October 24 2012 Jonathan Wiens PhD Accountability and Reporting Oregon Department of Education Growth Model Overview Oregons Growth Model Uses the Colorado Growth Model Includes all students having two consecutive years of standard OAKS assessments regardless of whether or not th ID: 713751
Download Presentation The PPT/PDF document "Growth Targets for 2013-2014" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Growth Targets for 2013-2014
October 24, 2012
Jonathan Wiens, PhD
Accountability and Reporting
Oregon Department of EducationSlide2
Growth Model
OverviewSlide3
Oregon’s Growth Model
Uses the Colorado Growth Model.
Includes all students having two consecutive years of standard OAKS assessments, regardless of whether or not they are meeting standard.
A student’s growth is compared to the growth of other students in the state having the same prior test scores (“Academic Peers”)
Student Growth is expressed as a percentile. A growth percentile of 75 would mean the students growth was as high or higher than 75 percent of his/her academic peers.Computes Growth Targets – growth percentiles that put a student on track to be at standard in three years.
3Slide4
Growth Model Points to Remember
Growth is based on comparing a student to his or her Academic Peers, who are students with the similar test score histories.
Growth
of low performing students is compared to that of other low performing students in the state.
Growth of high performing students is compared to that of other high performing students in the state.Growth model applies to students in grades 4 to 8, and 11.Uses up to four years of test data for each student.** - High school students use only the 7th, 8th
and 11
th
grade scores to evaluate growth.
4Slide5
Growth and School Accountability
Schools are given “Levels” in reading and math growth:
Level 5
: Schools with high growth
Level 4: Average to above average growthLevel 3: Below average, but not low, growthLevel 2: Low growthLevel 1: Very low growthReading and math growth are combined into a Growth Rating.
5Slide6
School Growth Ratings
School accountability uses the median growth percentile.
Median growth is the “middle” growth percentile.
This is the “typical” growth at the school.
We also report the median target growth percentile.A school has “On Track” growth if the median growth percentile is as high as the median target percentile. “On Track” growth indicates that a typical student is meeting his/her growth target.Requirements to reach Level 5, Level 4, etc., are lower for schools with “On Track” growth.
6Slide7
Growth Example - Elementary
7
Here
is an example of how growth is reported on the School Ratings detail reports. Subgroup growth is also reported. Slide8
Growth Percentiles in Oregon
8Slide9
Sample Growth Percentiles
9
Grade 5
Grade
6
Gain
Growth Percentile
Growth Target
195
200
5
16
80
209
208
-1
8
72
209
214
5
34
64
209
218
9615820922617964822723255727
This sample shows various growth percentiles. Note that the middle four students all had the same starting point in 5
th
grade. The students in red are shown to emphasize that growth is evaluated relative to academic peers, not on absolute gains in test scores.Slide10
SGPs – Higher Order Growth
10
Student
Grade
3
Grade 4
Grade 5
Growth Percentile
Affect of using 3
Yea
rs of data
A
--
211
215
34
--
B
190
211
215
66
+32
C
195
21121555+21D20021121542+8E20621121527-7
F
212
211
21517-17G2182112158-26
This data show how 3 years of test scores can affect growth percentiles. Data are taken from Math growth in 2011-12.Slide11
Growth Targets for
School Accountability Slide12
Growth Targets
Growth targets are
forward looking
.
They estimate the growth necessary to meet standard in three more years, or by grade 11.Targets are provided both as percentiles and as RIT scores.The target RIT score represents the typical score attained by students who grew to
standard in the past.
The target percentile should be viewed as an estimate of the difficulty of attaining the goal of proficiency in three years.
12
Grade
Target Grade
3
6
4
7
5
8
6
11
7
11
8
11Slide13
Target Growth
13
Student
Grade
3
Grade 4
Grade 5
Growth
Percen
tile
Target Percentile
6
th
Grade Target RIT
B
190
211
215
66
69
220
C
195
211
215
5568221D2002112154266221E2062112152764222F
212
211
215
1763222G2182112158
62
223
This data show how
using three years
of test scores
affect
growth
percentiles and create a more complete view of growth.
Data are taken from 5
th
grade Math growth in 2011-12.Slide14
Growth Targets, continued
14
The data below shows how important it is to remember that the growth data is based on “academic
peers,”
who are students with similar score histories.
Targets
for Students with Same Prior Test Score
Student
Math Score Histories
7th Grade Growth
Percentile
8th Grade Growth Targets
4th
5th
6th
7th
Percentile
RIT
A
205
212
213
225
66
73
229
B
225
228
222225459233Slide15
2012-13State Level
Data Slide16
Reading Growth by Subgroup
16
Growth does vary by subgroup, but by far less than status varies. The differences below amount to only one or two RITs points.Slide17
Mathematics Growth by Subgroup
17
Growth does vary by subgroup, but by far less than status varies. The differences below amount to only one or two RITs points.Slide18
Interpreting Growth
at the
Student LevelSlide19
Typical Growth
While growth targets are used for school accountability, it is important to remember that growth for individual students should be a much more nuanced conversation.
To help teachers and parents interpret growth, we classify growth in one of three categories:
Low Growth
: growth below the 35th percentile.Typical Growth: growth between the 35th and 65th percentile.High Growth: growth above the 65th percentile.At this point we have not had the chance to produce individual student growth reports, but if we did …
19Slide20
Student Growth Report (forthcoming)
This is a sample that would show (in graphic format) the data you have access to today.
Typical Growth
: growth between the 35
th and 65th percentile.High Growth: growth above the 65th percentile.At this point we have not had the chance to produce individual student growth reports, but if we did …
20Slide21
Growth Goals
A thorough discussion of individual growth goals should take into account:
The growth target (i.e., target for proficiency in three years)
The
Low, Typical, and High growth ranges.Remember that student growth occurs in a range and that, by definition:about one third of students are likely to show low growth;about one third of students are likely to show typical growth; andabout one third of students are likely to show high growth.For students well above standard the “target” represents fairly low growth.
For students well below standard the “target” represents high growth.
21Slide22
Growth Goals, continued
Even though average growth does differ slightly by subgroup, the ranges for low, typical, and high growth are still reasonable, regardless of subgroup.
However, the fact that the English Learner (EL) and Students with Disabilities (SWD) subgroups have lower growth shows that the Oregon Growth Model is not a true value-added model.
Teachers with high percentages of EL or SWD students might want to take this into account when setting growth goals using this growth model.
22Slide23
Review of Data File
Includes all students enrolled in grades 3 to 8 on May 1, 2013.
Includes resident school and district from May 1.
Includes resident and attending school and district from SSID (as of October 23).
Extended assessment are not included. Students taking extended assessments will not have any growth or target data.A simplified table of growth targets is also available at: http://www.ode.state.or.us/search/page/?id=3797
23Slide24
How to get your data
Contact your ESD Partners
:
http://
www.ode.state.or.us/wma/teachlearn/testing/oaks/esdpartners1314.pdfIndividual school requests are discouraged, it creates a huge workload issue.Instead, District Test Coordinators should contact the ESD Partners to request data.DTCs can then send the data to internal district staff.Remember: these files contain FERPA protected data!
24Slide25
Contact Information and Links
Main report card page:
http://www.ode.state.or.us/search/page/?
id=1786
Details on priority, focus, and model schools: http://www.ode.state.or.us/search/page/?id=3742Report Cards and Report Card ratings details: http://www.ode.state.or.us/data/reportcard/reports.aspxRepot Card resources: http://www.ode.state.or.us/search/page/?id=3881Contacts:Jon Wiens: email: jon.wiens@state.or.us phone: 503-947-5764Josh Rew: email
josh.rew@state.or.us
phone: 503-947-5845
25