine for sure, the sample questions
59
leave us optimistic that rigorous skills will
be tested, and low-level multiple-choice tests will be deprioritized.
It is very important that PARCC continuously involves teachers in the creation and revision of the
exams. PARCC has already shown evidence of having engaged teachers throughout this process,
and we are pleased to see such a clear commitment to teacher input.
60
Moreover, we recommend
that PARCC distribute surveys to teachers at the end of each year to garner feedback on the
years assessments.
Although PARCC tests will be completed using computers, they will not be computer adaptive,
61

with the important exception of optional diagnostic exams. It is disappointing that this valuable
technology will not be utilized for the summative assessments, as PARCC is missing an opportunity
to get accurate growth measures of high- and low-achieving students. Although a PARCC frequently
asked questions document
62
claims that the assessments will measure the full range of student
performance, including the performance of high- and low-achieving students, it is not clear how
they will manage to do so. We urge PARCC to consider moving to computer-adaptive assessments,
particularly in light of the fact that SBAC will be utilizing this technology.
63

An advantage of computer-based assessments is that cheating will be more difcult, since school staff
will not handle or transport physical testing materials.
64
However, new threats to testing security
such as access to the Internetmay exist, and PARCC, in partnership with schools and districts,
must ensure teachers and school leaders are prepared to administer the tests fairly and monitor
for irregularities.
An additional advantage of using computer-based assessments is timely feedback to schools,
teachers, and students. For many questionsones that have clear right or wrong answersthe
data should be available almost immediately. Though for othersperformance tasks, essays, or any
items that require manual gradingthe turnaround will understandably be longer. However, we are
glad that PARCC has stated that its goal is to have data from the performance-based assessments
returned before the end of the school year.
65
It is crucial that PARCC ensures that teachers receive
timely, disaggregated, and user-friendly data.
As we have argued, transparency is a necessary aspect for all important exams, in part to ensure
that the public is given an opportunity to offer feedback on the content and quality of assessments,
and in part to ensure public trust in such assessments. So far, we are encouraged that PARCC has
already released sample tests
66
and plans to release 40 percent of test items each year. We hope the
commitment to transparency continues and expands as full-scale tests are implemented.
20
I NCLUDE DATA I N CRI TI CAL DECI SI ONS
KEY TAKEAWAYS FROM RESEARCH AND EXPERIENCE
There is now abundant evidence that using test score growth as part of a multiple-measure
evaluation and accountability system can benet students. Multiple peer-reviewed studies
67, 68, 69

have found that students benet when adults are held accountable for results.
70
There is also
research showing that teacher evaluation that considers evidence of student learning can be
benecial to students.
71
Finally, and most importantly, evidence suggests that, when designed and
implemented well, accountability systems can impact school quality in a way that leads to long-term
positive effects on students adult incomes.
72
All that being said, the current way that test scores are
used to make important decisions needs to be improved to ensure they are fair to students, teachers,
and schools.
RECOMMENDATION:
ISOLATE THE EFFECTS OF TEACHERS AND
SCHOOLS TO ENSURE THAT THOSE SERVING
AT-RISK STUDENT POPULATIONS ARE NOT
PENALIZED BY OUT-OF-SCHOOL FACTORS.
One of the most difcult, but most important, aspects of
using student test score growth in an evaluation system is
isolating the efects of schools and teachers. After all, many
factorsincluding poverty and parental involvement
afect a given students achievement, and only a fraction
can be attributed to his teachers or his school. Indeed,
only about one-ffth to one-quarter of student test scores
are explained by the quality of their schools, and of that,
about one-half to two-thirds are the result of the students
individual teachers.
73
SUMMARY OF RECOMMENDATIONS
Isolate the efects of teachers and schools to ensure
that those serving at-risk student populations are not
penalized by out-of-school factors.
Evaluate teachers of non-tested subjects based on
authentic assessmentsdeveloped and validated
by teachersusing growth measures or student
learning objectives.
Make high-stakes decisions based on multiple
sources and multiple years of evidence.
=STATE =DI STRI CTS =SCHOOLS
We are not saying that teachers and schools do not matter.
But we also cannot blame those same teachers and schools for
all the factors that can contribute to low student achievement.
If we simply look at absolute test scores, as often happens,
74
with
no accounting for growth or student background, the schools
and teachers working with our most challenging students will
be unfairly penalized. Moreover, some struggling schools
and teachers who work with high-achieving students will
be overlooked.
75
With the use of value-added modeling,
76
we can go a long
way toward isolating teachers and schools efects by controlling
for students prior tests scores, as well as other factors outside
teachers control.
WHAT IS VALUE ADDED?
Value added is a statistical method that
attempts to isolate teachers inuence
on their students test score growth.
Value-added models can take into
account a variety of variables that afect
students performance, including prior
achievement, socioeconomic status,
disability status, special education
status, attendance, disciplinary record,
and class size.
77
Although some critics
of value-added measures correctly point
out that teachers ratings can vary from
year to year,
78
others respond that this
can be ameliorated through multiple
years of data, and that similar variance
exists in performance metrics of other
professions.
79
Value-added scores are
particularly reliable for teachers at the
extremes of the distribution.
80
Research
also suggests that teachers value-added
scores predict their efects on students
long-term outcomes such as income and
college attendance.
81
I N-SCHOOL
FACTORS
At l east hal f
of i n-school
efect i s
based on
students
i ndi vi dual
teachers.
OUT-OF-SCHOOL
FACTORS
UNEXPLAI NED
VARI ATI ON
20%
20%
60%
FACTORS
CONTRIBUTING TO
STUDENT ACHIEVEMENT
Source: Di Carlo, M. (2010, July 14) Teachers Matter, But So Do Words.
Shanker Blog. Retrieved from http://shankerblog.org/?p=74. (Note that these
percentages are approximations.)
22
EXAMPLE: TWO-STEP VALUE-ADDED MODEL
In recent years, as New York has started using a
student growth model to evaluate teachers, concerns
have been raised about the extent to which it fairly
accounts for factors outside of educators and
schools control.
82
A report on the subject found
evidence that the 20122013 New York State growth
measure may have been partially biased against
some teachers and principals who serve certain
student populations.
83
With New York State likely
to use value-added scores as 25 percent of teacher
evaluation in the 20142015 school year,
84
now is the
time to consider the ideal model.
We recommend an approach that more fully
accounts for factors outside teachers and schools
control. This methodknown as a two-step value-
added model, or proportionalityis designed to
make apples-to-apples comparisons.
85
In other
words, this model eliminates any correlation
between teachers and schools value-added
scores and the student populations they teachit
guarantees that educators of, for example, students
in poverty or students with disabilities will not
receive disproportionately low ratings.
This will address the concern that student
achievement measures penalize teachers and
schools who serve certain student populations.
It will also ensure that evaluation measures will not
exacerbate persistent inequities in those schools
high-poverty schools will have a tougher time
recruiting and retaining teachers if those educators
face a higher chance of a low evaluation score.
We recognize that genuine inequalities persist
between and within our schools,
86
and that
correlations between teacher efectiveness scores
and student populations likely reect some genuine
diferences in teacher quality. But our goal in an
evaluation system is not just to get an accurate
picture of teacher quality, but also to design a system
that provides useful information to support teacher
and school improvement, while helping districts and
principals make retention and dismissal decisions.
We are convinced that the two-step model does
just that.
87
23
Sample for all growth measures is 1,846 schools
Source: Ehlert, M., Koedel, C., Parsons, E., et al. Selecting Growth Measures for School and Teacher Evaluations: Should Proportionality Matter? National
Center for Analysis of Longitudinal Data in Education Research. Retrieved from http://www.caldercenter.org/publications/upload/wp-80-updated-v3.pdf
The following graphs show three diferent ways of measuring schools student achievement growth. The x-axis
is a measure of school poverty, while the y-axis is a measure of school efectiveness based on the given growth
measure. The shaded areas are scatter plots showing the range of schools scores. The line shows the correlation
between schools level of poverty and their level of efectiveness. Note that these are examples based on schools
in Missouri, so representations of New York schools may vary in certain ways.
A
C
C
O
U
N
T
A
B
I
L
I
T
Y
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
MEDIAN
STUDENT GROWTH
PERCENTILE SGP
ONESTEP
VALUEADDED MODEL
TWOSTEP
VALUEADDED MODEL
M
E
D
I
A
N

S
G
P
75
50
25
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
0. 5
0
0. 5
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S

S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S
0. 5
0
0. 5
0 20 40 60 80 100
0 20 40 60 80 100
0 20 40 60 80 100
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
MEDIAN
STUDENT GROWTH
PERCENTILE SGP
ONESTEP
VALUEADDED MODEL
TWOSTEP
VALUEADDED MODEL
M
E
D
I
A
N

S
G
P
75
50
25
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
0. 5
0
0. 5
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S

S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S
0. 5
0
0. 5
0 20 40 60 80 100
0 20 40 60 80 100
0 20 40 60 80 100
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
MEDIAN
STUDENT GROWTH
PERCENTILE SGP
ONESTEP
VALUEADDED MODEL
TWOSTEP
VALUEADDED MODEL
M
E
D
I
A
N

S
G
P
75
50
25
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
0. 5
0
0. 5
PERCENT OF STUDENTS
ELI GI BLE FOR FREE OR
REDUCED- PRI CE LUNCH
S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S

S
T
A
N
D
A
R
D

D
E
V
I
A
T
I
O
N
S
0. 5
0
0. 5
0 20 40 60 80 100
0 20 40 60 80 100
0 20 40 60 80 100
COMPARING DIFFERENT GROWTH MEASURES
RECOMMENDATION:
EVALUATE TEACHERS OF NON-TESTED
SUBJECTS BASED ON AUTHENTIC ASSESSMENTS
DEVELOPED AND VALIDATED BY TEACHERS
USING GROWTH MEASURES OR STUDENT
LEARNING OBJECTIVES.
Many educators do not teach in grades or subjects that
have annual state tests, and therefore cannot be evaluated
using value-added measures. In order to comply with the
new evaluation law, some teachers are being rated based
on students or subjects they do not teachfor example,
in some cases gym teachers are being rated on English
scores.
88
This practice must stop, because it violates a core
tenet
89
of any accountability system: Teachers should not
be held accountable for outcomes outside of our control.
90

We are glad that the New York City teachers contract will
move accountability in this direction.
91
Districts need to invest in authentic performance measures
for teachers in non-tested subjects, particularly ones
like music, art, and physical education. In many cases,
these performance assessments may be combined with
more traditional written tests. Results should not only
be considered in individual teachers evaluations, but
school evaluations as well. The creation of standardized
performance assessments for these subjects has been
experimented with,
92
though the evidence is limited on
how successful such programs have been. In all non-tested
subjects, evaluations should be based on student learning
objectives
93
or measures of student growth that ensure fair
comparisons are being made across classrooms.
Our top priority is to ensure that any such assessments
are designed by and with teachers, and are validated by
teachers. Educators should have a hand in the design, the
administration, and the revision of these assessments. This
is absolutely essential. When teachers are involved in the
creation of exams, the tests are more likely to refect what
is being taught in the classroom.
EXAMPLE: HIGH SCHOOL GRADUATION EXAMS
The current requirement that all New York State students pass a series of exams in order to
graduate high school is an example of a policy that fails to consider multiple measures. Under the
current system, students will only receive a high school diploma if they pass ve state-mandated
Regents exams.
98
(Students with disabilities or IEPs have some limited additional options.)
This policy is designed to create high expectations for studentsan admirable goalbut it ends
up harming some of them. Anecdotal
99
and empirical
100, 101, 102, 103
evidence show that high school
graduation exams have little or no positive efects and signicant negative consequences for
students who fail such tests. There is even alarming research showing that mandated graduation
exams can lead to increased incarceration rates.
104
With this evidence in mind, we take the position that high school graduation exams should never
be the sole basis for denying students their diplomas. It is appropriate for such tests to be part
of a multiple-measure graduation system, but not as inexible roadblocks for students trying to
graduate. It is outside our scope to discuss what precisely such a system should look like, but we
will note that, holistic multi-measure graduation models exist and should be studied.
105

RECOMMENDATION:
MAKE HIGH-STAKES DECISIONS BASED
ON MULTIPLE SOURCES AND MULTIPLE
YEARS OF EVIDENCE.
We believe in the value of test scores to inform and
evaluate students, teachers, principals, and schools, but we
also are convinced that a single test score should not be
the sole basis for any high-stakes decision. A broad array
of theory and evidence suggests that multiple measures are
always preferable in high-stakes circumstances.
94
We are
encouraged, then, that New York Citylike all districts
and states that have adopted the new wave of teacher
evaluation
95
has used a multiple measure system, with
student growth as one factor among others.
96
Similarly, we
are glad that the New York City Department of Education
recently adopted a multiple-measure system for student
promotion and retention decisions.
97
We think the city
and state have done a good job ensuring that important
decisions are based on multiple sources of evidence.
Nevertheless, there is room for improvement.
Using multiple measures for high-stakes decisions is
particularly important to me and my students because so many
ELLs often struggle on tests but are bright, capable students.
Maura N. Henry, sixth- to 12th-grade English as a Second Language teacher,
The Young Women's Leadership School of Astoria
UNIQUE STUDENT POPULATIONS
One important aspect of assessment that is not discussed enough is the
efect on unique populations of students, including those receiving special
education, students with disabilities, English-language learners, and
gifted and talented students. A thorough discussion of issues surrounding
testing with each of these student populations is beyond the scope of this
paper. However, we were very cognizant of these students while crafting
our recommendations. Here, we highlight and elaborate on how specic
components of our recommendations afect these students.
In the design of tests, the needs of unique populations of students must be
carefully considered. First and foremost, teachers of a variety of student
populations should be represented on the panel of educators who design
and review assessments. Particular care must be given in writing test
items to ensure that certain students are not disadvantaged. For example,
math tests should not, in most cases, include idioms that English-language
learners might not be familiar with, since such a question would not
measure those students mathematical ability.
As we have previously articulated, we believe in the value of computer-
adaptive testing. These assessments will benet unique student
populationsspecically those who are low- and high-achievingby
gauging their growth accurately. This needs to be a high priority. If we want
students and teachers to believe in the value of the assessments, we need to
make them useful to all students. Computer-adaptive tests will signicantly
help in this regard.
Our recommendation regarding the use of multiple measures in making
high-stakes decisionsspecically graduation decisionswill have a
positive effect on unique populations of students.
106
English-language
learners and special education students have long graduated at lower rates
than other students. The move to a multiple-measure system will not solve
this problem, but it will give all students multiple avenues to demonstrate
their knowledge of the content necessary to graduate.
When teacher input is sought out
and reflected in assessments and
their implementation, tests will
become an effective tool to accurately
gauge student achievement and
growth, as well as an empowering
tool for the teachers to improve
their teaching practices.
I RI S WON, ninth- to 12th-grade mathematics
and technology teacher, Renaissance High School
for Musical Theater & Technology
As teachers, this is our vision for making full use of standardized assessments
for taking advantage of a powerful tool that requires careful execution.
Tests can be a force for good, and we would be unwise to throw them out of our
toolbox. At the same time, they cannot be our only tool. We cannot use a hammer
when a wrench is necessary, and we will usually need both.
Improving how tests are used is a shared responsibility. As teachers, we must do
our partadminister tests with fdelity, use data to improve when it is available,
and advocate for better assessments when necessary. But policymakers must also
step upthey must provide us with the support we need, and they must make
wise decisions about how often tests are administered and how results are used.
This will take time, money, refection, and a lot of work.
Lets get started.
28
KEY TAKEAWAYS
Tests are useful, though imperfect,
measures of students learning and
teachers efectiveness.
The accuracy of tests is directly
related to test qualitywell-
designed assessments provide
important information, but poorly
designed tests have little to no use.
RECOMMENDATIONS
When designing tests, follow
best practices such as ensuring
alignment to standards, testing
higher-order thinking, and actively
soliciting teacher input.
Prioritize higher-order instruction,
and eliminate excessive test
preparation that does not
contribute to meaningful learning.
Use computer-adaptive
assessments, which improve
tests accuracy by measuring
the growth of low- and high-
performing students.
Release the vast majority of
state test items publicly after the
assessment window has closed so
that all stakeholders can monitor
the quality of the exams.
CULTURE
CREATE AND MAINTAIN
A POSITIVE TESTING
ENVIRONMENT IN
SCHOOLS
TEACHING
USE DATA TO IMPROVE
INSTRUCTION
DESIGN
IMPROVE THE ACCURACY
OF STANDARDIZED
ASSESSMENTS
ACCOUNTABILITY
INCLUDE DATA IN
CRITICAL DECISIONS
CULTURE
CREATE AND MAINTAIN
A POSITIVE TESTING
ENVIRONMENT IN
SCHOOLS
TEACHING
USE DATA TO IMPROVE
INSTRUCTION
DESIGN
IMPROVE THE ACCURACY
OF STANDARDIZED
ASSESSMENTS
ACCOUNTABILITY
INCLUDE DATA IN
CRITICAL DECISIONS
KEY TAKEAWAYS
Student achievement is a useful
measure that should be a part
of a multi-measure evaluation
framework that holds teachers
and schools accountable for
student performance.
Holding schools and teachers
accountable for students
performance produces
positive results.
RECOMMENDATIONS
Isolate the efects of teachers and
schools to ensure that those serving
at-risk student populations are not
penalized by out-of-school factors.
Make high-stakes decisions based
on multiple sources and multiple
years of evidence.
Evaluate teachers of non-tested
subjects based on authentic
assessmentsdeveloped and
validated by teachersusing
growth measures or student
learning objectives.
KEY RESEARCH TAKEAWAYS AND OVERVI EW OF RECOMMENDATI ONS
29
KEY TAKEAWAYS
The toxic culture of testing
that pervades some schools
undermines the value of
assessments and harms
teachers morale.
A positive culture begins
with viewing assessments as
opportunities for growth, and also
requires policymakers to create an
environmentthrough support and
thoughtful decision-makingthat
encourages a healthy culture.
RECOMMENDATIONS
Measure time spent, by both
students and teachers, on testing,
and eliminate unnecessary and
redundant exams.
Implement best practices, such as
administering tests in controlled
environments and monitoring for
test irregularities, to prevent and
detect cheating.
Create or expand pilot programs
of schools using nontraditional
tests to determine whether they
lead to positive results for students,
and can be used to evaluate and
support teachers and schools.
CULTURE
CREATE AND MAINTAIN
A POSITIVE TESTING
ENVIRONMENT IN
SCHOOLS
TEACHING
USE DATA TO IMPROVE
INSTRUCTION
DESIGN
IMPROVE THE ACCURACY
OF STANDARDIZED
ASSESSMENTS
ACCOUNTABILITY
INCLUDE DATA IN
CRITICAL DECISIONS
CULTURE
CREATE AND MAINTAIN
A POSITIVE TESTING
ENVIRONMENT IN
SCHOOLS
TEACHING
USE DATA TO IMPROVE
INSTRUCTION
DESIGN
IMPROVE THE ACCURACY
OF STANDARDIZED
ASSESSMENTS
ACCOUNTABILITY
INCLUDE DATA IN
CRITICAL DECISIONS
KEY TAKEAWAYS
When used properly,
assessment data is valuable for
improving teachers practice,
and provides helpful information
to administrators, parents,
and students.
Teachers and administrators
need more support in using data to
inform their practice and ensure it
is meaningful.
RECOMMENDATIONS
Ofer high-quality training
throughout the year for teachers on
how to improve instruction using
assessment data.
Provide each school with a teacher
who serves as a data specialist.
Ensure that teachers and
administrators receive timely,
detailed, and disaggregated data in
a transparent, accessible format.
KEY RESEARCH TAKEAWAYS AND OVERVI EW OF RECOMMENDATI ONS
30
IDENTIFYING E4ES POLICY FOCUS
E4E surveyed members and held focus groups with
E4E-NY members to determine the most important
policy issues from teachers perspective.
OUR PROCESS
We met for eight weeks to review research on diferent
facets of testing and assessment, particularly as they relate
to New York City and State. We considered evidence
from diferent perspectives, held small and large group
discussions, and regularly challenged each others thinking.
We ended up with four main categories under which we
elaborate upon specifc recommendations.
P
R
O
C
E
S
S

A
N
D

M
E
T
H
O
D
O
L
O
G
Y
31
N
O
T
E
S
1
For one example, see: D. Ravitch. (2014, January 18). Do International
Test Scores Matter? (Weblog post). Retrieved from http://dianeravitch.
net/2014/01/18/do-international-test-scores-matter/ (Readers of this
blog know that I have repeatedly argued that standardized scores on
international tests predict nothing about the future.)
2
Short, A., Campanile. C. (2014, April 9). Bloomberg-era tests no longer
top criteria for student promotion: Faria. New York Post. Retrieved
from http://nypost.com/2014/04/09/city-scraps-bloombergs-
standardized-tests/
3
Sackett. P.R., Kuncel, N.R., Beatty, A.S., et al. (2012, April 2). The Role
of Socioeconomic Status in SAT-Grade Relationships and in College
Admissions Decisions. Psychological Science, 23(9), 1000-1007. doi:
10.1177/0956797612438732
4
Schmitt, N., Keeney, J., Oswald, F.L., et al. (2009, November). Prediction
of 4-year college student performance using cognitive and noncognitive
predictors and the impact on demographic status of admitted students.
Journal of Applied Psychology, 94(6), 1479-97. doi: 10.1037/a0016810.
5
Robertson, K.F., Smeets, S., Lubinski, D., et al. (2010, December).
Beyond the Threshold Hypothesis Even Among the Gifted and Top
Math/Science Graduate Students, Cognitive Abilities, Vocational Interests,
and Lifestyle Preferences Matter for Career Choice, Performance, and
Persistence. Current Directions in Psychological Science, 19(6), 346-51.
doi: 10.1177/0963721410391442
6
Kuncel, N.R., Hezlett, S.A. (February, 2007). Standardized Tests Predict
Graduate Students Success.315(5815).DOI: 10.1126/science.1136618
7
Hanushek, E.A., Jamison, D.T., Jamison, E.A., et al. (2008, Spring).
Education and Economic Growth. Education Next, 8(2). Retrieved from
http://educationnext.org/education-and-economic-growth/
8
Chetty, R., Friedman, J.N, Rockof, J.E. (2011). The Long-Term Impact
of Teachers: Teacher Value-Added and Student Outcomes in Adulthood.
American Economic Review. Retrieved from http://obs.rc.fas.harvard.
edu/chetty/value_added.html
9
Master, J. (2014, June). Stafng for Success. Education Evaluation
and Policy Analysis. Retrieved from http://epa.sagepub.com/
content/36/2/207.abstract?rss=1
10
Jennings, J.L., DiPrete, T.A. (2009, March 15). Teacher Efects on Social/
Behavioral Skills in Early Elementary School. Retrieved from http://
www.columbia.edu/~tad61/Jennings%20and%20DiPrete_3_15_2009_
Final.pdf
11
Adland, J., Braslow, D., Brosbe, R., et al. (Spring, 2011). Beyond
Satisfactory: A New Teacher Evaluation System for New York. Retrieved
from http://educators4excellence.s3.amazonaws.com/8/3f/b/1362/
E4E_Evaluation_Paper_Final.pdf
12
Barraclough, N., Farnum, C., Loeb, M., et al. (Spring, 2014). A
Path Forward: Recommendations from the classroom for efectively
implementing the Common Core. Retrieved from http://
educators4excellence.s3.amazonaws.com/8/0b/a/2258/03.24.14_TAT_
CCSS_Memo.pdf
13
New York State Department of Education. (2014, July 9). New York
State Education Department Test Development Process. Retrieved from
http://www.p12.nysed.gov/assessment/teacher/home.html#process
14
See for example: Phillips. E. (2014, April 9). We Need to Talk About
the Test: A Problem With the Common Core. The New York Times.
Retrieved from http://www.nytimes.com/2014/04/10/opinion/the-
problem-with-the-common-core.html; and Hartocollis, A., (2012, April
20). When Pineapple Races Hare, Students Lose, Critics of Standardized
Tests Say. The New York Times. Retrieved from http://www.nytimes.
com/2012/04/21/nyregion/standardized-testing-is-blamed-for-question-
about-a-sleeveless-pineapple.html?pagewanted=all
15
King, F.J., Goodson, L., Rohani, F. Higher Order Thinking Skills.
Center for Advancement of Learning and Assessment. Retrieved from
http://www.cala.fsu.edu/fles/higher_order_thinking_skills.pdf
16
Newmann, F.M., Bryk, A.S., Nagaoka, J. (2001, January). Authentic
Intellectual Work and Standardized Tests: Confict or Coexistence?
Retrieved from http://ccsr.uchicago.edu/publications/authentic-
intellectual-work-and-standardized-tests-confict-or-coexistence
17
UChicagoNews. (2008, May 27). Intensive ACT test prep during class
leads to lower scores; students dont connect grades, study habits to exam
scores. Retrieved from http://news.uchicago.edu/article/2008/05/27/
intensive-act-test-prep-during-class-leads-lower-scores-students-don-t-
connect-gr
18
Rafter, D. (2014, January 2). De Blasio picks a schools chancellor.
Queens Chronicle. Retrieved from http://www.qchron.com/editions/
queenswide/de-blasio-picks-a-schools-chancellor/article_687e9c54-
a168-54a6-9df7-ebed13034cc2.html
19
Spector, J. (2014, March 24). John King on upcoming Common Core
tests: The best preparation for testing is good teaching. Politics on the
Hudson. Retrieved from http://polhudson.lohudblogs.com/2014/03/24/
john-king-upcoming-common-core-tests-best-preparation-testing-good-
teaching/
20
Graduate Record Examinations. How the test is scored. Retrieved
from https://www.ets.org/gre/revised_general/scores/how/
21
Graduate Management Admission Test. (2010, January 13). The CAT in
the GMAT. Retrieved from http://www.mba.com/us/the-gmat-blog-
hub/the-ofcial-gmat-blog/2010/jan/the-cat-in-the-gmat.aspx
22
Smarter Balanced Assessment Consortium. Computer Adaptive Testing.
Retrieved from http://www.smarterbalanced.org/wordpress/wp-
content/uploads/2011/12/Smarter-Balanced-CAT.pdf
23
Brown, E. (2014, March 2). D.C. mulling over Common Core
test switch. The Washington Post. Retrieved from http://www.
washingtonpost.com/local/education/dc-mulling-over-common-core-
test-switch/2014/03/02/29478710-a0b3-11e3-a050-dc3322a94fa7_story.
html?wprss=rss_education
24
Strauss, V. (2014, April 25). AFT asks Pearson to stop gag order barring
educators from talking about tests. The Washington Post. Retrieved from
http://www.washingtonpost.com/blogs/answer-sheet/wp/2014/04/25/
aft-asks-pearson-to-stop-gag-order-barring-educators-from-talking-
about-tests/
25
Times Union. (2014). Times Union/Siena College Poll [Data File].
Retrieved from http://www.timesunion.com/7dayarchive/item/Times-
Union-Siena-College-education-poll-30096.php
26
We say this with the understanding that it may not be possible for
100% of all items to be released publicly. We are comfortable with a small
number of itemsno more than 10%being held from public view to
ensure comparability from tests year to year.
27
McIntire, M.E.. (2014, June 11) As Pearsons annual feld testing ends,
some want them never to start again. Chalkbeat. Retrieved from http://
ny.chalkbeat.org/2014/06/11/as-pearsons-annual-feld-testing-ends-
some-want-them-never-to-start-again/#.U8Le9FNyjec
28
Roediger, H.L. and Karpicke, J.D. (2006). Test-enhanced learning:
Taking memory tests improves long-term retention. Psychological
Science. Retrieved from http://learninglab.psych.purdue.edu/
downloads/2006_Roediger_Karpicke_PsychSci.pdf
29
Litow, S.S., Flanagan, J., Nolan, C., et al. (2014, March). Putting
Students First: Common Core Implementation Panel Recommendation
Report to Governor Andrew M. Cuomo. Retrieved from http://www.
governor.ny.gov/sites/default/fles/Common_Core_Implementation_
Panel_3-10-14.pdf
30
S. 6356D, (2013). Retrieved from http://open.nysenate.gov/legislation/
bill/A8556d-2013
31
Ibid.
32
Resmovits. J. (2011, August 8). Schools Caught Cheating in Atlanta,
Around the Country. The Hufngton Post. Retrieved from http://www.
hufngtonpost.com/2011/08/08/atlanta-schools-cheating-scandal-
ripples-across-country_n_919509.html
33
Alpert, T., Amrein-Beardsley, A., Bruce, W., et al. (2013). Testing Integrity
Symposium: Issues and Recommendations for Best Practice. Symposium
conducted at meeting of U.S. Department of Education.
32
34
Alpert et al. (2013)
35
Alpert et al. (2013)
36
Alpert et al. (2013)
37
New York Performance Standards Consortium. Retrieved from http://
performanceassessment.org/index.html
38
Educating for the 21st Century: Data Report on the New York
Performance Standards Consortium. Retrieved from http://www.nyclu.
org/fles/releases/testing_consortium_report.pdf
39
Mathews, J. (2004, Summer). Portfolio Assessment: Can it be used to
hold schools accountable? Education Next, 4(3). Retrieved from http://
educationnext.org/portfolio-assessment/
40
Wayman, J.C. (2005). Involving Teachers in Data-Driven Decision
Making: Using Computer Data Systems to Support Teacher Inquiry and
Refection. Journal of Education for Students Placed at Risk, 10(3), 295
308. Retrieved from http://myclass.nl.edu/tie/tie533/teacherdatause.pdf
41
Wayman. (2005)
42
Light, D., Honey, M., Heinze, J. (2005, January). Linking Data and
Learning: The Grow Network Study. Center for Children and Technology.
Retrieved from http://cct.edc.org/publications/linking-data-and-
learning-grow-network-study
43
Chen, E., Heritage, M., Lee, J. (2005). Identifying and Monitoring
Students Learning Needs with Technology. Journal of Education for
Students Placed at Risk, 10(3), 3o9322. Retrieved from http://www.
tandfonline.com/doi/abs/10.1207/s15327671espr1003_6#.U4ijD1Nyjec
44
Datnow, A., Park. V., Wohlstetter, P. (2007). Achieving with Data: How
high-performing school systems use data to improve instruction for
elementary students. Retrieved from http://www.newschools.org/fles/
AchievingWithData.pdf
45
Fryer, R.G. (2012, September). Learning from the Successes and
Failures of Charter Schools. Retrieved from http://scholar.harvard.edu/
fles/fryer/fles/haUsingmilton_project_paper_2012.pdf
46
Data Quality Campaign. (2012, January). Retrieved from http://www.
dataqualitycampaign.org/fles/1357_DQC-TE-primer.pdf
47
Tyler, J.H. (2013). If You Build it Will They Come? Teachers Online
Use of Student Performance Data. Education Finance and Policy,
8(2), 168-207. http://www.mitpressjournals.org/doi/abs/10.1162/
EDFP_a_00089#.U4intlNyjec
48
McCafrey, D.F., Hamilton, L.S. (2007). Value-Added Assessment in
Practice: Lessons from the Pennsylvania Value-Added Assessment System
Pilot Project. Retrieved from http://www.rand.org/content/dam/rand/
pubs/technical_reports/2007/RAND_TR506.sum.pdf
49
United Federation of Teachers. Repurposed workday. Retrieved from
http://www.uft.org/proposed-contract/repurposed-workday
50
Consentino, L., DAmico, J., Fazio, C., et al. (Spring 2014). A Passing
Grade: Teachers Evaluate the NYC Contract. Retrieved from http://
www.educators4excellence.org/nycontract/report
51
Data Quality Campaign. (2014, February). Teacher Data Literacy: Its
About Time. Retrieved from http://www.dataqualitycampaign.org/fles/
DQC-Data%20Literacy%20Brief.pdf
52
Smarter Balanced Assessment Consortium. Retrieved from http://
www.smarterbalanced.org/
53
Smarter Balanced Assessment Consortium. Member States. Retrieved
from http://www.smarterbalanced.org/about/member-states/
54
PARCC. PARCC Online. Retrieved from http://www.parcconline.
org/
55
PARCC. PARCC States. Retrieved from http://www.parcconline.org/
parcc-states
56
Gewertz, C. (2014, March 21). Field-testing Set to Begin on Common
Core Exams. Education Week. Retrieved from http://www.edweek.org/
ew/articles/2014/03/21/26feldtests_ep.h33.html
57
PARCC. New York. Retrieved from https://www.parcconline.org/
new-york
58
Ed Week. (2014, May 19). The National K12 Testing Landscape.
Retrieved from http://www.edweek.org/ew/section/multimedia/map-
the-national-k-12-testing-landscape.html
59
PARCC. PARCC Task Prototypes and Sample Questions. Retrieved
from http://www.parcconline.org/samples/item-task-prototypes
60
PARCC. Item Development. Retrieved from http://www.parcconline.
org/assessment-development"
61
Brown, E. (2014, March 2). D.C. Mulling Over Common Core
Test Switch. The Washington Post. Retrieved from http://www.
washingtonpost.com/local/education/dc-mulling-over-common-core-
test-switch/2014/03/02/29478710-a0b3-11e3-a050-dc3322a94fa7_story.
html?wprss=rss_education
62
PARCC. (2013, August). PARCC Fact Sheet and FAQs.
Retrieved from http://www.parcconline.org/sites/parcc/fles/
PARCCFactSheetandFAQsBackgrounder_FINAL.pdf
63
Smarter Balanced Assessment Consortium. Computer Adaptive Testing.
Retrieved from http://www.smarterbalanced.org/smarter-balanced-
assessments/computer-adaptive-testing/
64
Alpert et al. (2013)
65
PARCC. (2013, August). PARCC Fact Sheet and FAQs.
Retrieved from http://www.parcconline.org/sites/parcc/fles/
PARCCFactSheetandFAQsBackgrounder_FINAL.pdf
66
PARCC. Practice Tests. Retrieved from http://www.parcconline.org/
practice-tests
67
Hanushek, E.A., Raymond, M.E. (2005). Does School Accountability
Lead to Improved Student Performance? Journal of Policy Analysis and
Management, 24(2), 297327. Retrieved from http://hanushek.stanford.
edu/sites/default/fles/publications/hanushek%2Braymond.2005%20
jpam%202 4-2.pdf
68
Chiang, H. (2009, October). How accountability pressure on failing
schools afects student achievement. Journal of Public Economics, 93(9-
10), 104557. Retrieved from http://www.sciencedirect.com/science/
article/pii/S0047272709000693
69
Rouse, C.E., Hannaway, J., Goldhaber, D., et al. (2013, May). Feeling
the Florida Heat? How Low-Performing Schools Respond to Voucher
and Accountability Pressure. American Economic Journal: Economic
Policy, 5(2), 251-81. Retrieved from http://www.aeaweb.org/articles.
php?doi=10.1257/pol.5.2.251
70
All of these studies measure scores based on assessments other than the
state exam, so cheating, gaming, or test prep cannot explain these results.
71
Rockof, J.E., Staiger, D.O., Kane, T.J, et al. (2010, July). Information
and Employee Evaluation: Evidence from a Randomized Intervention
in Public Schools. The National Bureau of Economic Research Working
Paper No. 16240. Retrieved from http://www.nber.org/papers/w16240
72
Deming, D.J., Cohodes, S., Jennings, J., et al. (2013, September). School
Accountability, Postsecondary Attainment and Earnings. The National
Bureau of Economic Research Working Paper No. 19444. Retrieved
from http://www.nber.org/papers/w19444
73
DiCarlo, M. (2010, July 14). Teachers Matter, But So Do Words.
(Weblog). Retrieved from http://shankerblog.org/?p=74
74
Di Carlo, M. (2012, February 2). The Perilous Confation of Student
and School Performance. (Weblog). Retrieved from http://shankerblog.
org/?p=4980
75
Di Carlo, M. (2013, October 3). Are There Low-Performing Schools
With High-Performing Students? (Weblog). Retrieved from http://
shankerblog.org/?p=8887
76
Value-Added Modeling 101. (2012, September). Rand Education.
Retrieved from www.rand.org/education/projects/measuring-teacher-
efectiveness/value-added-modeling.html
77
McCafrey, D. (2012, October 15). Do Value-Added Methods Level the
Playing Field for Teachers? Carnegie Knowledge Network. Retrieved
from http://www.carnegieknowledgenetwork.org/briefs/value-added/
level-playing-feld/
N
O
T
E
S
33
78
Baker, E., Barton, P., et al. (2010, August 27). Problems with the use
of student test scores to evaluate teachers. Economic Policy Institute.
Retrieved from http://www.epi.org/publication/bp278/
79
Glazerman, S., Loeb, S., et al. (2010, November 17). Evaluating
Teachers: The Important Role of Value-Added. Brown Center on
Education Policy at Brookings. Retrieved from http://www.brookings.
edu/~/media/research/fles/reports/2010/11/17%20evaluating%20
teachers/1117_evaluating_teachers.pdf
80
Di Carlo, M. (2010, December 7). The War on Error. (Weblog)
Retrieved from http://shankerblog.org/?p=1383
81
Chetty, R., Friedman, J., Rockof, J. (2011, December). The Long-
Term Impacts of Teachers: Teacher Value-Added and Student Outcomes
in Adulthood. National Bureau of Economic Research. Retrieved from
http://www.nber.org/papers/w17699
82
Stern, G. (2013, October 15). N.Y.s Teacher Evaluation Faulted in
Study. The Journal News. Retrieved from http://archive.lohud.com/
article/20131015/NEWS/310150042/N-Y-s-teacher-evaluations-
faulted-study
83
Lower Hudson Council of School Superintendents. (2013, October).
Review and Analysis of the New York State Growth Model. Retrieved
from http://www.lhcss.org/positionpapers/nysgrowthmodel.pdf
84
Decker, G. (2013, June 18). State to Use Value-Added Growth Model
without Calling it That. Chalkbeat. Retrieved from http://ny.chalkbeat.
org/2013/06/18/state-to-use-a-value-added-growth-model-without-
calling-it-that/#.U61JpVNyjec
85
Ehlert, M., Koedel, C., Parsons, E., et al. Selecting Growth Measures
for School and Teacher Evaluations: Should Proportionality Matter?
National Center for Analysis of Longitudinal Data in Education Research.
Retrieved from http://www.caldercenter.org/publications/upload/wp-
80-updated-v3.pdf
86
Lankford, H., Loeb, S., Wyckof, J. (2002, March). Teacher Sorting and
the Plight of Urban Schools. Education Evaluation and Policy Analysis.
Retrieved from http://epa.sagepub.com/content/24/1/37.short
87
Koedel, C. (2014, May 27). The Proportionality Principle in
Teacher Evaluation. Shanker Blog. Retrieved from http://shankerblog.
org/?p=9924
88
Cramer, P., Decker, G. (2013, September 16). Instead of Telling Teachers
Apart, New Eval Lumps Some Together. Chalkbeat. Retrieved from
http://ny.chalkbeat.org/2013/09/16/instead-of-telling-teachers-apart-
new-evals-lump-some-together/#.U4-VpFNyjec
89
Di Carlo, M. (2012, May 29). We Should Only Hold Schools
Accountable for Outcomes They Can Control. (Weblog). Retrieved from
http://shankerblog.org/?p=5959
90
We distinguish this practice from evaluation systems that have school-
wide rating components, meaning that all teachers in a school are judged
by a schools overall components. This practice has several pros and cons;
in this paper, we do not take a position on it.
91
Decker, G. (2014, May 14). Appeal Process in New Evaluation Plan
Shifts Weight from Student Scores for Some. Chalkbeat. Retrieved from
http://ny.chalkbeat.org/2014/05/14/appeal-process-in-new-evaluation-
plan-shifts-weight-from-student-scores-for-some/#.U4-ViVNyjec
92
Goldstein, D. (2012, June 13). No More Ditching Gym Class.
Slate. Retrieved from http://www.slate.com/articles/double_x/
doublex/2012/06/standardized_tests_for_the_arts_is_that_a_good_idea_.
html
93
EngageNY. Overview of Student Learning Objectives. Retrieved from
http://www.engageny.org/sites/default/fles/resource/attachments/
overview_of_student_learning_objectives.pdf
94
For one example, among many others, of this argument, see: https://
www.aft.org/pdfs/teachers/devmultiplemeasures.pdf
95
Worrell, C. (2013, October 25). In Teacher Evaluations, Student
Data and Multiple Measures Show Progress. Data Quality Campaign.
Retrieved from http://www.dataqualitycampaign.org/blog/2013/10/in-
teacher-evaluations-student-data-show-progress/
96
New York City Department of Education. NY State Policy Context:
Education Law 3012-c. Retrieved from http://schools.nyc.gov/Ofces/
advance/Background/Policy+Context/default.htm
97
New York City Department of Education. (2014, April 9). Chancellor
Faria Announces New Promotion Policy for Students in Grades
3-8. Retrieved from http://schools.nyc.gov/Ofces/mediarelations/
NewsandSpeeches/2013-2014Chancellor+Fari%C3%B1a+Announces+
New+Promotion+Policy+for+Students+in+Grades+3-8.htm
98
New York State Department of Education. (2013, June). Diploma/
Credential Requirements. Retrieved from http://www.p12.nysed.gov/
ciai/gradreq/diploma-credential-summary.pdf
99
Wall, P. (2013, November 14). Tougher Diploma Rules Leave Some
Students in Graduation Limbo. Chalkbeat. Retrieved from http://
ny.chalkbeat.org/2013/11/14/tougher-diploma-rules-leave-some-
students-in-graduation-limbo/#.U44Jm1Nyjec
100
Jacob, B. (2001, June). Getting Tough: The Impact of High School
Graduation Exam. Educational Evaluation and Policy Analysis. Retrieved
from http://epa.sagepub.com/content/23/2/99.short
101
Marchant, G., Paulson, S., (2005, January). The Relationship of
High School Graduation Exams to Graduation Rates and SAT Scores.
Education Policy Analysis Archives. Retrieved from http://fles.eric.
ed.gov/fulltext/EJ846516.pdf
102
Grodsky, E., Warren, J., Kalogrides, D. (2009, May). State High School
Exit Examinations and NAEP Long-Term Trends in Reading and
Mathematics, 19712004. Education Policy. Retrieved from http://epx.
sagepub.com/content/early/2008/06/13/0895904808320678.abstract
103
Reardon, S., Arshan, N., Atteberry, A., Kurlaender, M. (2010,
December). Efects of Failing a High School Exit Exam on Course
Taking, Achievement, Persistence, and Graduation. Education Evaluation
and Policy Analysis. Retrieved from http://epa.sagepub.com/
content/32/4/498
104
Baker, O., Lang, K. (2013, June). The Efect of High School Exit
Exams on Graduation, Employment, Wages, and Incarceration. National
Bureau of Economic Research. Retrieved from http://www.nber.org/
papers/w19182
105
Darling-Hammond, L., Rustique-Forrester, E., Pecheone, R. (2005).
Multiple Measures Approaches to High School Graduation. The School
Redesign Network at Stanford University. Retrieved from https://
edpolicy.stanford.edu/sites/default/fles/publications/multiple-measures-
approaches-high-school-graduation.pdf
106
New York State Education Department. (2014). Graduation Rate Data.
Retrieved from http://www.p12.nysed.gov/irs/pressRelease/20140623/
home.html
N
O
T
E
S
THE 2014 EDUCATORS 4 EXCELLENCE NEW YORK
TEACHER POLICY TEAM ON TESTING AND ASSESSMENT
Trevor Baisden
Founding Fifth-Grade ELA and History Lead Teacher
Success Academy Bronx 2 Middle School
Elizabeth Barrett-Zahn
Kindergarten to Fifth-Grade Science Facilitator
Columbus Elementary School, New Rochelle
Rachael Beseda
First-Grade Special Education Teacher
Global Community Charter School
Ezekiel Cruz
Ninth- to 12th-Grade Social Studies Teacher
Manhattan Bridges High School
Suraj Gopal
Ninth-Grade STEM Special Education Teacher
Hudson High School of Learning Technologies
Vivett Hemans
English and Language Arts Teacher
Eagle Academy for Young Men of Southeast Queens
Maura N. Henry
Sixth- to 12th-Grade ESL Teacher
The Young Women's Leadership School of Astoria
Michelle Knifn
Ninth- to 12th-Grade Math Teacher
High School of Telecommunication Arts and Technology
Jason Koo
Math Teacher
Albert Einstein Junior High School I.S. 131
Christine Montera
Social Studies Teacher
East Bronx Academy for the Future
Liliana Ruiz
Sixth- to Eighth-Grade Special Education Teacher
Bea Fuller Rodgers School I.S. 528
Charlotte Steel
Seventh-Grade Math Teacher
Booker T. Washington M.S. 54
Blackfoot U-Ahk
Fourth- and Fifth-Grade Teacher of Students
with Severe Emotional Disabilities
Coy L. Cox School P.369k
Iris Won
Ninth- to 12th-Grade Mathematics and Technology Teacher
Renaissance High School for Musical Theater & Technology
This report, graphics, and fgures were designed by Kristin Girvin Redman and Tracy Harris at Cricket Design Works in
Madison, Wisconsin.
The text face is Bembo Regular, designed by Stanley Morison in 1929. The typefaces used for headers, subheaders, and pull quotes are
Futura Bold, designed by Paul Renner, and Museo Slab, designed by Jos Buivenga. Figure labels are set in Futura Regular, and fgure
callouts are set in Museo Slab.
For far too long, education policy has been created
without a critical voice at the tablethe voice of classroom teachers.
Educators 4 Excellence (E4E), a teacher-led organization, is changing
this dynamic by placing the voices of teachers at the forefront of the
conversations that shape our classrooms and careers.
E4E has a quickly growing national network of educators united by
our Declaration of Teachers Principles and Be liefs. E4E members
can learn about education policy and re search, network with like-
minded peers and policymakers, and take action by advocating
for teacher-created policies that lift student achievement and the
teaching profession.
Learn more at Educators4Excellence.org.