DDM Webinar Part 6: Determining How to Integrate Assessments

Report
Waiting Room
Today’s webinar will begin shortly.
REMINDERS:
• Dial 800-503-2899 and enter the passcode
6496612# to hear the audio portion of the
presentation
• Download today’s materials from the sign-in
page:
• Webinar Series Part 6 PowerPoint slides
• Correlation Example Excel file
Determining How to
Integrate Assessments into
Educator Evaluation:
Developing Business Rules
and Engaging Staff
Webinar Series Part 6
Webinar Series
Title
Date
Length
Time
1
Introduction: District-Determined Measures and
Assessment Literacy
3/14
60 minutes
4-5pm
2
Basics of Assessment
4/4
90 minutes
4-5:30pm
3
Assessment Options
4/25
60 minutes
4-5pm
TA and Networking Session I
7/11
3 hours
9am-12pm
4
Determining the Best Approach to DistrictDetermined Measures
7/18
60 minutes
4-5pm
5
Measuring Student Growth and Piloting DistrictDetermined Measures
8/15
60 minutes
4-5pm
TA and Networking Session II
9/19
3 hours
2:30pm5:30pm
6
Integrating Assessments into Educator
Evaluation: Developing Business Rules and
Engaging Staff
10/24
60 minutes
4-5pm
7
Communicating Results
12/5
60 minutes
4-5pm
TA and Networking Session III
12/12
3 hours
2:30pm5:30pm
Sustainability
1/23
60 minutes
4-5pm
8
2
Audience & Purpose
Target audience
District teams that will be engaged in the work of
identifying, selecting, and piloting DistrictDetermined Measures.
After today participants will understand:
Examples of practical solutions to issues of fairness
in using District-Determined Measures (DDMs).
Practical examples of engaging educators in the
process of implementing DDMs.
3
Agenda
Student Impact Rating Rollout Reminder
DDM Comparability
Identifying Bias
Standardizing DDMs
Ensuring Sufficient Variability
Q&A and Next Steps
4
Student Impact Rating Rollout:
Date
Action
Sept. 2013:
Decide which DDMs to pilot and submit list to ESE.
Sept. 2013 –
June 2014:
Pilot DDMs in at least the five required areas and research
DDMs in additional areas.
June 2014:
Submit final plans, including any extension requests, for
implementing DDMs during the 2014-15 school year*.
SY 2014-2015
Implement DDMs and collect Year 1 Student Impact
Rating data for all educators (with the exception of
educators who teach the particular grades/subjects or
courses for which an extension has been granted).
SY 2015-2016
Implement DDMs, collect Year 2 Student Impact Rating,
and determine and report Student Impact Ratings for all
educators (with the exception of educators who teach the
particular grades/subjects or courses for which a district
has received an extension).
*ESE will release the June 2014 submission template and DDM implementation
extension request form in December 2013.
5
DDM Key Questions
Is the measure aligned to content?
Does it assess what the educators intend to teach
and what’s most important for students to learn?
Is the measure informative?
Do the results tell educators whether students are
making the desired progress, falling short, or
excelling?
Do the results provide valuable information to
schools and districts about their educators?
6
Refining your Pilot DDMs
 Districts will employ a variety of approaches to identify
pilot DDMs (e.g., build, borrow, buy).
 Key considerations:
1. How well does the assessment measure growth?
2. Is there a common administration protocol?
3. Is there a common scoring process?
4. How do results correspond to low, moderate, of
high growth?
5. Is the assessment comparable to other
DDMs?
 Use the DDM Key Questions and these considerations
to strengthen your assessments during the pilot year.
7
DDM Comparability: Two Types
DDMs must be “comparable across
schools, grades, and subject matter
district-wide.” (Per 603 CMR 35.09(2)a)
Comparability = Two types
(Type 1) Comparable across schools
(Type 2) Comparable across grades and
subject matter
Learn more in Technical Guide B, page 9
and appendix G
8
Comparability (Type 1)
Comparable across schools
Example: Teachers with the same job (e.g., all 5th
grade teachers)
Where possible, measures are identical
Easier to compare identical measures
Do identical measures provide meaningful information
about all students?
When might they not be identical?
Different content (different sections of Algebra I)
Differences in untested skills (reading and writing on math
test for ELL students)
Other accommodations (fewer questions to students who
need more time)
9
Error and Bias
 Error is the difference between true ability and a
student’s score.
Random error
Student sleeps poorly, lucky guess, … etc
Systematic error (bias)
 Error occurs for one type or group of students
 ELL student misreads a set of questions
 Systematic Error = Bias
 Why This matters?
 Error (OK) decreases with longer/additional measures
 Bias (BAD) does not decrease with longer/additional measures
 Even with identical DDM, bias threatens comparability
10
When does bias occur?
Situation: Students who score high on the pretest have less of an opportunity to grow
because they cannot get more than a top
score (Ceiling Effect).
Situation: Special education students gain
fewer points from pre-post test, and as a
result are less likely to be labeled as having
high growth.
11
Checking for Bias
Do all students have an equal chance to grow?
Is there a relationship between the initial score and
gain score?
We can do this in EXCEL using correlation
 We have
 Pre-Test Score
 Post-Test Score
 Gain Score
Correlation formula in Excel:
=CORREL(PRE-TEST SCORES, GAIN SCORES)
 Type “=correl”, click formula
 Highlight Pre-Test Scores, Press “Comma”
 Highlight Difference Scores, Close Parentheses, Press “Enter”
12
Interpreting Correlation
 Correlation is the degree to which two numbers are
related
 Correlation
 Number between -1 and 1.
 A zero correlation means numbers are unrelated
 Closer to 1 or -1 means strong correlation
 DDMs should provide all students an opportunity to
demonstrate growth
 We want to see little to no correlation between pre-test
scores and gain scores
 A correlation above .3 or below -.3 suggests that there are
systematic differences in gain for low and high ability students
13
Correlation Example
Demonstration of computing Correlation
between pre-test and gain
Very Low Correlation
students of all ability were equally likely to
demonstrate growth
Negative Correlation
Students of high ability systematically
demonstrated less growth (due to ceiling effect)
Positive Correlation
Students with lower scores generally grew less
(bias)
14
Interpreting Correlation
Strong correlation is an indication of a problem
A low correlation is not a guarantee of no bias!
Strong effect in small sub-population
Counteracting effects at both low and high end
Use common sense
Always look at a graph!
 Create a scatter-plot graph and look for patterns
15
Example of Bias at Teacher Level
Teacher A
Teacher B
Pre
Post
Gain
Pre
Post
Gain
3
4
1
3
4
1
3
4
1
8
14
6
3
4
1
8
14
6
3
4
1
8
14
6
8
14
6
8
14
6
Even though similar students gained the same amount
Teacher A’s average gain is 2
Teacher B’s average gain is 5
16
Solution: Grouping
Grouping allows teachers to be compared
based on similar students, even when the
number of those students is different
Low
Students
High
Students
Teacher
Average
Growth
A
1
B
1
A
6
B
6
17
Addressing Bias: Grouping
How many groups?
What bias are you addressing?
Enough students in each group?
Using Groups
Weighted average
Rule based (all groups must be above cut off)
Professional judgment
18
Comparability (Type 2)
Comparability across different DDMs
Across different grades and subject matter
Are different DDMs held to the same standard of
rigor?
Does not require identical number of students
in each of the three groups of low, moderate,
and high
Common sense judgment of fairness
19
One option: Standardization
Standardization is a process of putting
different measures on the same scale
For example
 Most cars cost $25,000 give or take $5,000
 Most apples costs $1.50 give or take $.50
 Getting a $5000 discount on a car is about equal to what
discount on an apple?
Technical terms
“Most are” = mean
“Give or take” = standard deviation
20
Guest Speaker
Jamie LaBillois –
Executive Director of Instruction,
Norwell Public Schools
21
Developing Local Norms
 Student A






English:
Math:
Art:
Social Studies:
Science:
Music:
15/20
22/25
116/150
6/10
70/150
35/35
 We learned early on that we needed a process
that would create one universal measurement
unit to discuss student progress.
22
Transform the Data…
23
How?
 Step One
 Calculated the difference between Post
and Pre (or any approach from Technical
Guide B)
 Step Two
 Find the mean (average) of the difference
scores
 Step Three
 Find the standard deviation of the
difference scores
24
How?

Now, we’re ready to “transform” the difference
scores into a universal measurement system.

Step Four
 Calculate the z-score of each individual
difference score
(observation – Mean)
Z = -----------------------------------Standard Deviation

Step Five
 Calculate percentile rank for each z-score
25
Developing Local Norms
 Student A






English:
Math:
Art:
Social Studies:
Science:
Music:
15/20
22/25
116/150
6/10
70/150
35/35
 Student A






English:
Math:
Art:
Social Studies:
Science:
Music:
62
72
59
71
70
61
%ile
%ile
%ile
%ile
%ile
%ile
26
Examining an Educator’s Impact
 Grade 4 DIBELS Oral Reading Fluency
 MEDIAN %ile per class:








Teacher
Teacher
Teacher
Teacher
Teacher
Teacher
Teacher
Teacher
1:
2:
3:
4:
5:
6:
7:
8:
65
71
59
59
62
57
29
50
%ile
%ile
%ile
%ile
%ile
%ile
%ile
%ile
Evaluator’s
Focus
27
Lessons Learned







Growth vs. Achievement
Robust Tool
Timely Analysis
Re-Assessment of Instruction
Re-Assessment of Ability vs. Disability
Development of Building-Based Evaluators
Educator Engagement is Essential
28
Ensuring Sufficient Variability
Technical Guide B’s two key questions:
Is DDM aligned to content?
Does the DDM provide information to educators
and evaluations?
Lack of variability reduces information
29
Looking for Variability
Problematic
200
200
150
150
# of students
# of students
Good
100
50
100
50
0
0
Low
Moderate
High
Low
Moderate
 The second graph is problematic because it doesn’t give us
information about the difference between average and high
growth because so many students fall into the “high” growth
category.
High
30
Guest Speaker
Experience with constructing measures with
greater variability
31
Wrap-Up
Today, we discussed three strategies for
evaluating the fairness of your DDMs
1. Check for bias by computing the correlation
between pre-test scores and gain scores.

Remember: Zero correlation indicates that all students have
an equal chance to demonstrate growth.
2. Standardization can help you compare DDMs in
different content areas.
3. Look for variability in student growth. A lack of
variability reduces the amount of information
available to educators about their students.
32
Resources
Available Now at http://www.doe.mass.edu/edeval/ddm/:
 Technical Guide B
 DDMs and Assessment Literacy Webinar Series
 Technical Assistance and Networking Sessions
 Core Course Objectives and Example DDMs
Coming Soon
 Using Current Assessments Guidance (Curriculum
Summit)
 Model Contract Language
 DDM Pilot Plan Cohorts
33
Register for
Webinar Series Part 7
Part 7: Communicating Results
Date: December 5th, 2013
Time: 4-5pm EST (60 minutes)
Register: https://air-event500.webex.com/airevent500/onstage/g.php?d=597905353&t=a
34
Questions
Contact
Craig Waterman at [email protected]
Ron Noble at [email protected]
Feedback
Tell us how we did:
http://www.surveygizmo.com/s3/1421848/Dist
rict-Determined-Measures-amp-AssessmentLiteracy-Webinar-6-Feedback
35

similar documents