Improving Schools
http://imp.sagepub.com
Creating small schools in Chicago: an early look at implementation and impact
Joseph E. Kahne, Susan E. Sporte and John Q. Easton
2005; 8; 7
Improving Schools
DOI: 10.1177/1365480205048929
The online version of this article can be found at:
http://imp.sagepub.com/cgi/content/abstract/8/1/7
Published by:
http://www.sagepublications.com
can be found at:
Improving Schools
Additional services and information for
http://imp.sagepub.com/cgi/alerts
Email Alerts:
http://imp.sagepub.com/subscriptions
Subscriptions:
http://www.sagepub.com/journalsReprints.nav
Reprints:
http://www.sagepub.co.uk/journalsPermissions.nav
Permissions:
http://imp.sagepub.com/cgi/content/refs/8/1/7
Citations
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Creating small schools in Chicago: an
early look at implementation and impact
Joseph E. Kahne®
Mills College, California
Susan E. Sporte and John Q. Easton®
Consortium on Chicago School Research
Abstract
There is widespread concern in many US cities about student alienation, dropout and under-
achievement in large urban high schools. Chicago, with support from the Gates Foundation
and other agencies, has embarked on a major reform involving the establishment of new
small schools and the division of larger schools into smaller ones. This article reports on
early evaluations of the project, alongside previous evaluations of small high schools. It
highlights early effects such as: improved relationships; cooperation between teachers;
awareness of personal difficulties; student engagement; and better attendance. Although it
is too early to demonstrate the result in terms of summative attainment measures, these
clearly provide a strong foundation.
Keywords community, ethos, small schools, trust
Increasingly, researchers, policymakers, school leaders, and concerned citizens are
recognizing that high schools in the United States are in need of major reform – or, as
some have put it, in need of being re-invented (Harvey and Housman, 2004). Current
research often highlights that high schools are not preparing students for college, work,
or life (American Diploma Project, 2004), and that they lead to increased alienation
(American Youth Policy Forum, 2001). The problem is especially severe in large urban
high schools, which disproportionately serve students of low socioeconomic status and
students of color. In Chicago’s public schools, for example, where more than 85 per cent
of students come from low income backgrounds and 91 per cent are students of color,
only 29 per cent of all 9th graders were reading at grade level in 2000 and the cumula-
tive dropout rate for the most recent cohort of students was 43 per cent (Allensworth,
2004: 10; Allensworth and Easton, 2001) As Daniels et al. write, ‘America’s high
schools are failing all of our kids some of the time and some of our kids all the time’
(2001: 22).
As a result of these concerns, many educational reformers are looking to small schools as
a possible response to what Powell et al. (1985) characterized as impersonal, incoherent,
Improving Schools © SAGE Publications
Volume 8 Number 1 March 2005 7–22
ISSN 1365-4802 DOI: 10.1177/1365480205048929
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
8 Improving Schools 8(1)
and ineffective comprehensive ‘shopping mall’ high schools. This focus reverses a
trend often associated with James Bryant Conant (1959) who, roughly 50 years ago,
argued that small rural schools would be less effective than larger comprehensive high
schools that could provide students with greater opportunities through an appropriately
differentiated curriculum.
Reform focused on smaller, more personal schools has been spurred by educators such
as Deborah Meier (1995), researchers such as Fred Newmann and Gary Wehlage
(1995), and by foundations. Most notably, the Annenberg Foundation’s $500 million
school reform effort emphasized reducing school size and the Bill and Melinda Gates
Foundation has now committed $647 million (Hendrie, 2004) to improving schools
through the creation and replication of small high schools. Energized by these efforts,
the city of Chicago and other urban districts are placing the creation of small schools
at the center of their high school improvement strategies.
Small school reform in Chicago
Chicago provides a valuable setting from which to consider the challenges and possi-
bilities of small school reform. It is the third largest district in the United States, serv-
ing some 439,000 students. Slightly more than 100,000 of these students are in grades
9–12, attending one of 95 high schools, including roughly a dozen small schools and
charter schools. In addition, as noted above, overall student performance in high
schools is unacceptably low. In response to this problem, the district, like many across
the country, has instituted a wide range of policies to improve educational opportuni-
ties and outcomes for these students. The district has made schools accountable for
student performance, sanctioning those who do not meet test-score performance levels;
has created a system of selective enrollment schools intended to attract and retain high-
achieving students; has strengthened the list of courses required for graduation, and has
created both opportunities and requirements related to professional and curricular
development (Lee, 2002).
As one part of their strategy for high school reform, the district, the Gates Foundation,
and local funders recently launched the Chicago High School Redesign Initiative
(CHSRI). CHSRI aims to support redesign by developing almost three dozen new
small high schools.1 It began in September 2001 through a $12 million grant from the
Bill and Melinda Gates Foundation. This grant was matched by $6 million from foun-
dations in Chicago. The goal of the first phase of this initiative was to convert up to
five large high schools into 15–20 autonomous small schools over five years. In April
2003, CHSRI received a second grant of almost $8 million from the Gates Foundation
to open 12 new (rather than converted) small high schools over five years (Office of
Small Schools, CPS). The first five autonomous small schools opened within three
converting buildings in the fall of 2002.
To learn from and support this process, we have begun a three-year study of this effort.
Our goal is to provide formative and summative analyses for the Initiative and to add
to the broader dialogues on the reform of low-performing urban high schools. Now one
year into this study, this article focuses on the five converting small schools that opened
in the fall of 2002. We report findings regarding both implementation and impact as a
means of raising issues and considering possibilities – since the initiative is just begin-
ning, it is far too soon for any summative judgments.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 9
What we know from prior research and what we need to know
Various reviews of the literature on student achievement and school size indicate that
this strategy does indeed improve student outcomes (Cotton, 1996; Darling-Hammond
et al., 2002; Haller, 1993; Holland, 2002; Howley, 1989; Lee, 2002). There is also
evidence that small schools can promote more equitable access to academically
demanding courses (Bryk et al., 1993), more equitable achievement gains (Darling-
Hammond et al., 2002; Lee & Smith, 1995), and lower dropout rates (Darling-Hammond
et al., 2002; Holland, 2002; Pittman & Haughwout, 1987). In addition, Barker and
Gump’s seminal study (1964) also demonstrated that small schools can dramatically
increase students’ opportunities for participation and leadership in a wide range of
extracurricular activities, so even marginal students get involved. Not all findings are
positive, however. Wasley et al. (2000) and Hess and Cytrynbaum (2002) both studied
small schools in Chicago and found enhanced engagement but not a consistent impact
on student achievement. More recently, when asked about the academic value of small
schools, Tom VanderArk, executive director of the Gates Foundation, admitted that
‘proof’ was hard to come by.2 Other research suggests that smallness alone is not
enough; for example, the beneficial value of small schools depends on the degree to
which they promote factors such as personalization and an interactive, authentic and
challenging curriculum while avoiding inequitable divisions in students’ opportunities
to learn (see Darling-Hammond et al., 2002; Lee, 2002).
Thus, although much has been written about small high schools, and although the foun-
dation community and many reformers believe that this strategy can improve the life
chances of high school students, the evidence base is quite limited. Indeed, the promo-
tional nature of much of the literature limits our confidence in some of the findings. In
addition, while current scholarship indicates that small schools are often associated
with improved outcomes, neither the magnitude nor the consistency of these effects are
clear, and the factors that can lead small schools to produce improved outcomes are not
well understood. Often identification of features responsible for success is based on
case studies of effective small schools and not on comparisons of effective and ineffec-
tive small schools with controls for demographic and other relevant variables. In some
of the qualitative literature, there is no or insufficient discussion of the methodological
approach that led to identification of features responsible for desired outcomes. Many
studies do not compare key elements of design strategy or small school structure –
schools within schools, freestanding small schools, newly created small schools,
schools created by breaking up large schools. Finally, we don’t currently know as much
as we need to about policy efforts to implement small schools on a large scale. And,
our understanding of these policy decisions and their relation to successful small
schools and successful implementation of small school reform is quite limited.
Our study represents an attempt to respond to these questions. We are interested in
understanding district-wide efforts to promote small school reform – we want to know
both about the implementation challenges posed by this policy direction and about the
impact of the reform on teachers and, most importantly, students. In addition, we
ultimately hope to better understand the relationship between changes in the contexts
created for teachers and students and ultimate outcomes for students. Since, at this
point in time, we are only able to report on the first year of the effort, our goals for
this article are much more limited. We aim to inform answers to three broad sets of
questions:
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
10 Improving Schools 8(1)
• How is Chicago’s effort to implement small schools on a large scale proceeding?
• Are small schools creating the contexts for principals, teachers, and students that
reformers believe will ultimately lead to desired reform and improved outcomes?
• What are early indications of the small schools’ impact on student outcomes?
To provide a framework for this discussion, the section that follows outlines a model
that details mechanisms through which the creation of semi-autonomous small schools
of choice is thought to lead to desired outcomes.
Our framework for inquiry
To respond to issues noted above, rather than testing whether outcomes at small schools
are better than those at larger ones, we believe it makes more sense to examine whether
small semi-autonomous schools of choice create conditions that enable improved
opportunities and supports for students and whether these, in turn, foster improved out-
comes. This framework, or theory of change, is detailed in Figure 1. It was developed
through consideration of relevant literature, analysis of documents related to the
Initiative, and through interviews and discussions with key stakeholders. The frame-
work or ‘Theory of Change’ (Weiss, 1995) portrays the mechanisms through which
varied features of small school reform are thought to promote desired contexts for
students and teachers. It also details how these contexts, in the presence of district, state
and federal influence, can promote both curricular change and desired outcomes.
The basic argument views small schools as enabling but not as guaranteeing improve-
ment. Proponents believe that small, largely autonomous, schools of choice will be
likely to foster desirable contexts for teachers and students – contexts characterized by,
for example, trust, coherent vision, and commitment. These contexts, in turn, will
enable the development of high quality instruction and personalized academic and
social support which will, in turn, lead to desired student outcomes. As Michele Fine
has written:
Small may be a necessary condition for a nonselective high school to excel. Small is neces-
sary if teachers are to have rich conversations with one another about practice, policy,
inquiry, and student work. Small is necessary if students are to feel attached to each other
and to faculty. (Fine, 1998: 4 as cited in Husbands & Beese, 2001)
Specifically, as outlined in Figure 1, those designing and working to implement small
schools situated within school districts posit that desirable student outcomes will be the
product of numerous factors. First, and primarily, they believe that if a district (Box 1;
box numbers refer to those in Figure 1) creates small, voluntary, relatively autonomous
schools and limits bureaucratic control (Box 2) that it will create schooling contexts
where trust, coherent vision, and commitment will be more likely to take root (Box 3).
Such schools will also have distributed leadership and be marked by strong and vibrant
professional communities, where teachers share in decision making, reflect on and
share practice, and collaborate with each other. Moreover, they believe that creating
such contexts for teachers and principals is fundamentally important as a means of
fostering a setting where productive reform can occur (Box 6). Clearly, both the
creation of this context for teachers and principals and the broader context for reform
are also being shaped by the provision of resources and supports from both CHSRI and
the district (Boxes 4 and 5). Federal and state curriculum standards and accountability
structures, as mediated by the district (Box 5), also help structure this setting. Ideally,
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
4. Chicago High School
Redesign Initiative
5. District, state, and federal
Curriculum standards and accountability
structures; Support from district (e.g. Office of
Small Schools, Area Instructional Offices)
Resources, supports, and
limited oversight
7. Curricular and instructional quality
2. Small school
features –
teachers and
principals
3. Small school
context –
teachers and
principals
6. Productive setting for
instructional reform
Aligning external resources,
support, and standards with
school’s coherent vision
Small size,
voluntary, some
autonomy,
limited
bureaucracy
Distributed
leadership,
trust, coherent
vision,
commitment,
professional
community
12. Student
experience
13. Student
outcomes
11. Productive affective
setting for students
Academic and
social personalism,
student
engagement,
academic press,
trust – teachers
and students, focus
on academic
standards
Test scores,
on-track rates,
graduation
rates,
attendance
1. District
Resources,
policies that
enable small
schools
Teacher commitment to student
learning in a personalized
atmosphere of respect and trust
8. Small school
features –
students and
parents
9. Small school context: students
Aligned with student interest,
commitment to school vision
Small size,
student choice
10. Small school context: parents
Trust, involvement/support for school’s vision
Figure 1: Theory of action: Chicago High School Redesign Initiative
12 Improving Schools 8(1)
this combination of internal vision and external forces will lead to improved curricular
and instructional quality (Box 7), which should have a positive impact on student
experience (Box 12), leading to improved student outcomes (Box 13). Those helping
to shape and implement this initiative must deftly balance their need to provide support
and direction with their need to protect the small schools’ autonomy. The primacy of
small school autonomy and flexibility must also be balanced with the need for account-
ability and bureaucratic structures so that the reform can function effectively and
responsibly on a large scale.
The rationale for small schools extends beyond their ability to foster curricular and
instructional quality. Also central to the design are beliefs about how small school
features will be experienced by students and parents. In particular, proponents of small
schools believe that the schools’small size and provision of student choice (Box 8) will
create a highly desirable context for students, one in which students choose a learning
environment based on their own interests, and are committed to the school’s vision
(Box 9). Such a context, the argument goes, will be bolstered by parental trust, support,
and involvement (Box 10) and will help provide an effective setting for reform marked
by mutual trust and respect between students, teachers, and parents (Box 11). This in
turn will further magnify the benefits of the curriculum to create an experience for stu-
dents characterized by features such as academic and social personalism, engagement,
academic press, and trusting relationships between students and teachers (Box 12).
These experiences, in turn, should make desirable student outcomes (Box 13) more
likely.
Methodology
This article draws on both quantitative and qualitative analyses to explore contexts for
students and teachers and their relation to student outcomes. The quantitative analyses
use both survey and record data, allowing us to compare the contexts and outcomes of
the small schools and the contexts and outcomes of traditional high schools serving
similar populations of students. The qualitative analyses rely on interviews and focus
group discussions to highlight key dynamics and to study implementation opportuni-
ties and challenges.
For the quantitative study we analyzed responses from the Consortium on Chicago
School Research’s biannual district-wide survey, including responses from some
29,000 9th and 10th graders and almost 3000 high school teachers (see Appendix A for
a more detailed description of our sample and method). We used Hierarchical Linear
Modeling to see whether students and teachers in these small schools reported higher
levels of measures believed to be present in improving schools than did similar students
and teachers in similar schools. For teachers these measures included indicators of
school leadership, professional community, and parent and community support; for
students the measures included indicators of parent and community support, quality
instructional program, and a student-centered learning climate. For the analysis of
student outcomes, we focused first on 9th graders. Again we used Hierarchical Linear
Modeling to compare attendance rates, drop-out rates,3 and on-track rates (a measure
of freshman student success)4 for students at the small schools with those of similar
students at large traditional high schools. We also compared 11th grade test scores for
the three small schools enrolling 11th graders with the 11th grade test scores of similar
students, since systematic testing of high school students in CPS occurs in the 11th grade.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 13
As part of the qualitative study, we interviewed each school’s principal/director and
met with a focus group of teachers and a focus group of students at each school. We
also interviewed the converting school principal in each building, attended parent
meetings at two of the schools, and met with program staff at two schools. These inter-
views were transcribed and coded, and the results were organized into categories.
Findings
As noted above, our analysis of data informs answers to three broad questions. Our
findings are discussed below.
Question 1: How is Chicago’s effort to implement small schools on
a large scale proceeding? Have CPS and CHSRI been able to
provide the support/resources and flexibility assumed to be
required for successful reform?
As anticipated, Chicago’s efforts to implement small schools on a large scale have
proved challenging. At the same time, these efforts have been aided by varied supports
from the CHSRI staff and the district as well as by teachers’creative visions and enthu-
siasm for the reform. We describe these supports and constraints on implementation
below.5
Problems with implementation took relatively predictable forms, both operational and
philosophical. The volume of tasks associated with creating a small school was enor-
mous – limits on time and resources were particularly challenging. As a principal of a
host school explained, ‘April 6th was the announcement [that we’d be converting our
high school into small schools] and then I had to open three [new small] schools in
August.’
Core infrastructure from the district (Box 1) was often problematic. For example, some
small schools did not get some of their ‘start-up’ monies from the district until schools
had been in operation for three months. Similarly, some small schools lacked science
labs, computers, telephones, office space, and overhead projectors. Working through
these and related challenges diverted time that could have been spent on programmatic
development. As a result, one teacher explained, ‘We were building on the fly. Things
of concern just slipped through the cracks.’
Implementation was further complicated by uncertainties regarding the level of auto-
nomy they would be granted by the district (Boxes 1 and 2). Specifically, those involved
with the small schools often expected more autonomy and flexibility than they feel they
received. As an external partner for one of the small schools put it, ‘The system wants
to standardize. Are the small schools the same as charter schools or not? Are they
autonomous, or not?’ Similarly, a teacher explained that there were often conflicts
between the small school’s curriculum design and the district’s instructional goals
(Boxes 5 and 6):
There was a mixed message that the new small schools should develop an idealized curricu-
lum of what they’d like to have. Then there was the other side saying, ‘No, no, you have to
meet our measurement.’ And so that was very unclear and so that’s really hard for teachers.
Similarly, at several schools, the goal of having teachers lead the reform conflicted
with district regulations and procedures regarding teacher evaluation and other
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
14 Improving Schools 8(1)
‘management’ issues (Boxes 2 and 6). This undermined efforts to create a productive
setting for reform. Teachers in one school, for example, wanted to debrief the results of
a ‘walk-through’ from the Area Instructional Officer in a meeting immediately follow-
ing the event. District policy, however, dictated that the results of such walk-throughs
are to be shared only with the principal, who then shares comments with the teaching
staff. As more than one angry teacher put it, ‘And we’re supposed to be teacher led!’
We didn’t have the teachers we needed to start [September], and we still don’t have the
teachers we need [May]. (Principal)
I don’t think CPS [Chicago Public Schools] knows what we are, what they want us to be.
(Teacher)
At the same time that challenges associated with implementation were substantial,
most teachers and small school leaders also said that they received valuable support
from both the district and the CHSRI staff (Boxes 4 and 5). ‘I couldn’t say enough
about the support we get from the Office of Small Schools . . . and the Initiative, and
the key list of others,’ one principal told us. Several others echoed these sentiments.
Most schools appreciated opportunities for professional development, although this
attitude was not shared by all:
The Initiative is very good at guiding us to professional development as well as providing
for us, which is great because it’s really something new for a lot of us. We have older staff,
and [they are] very set in their ways. So, we need professional development to get all the
teachers on-board with strategies to make us better teachers.
While professional development has begun, however, innovative changes in curriculum
and instruction were limited during the first year of implementation. To the extent that
there have been instructional changes, they have centered on incorporating out-of-
school experiences into the curriculum and promoting interdisciplinary projects (Box 7).
We go out into the city once a week and we tie it in with our curriculum. So, we’ve gone to
the Museum for science, we’ve gone to the Planetarium, we’ve gone on a walking architec-
tural tour . . . We’ve gone to the Museum of Science and Industry to look at the great train
exhibit. [We] tie that in with history.
At a different small school, one principal told of us, ‘Our commitment is to make sure
that every class they take integrates arts infused with the class, as a learning strategy,
as a tool to get them to advance their academic achievement.’
For the most part, principals and teachers agreed that the first year emphasis was on
launching the schools and they hoped to be able to focus more on teaching and learn-
ing in year two. As one teacher told us, ‘We have ideas of what we want the curricu-
lum and instruction to look like next year, and so we’re going to spend some time
[working on] that during the summer.’
This teacher’s perspective was common. Teachers and principals alike noted increased
energy and commitment for reform – but also were clear that the hard work of fostering
improved opportunities for teaching and learning had yet to occur. This is hardly surpris-
ing. As we detailed above, changes to curriculum and instruction are not inevitable out-
comes in small school reform. Rather, small schools can foster contexts and supports
that enable innovation related to teaching and learning. Thus, even though the first year
might be too soon to expect significant changes related to instruction, it would not be
too soon to see whether the context for meaningful change is developing.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 15
Question 2: Are small schools creating contexts for principals,
teachers, and students that can support desired reform and foster
improved outcomes?
Finding (a): teachers
We saw a great deal of evidence that small schools are creating desirable contexts for
teachers and principals to pursue reform (Box 3). For example, teachers at the small
schools told us:
I would say without qualification that every single one of our teachers loves the fact that
they participate in the governance of the school and there is shared decision-making.
(Principal)
One of the things that is happening, is that we have more contact [with each other] in
comparison with the host school. We meet every day almost, we see each other all the time.
(Teacher)
Last year I must have used ten sick days. This year I have used one. (Teacher)
Consistent with these and similar statements by teachers, our survey of teacher per-
spectives found that teachers at CHSRI schools ranked many aspects of their context
more positively than did other similar teachers at other similar Chicago high schools.6
To quantify teacher perspectives, we relied on the Consortium’s teacher measures related
to school leadership, professional capacity, and parent and community partnerships.
The Consortium has analyzed teacher responses to these measures on five survey
administrations over the last decade.7
As noted in Table 1, our survey results reveal sizable and statistically significant differ-
ences between CHSRI small high schools and other high schools in Chicago. Teachers
at CHSRI schools reported more participation in decision making, a greater sense of
collective responsibility for student learning, more opportunities for reflective dialog,
a greater sense of mutual trust and more collaboration among teachers. In addition,
teachers reported higher levels of parent participation at school. Marginally significant
(p < .1) differences related to program coherence and innovation were also noted. In
sum, we saw relatively consistent and strong evidence that the new small schools were
creating the kind of professional communities that might foster desired reform and
improved outcomes. Most importantly, the qualities small schools appear to have
promoted, such as teacher collaboration and influence, align with the rationales for
developing small schools (i.e. small schools were supposed to promote teacher collab-
oration and influence, program coherence, etc.).
No statistically significant differences were found with respect to several other teacher
measures of: school leadership (principal instructional leadership, teacher–principal
trust); professional capacity (access to new ideas, quality professional development,
school commitment); or parent and community partnerships (knowledge of student
culture, teacher outreach to parents, teacher–parent trust). However, while not statisti-
cally significant, in all of these cases, CHSRI schools scored higher than other high
schools in Chicago.
Finding (b): students
Students who attended the CHSRI small high schools rated many aspects of their
experience more positively than students from the rest of Chicago’s high schools (Box
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
16 Improving Schools 8(1)
Table 1: Differences in teacher survey measures between CHSRI schools
and non-CHSRI schools serving similar student populations
Survey measure Measure description
Difference1
Teacher influence# The extent of teachers’ involvement in school 2.75*
decision making
Collective responsibility# Extent of teachers’ shared commitment to 2.18*
improving the school so all students learn
Reflective dialogue# How often teachers discuss instruction and student 1.99*
learning with each other
Teacher–teacher trust# Extent to which teachers express respect for each 1.95*
other and for teachers who are leaders and who are
experts at their craft
Collaboration among The level of cooperation and collaboration among 1.88*
teachers# teachers
Parent support‡ Teachers’ perceptions of the level of parent 1.01*
involvement and support for the school
Program coherence# Extent to which teachers feel programs in their 1.73~
school are coordinated with each other and with
the school’s mission
Innovation# Teachers’ perceptions of whether they are 1.64~
continually learning and trying new things
Notes: 1Differences measured in standard deviation units; #: See Box 3; ‡: See Box 10;
*: significant at p < .05; ~: significant at p < .10.
12). ‘Teachers don’t pay attention to students at a large high school,’ a student told us.
‘Small school students get more attention.’ A different student explained it this way,
‘We have relationships with our teachers. They understand us and they help us with
everything.’
Our student survey responses corroborated these focus group findings. We analyzed
student responses to the Consortium’s student measures, testing whether students
attending CHSRI small high schools responded differently to measures of Student-
Centered Learning Climate, Quality Instructional Program, and Parent and Community
Support. As with the teacher measures, student responses have been analyzed for five
survey administrations over the past 10 years (see note 7).
As detailed in Table 2, students who attended CHSRI small high schools rated many
aspects of their experience more positively than those attending other high schools in
Chicago. Specifically, they reported greater academic personalism, a stronger press
toward academic achievement, more engagement, and higher levels of student–teacher
trust.8 These results are encouraging. Related research would lead us to expect a
relationship between student learning (Box 13) and the aspects of curriculum and
school climate reflected in measures of academic engagement, academic personalism,
academic press, and teacher–student trust (Box 12) (see for example Bryk & Schneider,
2002; Bryk et al., forthcoming; Lee et al., 1999).
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 17
Table 2: Differences in student survey measures between CHSRI schools
and non-CHSRI schools serving similar student populations
Survey measure Measure description
Difference1
Academic personalism# Degree to which students perceive that their 1.59*
teachers give individual attention to and are
concerned about their students.
Press toward academic Gauges whether students feel their teachers 2.18*
achievement# challenge them to reach high levels of academic
performance.
Academic engagement# Students’ reports about their interest and 1.99*
engagement in learning.
Student–teacher trust# Students’ perceptions about the quality of their 1.95*
relationships with teachers.
Notes: 1Differences measured in standard deviation units; #: See Box 12; *: significant at
p < .05.
Moreover, since these qualities of the student experience are all central goals of the
small school reform effort, it is encouraging that small schools appear to support such
goals.
No statistically significant differences were found between students at CHSRI small
high schools and students at other traditional high schools in other measures of parent
and community partnerships (human and social resources in the community, parent
support for student learning) and student-centered learning climate (peer support for
academic work, student classroom behavior, safety).
Some teachers get into it. They make you want to be included. It’s not like they force you.
It’s like the way they do it. They make it interesting. (Student)
I’m glad that I went to [small school] because I learned how to be a leader not a follower and
how to live your life instead of fighting. They really focus us here. (Student)
Question 3: What are early indications of the small schools’ impact
on student outcomes (Box 13)?
At the outset of this discussion, we should say that after only one year of operation, we
are hesitant to place too much emphasis on data related to outcomes. The first year of
a reform is often spent ‘setting-up’: hiring staff; outfitting classrooms; negotiating how
to share a building; working through alternative schedules; and preparing for curricu-
lar change through staff discussions and professional development. Actual changes to
curriculum and instruction therefore may not occur to a large degree during year one
so there may be a lag in changes to student outcomes. We therefore place greater stock
in the analysis of school context presented above than in analysis of student outcomes.
Still, given the importance of outcome data, it makes sense to share it, as long as we
resist the urge to judge the Initiative by its perceived impact on student outcomes at this
point in time.9
Tentative finding (a)
Small schools appear to be promoting improved attendance.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
18 Improving Schools 8(1)
Students at CHSRI small high schools were absent 29.3 per cent fewer days than similar
students at similar high schools. This difference was statistically significant (p < .05).
Tentative finding (b)
Small schools may be promoting higher numbers of students who are on track to
graduate in four years (see note 3).
There are indications that students at CHSRI schools are more likely to be ‘on-track’
than other similar students. This difference is not statistically significant (p < .19),
however.
Tentative finding (c)
Students at the CHSRI small high schools performed no differently on standardized
tests than similar students at traditional high schools.
For this analysis we used only those three schools having 11th graders, which is the
only high school group that takes a standardized test. Student test performance at these
three schools was not different than test performance at schools serving demographi-
cally similar students. See note 9 for a list of control variables.
Assessments regarding the impact of small schools on the kinds of outcomes noted
above will become more meaningful as the Initiative matures.
Discussion and implications
Our review of the data collected on the reform’s first year highlights both challenges
associated with implementation and some meaningful accomplishments. Without a
doubt, there have been significant challenges associated with implementation, but such
challenges should be expected and do not appear to have prevented the creation of a
generally desirable context for reform. While it is too soon to tell whether these new
small schools enhance student performance in Chicago, the early results indicate that
some important pre-conditions are in place. The developing contexts for students and
teachers in these small schools appear to be more conducive to meaningful change in
teaching and learning than the contexts of larger schools that serve similar populations
of students.
For the reasons stated above, we put less stock in early indicators of student outcomes.
One year simply isn’t enough time to expect implementation of enhanced curriculum
and instruction. That said, schools appear to have fostered a sizable improvement in
attendance and they may be improving students’ likelihood of being on-track. We have
not yet seen evidence that attending small schools impacts test scores. We expect
greater clarity on all of these outcomes as the Initiative matures and the sample size
grows.
As the study continues, we should also be better able to assess the reform’s theory of
change. For example, do changes in teacher and principal contexts lead to shifts in
practices and to improved outcomes? While we are not yet in a position to comment on
such relationships, it is already clear that factors associated with implementation deserve
careful attention and that fostering small schools on a large scale can be challenging for
districts. Indeed, there are specific challenges associated with the focus of this specific
effort, over and above the standard implementation challenges associated with any new
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 19
reform. For instance, attention to tensions between small schools’ autonomy and
flexibility on the one hand and the district’s need for accountability and bureaucratic
structures on the other is fundamentally important. The best ways to think about this or
other tensions embedded in the initiative’s theory of change are not yet clear.
What is clear is that both opportunities and challenges present themselves to districts
pursuing small high schools on a large scale. Given widespread efforts to move schools
in this direction, both in Chicago and around the nation, we will hopefully know much
more in the years to come.
Appendix A
Ta ble A1: Sample: small school demographics
Entering 9th School racial/ethnic
Grades Per cent grade mean composition: per cent
Name Number served Special Ed achievement1 Black Latino White
CHSRI high schools
B.E.S.T. 66 9 27.3 6 (225.6) 86.4 13.6
Chicago Discovery 387 9–12 14.7 8 – (242.9) 48.0 51.7
Academy
Phoenix 232 9–12 27.2 6 (227.3) 80.5 19.0
School of the Arts 261 9–11 21.8 6 (226.0) 100
School of
Entrepreneurship 250 9–10 20.0 7 – (235.2) 100
All CPS
non-alternative 97,879 16.1 8 – ( 246.3) 50.4 36.8 9.3
Note: 1Achievement based on 8th grade ITBS scores. First number gives approximate grade
level based on national norms at the end of the year. Number in parentheses gives scale score
average. In general, national norms by end-of-grade level are as follows: 6th grade = 227,
7th grade = 239, 8th grade = 250, 9th grade = 260.
Survey sample10
1. CPS high school teachers: 67 schools; responses from 47 per cent of high school
teachers.
2. CHSRI teachers: four schools; responses from 59 per cent of eligible teachers at all
five schools. Response rates at three of these schools exceeded 87 per cent. In the
fourth school, three of 17 teachers responded, for a response rate of 73 per cent
among the participating four schools.
3. CPS 9th and 10th grade students: 67 schools, responses from 50 per cent of sur-
vey-eligible students.
4. CHSRI students: four schools; responses from 55 per cent of eligible students at all
five schools. Response rates in three schools ranged from 69 per cent to 91 per
cent. In the fourth school, 39 per cent of the students completed the survey, for a
response rate of 65 per cent among the participating schools.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
20 Improving Schools 8(1)
Record data
Data from 23,666 freshmen for whom we have CPS administrative records. Data from
10,551 juniors for whom we have both prior 8th grade test scores and standardized test
scores from the spring of 2003.
Methodology
Analyzing the contexts for students and teachers
We used teacher and student responses to the May 2003 Consortium on Chicago
School Research Survey to estimate whether the contexts for teachers and students
were different at CHSRI schools than they were for similar teachers and students at
similar non-CHSRI schools. Ninth and 10th grade students and teachers of all grade
levels take this survey.
We used the Consortium’s teacher and student survey measures as our outcome
variables. We focused on those measures with theoretical or practical implications.
We used 3-level Hierarchical Linear Modeling to conduct these analyses. This
approach allows us to adjust for measurement error, to control for individual character-
istics of students and teachers, and to control for aggregate characteristics of the schools
themselves. The set of controls we used depended on both theory and initial analysis.
Analyzing student outcomes
We used CPS administrative data for freshman students in academic year 2002/2003 to
estimate whether student outcomes were different for students at CHSRI schools than
they were for similar students at similar non-CHSRI high schools. We chose to focus
on freshmen for three reasons. First, 9th grade is a pivotal year for students, and their
experiences during that year are increasingly important for their ultimate success in
school. Second, students in the other grades spent part of their academic career at a
non-small school, potentially confounding the results. Finally, 9th grade is the only
grade that all of the schools have in common.
We used the number of days a student was absent and whether a student was on-track
by the fall of 2003. To be on-track a student must have earned enough credits to become
a sophomore and must not have more than one failing grade in a core course (English,
math, science, social science).
Again, we used Hierarchical Linear Modeling to estimate whether the average number
of days absent at a CHSRI school differed from the average number of days absent for
similar students at a non-CHSRI school and whether the odds of a student’s being
on-track was different for students in CHSRI schools than for similar students in
similar non-CHSRI schools.
We used Hierarchical Linear Modeling to estimate whether the average 11th grade
student scores on the Prairie State Achievement Exam differed from the average 11th
grade student scores at non-CHSRI schools serving similar students. Because students
do not take standardized tests in grades 9 or 10, we used 8th grade test scores on the
Iowa Test of Basic Skills as our measure of prior academic achievement.
The authors can be contacted by email at: jkahne@mills.edu, s-sporte@consortium-chicago.org and
jqeaston@uchicago.edu
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
Kahne et al.: Creating small schools in Chicago 21
Notes
1 This focus is not new to Chicago. In 1995, the Chicago Board of Education endorsed a resolution
(95–0829-RS2) which affirmed the value of developing small schools, and six small charter schools as well
as one district small school were in operation prior to CHSRI’s launch.
2 He also commented ‘neither is there proof of the academic value of large schools’ (cited in Harvey and
Housman, 2004: 13).
3 This analysis uncovered data collection flaws that make it inappropriate to analyze drop-out rates at small
schools this year.
4 Students who are on-track at the end of their freshman year have accumulated enough credits to advance
to 10th grade and no more than one failing grade in a core course (English, math, science, social studies).
Students who are on-track after their freshman year are five times more likely to graduate within four years
than those who are not (Miller et al., 2002: 21).
5 For more details on these implementation challenges, see ‘A Snapshot of the First Year of Implementation’
(Sporte et al., 2003).
6 When making these comparisons, we controlled for individual and school-level characteristics. At the
school level we controlled for mean student academic achievement, mean student socioeconomic status, the
racial/ethnic composition of the school’s student body, and whether or not the school was on probation. At
the teacher level we controlled for gender, race/ethnicity, whether they were new to the school, their years of
experience, their education level, and the subject matter they taught.
7 For a thorough description of the measures, including a list of the items on each and reliability statistics
for 2003, see Key Measures for School Development at www.consortium-chicago.org
8 In carrying out this analysis, we controlled for school characteristics including mean academic achieve-
ment and socioeconomic status of the student body as well as the school’s racial/ethnic composition. We also
controlled for students’ gender, race/ethnicity, socioeconomic status, special education status, whether they
had ever been held back, and their prior academic test scores.
9 As noted in our methodology section, these findings are based on analysis of record data for students in
the 9th grade. Our use of hierarchical linear modeling enabled us to control for students’ gender, race/ethnic-
ity, socioeconomic status, special education status, whether they had been held back, and their prior academ-
ic test scores. We also controlled for the school’s average achievement, the school’s mean socioeconomic
status, and the school’s racial/ethnic composition.
10 A complete discussion of the 2001 survey, including sampling and response rates can be found in the
survey manual, which can be accessed at www.consortium-chicago.org/surveys/s0002.html. The 2003
manual will be posted in the fall of 2004; most details are similar to those found in 2001.
References
Allensworth, E. M. (2004) Ending Social Promotion: Dropout Rates in Chicago after Implementation of the
Eighth Grade Promotion Gate. Chicago: Consortium on Chicago School Research
Allensworth, E. M. & Easton, J. Q. (2001) Calculating a Cohort Dropout Rate for the Chicago Public
Schools: A Technical Research Report. Chicago, IL: Consortium on Chicago School Research.
American Diploma Project (2004) Ready or not: creating a high school diploma that counts. Online:
http://www.achieve.org/dstore.nsf/lookup/ADPsummary/$file/ADPsummary.pdf [accessed 5 July 2004].
American Youth Policy Forum (2002) High schools of the millennium: report of the work group. Online:
http:// www.aypf.org/publicationsHSchools_round_3.pd [accessed 5 July 2004].
Barker, R. & Gump, P. (1964) Big School, Small School: High School Size and Student Behavior. Palo Alto,
CA: Stanford University Press.
Bryk, A. S. & Schneider, B. (2002) Trust in Schools: A Core Resource for Improvement. New York: Russell
Sage Foundation.
Bryk, A. S., Lee, V. E. & Holland, P. B. (1993) Catholic School and the Common Good. Cambridge, MA:
Harvard University Press.
Bryk, A. S., Sebring, P. B., Allensworth, E. M., Luppescu, S. & Easton, J.Q. (forthcoming) Organizing
Schools for Improvement.
Conant, J. B. (1959) The American High School Today: A First Report to Interested Citizens. New York:
McGraw-Hill.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
22 Improving Schools 8(1)
Cotton, K. (1996) School size, school climate, and student performance. Northwest Regional Educational
Laboratory. Online: http://www.nwrel.org/scpd/sirs/10/c020.html [accessed 29 July 2004].
Daniels, H., Bizar, M. & Zemelman, S. (2001) Rethinking High School: Best Practice in Teaching, Learning,
and Leadership. Portsmouth, NH: Heinemann.
Darling-Hammond, L., Ancess, J. & Wichterle Ort, S. (2002) Reinventing high school: outcomes of the coali-
tion campus schools project. American Educational Research Journal, 39(3), 639–73.
Haller, E. J. (1993) Small schools and higher-order thinking skills. Journal of Research in Rural Education,
9(2), 66–73.
Harvey, J. & Housman, N. (2004) Crisis or Possibility: Conversations about the American High School.
Washington, DC: Institute for Educational Leadership.
Hendrie, C. (2004) High schools nationwide paring down. Education Week, 23(40), 128–30.
Hess, G. A. & Cytrynbaum, S. (2002) The effort to redesign Chicago high schools: effects on schools and
achievement. In V. Lee (ed.) Reforming Chicago’s High Schools: Research Perspectives on School and
System Level Change, pp. 190–490. Chicago, IL: Consortium on Chicago School Research.
Holland, N. E. (2002) Small schools transforming teacher and student experiences in urban high schools. In
V. Lee (ed.) Reforming Chicago’s High Schools: Research Perspectives on School and System Level Change,
pp. 89–124. Chicago, IL: Consortium on Chicago School Research.
Howley, C. B. (1989) Synthesis of the effects of school and district size: what research says about achieve-
ment in small schools and school districts. Journal of Rural and Small Schools, 4(1), 2–12.
Husbands, J. & Beese, S. (2001) Review of selected high school reform strategies. Paper prepared for the
Aspen Program on Education. Workshop on High School Transformation. Aspen, CO, USA, 17–22 July.
Lee, V. (2002) Setting Chicago high school reform within the national context. In V. Lee (ed.) Reforming
Chicago’s High Schools: Research Perspectives on School and System Level Change, pp. 7–17. Chicago, IL:
Consortium on Chicago School Research.
Lee, V. E. & Smith, J. (1995) Effects of high school restructuring and size gains in achievement and engage-
ment for early secondary school students. Sociology of Education, 68(4), 241–70.
Lee, V. E., Smith, J. B, Perry, T. E. & Smylie, M. A. (1999) Social Support, Academic Press, and Student
Achievement: A View from the Middle Grades in Chicago. Chicago, IL: Consortium on Chicago School
Research.
Meier, D. (1995) The Power of Their Ideas: Lessons for America from a Small School in Harlem. Boston,
MA: Beacon Press.
Miller, S. R., Allensworth, E. M. & Kochanek, J. R. (2002) Student Performance: Course Taking, Test
Scores, and Outcomes. Chicago, IL: Consortium on Chicago School Research.
Newmann, F. & Wehlage, G. (1995) Successful School Restructuring. Madson, WI: Center on Organization
and Restructuring of Schools.
Office of Small Schools, CPS (2004) Online: http: www.smallschools.cps.k12.il.us/grants.html [accessed 5
July 2004].
Pittman, R. & Haughwout, P. (1987) Influence of high school size on dropout rate. Educational Evaluation
and Policy Analysis, 9(4), 337–43.
Powell, A., Farrar, E. & Cohen, D. (1985) The Shopping Mall High School: Winners and Losers in the
Educational Marketplace. Boston, MA: Houghton Mifflin.
Sporte, S., Correa, M., Kahne, J. & Easton, J. Q. (2003). Chicago High School Redesign Initiative: A
Snapshot of the First Year of Implementation. Chicago, IL: Consortium on Chicago School Research.
Wasley, P., Fine, M., Gladden, M., Holland, N., King, S., Mosak, E. & Powell, L. (2000) Small Schools:
Great Strides. New York: Bank Street College of Education.
Weiss, C. (1995) Nothing as practical as good theory: exploring theory-based evaluation for comprehensive
community initiatives for children and families. In New Approaches to Evaluating Community Initiatives.
Washington, DC: The Aspen Institute.
b y amr il mu ha mm ad on O c to b e r 29 , 2 0 08
h t tp :/ /imp .s a ge p u b .c om
D ow n lo ad e d fr o m
