Nothing Matters More than a Quality Teacher - PowerPoint PPT Presentation

1 / 61
About This Presentation
Title:

Nothing Matters More than a Quality Teacher

Description:

Title: PowerPoint Presentation Author: jwiederhorn Last modified by: Eileen M. Kellor Created Date: 1/9/2002 12:27:37 AM Document presentation format – PowerPoint PPT presentation

Number of Views:235
Avg rating:3.0/5.0
Slides: 62
Provided by: jwi86
Category:

less

Transcript and Presenter's Notes

Title: Nothing Matters More than a Quality Teacher


1
(No Transcript)
2
Nothing Matters More than a Quality Teacher
  • Sanders cross-sectional work found that 
    effective teachers boosted the math scores of
    low-achieving students 39 percentile points more
    than students assigned to the ineffective
    teachers.

3
Nothing Matters More than a Quality Teacher
  • Sanders longitudinal work demonstrated that
    students who performed equally well in math in
    second grade showed enormous performance
    differences three years later, depending on the
    quality of the teachers.

4
Nothing Matters More than a Quality Teacher
  • Rivers longitudinal work found that average
    achieving students assigned to 4 years of
    ineffective teachers had only a 40 percent chance
    of passing the Tennessee high school exit
    examination. The same students assigned to 4
    years of effective teachers had an 80 percent
    chance of passing.

5
Salaries not competitive
Everyone gets same pay
Costs of training not warranted by salary
Women have more career opportunities now
Little collegiality
Little respect from community
Often unpleasant, dangerous environment
6
(No Transcript)
7
(No Transcript)
8
New Ways to Attract Teachers
  • Increase Salaries
  • School Debt Forgiveness
  • Housing Subsidies
  • Perks
  • PR Campaign
  • New Recruitment Strategies
  • Accelerated Teacher Education
  • More Rigorous Training

9
Programs to Attract Retain High Quality Teachers
  • small
  • isolated efforts
  • not school-centered
  • poorly designed
  • poorly implemented
  • rather than systemic reforms
  • solve one problem only to create another

10
(No Transcript)
11
(No Transcript)
12
TAP Both Old and New
Multiple Career Paths Career Ladders Performance
Pay
Odden,Denver,Cincinnati Assessment
Sanders,
Danielson, NBPTS Ongoing, Applied
??? Professional
Development Expanding the Pool
N.J. Alt. Cert, Troops to
Teaches, Teach for America TAP is unique
because it ties both teacher performance
assessment student value added to teacher
compensation supports that with a unique
professional development tool.
13
Models for the Teaching Profession Career
Advancement
Traditional Model
TAP Model
  • Multiple Career Paths
  • Senior, Mentor Lead Positions
  • Requiring increasing levels of
  • professional qualifications
  • responsibilities
  • authority
  • assessment rigor
  • Single Career Path
  • Teacher Position Only
  • Requiring the same level of
  • professional qualifications
  • responsibility
  • authority
  • assessment rigor

14
Models for the Teaching Profession Compensation
Traditional Model
TAP Model
Performance and Responsibility Drive
Compensation Salary determined by level of
responsibilities and effectiveness of
performance.
Salary Schedule Drives Compensation Lock-step
salary determined only by years of experience and
training units accrued.
15
(No Transcript)
16
Models for the Teaching Profession Professional
Accountability
Traditional Model
TAP Model
  • Performance-based Accountability
  • TAP standards, procedures and performance rubrics
  • Hiring, advancement and compensation tied to
    evaluation
  • Support provided for growth
  • Uneven Accountability
  • Idiosyncratic evaluation standards procedures
  • Rewards and sanction unrelated to evaluation
    outcomes
  • Support provided for deficiencies only

17
Models for the Teaching Profession Professional
Growth
Traditional Model
TAP Model
  • Ongoing Applied Professional Growth
  • Schoolwide commitment, weekly, site-based,
    teacher lead activities
  • Goals and activities tied to standards, local SIP
    analysis of student learning outcomes
  • Used to support and reinforce evaluation growth
    goals
  • Inservice/Course-based Professional Development
  • Individual commitment, intermittent activities
  • Goals and activities tied to personal and
    financial interests of the individual
  • Unconnected to evaluation

18
Improving Teacher Quality Through
Performance-based Accountability
  • Evaluating teacher performance means
  • observing what teachers do in their classrooms,
  • rating how well they do it, and
  • assessing each teachers and the schools impact
    on student achievement.

19
TAP Teacher Accountability Versus Past Teacher
Accountability Efforts
TAP
Past Efforts
  • Checklist of Teaching Behaviors
  • Two Performance Levels
  • Evaluation excludes student achievement
  • One Evaluator
  • Evaluation supports deficiencies only
  • Performance independent of compensation
  • Teaching Performance Standards
  • Five Performance Levels
  • Evaluation includes school and classroom
    achievement
  • Multiple Evaluators
  • Evaluation support professional growth
  • Performance tied to compensation


20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
TAP Teaching Performance Standards Skills
Knowledge Responsibilities
  • What do the teacher performance standards look
    like?
  • 5 levels of performance ranging from exemplary to
    proficient to unsatisfactory
  • Descriptive narratives of expected performance
  • Teacher Performance Standards Examples
  • Communicating Directions Instructional Content
  • Questioning
  • Thinking

24
TAP Teaching Performance Standards
  • Who rates teacher performance?
  • Four Qualified School-based Evaluators
  • Administrator
  • Master Teacher
  • Mentor Teacher
  • Self
  • How often is each teacher observed?
  • 10 times per year
  • Administrator - at least 2 times per year
  • Master teacher - at least 6 times per year
  • Mentor teacher - at least 2 times per year

25
TAP Classroom and School Value-Added Achievement
  • Base decisions on value-added gains
  • Use the TAP value-added statistical model
  • Set leveled criteria for school gains, and
    classroom gains (13, 8, 4, Years Growth,
    Negative gain)
  • Test every year
  • Use reliable and valid tests
  • Tie student level data to teacher each year

26
Teacher Compensation
  • Abysmally low
  • Not differentiated
  • Not tied to teaching performance
  • Not tied to student achievement
  • Based on teachers years experience and units
    earned which are both poor predictors of student
    achievement.

27
(No Transcript)
28
(No Transcript)
29
TAP Compensation
  • Pays teachers based on performance
  • Pays teachers differently based on performance.
  • Three types of performances count
  • Teaching Performance Standards Skills, Knowledge
    Responsibilities 50
  • Classroom Achievement - 20
  • School Achievement 30

30
(No Transcript)
31
  • Augment not supplant single salary schedule
  • Reward teachers for achieving school wide goals
  • Financial incentives for low performing schools
  • Bonuses for passing National Board

32
(No Transcript)
33
(No Transcript)
34
How TAP Compensation System Has Evolved
  • Performance awards are given as a bonus to be
    earned each year.
  • Performance awards supplement rather than
    replace traditional step and column pay scale.
  • No one earns less than they would have earned
    under the traditional compensation system.
  • Performance awards are not cumulative and must
    be constrained by available funds, usually
    augmenting salaries by 5,000 or less
  • 50 of the bonus is awarded for skills and
    knowledge
  • 50 is based upon student achievement
    (value-added)
  • 30 school-wide for all teachers
  • 20 based on achievement of individual
    teachers students

35
How TAP Compensation System Has Evolved
  • Since subject specific tests are often
    unavailable, the student achievement element of
    the bonus for high school teachers is
    complicated.
  • Bonuses are criterion referenced rather than
    relative, which means that any teacher who meets
    a standard receives the bonus.
  • No one loses money (compared to what they would
    have earned without TAP) for poor performance.
  • Teachers who score well on the skills and
    knowledge part can earn bonuses even if their
    students scores do not improve.
  • There must be the opportunity for all teachers
    to get a bonus of some amount. We cannot say
    only the top X will receive bonuses.
  • If teacher work more days, they must get paid
    for them at least at their former daily rate.

36
Implementating TPBA System
  • Develop Teacher Evaluation System
  • Establish career, mentor, and master teaching
    performance standards
  • Establish career, mentor, and master teacher
    responsibilities performance standards
  • Review the TAP T-PBA policies
  • Establish the criteria for classroom achievement
    gains attributed to the teacher
  • Establish the criteria for the school achievement
    gains

37
Implementation Compensation
  • Establish the school or district compensation
    model
  • Establish TAP Salary Augmentations
  • Establish TAP Performance Award Fund
  • Establish Performance Award Weights

38
Implementation Teacher Evaluation Training
  • Participate in the teacher performance-based
    accountability training
  • Train all evaluators using the TAP video training
    materials
  • Certify all evaluators
  • Continue training throughout the year

39
We Have Revised Our Model as Follows
  • Most masters and mentors come from existing
    staff .
  • New teacher salaries are higher or not lower
    than before TAP.
  • New teachers probably will not be hired at the
    bottom of the associate range.
  • Significant time off from school for masters is
    unrealistic at the start.
  • Master teachers will have to spend more time
    teaching than anticipated.
  • Tenure not eliminated.
  • Ratios of mentor and masters vary with available
    resources.
  • Unanticipated costs paying for tests not now
    given, travel expenses for teachers to get
    training, training materials.

40
Intermediate Outcomes
  • Teachers opt for new system vs. existing system
  • Changes in types of individuals applying
  • Number of applicants
  • Differences in characteristics of people hired
  • Changes in teacher retention rates
  • Changes in which teachers stay in classroom
  • Survival rates in the first five years
  • Changing nature of collective bargaining
  • Stakeholder perceptions of staff quality
    professionalism
  • Teacher satisfaction data

41
Process Issues
  • Number of schools using the model
  • Distribution of TAP schools by location, SES,
    ethnicity
  • Amount of discretionary funds to support TAP
  • Support from private foundations and
    corporations
  • Extent to which schools adhere to the five
    principles

42
The Cost of TAP
  • Incremental costs 6 of budget OR
    400/student
  • No current teacher worse off
  • Salary supplements for Master Mentor teachers
  • New teacher positions
  • New specialists hired
  • Associate teachers summer professional growth
  • Turnover savings not kept by school
  • Traditional salary schedule increases in place

43
We Can Cover Costs By
  • Changing parameters (master mentor ratios)
  • Keeping turnover savings in the schools
  • Converting regular salary schedule increases
    into performance pool
  • Utilizing existing grants (Title I for aides or
    lead teachers)
  • New money

44
Overview of AZTAP School Characteristics
Total Number of AZTAP School 7 Number in year
one implementation 2 Number in year two
implementation 5 School Characteristics Range in
TAP School Student Population 400 to 1,350
students Ratio of Master to Career Teachers 9/1
to 17/1 Ratio of Mentor to Career Teachers 3/1 to
8/1 Number of Cluster Groups 4 to 10 Time Cluster
Groups Meet Weekly 1 hour to 4 hours Grades
Represented K-2 -K-4 -K-6 -K-12 - 3-8
45
(No Transcript)
46
Why the Slow Growth of Performance Pay?
  • Steps on the special interests of powerful
    education groups.
  • Unions do not want system where any members are
    disappointed.
  • Parents always want smaller classes for their
    kids despite lack of evidence that class size
    matters.
  • Smaller classes make things easier for teachers
    in their classrooms.
  • Standards, accountability and compensation based
    on performance put more pressure on teachers.
  • School boards districts that see teacher
    quality initiatives as devolving power from
    districts to individual schools.
  • Education schools see their market for
    credentials and professional development courses
    getting smaller

47
Why the Slow Growth of Performance Pay?
  • Strong vested interests oppose having teachers
    held accountable evaluated by student
    performance.
  • They disparage research that concludes that
    teachers are responsible for student achievement
    and should be rewarded accordingly
  • Some people do not find the research compelling
    because it depends upon test scores as the
    measure of student performance.
  • They claim test scores may be a fatally flawed
    measure of student learning
  • Teachers think using test scores for evaluation
    is unfair because they depend upon who is in
    their class THEY DO NOT UNDERSTAND OR
    ACKNOWLEDGE VALUE ADDED (GAINS RATHER THAN
    LEVELS)

48
Why the Slow Growth of Performance Pay?
  • Schools have other responsibilities than student
    achievement.
  • In earlier generations, kids who could learn
    were in school and those who couldnt learn left
    to pursue unskilled work.
  • Then it was decided that in the name of equity
    and social justice (and to keep delinquent kids
    off the street), all kids should stay in school,
    regardless of what they could learn.
  • Schools were distracted from main goal of
    helping kids learn.
  • To show effectiveness, teachers needed more
    achievable/ abstract/different goals than
    student learning to justify efforts.
  • Focus moved away from what kids learn to how
    they feel.
  • K-12 education has focused on affective domain
    that what kids learn has gotten short shrift
  • Extremely radical to now change entire focus of
    education

49
Why the Slow Growth of Performance Pay?
  • Effective teachers are defined as raising
    student test scores, but there is little
    evidence on the characteristics of such
    teachers.
  • We still do not know what these effective
    teachers do that is different or how to get
    others to do those things.
  • Little agreement about what expert pedagogy
    means, and little empirical confirmation of
    effective practice.
  • Traditionally, teachers have been expected to
    demonstrate their effectiveness-however
    defined-only while in school one time through
    National Board certification, not annually.

50
Why the Slow Growth of Performance Pay?
  • Systemic policy change requires leadership
    support at the school, district and state
    levels. Those in positions of leadership may
    resist because
  • School principals have rarely been
    instructional leaders
  • Most principals are more often building and
    personnel managers
  • Such efforts are controversial
  • They have other priorities
  • They require a huge commitment
  • It is hard work

51
Why the Slow Growth of Performance Pay?
  • Some dont truly believe that there are many
    things wrong with American Schools.
  • If schools are ineffective, that is due to
    poverty, race, geography, or lack of funding
    but not due to the need for systemic change in
    the schools.
  • Unfortunately, there is not universal belief all
    kids can learn.
  • There appear to be simpler solutions for
    improving schools like Nat.Board, across board
    raises, increasing standards for teachers or
    class size reduction.
  • Education is generally viewed as a state or
    local responsibility, so we must convince 17,000
    separate organizations of its merit.

52
Why the Slow Growth of TAP?
  • Other priorities
  • Unwilling to fight teachers unions
  • Belief that pluses of Perf. Pay outweighed by
    the negatives
  • Reluctance of school boards to support site
    based programs
  • Reform is not likely to be pivotal in
    reelection/reappointment
  • No evidence yet that TAP improves student test
    scores
  • Sticker shock high cost compared to other
    reforms
  • Lack of new funds state budget problems
  • Belief that political/educational payoff would
    be better if discretionary funds were spent
    elsewhere
  • Unwillingness to spend disproportionate share of
    discretionary funds on small of TAP schools

53
What It Takes to Successfully Implement
Whole-School Reform
  • Commitment to all components of reform model.
  • Buy-in by entire staff
  • Minimum of 80 buy-in.
  • Consensus of teacher support
  • Implementation must be by choice rather than
    mandated
  • Low teacher turnover
  • Commitment by staff to work beyond traditional
    levels to shift school paradigm.

54
What It Takes to Successfully Implement
Whole-School Reform
  • Political support by
  • School board
  • District leadership
  • Teachers associations/unions
  • Community
  • State education agency
  • State legislature

55
What It Takes to Successfully Implement
Whole-School Reform
  • Teacher perceptions
  • Need for reform
  • Readiness of students to learn
  • Support for all reform elements
  • Limited classroom management issues

56
What It Takes to Successfully Implement
Whole-School Reform
  • Consistent and stable leadership at school and
    district levels
  • Implement in well-managed schools
  • Strong principal with authority to make changes
    at the school level
  • Strong superintendent with commitment to program
  • Leadership communicates well with staff
  • Personal commitment to program

57
What It Takes to Successfully Implement
Whole-School Reform
  • Support of program developers and/or design team
  • Clear communication of program
  • Flexibility to adapt to local context
  • Ongoing training
  • Research support
  • Technical support
  • Ongoing teacher training
  • Orient new teachers
  • Provide ongoing support to continuing teachers

58
What It Takes to Successfully Implement
Whole-School Reform
  • Ongoing financial support
  • Appropriate reallocation of resources
  • Acquisition of new, ongoing funding

59
What It Takes to Successfully Implement
Whole-School Reform
  • Compatibility of assessments to curriculum
    standards and teaching strategies
  • Confidence that achievement gains will be
    captured by assessments used.
  • Appropriate use of student-level, teacher-level
    and aggregated school-level data
  • Consistency of implementation within and
    across schools
  • Absence of conflicting or distracting reforms
    being implemented at school
  • Absence of pressure from outside groups to
    implement alternative or additional reforms

60
What It Takes to Successfully Implement
Whole-School Reform
  • Implement on a small-scale before trying to
    implement on a large-scale
  • implement in 2 or 3 schools in a district,
    adding more schools gradually
  • Continued increase in schools implementing
    reform
  • Smaller schools easier to implement than
    secondary schools
  • Elementary schools easier to implement than
    secondary schools
  • Allow time for implementation 3-5 years minimum

61
What It Takes to Successfully Implement
Whole-School Reform
  • Long term commitment by school and district
    leadership 5-20 years
  • Research to study impact of program on student
    learning must be long-term
  • Results cannot be expected until 2-3 years after
    full implementation
  • Longitudinal data is necessary
  • Track students and staff over time
Write a Comment
User Comments (0)
About PowerShow.com