from Four Years of
Implementation in
DC Public Schools:
An Overview
Background on Washington D.C. Public Schools (DCPS)
46,000 students
•
77% high-poverty
•
69% Black, 16% Latino, 11% White
113 schools, 4,000 teachers, 7,000 total school-based staff
Historically one of the lowest-performing school districts in the United States
•
In 2007, only 12% of 8
thgraders were proficient at reading according to the National Assessment
of Educational Progress
•
Around half of students graduate from high school
Yet DCPS also spends more per student than most other districts
In 2007, the school district was placed under direct control of the Mayor
For the past six years, the district has been led by reform-oriented Chancellors
Michelle Rhee and Kaya Henderson
The Foundation: Our teachers’ contract
Key principle: Teacher performance should drive all human capital decisions
Three key changes:
1.
Performance-based termination and excessing – end of tenure as guarantee of a job
regardless of performance and end of seniority as primary driver of decisions
2.
Mutual consent: Teachers are no longer placed at schools. If a teacher cannot secure
another position:
o
Teachers who received an evaluation rating of Effective or higher in the previous year are
able to choose from three options: a $25,000 buyout, early retirement, or one extra year
placement at full salary. Teachers who select the extra year but still do not secure a new
position by the end of that year are terminated.
o
Teachers who received an evaluation below Effective in the previous year have 60 days to
secure another position, and are subject to termination after the 60 day period.
3.
Performance-based compensation
Teachers are the most important resource in our schools
Underperforming, low-income children who
have great teachers for three years in a row will
catch up to their higher-income peers. In short,
great teachers
can close the
achievement gap
.
-- Eric Hanushek in Teacher Quality, 2002
4
Where we were in 2007
12%
vs.
95%
8th Grade Reading
Proficiency (2007 NAEP)
Teachers Meeting or
Exceeding Expectations
5Implications for performance assessment
We needed a performance assessment system that:
Provided clear expectations and differentiated, relevant feedback
Provided clear opportunities for growth and professional development
Provided a means of identifying our high performers
Provided a fair, efficient process for moving out low performers
6
Simple goal for our teacher effectiveness work
Create a system in which all parents would be
satisfied
randomly
assigning their children to
any
classroom in DCPS.
District of Columbia Public Schools | May 29, 2014
Identify our best teachers and then do everything possible to keep them:
•
Shower them with praise
•
Pay them more
•
Give them new opportunities and responsibilities
Sub-Goal 2: Clear Expectations and Targeted Support
District of Columbia Public Schools | May 29, 2014
Set clear expectations; provide feedback and opportunities for growth
Identify which teachers need help and provide them with support:
•
Multiple observations
•
100 school-based instructional coaches
•
45 master educators
Janice Carter-Bowden
Coach, LaSalle-Backus EC
Camia Hoard
10
2%
2%
1%
1%
16%
13%
9%
5%
20%
66%
69%
68%
44%
16%
16%
21%
29%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
09-10
10-11
11-12
12-13
Ineffective
Minimally Effective
Developing (New in 2012-2013)
Effective
Highly Effective
Roughly 1% of the teaching force was rating “Ineffective” last year.
Roughly 1/3 of teachers rated “Minimally Effective” (second lowest category) resign
voluntarily from the school system.
IMPACT Timeline
June 12, 2007: Appointment of Chancellor Rhee
August 2007 – August 2008: Research phase
September 2008 – August 2009: Design phase
September 2009: Launch of IMPACT 1.0
August 2010: Launch of IMPACT 2.0
June 2011: Appointment of
Chancellor Henderson
August 2011: Launch of IMPACT 2.1
August 2012: Launch of IMPACT 3.0
August 2013: Launch of IMPACT 3.1
School Leader Task Force:
Biweekly meetings over five months with 12 school leaders,
representing a variety of experience levels, school types, and wards
Surveys:
Online surveys distributed to all principals, assistant principals, teachers, and staff
members (525 total responses)
Teacher-Led Feedback Sessions for Educators:
17 sessions in schools facilitated by members of
the Chancellor’s Teachers’ Cabinet
Central Office-Led Feedback Sessions for Educators:
12 sessions, including one in each ward, to
gather feedback on draft changes
Instructional Superintendent Meetings:
One-on-one meetings with each superintendent in
November, presentations at superintendent meetings in December and February to discuss draft
changes, and additional one-on-one meetings in June
Principal Phone Calls:
10 calls conducted with principals in June to review changes
Central Office Task Force:
Biweekly meetings over five months with 13 senior staff members
from across OHC, OCAO, and OSE
WTU meetings:
Initial meeting in October; regular check-ins with during the year
12
Lesson 1: IMPACT has allowed us to distinguish between our
outstanding, good, and poor teachers
13
Ineffective
0% 2% 4% 6% 8% 10% 12% 15 0 16 0 17 0 18 0 19 0 20 0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0 34 0 35 0 36 0 37 0 38 0 39 0 40 0Distribution of Overall IMPACT Scores, 2010-11
First Years 2 or more Years
Average first year teacher = 283
Average for teachers with 2+ years experience = 299
Minimally
Effective
Effective
Highly
Effective
IMPACT Score
%
of
Tea
che
rs
First years 2 or more years6 Full-Time Staff Members
Operating Costs, SY 11-12
$1.8 Million
IMPACT Operations
Phone and Email Helplines
7.5 hours of phone calls
on a busy day
Trainings
250 hours per year
Lesson 2: Implementation is everything
14
Lesson 3: Evaluation is stressful for teachers – and no
communications strategy can fix that entirely
“IMPACT is stressful. I never
know when someone is going
to walk into my classroom to
evaluate me.”
– Elementary Teacher,
Murch Elementary School
Feedback sessions
Fewer observations for our highest
performers
Collecting contextual information about
teachers’ classes to share with master
educators
A focus on relationship-building between
teachers and master educators
Informal observations
Teacher career ladders
How We’re Working to Address This:
Lesson 4: Stakeholder input can only get you so far
Excerpted feedback from educators during IMPACT focus groups:
“I want someone other than my
principal to observe me.”
“I don’t trust outsiders to observe me
fairly.”
“Teamwork is really important.”
“I should only be evaluated on the
things that I do.”
“Student achievement is the most
important thing.”
“Student achievement shouldn’t be
part of my evaluation.”
“You can’t hold me accountable for
anything beyond the contract.”
“I should get credit for coming to school
at 6:00 a.m. and leaving at 6:00 p.m.”
16
Lesson 5: Evaluation is a foundation, but it needs to be part of a
broader human capital strategy
17 Recruit and
Select the Best Possible Talent Rigorously Evaluate Performance Retain the Best and Extend Their Reach Provide Targeted Support Remove the Ineffective Increase Student Achievement Set New Hires
Up for Success
So is it working?
IMPACT causes teachers to improve – both low and high performers.
DCPS is retaining its best teachers at very high rates - 92% of Highly Effective
teachers were retained in 2010-2011 and 2011-2012, while in contrast, only
59% of teachers rated Minimally Effective were retained.
IMPACT causes many low-performing teachers to leave on their own; those
hired to replace them perform better - teachers hired in the 2011-2012
school year substantially outperformed those who left DCPS in 2010-2011.
Initial research results are promising
18
34.7 28.5 43.3 39.0 45.5 44.8 43.9 43.5 43.5 43.9 33.2 43.4 45.9 45.7 47.4 49.5 50.1
Reading
Math
Composition
Percent Proficient by Year
2007
2008
2009
2010
2011
2012
2013
Over the past six years, our schools have made dramatic gains;
2013 saw the highest proficiency rates ever
19
+ 3.9
+ 3.6
+12.7
+ 21.0
+ 4.4
+16.9
our gains were the best in the nation
-10 -5 0 5 10 15 20 25 Dis trict o f C o lu m b ia Te n n es se e In d ian a DoDE A Wa sh in gt o n H awaii Flori d a Ore gon Iow a Cali fo rn ia Min n es o ta N ew H amp sh ire Arizo n a N eb ra sk a U ta h Wy o m in g N ew York Arka n sas G eo rgia We st Virg in ia Pe n n sy lv an ia N at ion al p u b lic Ve rm o n t N ev ad a Col o ra d o Ma in e Oh io Lo u is ian a De laware Rh o d e I slan d Mis so u ri Illino is Virg in ia N o rth C ar o lina Mis sis sip p i Con n ec ticu t Ok lah o m a Id ah o Ma ry lan d Wis con si n Ke n tu ck y Mic h igan Te xa s N ew Je rs e y Alas ka Kan sas Alab am a Ma ss ac h u se tt s Sou th Car o lina N o rth D ak o ta N ew M exi co Sou th Da ko ta Mo n ta n aAppendix
22
Who Does IMPACT Cover?
All school-based staff including principals and Assistant Principals
Nearly
7,000
people
Approximately
1/3
are not teachers
IMPACT Groups
Group 1: Teachers with DC CAS Data
Group 2: Teachers without DC CAS Data
Group 2a: Early Childhood Education Teachers
Group 3: Special Education Teachers
Group 3a: Special Education Teachers – Autism
Group 3b: Special Education Teachers – Early
Childhood
Group 4: Non-Itinerant ELL Teachers
Group 5: Itinerant ELL Teachers
Group 6: Shared Teachers
Group 7: Visiting Instruction Service Teachers
Group 8: Student Support Professionals
Group 9: Librarians
Group 10: Counselors
Group 11: School-based Social Workers
Group 11a: School-based Psychologists
Group 12: Related Service Providers
Group 13: Special Education Coordinators
Group 14: Program Coordinators and Deans
Group 15: Instructional Coaches
Group 16: Mentor Teachers
Group 17: Educational Aides
Group 18: Office Staff
Group 19: Custodial Staff
Group 20: Other
24
IMPACT Components
Different components
are used to evaluate different groups
Each component is assessed
multiple times
over the course of the year
Group 2 –
General Ed
Teachers
Group 3a –
Special Ed Autism
Teachers
Group 17 –
Educational Aides
25IMPACT Rubrics
Each component has a
scoring rubric
26
IMPACT Guidebooks
Each group has a
unique guidebook
which contains information about
all
components
,
rubrics
, and
additional group specific information
Group 1 - General Education Teachers with IVA Data
28
Component 1: Individual Value-Added (IVA)
35%
Component 1: Individual Value-Added (IVA)
A
measure of a
teacher’s impact
on student learning
Measures how a teacher’s students performed on the
DC CAS in comparison
with
how similar children performed (that is, children with the same skill
level at the beginning of year)
DOES
take into account different starting points of different children
DOES take into account other factors that are outside the teacher’s control:
for example, poverty, special education status, etc.
Does NOT require teachers to meet Adequate Yearly Progress (AYP)
30
Component 1: How Does Value-Added Work?
Teacher’s Level of Expectations
Teacher’s Pedagogical Expertise
Teacher’s Ability to Motivate
Teacher’s Content Knowledge
Student’s Prior Learning
Student’s Disability (If Any)
Student’s English Proficiency
Student’s Resources at Home
Student
Achievement
As Measured
by the DC CAS
Value-added
isolates
the teacher’s
impact on student
achievement.
Component 1: Who Has Individual Value-Added (IVA)?
Group 1 teachers
•
4
th
– 8
th
grade,
reading
and
math
•
9
th
- 10
th
grade,
ELA only
About 15% of teaching force
32
Component 2: Teacher-Assessed Student Achievement Data (TAS)
15%
Component 2: Teacher-Assessed Student Achievement Data (TAS)
A
measure of a
teacher’s impact
on student learning
Based on a grade and subject area
appropriate assessment
Goals are unique and developed based on
baseline data
School leaders and teachers determine the most appropriate assessment
and goals during a beginning of the year
TAS conference
Teachers
collect data
over the course of the year and meet with their
principal again at the end of the year to discuss and assign a
final TAS score
34
Component 3: Teaching and Learning Framework (TLF)
40%
36
District of Columbia Public Schools | May 29, 2014
The Framework defines the actions that effective
teachers take to deliver high-quality instruction
within a safe, engaging, and productive classroom.
Recognizing the significance of teachers in
increasing student achievement, the framework:
Empowers teachers to drive instructional design and
delivery.
Provides a common language to discuss instructional
practices.
Drives robust, coherent professional development
opportunities.
The Framework provides a process for approaching
instructional design and delivery; it is not a
Component 3: Teaching and Learning Framework (TLF)
Teach 1: Lead
well-organized, objective-driven
lessons
Teach 2: Explain
content clearly
Teach 3: Engage students at all learning levels in
accessible and challenging work
Teach 4: Provide students
multiple ways to move toward mastery
Teach 5:
Check
for understanding
Teach 6:
Respond
to student understanding
Teach 7: Develop
higher-level understanding
through effective questioning
Teach 8:
Maximize
instructional time
Teach 9: Build a supportive,
learning-focused classroom
community
Component 3: Teaching and Learning Framework (TLF)
Objective
Importance
38
Component 3: Teaching and Learning Framework (TLF)
Evaluators use the TLF rubric to conduct both
formal
and
informal
observations
Most teachers receive 4 formal and 1 informal observation throughout the
year
Formal
Observations
Informal
Observations
Length
30 minutes
At least 15 minutes
Scores
Scores for all
standards
No scores
Comments
Full written report, 5
comments across 9
standards
Brief comments on
strengths and areas
for growth
Conference
Yes
No
Component 3: Teaching and Learning Framework (TLF)
Master Educators:
•
Total of
45
(approximately 1,500 applications from DCPS and around the country)
•
Go through an extensive
4-part application
process
•
Receive
6 weeks of initial training
, more throughout the year
•
Are
subject-specific
(not school-based)
•
Are
full-time
(Salary: $92,000-$102,000)
•
Conduct about
200 observations
per year
•
Provide
targeted professional development
through post-observation
conferences, subject area “inquiry groups,” new teacher visits, and other venues
Component 4: Commitment to School Community (CSC)
10%
Component 4: Commitment to School Community (CSC)
Standard
1
: Support of Local School Initiatives
Standard 2: Support of Special Education and English Language Learner Teams
Standard 3: High Expectations
Standard 4: Partnership with Families
Standard 5: Instructional Collaboration
42
Additional Component: Core Professionalism
Standards
•
Attendance
•
Tardiness
•
Policies and procedures
•
Respect
Ratings
•
Meets Standard = No change in final score
•
Slightly Below Standard = 10 point deduction
•
Significantly Below Standard = 20 point deduction
43
How is the Data Captured?
impactdcps.dc.gov
44
Sample Final Score
Overall IMPACT Scale
Overall IMPACT Scale
100
Points –
400
Points
Five Ratings:
Ineffective
,
Minimally Effective
,
Developing
,
Effective
, and
Highly
Effective
.
46
The Ratings
The Ratings
Ineffective (100-199)
Unacceptable performance
1 year = separation
48
The Ratings
Ineffective (100-199)
Unacceptable performance
1 year = separation
Minimally Effective (200-249)
Performance that is below
expectations
Salary “hold”
2 consecutive years = separation
The Ratings
Ineffective (100-199)
Unacceptable performance
1 year = separation
Minimally Effective (200-249)
Performance that is below
expectations
Salary “hold”
2 consecutive years = separation
Developing (250-299)
•
Performance is below expectations;
prioritized professional development
•
Salary “hold”
•
3 consecutive years = separation
50
The Ratings
Effective (300-349)
Solid performance
Normal salary advancement
The Ratings
Effective (300-349)
Solid performance
Normal salary advancement
Highly Effective (350-400)
Outstanding performance
Additional compensation
52
