7/10/13 Charter School Performance in Illinois
Jun 19, 2015
7/10/13
Charter School Performance in Illinois
2
Overview
• Acknowledgements – ISBE (especially Brandon Williams and Peter Godard) – Joyce Foundation
• Study Approach
• Findings • Summary & Implications
Study Approach
3
4
Research Questions
• How do charter schools compare to TPS in their academic achievement gains?
• What characteristics of charter schools are associated with better achievement gains?
• Do charter schools have more success than
TPS working with certain student subgroups?
5
Impact Analysis: Years of Study
Test Administration
Spring 2009
Spring 2010
Spring 2011
Spring 2012
Academic Year Covered by Test
2008-09 2009-10 2010-11 2011-12
Grades 3-8 3-8 3-8 3-8
Growth Period 2010 2011 2012
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Demographics
19
Demographic Comparison
20
2010-2011
Source: Common Core of Data, National Center for Education Statistics
TPS Feeders Charters
Number of schools 4362 618 43Average enrollment per school 472 610 842Total number of students enrolled 2,060,340 376,985 36,201Students in Poverty 44% 82% 71%English Language Learners 7% 13% 6%Special Education Students 13% 13% 10%White Students 53% 15% 5%Black Students 18% 42% 62%Hispanic Students 21% 39% 31%Asian/Pacific Islander Students 4% 3% 1%Native American Students 0.2% 0.2% 0.1%
Charter Student Population
21
All Charter School Students 36,201
Tested Students 20,349
Matched Students 18,689
Average one-year student growth
(based on 3 growth periods)
Results
• Student-level Findings • School-level Findings
22
Student-level Findings
23
Impact of Charter Schools
24
Charter students outperform traditional public school (TPS) students in reading and math; results are slightly stronger for math than for reading.
.01**.01**
-.03**
.03** .03**
.01
-.1
.0
.1
Overall Illinois Chicago Not Chicago
Gro
wth
(in
sta
nd
ard
de
via
tio
ns)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Charter School Impact with Original School Cohort
25
Students in Chicago schools from the 2009 report outperform TPS in both subjects in the new time period, an improvement over the original results.
.00
.02**.02**
.03**
.00
.05
Overall Chicago 2009 Overall Chicago 2012
Gro
wth
(in
sta
nd
ard
de
via
tio
ns)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Charter School Impact by Growth Period
26
Note: Count of new schools is from student test data.
Charter students outperformed TPS in math in all growth periods; in reading, results were positive in two of three periods.
.02**
.04**
-.02**
.04**
.02**.03**
-.1
.0
.1
2010 2011 2012
Gro
wth
(in
sta
nd
ard
de
via
tio
ns)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
ReadingMath
New schools: 3 6 4Persisting : 46 51 61
Charter School Impact by CMO Affiliation
27
Charter students at CMO-affiliated and non-CMO schools perform better than TPS in math, but only CMO-affiliates do so in reading.
.01**.01
.03**
.04**
.00
.05
CMO Non-CMO
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Charter School Impact by Race/Ethnicity
28
Black students have similar learning gains in reading and math at TPS and charters. Hispanic students have similar learning gains in reading but better learning gains in math when they attend a charter compared to TPS.
-.18**-.20**
-.04*
-.06**
-.09** -.09**
-.03*
.04**
-.25
-.20
-.15
-.10
-.05
.00
.05
.10
Traditional Public School
Charter Traditional Public School
Charter
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Black Hispanic
Charter School Impact with Students in Poverty
29
Students in poverty have an advantage in charter schools in reading but similar gains in math compared to TPS students in poverty.
-.11**
-.08**
-.05** -.05**
-.15
-.10
-.05
.00
Traditional Public School Charter
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Charter School Impact by Race/Ethnicity and Poverty
30
Black students in poverty have better learning gains in reading at charters than at TPS. Hispanic students in poverty have better learning gains in math at charter schools than at TPS.
-.30**-.28**
-.15** -.14**-.14** -.13**
-.07**
-.004
-.35
-.30
-.25
-.20
-.15
-.10
-.05
.00
Traditional Public School
Charter Traditional Public School
Charter
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Black / Poverty Hispanic / Poverty
Charter School Impact with Special Education Students
31
Special education students in charter schools have similar learning gains as their TPS counterparts in both reading and math.
-.36** -.36**
-.19** -.19**
-.4
-.3
-.2
-.1
.0
Traditional Public School Charter
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
Charter School Impact with English Language Learners
32
There is no difference in learning gains for ELLs in charters and TPS.
-.27**-.28**
-.12** -.12**
-.3
-.2
-.1
.0
Traditional Public School Charter
Gro
wth
(in
sta
nd
ard
dev
iati
ons)
* Significant at p ≤ 0.05 ** Significant at p ≤ 0.01
Reading
Math
33
Charter Impact by City
.34**
.01**
.09**
.06**
.22**
.03**
.36**
.03**
.08** .08**
.26**
.14**
.00
.10
.20
.30
.40
Boston Chicago Detroit Indianapolis Newark New York City
Gro
wth
(in
sta
nd
ard
de
via
tio
ns)
** Significant at p ≤ 0.01 Reading Math
School-level Findings
34
Distribution of Charter School Impacts
35
Subject Number Percent Number Percent Number PercentReading 13 21.3% 36 59.0% 12 19.7%
Math 13 21.0% 26 41.9% 23 37.1%
Significantly Worse Not Significant Significantly
Better
Impact of Growth on Achievement - Reading
36
Growth(in Standard Deviations) 1.6%
70th Percentile
3.3% 8.2%50th Percentile
3.3% 23.0% 37.7% 3.3%30th Percentile
14.8% 4.9%
-0.15 0.15
Low Growth,High Achievement
High Growth,High Achievement
Low Growth,Low Achievement
High Growth,Low Achievement
0
Impact of Growth on Achievement - Math
37
Growth(in Standard Deviations) 1.6%
70th Percentile
1.6% 3.2% 4.8%50th Percentile
21.0% 41.9% 4.8%30th Percentile
3.2% 12.9% 4.8%
-0.15 0.15
Low Growth,High Achievement
High Growth,High Achievement
Low Growth,Low Achievement
High Growth,Low Achievement
0
Summary & Implications
38
39
Summary of Findings
• Typical charter student has greater learning gains than TPS – 2 weeks in reading – 1 month in math
• No consistent trend of charter benefits for specific student groups – Math: Hispanic students, Hispanic students in poverty – Reading: Students in poverty, black students in poverty
• Some charters have below-average growth & below-average achievement – Math: 37% – Reading: 41%
VULNERABLE
Implications
• Decent baseline for Chicago, less so for rest of state
• Substantial need for stronger oversight – low quality has ripples across sector
• Positive aspects of charter experience – role for INCS to find & leverage these?
• Strong performance may not sustain in reading
40
Questions?
41
Thank You
42