PowerPoint Presentation - MyData Portal - Dallas ISD

Document Sample
PowerPoint Presentation - MyData Portal - Dallas ISD Powered By Docstoc
					Dallas ISD’s Value-Added Model

   School Effectiveness Index (SEI)
 Classroom Effectiveness Index (CEI)

    Data Analysis, Reporting, and
         Research Services
       Why Dallas ISD Uses Indices


 To gauge students’ progress in relation
  to their peers
 To hold schools and teachers
  accountable for the improvement of all
  students, both those who are not passing
  and those who are
 To reward improvement, not just passing
  rates

                                             2
          Objectives of this Segment


 Explain the Indices (SEIs and CEIs)
  without complex formulas and statistics
     Basics of value-added models
     Computation of the Indices
 Address common concerns




                                            3
                    Passing Rates


 Passing rates are important
     Demonstrate efforts of schools and teachers
     Main components of state (AEIS) and federal (AYP)
      accountability systems
     Reflect a necessary minimum standard of
      achievement
 Passing rates are insufficient
     Innate student differences are ignored
     Performance alone tells little about growth and
      effect of instruction
     “Setting the bar” fails to challenge proficient and
      excellent students

                                                            4
             Value-Added Measures


 Measures based on value-added models
  address these issues
     By factoring in characteristics that may
      impact students’ learning (gender, ethnicity,
      language proficiency, socio-economic
      status, etc.)
     By measuring a student’s change in
      performance relative to her peers’
     By creating comparison measures for all
      students, every year
                                                      5
            What is “Value-Added”?


 A value-added model measures the
  “academic value” added to students after
  a year of instruction
 Components:
     Previous level of achievement (academic
      value at the end of the prior school year)
     Current level of achievement (academic
      value at the end of this school year)
     Difference (change, growth, gain, etc.)

                                                   6
           Value-added “Growth”


A note to which we will return…

In the Dallas ISD,

  “Growth” ≠
    Current score – Previous score




                                     7
           Are the Indices “Fair”?


 “Our (My) students were struggling students
  to start with. That’s why they didn’t do as
  well as other students.”
 “Most of our (my) students were limited
  English proficient. We (I) can’t be compared
  to schools (teachers) that had only non-LEP
  students!”
 “Our (My) students didn’t pass, but they did
  much better than last year. Shouldn’t that
  count?”
                                                 8
           Are the Indices “Fair”?


 “Our (My) students had high scores last
  year. They didn’t have much room to ‘grow,’
  not like students with low scores.”
 “We (I) had many of our (my) students for
  only a few months. How can we (I) be held
  accountable for their progress?”




                                                9
               Fairness Variables


 These questions are valid
 A value-added model for accountability
  does address these fairness issues
     Compare students with the same previous
      level of achievement
     Compare students with the same
      demographic characteristics
     Evaluate change in achievement without
      regard to arbitrary standards
     Include only students at the school/in the
      classroom for most of school year
                                                   10
               Fairness Variables


Student performance controlled for:
     Previous level of achievement
     Gender
     Ethnicity
     English-language proficiency
     Free or reduced-price lunch status
     Neighborhood family income
     Neighborhood education level
     Neighborhood poverty index
                                           11
       Grouping Students on Fairness Variables
     Identify unique groups of students districtwide

                                        Grade 3 students

                         Male                                       Female

            LEP                 Non-LEP                       LEP           Non-LEP

 Student                     Student    Student    Student      Student    Student    Student
                  Student
 Variable                    Variable   Variable   Variable     Variable   Variable   Variable
                  Variable




Group A                 Group B         Group C           Group D ... Group n

                                                                                             12
          Criteria for “Eligibility”

 Continuously enrolled: in attendance for a
  minimum number of instructional days
 For measures based on the TAKS, not
  retained in either of last two years
 Appropriate scores from last year and
  current year
 (CEIs) Received instruction in all six-week
  grading periods

                                            13
             Value-added “Growth”


 Growth ≠ Current score – Previous score
 Computation of “growth”
     For each unique group, determine the
      expected score on the current year test
     Evaluate a student’s performance based on
      how far from expected it was
 Potential consequences:
     Student’s score “falls”  above expectation
     Student’s score “rises”  below expectation
                                                    14
                                                                  Value-Added Growth
                                                 Students in a Group with 40% of Items Correct Last Year

                                      50

                                      45

                                      40
# Items Correct (current-year test)




                                                                                         Students scoring above the district
                                      35                                                 average exceed expectation

                                      30
                                               50% Correct
                                      25                                            District students that started with 40% of items correct,
                                                                                    on average got 50% of items correct this year.
                                      20
                                                                                         Students scoring below the district average
                                      15                                                 did not meet expectation

                                      10

                                      5
                                                                       40% Correct
                                      0
                                           0       5         10   15        20        25         30        35       40         45       50
                                                                       # Items Correct (one prior-year test)
                                                                                                                                         15
                                                                       Value-Added Growth
                                                   Students In a Group with 80% of Items Correct Last Year

                                      50
                                                             Students scoring above the district
                                      45                     average exceed expectation

                                      40
# Items Correct (current-year test)




                                               70% Correct
                                      35

                                      30        District students that started with 80% of items correct,
                                                on average got 70% of items correct this year.
                                      25

                                      20                     Students scoring below the district average
                                                             did not meet expectation
                                      15

                                      10

                                      5
                                                                                                                           80% Correct
                                      0
                                           0         5         10        15         20         25        30           35      40         45   50
                                                                              # Items Correct (one prior-year test)
                                                                                                                                              16
                                                                            Value-Added Growth
                                                       Two Entities with Passing Rate Increase (20% to 30%)

                                      50
                                               “Fail”  “Pass”                                                                     “Pass”  “Pass”
                                      45


                                      40
                                                   School/Teacher A: All but two students
# Items Correct (current-year test)




                                      35           exceeded expectation = high Index

                                      30


                                      25


                                      20

                                                                                                         School/Teacher B: All students failed
                                      15
                                                                                                         to meet expectation = low Index
                                                                 Student starting with
                                      10
                                                                   20 items correct
                                      5
                                               “Fail”  “Fail”                                                                     “Pass”  “Fail”
                                      0
                                           0       5         10           15           20           25            30       35        40          45   50

                                                                          # Items Correct (one prior-year test)


                                                                                                                                                      17
                         Value-Added Growth
             Two Entities with Passing Rate Decrease (80% to 70%)

50


45
                        School/Teacher D: 6/10 students
                        exceeded expectation = high Index
40


35


30


25
                                                                    Student starting with
                                                                      40 items correct
20                           School/Teacher C: Only 1/10 students
                             exceeded expectation = low Index
15


10


 5


 0
     0   5        10    15       20       25       30       35      40       45        50




                                                                                       18
                   Indicators

 CEIs
     TAKS
     Norm-referenced
     Assessment of Course Performance (ACPs)
 SEIs, above plus
     graduation rate
     SAT/PSAT/ACT participation
     PSAT averages
     Percentage passing AP exams
     Percentage enrolled in AP courses
                                                19
            Summary: Indices


 Measure amount of academic progress
  after receiving a year of instruction
 High Indices indicate more progress than
  similar students across the district
 Provide an additional tool to determine
  progress
 Cannot be used in isolation from other
  tools: they don’t tell how to effect the
  change, just provide evidence
                                             20
                           Contact

Evaluation & Accountability
  972-925-3503
  http://www.dallasisd.org/eval/

Data Analysis, Reporting, and Research Services
  972-925-6446
  OIR@dallasisd.org
  MyData Portal: https://portal.dallasisd.org/mydata

Teaching & Learning
  ccentral@dallasisd.org
  Curriculum Central: https://portal.dallasisd.org/curriculum
                                                             21

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:12
posted:1/20/2012
language:English
pages:21