 

 

. g1. ‘
. 2.3!. 6 : .5.
Hwﬁiv. 3.56... ~55...

: .
.2 a! .
35:355.!

5.25..” a.
o is...»

3...: 3....
I I. J... Nu
. ‘5

2.31:: 1- 7.3l291ltitx:
) 3'21! 1 15.9.9... bs\|v I
t. _ xntuvxcvut

{II-«Ii 1 :
ﬂy!)-
.
t \E
II

53v
viiiijtalﬁlgil? I‘ll

 

 

 

 

;a!v-.A$ev 11!. thhthﬁQIﬁl... an. 61‘
A ohttsylﬂl'\ 31.0: [-
iiivbvtlhﬂ

If. 11.0091}
9. II

3
tar... giant};
.. .‘.......h......uz.-.
1:. 1.14.2...
5.1L“.
2. I: .5
21,

 

 

LIBRARY
Micl'uym. State

University

 

 

 

This is to certify that the
dissertation entitled

HOW SCHOOL ADMINISTRATORS AND TEACHERS
RESPOND TO SCHOOL CHOICE POLICIES

presented by

Wang Jun Kim

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Educational Policy

 

OWL 0- W

Major Professor’s Signature
[)8 Lem 53V Q m ‘ ZOOq

 

Date

MS U is an Affirmative Action/Equal Opportunity Employer

FA- -.-‘-.-n—«- -i—-.---

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE ' DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lProjIAa:&Pres/ClRC/DateDue.indd

 

 

 

   

._- (z-Ihrlc?a.’1NIn.P

,, . ... _.. $53. 49:32

 
    

., .xfn . . 21...... , ..‘..Jf

 

.

 

HOW SCHOOL ADMIN ISTRATORS AND TEACHERS RESPOND
TO SCHOOL CHOICE POLICIES

By

Wang Jun Kim

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirement
for the degree of

DOCTOR OF PHILOSOPHY
Educational Policy

2009

ABSTRACT

HOW SCHOOL ADMIN ISTRATORS AND TEACHERS RESPOND TO
SCHOOL CHOICE POLICIES

By

Wang Jun Kim

Based on surveys of and interviews with school teachers and administrators in
Michigan, this dissertation study examined how school administrators and teachers in
traditional public schools respond to school choice policies. Two competing economics
perspectives provided useful insights into examining the issue: the neoclassical
economics (N CE) perspective and the institutional and behavioral economics (IBE)
perspective (Hirschman, 1970; Schmid, 2004). Regarding educators’ responses to school
choice policies, the NCE and IBE perspectives have different expectations. School choice
policies or market approaches to education build on theory from neoclassical economics.
According to the NCE perspective, school administrators and teachers will be responsive
to the needs or demands of parents and will create effective schools in order to attract
students and survive in the market. Questioning the NCE perspective, the IBE perspective
suggests the possible effects of school choice policies on educators as follows. First, '
educators might fail to be aware of student enrollment changes. Second, even though they
perceive student enrollment changes, they might have no cencerns about them. Third, it is
possible that they might not know what to do in order to increase student enrollment or
they might employ inadequate strategies. Finally, they may know what to do in order to

increase student enrollment but they might be engaged in socially inappropriate market

behaviors such as cream skimming. The paths educators take will depend on their beliefs
or knowledge, capacity, and learning opportunities.

Many ﬁndings from this study did not seem to support the NCE perspective. First,
educators did not know a lot about student enrollment changes and the percentage of
choice students within their schools. To explain this, this study examined the association
between incentive systems for educators and their awareness. However, the association
was found to be statistically insigniﬁcant. In addition, incentive systems for educators
were not aligned with changes in student enrollment. Second, schools and districts did
not make much effort to attract more students, and no signiﬁcant associations were
observed between activities for attracting more students and the intensiveness of
competition among schools. Third, educators tended to view the problems related to
students or parents as being more serious than the problems related to educators
themselves. In addition, school activities for attracting students did not seem to be
commensurate with the seriousness of school problems perceived by educators. Finally,
educators tended to evaluate the effects of school choice programs on their practices and
schools negatively. The IBE perspective would be of help in explaining these ﬁndings.

This study challenged the basic assumptions made by proponents of school choice
policies by addressing educators’ responses to such policies. In addition, by introducing
the IBE perspective to research on such policies, this study encourages scholars of school
choice policies to pay attention to educators’ limited information, their capacity for

information processing, and their beliefs.

Copyright by
Wang Jun Kim
2009

ACKNOWLEDGEMENTS

I would like to express my deepest gratitude to my dissertation committee chair
and advisor, Dr. Peter Youngs, for his mentoring both academically and personally
throughout my doctoral studies at Michigan State University. This study would not have
been possible without his help and guidance. Dr. Youngs devoted numerous hours to
reading every single page of this work and providing detailed and insightful comments.
His encouragement inspired conﬁdence in me and made me overcome difﬁcult times.

My dissertation committee, Dr. Susan Printy, Dr. Melinda Mangin, and Dr.
Kimberly Maier, provided invaluable advice and comments. I would like to thank them
for their support, encouragement, and trust in me. It has been a great honor to have them
serve on my dissertation committee.

In addition, I would like to thank my mentors: Dr. Chong Jae Lee, Dr. Chung-i1
Yun, and Dr. Dong-Seop J in at Seoul National University. They introduced me to
educational administration and policy studies and encouraged me to study further. I also
would like to extend my thanks to faculty members at the Department of Education at
Seoul National University who provided me with foundational knowledge of education.

This dissertation would not have been possible without the help of many
interview participants and survey respondents. I am grateful for their generosity in giving
me their time and thoughts about school choice policies.

I would like to extend many thanks to my colleagues, Todd Drummond, Paul
Tanner, Min Sun, Timothy Ford, Richard T. Holdgreve-Resendez, Chong Min Kim,

Jonghwan Lee, Dr. Soo-yong Byun, Hosun Kang, and Tae Seob Shin for their support

and valuable comments on my work. They were my “resource persons” 1 could readily
reach out to when I had questions. I also thank Kari Selleck for her arrangement of
interviews with educators and her support for this study. I also thank Hiep Thien Chau at
the Writing Center. He read all the chapters of this dissertation and helped me clarify my
arguments. Special thanks go to Seung-Hwan Ham who not only read and commented on
multiple versions of all the chapters of this dissertation, but also provided many insightful
suggestions to help me improve my writing and clarify my arguments. Sometimes I
thought that he knew what I was trying to say better than I did. I learned much from him.

I would like to thank my parents (-in-law) and brothers and sisters (-in-law). They
might not know about what I have been studying, but they trusted me and supported me
in many ways.

Finally, I would like to thank my wife, Eunjung, for her unconditional support and
trust in me. She had to leave her parents and friends in Korea and endure lonely and poor
living in a faraway foreign country during my doctoral studies at Michigan State

University.

vi

TABLE OF CONTENTS

 

 

 

 

LIST OF TABLES ix
LIST OF FIGURES xiii
PART I
CHAPTER 1: TRENDS IN SCHOOL CHOICE POLICY STUDIES 1
Introduction
Literature Review 5

 

CHAPTER 2: THEORETICAL FRAMEWORK AND RESEARCH QUESTIONS --- 15

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Theoretical Framework 15
Research Questions and Hypotheses 24
CHAPTER 3: METHODOLOGY 27
Research Sites and Policy Context 27
Interviews with School Administrators and Teachers for a Pilot Study ---------- 28
Sampling and Response Rates 31
Key Constructs of Surveys and Interviews 36
Analysis Scheme 38
PART 11
CHAPTER 4: AWARENESS OF ENROLLMENT CHANGES 43
Educators’ Awareness of Enrollment Changes and the Percentage of Choice
Students 45
Educators’ Job Security, Salaries, and Benefits 50
Discussion 60
CHAPTER 5: TEACHERS’ WORKING CONDITIONS 63
Educators’ Perceptions of Working Conditions ‘ 65
Differences in Subgroups 67
Associations between Competition among Schools and Teachers’ Working
Conditions 70

 

 

Discussion 74

vii

 

CHAPTER 6: EFFORTS FOR ATTRACTING STUDENTS 77
Activities for Attracting Students 79
School Problems 91
Educators’ Perceptions of the Reasons for Parents’ School Choice Decisions -- 99
Associations between School Problems and Activities for Attracting Students

 

 

 

 

106
Associations between Choice Reasons and Activities for Attracting Students

109
Discussion 1 12

 

CHAPTER 7: IMPACT OF SCHOOL CHOICE POLICES ON EDUCATORS
AND SCHOOLS 117
Educators’ Perceptions of Impact of School Choice Policies on Their Practices

 

 

 

 

 

 

 

 

 

 

 

 

 

 

119
Educators’ Perceptions of the Impact of School Choice Policies on Their Schools
128
Discussion 143
PART 111
CHAPTER 8: IMPLICATIONS FOR POLICY AND FUTURE RESEARCH --------- 146
Background of This Research, Theoretical Framework, and Research Questions
146
Summary of the Findings and Discussion 148
Implications for Policy and Practice 156
Limitations of the Study and Implications for Future Research 158
APPENDICES 163
A: ADMINISTRATOR SURVEY INSTRUMENT 164
B: TEACHER SURVEY INSTRUMENT 176
C: FACTOR ANALYSIS AND RELIABILITY CHECK FOR
TEACHERS’ WORKING CONDITIONS 190
D: INTRA-CLASS CORRELATION COEFFICIENTS (ICCS) FOR
TEACHERS’ WORKING CONDITIONS ‘ 192
E: MISSING DATA ANALYSIS FOR TEACHER SURVEY 193
F: IMPACT OF ENROLLMENT CHANGES ON TEACHERS’
JOB SECURITY, SALARIES, AND BENEFITS BY SCHOOL ------ 196
G: EDUCATORS’ PERCEPTIONS OF CHOICE STUDENTS 199

 

 

REFERENCES 200

viii

LIST OF TABLES

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3-1: School District Characteristics 32
Table 3-2: Survey Sample and Response Rates 33
Table 3-3: Teacher Characteristics 34
Table 3-4: Administrator Characteristics 35
Table 3-5: Key Constructs of Teacher and Administrator Surveys 37
Table 3-6: School-Level Variable Descriptive Statistics 40
Table 4-1: Educators’ Awareness of Student Enrollment Change 46
Table 4-2: Awareness of the Percentage of Choice Students 46
Table 4-3: Differences between Subgroups in Awareness of Student Enrollment Changjls8
Table 4-4: Variance Components in the Differences between Subgroups in Awareness of
Student Enrollment Changes 49
Table 4-5: Differences between Subgroups in Awareness of the Percentage of Choice
Students 49
Table 4-6: Variance Components in the Differences between Subgroups in Awareness of
the Percentage of Choice Students 50
Table 4-7: Impact of Student Enrollment Changes on Teachers (%) 51
Table 4-8 Impact of Student Enrollment Changes on Administrators (%) --------------- 51
Table 4-9 Differences between Subgroups in Educators’ Perceptions of Job Security - 53
Table 4-10: Differences between Subgroups in Educators’ Perceptions of Salary ------- 54
Table 4-11: Differences between Subgroups in Educators’ Perceptions of Beneﬁts ---- 54

Table 4-12: Associations between Educators’ Job Security, Salary, and Benefits and

 

their Awareness of Enrollment Changes 58

Table 4-13: Associations between Educators’ Job Security, Salary, and Benefits and

 

their Awareness of the Percentage of Choice Students 59

ix

 

Table 5-1. Teachers' Perceptions of Working Conditions 66

 

 

 

 

 

 

Table 5-2 Administrators’ Perceptions of Working Conditions 66
Table 5-3. Differences between Teachers’ and Administrators’ Perceptions of Teacher
Working Conditions 67
Table 5-4: Differences between Groups with regard to Educators’ Perceptions of
Working Conditions 69
Table 5-5: Associations between School Competition and Teachers’ Working Conditions
73
Table 6-1: Teachers’ Perceptions of Schools’ Activities for Increasing Student
Enrollment 80
Table 6-2: Administrators’ Perceptions of Schools’ Activities for Increasing Student
Enrollment 81
Table 6-3: Differences in Perceptions of Schools’ Activities for Increasing Student
Enrollment Between Teachers and Administrators 81

 

Table 6-4: Differences in Perceptions of Schools’ Activities for Increasing Student
Enrollment Between Teacher and Administrators in Gaining and Losing
School Districts 82

 

Table 6-5: Differences between Subgroups in the Perceptions of Schools’ Activities for

 

 

 

 

 

Attracting Students 85
Table 6-6: Associations between School Competition and Schools’ Activities for

Attracting Students 89
Table 6-7: Associations between School Competition and School Activities for Attracting

Students after Controlling for School Characteristics 90
Table 6-8: Teachers’ Perceptions of School Problems 92
Table 6-9: Administrators’ Perceptions of School Problems 93

Table 6-10: Differences in Perceptions of School Problems between Teachers and
Administrators 94

 

Table 6-11: Differences in Perceptions of School Problems between Teachers and
Administrators in Gaining and Losing School Districts 95

 

Table 6-12: Difference between Subgroups regarding Educators’ Perceptions of School

 

Problems 98
Table 6-13: Teachers’ Perspectives on Parents’ School Choice Reasons ---------------- 100
Table 6-14: Administrators’ Perspectives on Parents’ School Choice Reasons --------- 101

Table 6-15: Differences between Teachers’ and Administrators’ Perspectives on Parents’

School Choice Reasons 101

 

Table 6-16: Differences between Teachers’ and Administrators’ Perspectives on Parents’

School Choice Reasons in Gaining and Losing School Districts ------- 102

Table 6-17: Differences between Subgroups in Educators’ Perceptions of School Choice

 

Reasons 105

Table 6-18: Associations between School Problems and Activities for Attracting Students

 

108

Table 6-19: Associations between Choice Reasons and Activities for Attracting Students

Table 7-1:

Table 7-2:

Table 7-3:

Table 7-4:

Table 7-5:

Table 7-6:

Table 7-7:

Table 7-8:

Table 7-9:

111

 

Impact of School Choice Programs on Teachers (%) 120

 

Impact of School Choice Programs on Teachers in Gaining and Losing School
Districts 120

 

 

Impact of School Choice Programs on Administrators 121

Impact of School Choice Programs on Administrators in Gaining and Losing
School Districts 12 1

 

Differences between Subgroups in Educators’ Perceptions of the Impact of
School Choice Programs on their Practices 124

 

Association between School Competition and Educators’ Perceptions of
the Impact of School Choice Programs on their Practices (Model 1) --- 127

Association between School Competition and Educators’ Perceptions of
the Impact of School Choice Programs on their Practices (Model 2) --- 127

Teachers’ Perceptions of the Impact of School Choice Programs on Schools
129

 

Administrators’ Perceptions of the Impact of School Choice Programs on
Schools 130

 

xi

Table 7-10: Differences in the Perceptions of the Impact of School Choice Programs on
Schools between Teachers and Administrators 131

 

Table 7-11: Differences in the Perceptions of the Impact of School Choice Programs on
Schools between Teachers and Administrators in Gaining and Losing
School Districts 1 3 1

 

Table 7-12: Differences between Subgroups in Educators’ Perceptions of the Impact of
School Choice Programs on their Schools and Education in General - 134

Table 7-13: Associations between School Competition and Perceptions of the Impact of
School Choice Programs on Schools and Education (Model 1) --------- 138

Table 7-14: Associations between School Competition and Perceptions of the Impact of
School Choice Programs on Schools and Education (Model 2) --------- 139

Table 7-15: Associations between Educators’ Perceptions of School Choice Reasons and
their Perceptions of the Impact of School Choice Programs on Schools
142

 

xii

LIST OF FIGURES

Figure 2-1: Neoclassical economics perspective vs. Institutional economics perspective
- 23

 

xiii

CHAPTER 1

TRENDS IN SCHOOL CHOICE POLICY STUDIES

Introduction

Among the popular education reform initiatives in recent years are school choice
policies that emphasize parental choice and competition among schools. School choice
policies or market approaches to education build on theory from neoclassical economics.
The neoclassical economics perspective assumes that humans as economic actors seek
their maximum interest, make efforts to obtain perfect information and make rational
choices to satisfy their interest. They argue that since a free market system based on
choice and competition will guarantee effectiveness and efﬁciency of goods or service
production, government should not regulate or control the free choice of economic actors
(Friedman, 1962). According to the neoclassical economics perspective, parents’ choice
of schools for their children is an economic action and they will choose the most
appropriate schools for their children with perfect information about available schools in
order to maximize their interests. Further, school administrators and teachers will be
responsive to the needs or demands of parents and will create effective schools in order to
attract students and survive in the market.

There are many research studies on school choice programs, but they do not
consistently support arguments from neoclassical economics. Research studies on the
impact of competition among schools on academic achievement and school culture have
produced mixed results (Hoxby, 2004; Shanker & Rosenberg, 1992; Witte, 1996). For

instance, Hoxby (2004) and Holmes, DeSimone, and Rupp (2003) argue that school

competition due to charter schools contributes to improving charter schools or traditional
public schools, whereas Shanker and Rosenberg (1992) maintain that private schools
under competition do not outperform public schools when parental socioeconomic
backgrounds are taken into account. In terms of school climate, Chubb and Moe (1988)
argue that climates are better in private schools than in public schools, while Benveniste,
Camoy, and Rothstein (2003) argue that public schools and private schools do not differ
in their organizational cultures when family background and school community are taken
into account.

In terms of parental preferences for schools, there are many preferences that are
not related to academic achievement or that exacerbate social inequalities in schools
(Abernathy, 2008; Fuller, Elmore, & Orﬁeld, 1996; Holme, 2002; Martinez, Godwin, &
Kemerer, 1996; Wells, 1996). For example, some middle class parents choose schools for
their children based on status ideologies that emphasize race and class (Holme, 2002;
Schneider & Buckley, 2002). Lee, Croninger, and Smith (1996) showed that Detroit-area
residents rated two nonacademic qualities of schools as most important to them: school
safety and whether schools supported their moral and ethical values. Martinez, Godwin,
and Kemerer (1996) argue that the utilization of a public choice program in San Antonio
was affected by family backgrounds such as parent education, family income, the number
of family members, and children’s future education.

The effects of school choice polices depend on how schools (i.e., the supply side)
and parents (i.e., the demand side) respond to choice and competition, but there are only a
few research studies on how school administrators and EMOs (educational management

organizations) respond to school choice policies (Abernathy, 2008; Arsen, Plank, &

Sykes, 2002; Miron & Nelson, 2002). Based on surveys of and interviews with school
teachers and administrators in Michigan, this dissertation study will examine how school
administrators and teachers in traditional public schools respond to school choice policies.
In particular, this study will address the following research questions:
1. How much are teachers and school administrators aware of changes in
enrollment? Do schools have incentive systems aligned with decreases or
increases in student enrollment?
2. How are schools/districts that lose students as part of school choice policies
(i.e., losing schools/districts) different from schools/districts that gain students
(i.e., gaining schools/districts) in terms of teachers’ working conditions such as
available time and resources for teachers, leadership, and opportunities for
professional development? Are teachers’ working conditions associated with the
intensiveness of school competition?
3. What are schools doing to attract students? Is the degree of organizing activities
for attracting more students associated with school competition? Are the activities
associated with school problems or parents’ school choice reasons?
4. How do teachers and administrators perceive the impact of school choice
policies on teachers and administrators themselves and their schools? Are their
perceptions of the impact associated with the degree of competition among
schools?
Two competing economics theories on markets guide this study: the neoclassical
economics perspective and the institutional and behavioral economics perspective. They

provide useful frameworks for examining how parents, school administrators and

teachers respond to school choice. These theories have different assumptions with regard
to human actions, the roles of a market and government (or institutions), and the
performance of a market. From the neoclassical economics perspective, whether and how
school administrators and teachers respond to school choice programs depend on whether
and how changes in enrollment due to school choice policies affect their interests,
including their job security or beneﬁts. To examine whether school choice policies bring
competition among schools and motivate school administrators and teachers to improve
their practice, we need to examine the nature of their job security and beneﬁts and how
these are affected by changes in student enrollment due to school choice.

In contrast to the neoclassical economics theory, the institutional and behavioral
economics (IBE) perspective questions basic assumptions of neoclassical economics and
posits that the behaviors of actors in a market are determined not only by the market
system but also by social institutions (i.e., social arrangements) and the capacity of actors
to behave in certain ways (Schmid, 2004). Institutions are characterized by laws, norms,
ideologies, and beliefs. The IBE perspective builds on institutional approaches in
sociology, behavioral science, and cognitive science. This perspective can help
policymakers attend to the importance of the norms, ideologies, and beliefs that school
administrators and teachers have, and how these are affected by broad institutional
arrangements.

Examining how competition among schools is affected by the nature of schools’
incentive systems, school administrators’ and teachers’ perceptions and beliefs, and their

capacity to cope with competition, this study examines whether it is important for policy

makers to consider educators’ beliefs and capacity when they design school choice
policies.
Literature Review

Previous studies on school choice focused on 1) the impact of school choice
programs on student academic achievement (Hoxby, 2004; Lubienski, Crane, &
Lubienski, 2008; Shanker & Rosenberg, 1992; Witte, 1996) and 2) factors that inﬂuence
parental choice (Fuller etal., 1996; Holme, 2002; Lee et al., 1996; Schneider & Buckley,
2002) and socioeconomic segregation due to school choice programs (Abernathy, 2008;
Martinez et al., 1996; Wells, 1996). However, there are only a few research studies on
how school administrators and EMOs (educational management organizations) respond
to school choice policies (Abernathy, 2008; Arsen et al., 2002; Miron & Nelson, 2002).
Impact of School Competition on Academic Achievement

Research ﬁndings about the effect of school choice on academic achievement are
mixed. Hobxy (2004) compares the reading and mathematics proﬁciency rate of charter
school students to that of their fellow students in neighboring public schools. She
assumes that greater levels of proﬁciency among students in charter schools result from
the better performance of charter schools (compared to traditional public schools) which
are operated in education markets. She found across the nation that the proportion of
students rated as proﬁcient on state exams in charter schools was larger than that in the
matched schools by 5.2% in reading and 3.2% in math. For example, charter schools in
Alaska and Louisiana had more proﬁcient students than matched public schools by about
20% and 30%, respectively. However, charter schools in North California and Texas had

fewer proﬁcient students than public schools by 4.3% and 6.8%, respectively.

Holmes, DeSimone, and Rupp (2003) also assume that the expansion of the
charter school system has encouraged traditional schools to increase achievement. They
investigate how the introduction of school choice in North Carolina, via a dramatic
increase in the number of charter schools across the state, affected the performance of
traditional public schools on statewide tests. By analyzing the relationship between 1) test
scores and 2) the distance between public schools and charter schools, they found that the
introduction of charter school competition caused an approximate one percent increase in
achievement scores, which constituted about one quarter of the average yearly growth in
North Carolina.

Many research studies which address the impact of school choice policies on
student achievement share the common assumption that school choice policies will create
competition among schools and increase student achievement. Therefore, when charter
schools and traditional public schools adjacent to them get higher scores or more
proﬁcient students than other schools, they interpret this to mean that competition among

schools improved student achievement. However, alternative explanations such as a

selection effect or a cream skrmmmg effect can invalidate the competition effect. For
example, in the case of charter schools in North Carolina, it is also plausible that public
schools came to have relatively more proﬁcient students without improving student

achievement because lower achievers moved from public schools to charter schools. The

 

The concept of cream skimming was drawn from the ﬁeld of business. The "cream" represents high-
value or low-cost customers, who are more proﬁtable to serve. When cream skimming occurs (i.e., when
these customers are the focus of a given business), more expensive or harder-to-service customers are left
without the desired product or service at all or they are "dumped" on some default provider
(http://en.wikipedia.org[wiki/Cream skimming). In the context of charter schools, cream skimming refers
to charter schools’ or publics schools’ behavior when they try to recruit high-achieving or low-cost students
without improving student achievement or reducing the cost for education (Lacireno-Paquet, Holyoke,
Moser, & Henig, 2002).

introduction of a charter school system might redistribute students between traditional
public schools and charter schools rather than increase public schools’ achievement to
survive competition.

Parents as Consumers

Research studies on parents’ responses to school choice policies address the
factors parents consider when they choose schools for their children, whether there are
any differences in utilizing choice options between socioeconomic classes, and the
impact of parental choice on equity in education or social desegregation.

What parents really prefer when they choose schools is not clear. Most surveys
show that parents endorse the “right” academic values such as student achievement,
school curriculum, and teacher quality when asked what they consider in choosing
schools (Armor & Peiser, 1998; Greene, Howell, & Peterson, 1998; Kleitz, Weiher,
Tedin, & Matland, 2000; Public Policy Forum, 1998; Schneider, Marschall, Teske, &
Roch, 1998; Vanourek, Manno, & Firm, 1998). However, in-depth interviews with
parents and analyses of information search behavior reveal that parents often choose
schools based on status ideology that emphasizes race and class (Holme, 2002; Schneider
& Buckley, 2002). The parents in Holme’s (2002) study believed that schools serving
higher-status (Whiter and/or wealthier) students were presumed to be good, while schools
serving lower status students (lower income and/or students of color) were presumed to
be unsatisfactory. Status concerns were far more salient to them than a school’s

curriculum and instructional quality. Analyzing the school information search behaviors

of parents in DCSchoolSearch.com2, Schneider and Buckley (2002) found a strong bias
toward accessing the demographic characteristics of the student population, which is in
marked contrast to verbal reports about the importance of race.

The utilization of public choice programs can be affected by family backgrounds
such as parent education, family income, the number of family members, and children’s
achievement level (Martinez et al., 1996). Parents who are better educated, more afﬂuent,
have higher expectations for their children, have fewer children, have children with
higher scores in tests, and/or have female students are more likely to utilize school choice
programs. Parents have limited information on school choice programs and there is
information asymmetry between races (Henig, 1996). Henig’s (1996) study of the magnet
school program in Montgomery County, Maryland, suggests that there were many
parents who had no information on the magnet school program. The study revealed the
differences in the amount of information among ethnic groups. While over 70% of White
parents had heard of the term “magnet schools” or “magnet programs,” only about 40%
of Hispanic parents had heard of one or both of these terms. Families with limited
information rely on anecdotal information or seemingly superﬁcial assessments of school
quality (Lee et al., 1996).

Even though parents appreciate quality differences among schools, some parents
might not utilize their choice options. After studying school choice in St. Louis, Wells

(1993) explained the differences in the interpretation of choice opportunities among

 

DCSchoolSearch.com presents data on all the “traditional” public schools in Washington, DC as well as
the more than 30 public charter schools that now enroll over 10% of the city’s school age children. The site
provides a host of information on each school, including location, test scores, student demographics,
mission statement, and academic programs. There is a core of information (e.g., test scores and student
demographics) available on each and every school.

parents and students from different races and classes. Not all low-income minority
students and parents exercised their choice options. Parents who feared competition or
failure in a higher status school or who did not have trust in the educational system were
most likely to choose not to exercise their choice option.

In sum, the research ﬁndings about parental choice and preference show that
parents tend to report that their choice of schools is motivated by academic quality when
asked but their actual choice or behaviors often favor status ideology or socioeconomic
segregation. Some parents who fear competition or do not have trust in schools might not
exercise their choice options. In addition, many parents do not have enough information
about school choice programs and available schools, and there is information asymmetry
between socioeconomic classes.

Educators’ Responses to School Choice Policies

With respect to administrators’ and teachers’ responses, the research studies were
conducted at institutional levels such as the school or district level rather thanthe
individual level such as the administrator or teacher level. Those studies addressed the
differences between public schools and private schools in terms of effective school
climate, the strategies of charter schools, and what school districts that had lost students
to other school districts or charter schools were actually doing to attract students.

The research ﬁndings about whether private school culture is different from
traditional public school culture are mixed. Chubb and Moe (1988) identiﬁed the factors
of effective schools from a literature review and examined whether a market system of
private schools or a system of public schools would lead schools to become effective.

They hypothesized that private schools which were operated in a market would be in a

better position to become effective schools. As they predicted, analysis of the
Administrator and Teacher Survey (ATS) data of 1984 also showed that private schools
earned better scores on each factor associated with effective schooling than public
schools.

However, Benveniste, Camoy, and Rothstein’s (2003) case study ﬁndings are
inconsistent with Chubb and Moe’s ﬁndings. They conducted case studies of eight public
and eight private schools in California, in which they conducted extensive interviews
with principals, other administrators, teachers, parents, and 8‘h-grade students. The
schools were selected to typify a range of socioeconomic characteristics and included
both sectarian and nonsectarian private schools. They found that there were few
differences between public and private schools in similar communities. Based on their
data collection, Benveniste et al. argue that “the social, cultural, and economic
backgrounds of the parents and the community in which the school was located seemed
to be the main determinant of variation, much more so than a school’s public or private
character” (2003, p. xiv).

To evaluate whether private schools are more effective than public schools, we
need to consider the context in which current private schools are situated. The effective
climate or culture of current private schools can result from scarcity of private schools,
high tuition cost, homogeneity of students and parents, and high socioeconomic
background of parents.

Maximizing proﬁts can result in deterioration in education. Under the market
system, some schools recruit low-cost students and exclude high-cost students (Arsen,

Plank, & Sykes). For instance, most charter schools in Michigan enroll only elementary

10

students. Public school districts participating in inter-district choice face a similar
incentive to open places for students in the lower grades. Charter schools enroll fewer
special education students than neighboring public schools. As a result, they enjoy lower
costs than traditional public school districts. For management companies, administration
costs are two to ﬁve times higher than the expenditure for similar purposes in school
districts. High administration costs appear to cut into instructional spending, not the cost
of facilities. Less spending on instruction reﬂects lower salaries for teachers, failure to
provide special services, and in some cases reluctance to reduce class size, all of which
might threaten the school’s ability to improve student achievement (Miron & Nelson,
2002)

In Arsen et al.’s (1999) study, school districts which have been losing students to
other school districts or charter schools due to school choice policies were implementing
various measures to retain students from their districts and attract students from other
districts, but they appeared to invest in advertisement or provide more of the same
programs, like all-day kindergarten, rather than making efforts to improve teaching
practice or teacher quality.

It is not clear whether competition among schools makes traditional public school
principals who are facing competition with charter schools more attentive to parental
involvement. According to Abernathy (2008), while charter school principals are more
likely to spend a great deal of time building relationships with the parent community than
traditional public school principals, there are almost no differences between public school
principals who are facing competition with charter schools and public school principals

who are not.

11

Increased competition might weaken rigorous programs. While there are few
research studies on how competition affects programs in K-12 schools, there are some
examples in the higher education market. In higher education, for example, as Hofstra
University increased its standards and requirements for its educational administration
program, they faced a student enrollment decrease in the program. Under this situation,
the fastest, easiest, and cheapest programs can ﬂourish rather than more rigorous, high-
quality programs (Young, Peterson, & Short, 2002). Aligning programs to the needs of
consumers does not necessarily mean quality improvement.

In sum, the empirical research ﬁndings reveal that it is not clear whether parental
choice and competition among schools are critical conditions for effective schools.
Rather, parents’ socioeconomic background appears to be a more important determinant
of school culture and climate. When schools are run by private ﬁrms like EMOs which
seek proﬁts, the investment in instruction might decrease but management costs might
increase. This might lead to an increase in education costs or a lower quality teaching
force. In addition, opportunities for special education students might be limited or special
education students might have to pay more tuition than general students in the education
market.

Implications for This Study

Teaching practice and leadership are critical in improving school quality and
student learning (Leithwood, Louis, Anderson, & Wahlstrom, 2004). However, many
research studies on the impact of school choice policies on student achievement have
failed to attend to how school choice policies affect teaching and leadership practices,

assuming that competition among schools would motivate educators to improve their

12

teaching and leadership practices in order to attract more students and enhance their
reputation. Inconsistent research ﬁndings on the effect of school choice policies on
student achievement imply that school choice polices might not contribute to improving
teaching and leadership practice. As an attempt to explain these inconsistent research
ﬁndings regarding the effect of school choice policies on student achievement, this study
addresses how school choice policies affect educators’ teaching and leadership practices.
Studies on parental school choice imply that parents’ school choice behaviors can
quantitatively limit competition among schools and might signal that educators do not
have to improve their teaching and leadership practice and programs in order to attract
more students. There is information asymmetry between socioeconomic classes and some
parents fear competition or failure in a higher status school. These ﬁndings imply that
parents might not exercise choice options enough to bring about competition among
schools. In terms of parental preferences for schools, most parents, when asked, seem to
say that they choose schools for their kids based on academic quality such as student
achievement, school curriculum, and teacher quality. At the same time, some studies
based on in-depth interviews with parents and parents’ information search patterns on the
internet revealed that some middle school parents choose schools for their children based
on status ideologies that emphasize race and class. These studies imply that school choice
might reinforce school segregation, and educators might not necessarily make serious
efforts to improve school quality. Assuming that educators as service providers will
attend to the factors which affect parental school choice and their responses will be

affected by how they frame parental choice, this study examines how educators frame

13

parental choice and how the framing of parental choice affects their responses to student
enrollment changes.

Research studies on the responses of parents and educators to school choice imply
that the introduction of a school choice policy itself does not bring about school
improvement. Effective school climates in private schools might result from the scarcity
of private schools and student selection, different priorities in education goals, high
tuition costs, homogeneity of students and parents, and high socioeconomic backgrounds
of parents. Some charter schools which are run with a given budget from districts or their
state government tend to maximize their proﬁt by recruiting low-cost students like lower
grade students and excluding high cost students like upper-grade students and special
education students. A study addressing higher education competition and program quality
implies that severe competition for students might not lead to rigorous or better programs
when students do not want rigorous or quality programs (Young et al., 2002). What really
matters might be how educators frame choice options, problems, and strategies and
whether they have capacity to cope with those issues.

Another issue in research on school choice is that prior research studies did not
focus much on traditional public schools. Considering the current education system
where traditional public schools account for a majority of schools and school choice
policies among public schools have been introduced broadly, it is important to examine
how school administrators and teachers in traditional public schools respond to school

choice policies.

14

CHAPTER 2

THEORETICAL FRAMEWORK AND RESEARCH QUESTIONS

Theoretical Framework

Two competing economics theories on markets provide useful insights into
examining how parents, school administrators and teachers respond to school choice: the
neoclassical economics perspective and the institutional and behavioral economics
perspective. These theories have different assumptions with regard to human actions, the
roles of a market and government (or institutions), and the performance of a market.
Neoclassical Economics Perspective

There is not complete agreement on what is meant by neoclassical economics and
as a result, there is a wide range of neoclassical approaches to various problem areas and
domains. Although certain branches of neoclassical theory may have different approaches,
neoclassical economics rests on the following three assumptions (Samuelson, 1973),

l. Pe0ple have rational preferences among outcomes that can be identiﬁed and
associated with a value.

2. Individuals maximize utility and ﬁrms maximize proﬁts.

3. People act independently on the basis of full and relevant information.

From these three assumptions, neoclassical economists have built a structure to
understand the allocation of scarce resources among alternative ends. In the context of
school choice, parental choice and responses of educators to parental choice can be
thoughtof as the allocation of scarce resources among alternative ends. Since

neoclassical economists assume that a free market system based on choice and

15

competition will guarantee effectiveness and efﬁciency of goods or service production,
they argue that government should not regulate or control the free choice of economic
actors (Friedman, 1962).

According to this perspective, in a free education market where parents choose
schools for their kids and schools make efforts to attract more students, parents will
choose the most appropriate schools for their children with perfect information about
available schools in order to maximize their interests or preferences. In addition, school
systems will create market incentives under which educators get rewards such as job
security, a bonus, or a raise when they attract more students and get sanctions or penalties
such as a wage cut or loss of their jobs when they lose students. In order to get rewards
by attracting more students, educators will make efforts to meet parents’ needs or
preferences, evaluate and compare their own programs with those of competing schools,
and reduce costs (see Figure 1). This study examines whether school systems create
market incentives for educators and how educators make efforts to attract more students.

In order to introduce school choice policies as a means to improve school quality,
neoclassical economists also assume that the preferences or interests of parents are
aligned with school quality and the social objectives of public education. However, it is
not easy to deﬁne school quality because it is related to various value judgments about
the goals of education, what to teach, and how to teach. What a society expects of schools
can be decided by the individual choices of parents and children or by collective
decision-making. While proponents of the neoclassical economics perspective argue that

individual choice is a way to address debates about how to deﬁne school quality,

16

opponents of school choice policies maintain that individual choice can exacerbate social
equity and social inclusion.
Institutional and Behavioral Economics Perspective

Questioning the basic assumptions of neoclassical economics, the institutional and
behavioral economics perspective postulates that the behaviors (including decisions or
choices) of actors in a market are inﬂuenced by social institutions (i.e., social
arrangements) such as laws, norms, ideologies, and beliefs (Schmid, 2004). Drawing on
behavioral science research on cognition and perception, institutional and behavioral
economists argues that human behaviors or responses to an event are shaped by social
institutions rather than just their rational preferences or maximization of utility or proﬁts.
Behavioral science indicates that the human brain features four aspects which limit
human capacity and lead humans to develop social institutions to overcome their limited
capacity; 1) bounded rationality, 2) emotions and evaluations, 3) behavioral regularities,
and 4) learning (Schmid, 2004). To predict the possible responses of educators in the
context of student enrollment changes due to school choice, I assume that these four
aspects of the human brain are applied to educators as humans.

First, school administrators and teachers have bounded rationality. The human
brain has limited information processing capacity and it features modularity. These two
characteristics of the human brain result in bounded rationality. Humans are procedurally
rational and use whatever reasoning power is available, but they know that they cannot be
substantially rational in the sense of considering everything that could possibly affect
their decisions (Simon, 1982). In the context of school choice, not all parents have

enough information about which schools are available or are best for their kids and no

17

parents have perfect information (Elmore, 1986; Henig, 1996; Lee et al., 1996; Pearson,
1993). Even those parents who are provided with enough information might not know
what the information means for them and their kids (Wells, 1996). For school
administrators and teachers, when there are decreases in student enrollment, we cannot
assume that they perceive such changes. In addition, even though they may perceive such
changes, they might not consider them as a problem, or if they consider the changes to be
a problem they might not know how to solve the problem.

An individual educator might have conﬂicting desires or interests. This implies
that he/she might not respond consistently to an event or signal. For example, even
though he/she is trying to improve their teaching practice to get rewards, he/ she might not
continue to do this at other times without any changes in the environment. This results
from the modularity of the human brain. The modularity of the human brain means that
different brain components have some ability to affect behavior independently of other
components. In other words, the brain is not unitary or necessarily internally consistent
(Carter, 1998). Since different parts of the brain can take in information from the
environment and issue orders for action without the participation of other parts, our
brains are not completely integrated and hierarchical.

Multiple beliefs or opinions originate from the modularity of the human brain and
this can create problems for public policy and collective decision making. For instance,
when asked, parents say that their choice of schools is motivated by academic quality
such as student achievement, school curriculum, or teacher quality (Armor & Peiser,

1998; Greene et al., 1998; Kleitz et al., 2000; Public Policy Forum, 1998; Schneider et al.,

1 998; Vanourek et al., 1998), but in reality they might choose schools based on status

18

ideology which emphasizes race and class (Fuller et al., 1996; Holme, 2002; Lee et al.,
1996). While teachers might think that they should improve their teaching practice to
attract more students into their schools, they might want to relax during their leisure time
rather than devote such time to trying to improve their teaching because another part of
them would prefer to relax. The modularity of the human brain implies that thinking and
action might not be consistent. Therefore, to examine how school choice policies work in
the real world, we need to examine whether educators perceive changes in student
enrollment, how they frame such changes, whether they have capacity to solve problems
associated with such changes, and how they act in practice.

Second, educators’ responses to events or other people are mediated by emotions
like pleasure or pain. “The brain continuously constructs an affective or hedonic
commentary on the current state of affairs. . .” (Kahneman, 1999, p. 7). Phenomena and
other people bring pleasure or pain and we seek or avoid them. However, not all
phenomena or people elicit so much emotion that it leads to a reaction. Events or
contingencies relevant to the actual or expected achievement of or harm to major goals,
motives, and values bring about emotions that can change behaviors (Frijda, 1999).

In relation to school choice, the key issue is whether phenomena like a student
enrollment change itself and/or resulting changes in rewards or job security can be
enough of an emotional stimulus to motivate teachers and administrators to improve
school quality. Emotions are triggered by beliefs which provide a meaning and sense of
direction to educators. As a result, educators’ beliefs strongly affect their behaviors
(Abelson, 1979; Bandura, 1986; Brown & Cooney, 1982; Clark & Peterson, 1986;

Eisenhart, Shrum, Harding, & Cuthbert, 1988; Nespor, 1987; Pajares, 1992; Rokeach,

l9

1968). For instance, whether student enrollment decreases can trigger educators’
emotions is determined by their beliefs. If educators believe that the student enrollment in
their school decreased because their school is located in a disadvantaged area and does
not have enough resources to attract students again, we cannot assume that the student
enrollment decrease will elicit teachers’ emotions enough or in ways that motivate them
to reﬂect on their teaching practices. The difﬁculty in policy formulation like school
choice is that the degree of pleasure versus pain is itself not calculated in any means-ends
relationship, but is the material of calculation or direct action. Emotions can replace
calculation and they can alter the values used in calculation (Elster, 1998). Therefore,
even though there are the same rewards or sanctions in terms of money or job security,
how they work depends on the beliefs possessed by individual educators.

Third, educators might simplify complex signals to overcome the limited
information processing capacity and modularity (i.e., signals in their environment or in
situations they encounter) and develop behavioral regularities. For example, humans
make mental accounts (Thaler, 1985), organize choices lexicographically (Earl, 1983),
engage in selective perception (Simon, 1991), develop habits and standard operating
procedures, and identify sub-goals (Simon, 1979). In curriculum reform, educators tend
to adhere to surface level understandings, which focus chieﬂy on the coverage and
sequencing of mathematics and science topics and fail to press any fundamental
reconceptualization of mathematics and science knowledge and teaching practice (Cohen
& Ball, 1990; Spillane, 2004).

When schools encounter decreasing student enrollment, school administrators and

teachers might not pay attention to the causes of the student enrollment decreases.

20

Though there might be many other reasons that their school is losing students, they might
simplify or selectively perceive that they are losing students because their schools are, for
example, located in low-income neighborhoods. As a result, they might fail to evaluate
their schools from different perspectives or to improve school quality.

‘ Finally, educators engage in learning and prior knowledge and beliefs affect their
learning. Educators as humans make sense out of their environment and act upon it
(Mantzavinos, 2001). Humans take in sensory information, act, take in more sensory
information and then rest, continue, or escape. There are two cognitive models to explain
human learning: a model of stimulus, behavior, and reinforcement (Skinner, 1971) and a
complementary model of cues and patterns. Stimuli include such things as changes in
student enrollment and changes in rewards and job securities. In response to a stimulus,
humans engage in behavior that may change something in their environment. Feedback
either reinforces the behavior or alters it. As emotions are affected by educators’ beliefs
or ideologies, their learning of and responses to student enrollment changes are also
affected by their beliefs or ideologies. In relation to school choice policies, it is important
to examine whether the rewards or sanctions as stimuli are enough to cause educators to
change their teaching or leadership practices and whether appropriate feedbacks are
provided for their efforts in order for them to continue to make efforts.

When the human brain tries to make sense out of complex new sensory inputs that
it cannot possibly compare in detail to prior experiences, it sees a pattern and in some
sense ﬁts an old understanding and habit to the new situation. This theory explains the

reasons that many curriculum reforms fail to reconceptualize educators’ prior knowledge

and practice (Cohen & Ball, 1990; Spillane, 2004). This theory also implies that there

21

should be considerable efforts to change the prior knowledge or beliefs. However, this
does not imply that educators’ beliefs or ideologies are ﬁxed and unchangeable. In
relation to the context of school choice, when educators analyze the causes of student
enrollment decreases in their schools, their prior understanding of parental school choice
can affect their analysis. If they believe or know that parents tend to choose schools for
their kids based on school location or student socioeconomic backgrounds, their analysis
is more likely to focus on those aspects rather than teaching practice or student discipline.

In sum, the institutional and behavioral economics perspective runs counter to the
basic assumptions of the neoclassical economics perspective (i.e., rational choice based
on perfect information, maximization of utility and proﬁts, and acting independently).
With limited information processing capacity, it is not easy for actors to identify all
available alternatives, compare and evaluate them, and ﬁnally choose best options.
Humans rely on behavioral regularities to overcome their limited information processing
capacity. Humans do not act independently on the basis of full and relevant information.
Instead, they make decisions or act based on various grounds such as emotions, beliefs,
ideologies, norms, or self interest. Self interest is just one of many references and may
not be critical in decision making and action.

The institutional and behavioral economics perspective suggests the possible
effects of school choice policies on educators as follows (see Figure 1). First, educators
might fail to be aware of student enrollment changes. Second, even though they perceive
student enrollment changes, they might have no concerns about them. Third, it is possible

that they might not know what to do in order to increase student enrollment or they might

employ inadequate strategies. Finally, they may know what to do in order to increase

22

student enrollment but they might be engaged in socially inappropriate market behaviors

such as cream skimming. The paths educators take will depend on their beliefs or

knowledge, capacity, and learning opportunities.

 

 

 

 

 

 

 

 

 

 

 

 

 

     
 

 

 

 

Neoclassical Basic assumption: Parents and educators as humans have rational preferences,
Economics maximize utility and proﬁts, and act independently with full and relevant
Perspective information
- Rational choice based - Survey parent needs - Contribute to
on school quality with - Evaluate their own and improving school
’ perfect information ¢ competitors’ programs _p quality
- Develop effective
organization
f
Market Educators Market
Approaches to S h l P f
Education ( c 00 3) er ormance
0
Institutional - Limited choice due to - Fail to be aware of - Improving school
Economics limited information student enrollment quality depends on
Perspective (capacity) or willing to changes social arrangements

 

avoid competition

- Choice based on various
preferences such as
school quality, status
ideology, or convenience

l.

 

 

- No concerns about
decreasing student
enrollment

- Employ inadequate
strategies

- Socially inappropriate
market behaviors: cost
creaming, cream
skimming, or watering
down curriculum

 

 

and educators’
capacity

 

 

 

 

 

Basic assumption: parents and educators as humans have bounded rationality,
emotions and evaluations, behavioral regularities, and learning. Behaviors are
determined by social institutions and limited human capacity.

 

Figure 2-1: Neoclassical economics perspective vs. Institutional economics perspective

23

 

 

Research Questions and Hypotheses

Research Question 1) How much are teachers and school administrators aware of
changes in enrollment? Do schools have incentive systems aligned with decreases or
increases in student enrollment?

Hypothesis 1a (NCE): When there are rewards or sanctions for student enrollment
changes, educators would be aware of enrollment changes. In addition, the degree of
educators’ awareness of student enrollment changes (or the percentage of choice students)
at their schools would be positively associated with the degree of the impact of student
enrollment changes on them.

Hypothesis 1b (IBE): Educators attach different meanings to the rewards and
sanctions depending on their values or belief systems. Thus, the impact of the rewards
and sanctions would be different among educators in a school. As a consequence, the
same rewards or sanctions could make some educators more sensitive than others to
enrollment changes.

Research Question 2) How are schools/districts that lose students as part of
school choice policies (i. e., losing schools/districts) different from schools/districts that
gain students (i. e., gaining schools/districts) in terms of teachers ’ working conditions
such as available time and resources for teachers, leadership, and opportunities for
professional development? Are teachers’ working conditions associated with the
intensiveness of school competition?

Hypothesis 2a (NCE): Schools in gaining school districts would have better

working conditions because rational parents would choose schools with better working

conditions and educators would make an effort to provide better working conditions. The

24

degree of the differences in working conditions between gaining and losing schools
would be associated with the degree of school competition (i.e., the degree of student
enrollment change). As competition among schools becomes more intensive, schools will
make more effort to improve working conditions.

Hypothesis 2b (IBE): There might be no signiﬁcant difference in teacher working
conditions between schools in gaining and losing school districts when educators do not
know the importance of teachers’ working conditions or when they underestimate the
effects of improving teachers’ working conditions. In addition, parents might not
consider teachers’ working conditions when they choose schools for their children. In this
case, parental choice would not lead educators to provide better working conditions.

Research Question 3) What are schools doing to attract students? Is the degree
of organizing activities for attracting more students associated with school competition?
Are the activities associated with school problems or parents ’ school choice reasons?

Hypothesis 3a (NCE): The schools would examine who choice students/families
are and what types of schooling options they are seeking. As competition among schools
becomes severe, schools would make more efforts to attract students. In addition, as a
given school problem becomes more serious, schools would make more efforts to address
that problem. Finally, schools’ strategies for attracting students would be closely
associated with the reasons for parents’ school choice decisions.

Hypothesis 3b (IBE): There might be no difference between schools when schools
fail to analyze problems, do not have enough capacity to come up with appropriate

strategies, or just adopt other schools’ strategies. Schools might rely only on their

25

perceptions of the reasons for parents’ school choice decisions without doing systematic
research about parents’ actual decisions.

Research Question 4) How do teachers and administrators perceive the impact of
school choice policies on teachers and administrators themselves and their schools? Are
their perceptions of the impact associated with the degree of competition among schools?

Hypothesis 4a (N CE): Educators would believe that school choice programs could
contribute to improving school quality through improving their teaching (leadership)
practices, academic programs, extra-curricular activities, collaboration with colleagues,
and relationships with parents. In addition, educators in schools under more intensive
competition would be more likely than those in schools under less intensive competition
to have these beliefs.

Hypothesis 4b (IBE): Educators might not believe that school choice programs
could contribute to improving school quality through improving their teaching
(leadership) practices, academic programs, extra-curricular activities, collaboration with
colleagues, or relationships with parents. In addition, there might be no associations
between the intensiveness of competition and their perceptions of the degree of the

impact of school choice programs on educators themselves and/or schools.

26

CHAPTER 3

METHODOLOGY

In order to examine in depth how school administrators and teachers in traditional
public schools respond to school choice policies, quantitative and qualitative research
methods were employed in this study.

Research Sites and Policy Context

The research sites for this study are in the state of Michigan. The schools of
choice state policy (an inter-district school choice program) in Michigan was made
possible by the approval of Proposal A in 1994, which radically changed the basis of
school funding. Before 1994, the amount of revenue in each school district depended
primarily on local property values and the local millage rates. When the sources of school
funding shifted primarily from districts to the state, Michigan also introduced charter
schools and schools of choice policies.

Since 1994, most local school districts have received their operating revenue from
state appropriations based on the number of students enrolled and local property tax
millage. The state appropriation per student in 2005-06 was about $6,100 to 6,800 per
student in the districts studied here. In addition, school districts receive funding from
local government and federal government. However, there is even more variance among
school districts in terms of the amount of funding from local and federal government. For
example, the amount of local funding in the districts in this study in 2005-06 ranged

approximately from $1,390 to $5,230. That of federal funding in the districts in this study

27

sites in 2005-06 ranged approximately from $170 to $1,320
(http://www.nces.ed.gov/ccd/districtsearch0.

In Michigan, students may enroll in any public schools within their intermediate
school districts (ISDs) or in contiguous school districts. Note that a school district may
decide whether or not to make spaces available for non-resident students. But a district
may not prevent students who reside within its boundaries from attending schools in
another district. This means that not all public schools participate in competition for
students. If there are more applicants than openings, districts must use a fair lottery to
make enrollment decisions.

Interviews with School Administrators and Teachers for a Pilot Study

Interviews were conducted during fall 2007 and spring 2008 in order to examine
whether the theoretical framework of this study was appropriate and to develop survey
questionnaires. Two school districts in the state of Michigan were selected. While one
school district was gaining students from neighboring school districts because of school
choice programs during the past ﬁve years, the other was losing students to neighboring
school districts. In the gaining school district, three schools (one at each school level)
were selected, and the principal and two teachers in each school participated in an
interview. In the losing school district, two schools (elementary and high school) were
selected, and the principal and two teachers in each school participated in an interview.
The participants in each district were recruited using chain-referral sampling (Glesne,
2006). Each interview lasted about 20 minutes to one hour.

The interviews addressed the following issues in depth: whether and how the

participant perceives changes in enrollment in his/her school or district; how changes in

28

enrollment have impacted his or her (colleagues’) salary, job security, and beneﬁts; how
the changes in enrollment have affected classroom teaching in his or her school or district;
why the enrollment in his/her school or his/her district has decreased or increased;
whether and how the participant, his/her school, and/or his/her district made (or currently
makes) efforts to attract more students; whether there have been changes in opportunities
for parents to provide input or feedback since the introduction of school choice policies;
and opinions about school choice policies.

With respect to analyzing the interview data, all the interviews were audio-taped
and transcribed, and a commercial computer-based qualitative coding program, NVivo,
was used to code all of the data. Using NVivo, emerging ideas and concepts from the data
were coded into free nodes that could be compared and related to each other, forming
larger “parent” nodes that could be stored in an index system that brought the different
components of the project together. These interviews made it possible to compare the
differences between the two districts and between school administrators and teachers.

Based on the interviews with school administrators and teachers, this study found
that educators did not seem to respond to competition in the ways that the classical
economics perspective predicted. First, educators in the losing district believed that they
were losing students to neighboring districts because neighboring districts were located in
more afﬂuent communities and had better facilities than their district. As a result, schools
that were losing students would not be expected to make efforts to improve their teaching
quality. Educators in the losing district might not have had appropriate strategies to keep

their district’s students from leaving for other districts.

29

Second, educators in both the losing and the gaining districts did not seem to
spend much time examining the reasons why they were losing students, what kinds of
students came to their districts and left for other districts, and what strengths their
competitors had. School principals and teachers felt that it was the responsibility of the
central ofﬁce to attract students to their schools. This implies that administrators and
teachers did not invest enough resources in making strategic plans to win competition. In
both districts, advertising their programs seemed to be the sole strategies for attracting or
maintaining student enrollment.

Third, the compensation and hiring/ﬁring systems for educators were not
consistent with market approaches. The changing enrollments affected job security,
especially for less experienced teachers, because more experienced teachers were
protected by seniority rules. Teacher salaries or beneﬁts were not affected by the
enrollment changes because of union contracts. The districts did not have appropriate
systems for screening out poor quality teachers.

Fourth, there were mixed responses about whether school choice programs
motivated teachers to be more responsive and improve their teaching. While some
participants reported that educators were motivated by work ethics or their own intrinsic
values rather than competition among schools or teachers, others responded that
competition among schools was an incentive to improve their practices.

Finally, community support in the gaining district seemed to be weakened due to
increasing choice students. The community members were reluctant to invest in schools
for choice students. In the gaining district, there were many people who did not

understand why they should attract choice students given that the parents of choice

30

students didn’t pay taxes to maintain or renovate schools. These individuals believed that
since choice students were using schools in their community, their parents should have
helped pay for the schools. But, instead, individuals who lived in the community were the
ones who were expected to pay for school renovations because the community could not
use state appropriations to build new schools or renovate them. The superintendent
reported that it was hard to gain support among community members for increased
spending on schools. Market competition might also have reduced collaboration among
districts because they may have felt that their programs and marketing skills were
business secrets.
Sampling and Response Rates

With respect to the sampling of participants for my dissertation study, I ﬁrst did
location sampling (Heckathom, 2002), assuming that school administrators and teachers
in districts with drastic changes in enrollment due to school choice policies would be
more responsive to such policies than other districts. Multiple districts that were faced
with drastic changes in enrollment from 2004 to 2008 due to school choice policies were
identiﬁed by analyzing the Michigan Department of Education Single Record Student
Database.

For the surveys, seven school districts in Michigan were selected from four areas
(see Table 3-1). By the time of this study, three of these were districts that had gained
students from neighboring school districts and four were districts that had lost students to
neighboring districts. For example, while school district A attracted about 28% of its total
enrollment from neighboring school districts, school district G lost about 19% of its

enrollment to neighboring school districts. The percentage of student enrollment changes

31

between 2004-2005 and 2008-2009 varied from -28% to 5% across the school districts.
The percentage of students eligible for free and reduced lunch in the districts ranged from
13% to 77%. The percentage of minority students in the districts varied from 2% to 67%.

Table 3-1: School District Characteristics

0
Total %Enrollment %Free/Reduced %Net Choice

 

 

08.2835. (Egggljggg) (22818859 )to (,gggfgg, (,gg‘gggtgg, (8355633)
A 2,433 5.19 36.87 27.78 3.74
B 4,102 -l.75 14.31 10.65 29.06
C 2,871 1.34 33.37 20.38 6.37
D 1,802 ~7.ll 41.84 -4.94 2.28
E 6,351 -6.27 70.10 -l6.34 47.80
F 4,766 -5.47 12.76 -6.67 37.94
G 1,997 -27.91 76.82 ,-18.58 67.35

 

Note. %Enrollment Change=(Enrollment in 2008-2009 — Enrollment in 2004-
2005)/Enrollment in 2004-2005 * 100. %Net Choice Students = (Incoming Choice
Students in 2008-2009 — Outgoing Choice Students in 2008-2009)/Enrollment in 2008-
2009 * 100.

Thirty schools in the seven school districts were selected. Depending on the
school level, 10 to 20 teachers and administrators in each school were selected. The
participants in each school were randomly selected among teachers who taught in the
core content areas: elementary education, English/language arts, mathematics, science,
and social studies. I selected core content area teachers because of the following two
reasons. One is that all the schools have core content area teachers, so it is possible that
the teacher groups in the sample are quite similar in terms of the subject matters they
teach. The other is that parents could be assumed to be more sensitive to the quality of
instruction of core content areas, including language arts and mathematics, than other

elective courses when they choose schools for their children. As shown in the Table 3-2,

surveys were administered to 49 school administrators and 552 teachers in spring 2009.

32

Among them, 32 school administrators and 332 teachers participated in the surveys, thus
response rates were 65% for administrators and 60% for teachers.

Table 3-2: Survey Sample and Response Rates

 

 

School Response/Target Response Rates
Admin. Teacher Admin. Teacher
Gaining School Districts
Elementary 5 3/5 42/70 60.0 60.0
Middle 4 8/8 47/78 100.0 60.3
High 3 6/8 59/87 75.0 67.8
Losing School Districts
Elementary l 1 7/12 93/162 58.3 57.4
Middle 4 6/10 52/85 60.0 61.2
High 3 2/6 36/70 33.3 51.4
Anonymous 3
Total 30 32/49 332/552 65.3 60.1

 

As shown in Table 3-3, female teachers accounted for 78% of the teacher
participants. Teachers with more than 13 years experiences in their current schools made
up 31.6% of the participants, teachers with 9-12 years of experience accounted for 21.3%,
teacher with 4-8 years made up 30.6%, and teachers with 3 years of experience or less
accounted for 16.6%. About 6% of teachers in the survey had worked less than 1 year in

their current schools.

33

Table 3-3: Teacher Characteristics

 

 

Gaining School Districts Losing School Districts Total
Gender

Female 1 10 124 234
77.5% 78.5% 78.0%
Male 32 34 66
22.5% 21.5% 22.0%
Total 142 158 300
100.0% 100.0% 100.0%

Work Years in Current School
1 year 5 14 19
3.5% 8.8% 6.3%
2-3 years 15 16 31
10.6% 10.0% 10.3%
4-8 years 52 40 92
36.9% 25.0% 30.6%
9-12 years 30 34 64
21.3% 21.3% 21.3%
More than 13 years 39 56 95
27.7% 35.0% 31.6%
Total 141 160 301
100.0% 100.0% 100.0%

 

As indicated in Table 3-4, female administrators accounted for 42% of the

participants in the administrator survey. About 45% of the administrators had worked less
than 3 years in their current schools. Administrators with more than 13 years experiences
in their current schools made up 22.6% of the participants, those with 9-12 years of

experience accounted for 16.1%, and those with 4 to 8 years made up 16.1%.

34

Table 3-4: Administrator Characteristics

 

 

Gaining School Districts Losing School Districts Total
Gender

Female 5 8 13
31.3% 53.3% 41.9%
Male 1 l 7 18
68.8% 46.7% 58.1%
Total 16 15 31
100.0% 100.0% 100.0%

Work Years In Current School
1-3 years 6 8 14
37.5% 53.3% 45.2%
4-8 years 3 2 5
18.8% 13.3% 16.1%
9-12 years 2 3 5
12.5% 20.0% 16.1%
More than 13 years 5 2 7
31.3% 13.3% 22.6%
Total 16 15 31
100.0% 100.0% 100.0%

 

Both electronic and hard copy surveys were administered to school administrators
and teachers. To ensure high response rates, I employed a ﬁve contact approach (Dillman,
2007). This method included sending each eligible administrator and teacher 1) an initial
letter introducing the study; 2) a cover letter, a consent form, a link to the electronic
survey, and a token incentive; 3) a postcard reminding eligible participants about the
survey/thank you letter to those who completed the survey; 4) a second mailing of the
cover letter, consent form, and electronic survey link to those who had not yet responded;
and e) a ﬁnal contact (hard copy survey with return envelope to those who had not yet
responded). For those who did not have access to the electronic version of the surveys,

hard copy surveys were sent in the ﬁnal contact. The ﬁve contacts (a through e) took

35

 

place about ten days apart from each other between March to May in 2009. At each step
(i.e., after each contact), about 15 to 20 percentage of the remaining sample completed
the survey.

Key Constructs of Surveys and Interviews

The survey instruments for teachers and administrators consisted of four key
constructs: 1) Teacher Working Conditions, 2) Teacher (Leadership) Practices and
Beliefs, 3) Perceptions of and Responses to School Choice, and 4) Demographics (see
Table 3-5). The items for evaluating Teacher Working Conditions included questions
about available time and resources for teachers, teacher empowerment, leadership,
collegial support, opportunities for professional development, and perceptions of school
problems. The items for examining Teacher or Leadership Practices and Beliefs included
questions about teaching practice, general beliefs about student ability and accountability,
participation in professional development, time commitment, and satisfaction.

The items for examining Perceptions of and Responses to School Choice included
questions about acquiring information about student enrollment; changes in enrollment
and awareness of such changes; information about choice students; impact of enrollment
changes on teachers’/administrators’ job security, salary, and beneﬁts; and reasons
behind the enrollment changes. In addition, these items also included evaluation of their
own schools in terms of competitiveness, endeavors at their schools to attract students,
evaluation of the endeavors, impact of school choice programs on teachers, and impact of
school choice programs on their schools. The items for Demographics included working
experiences, tenure, education, certiﬁcation, race/ethnicity, and school level (see Table 3-

2 and see Appendices 1 and 2).

36

Table 3-5: Key Constructs of Teacher and Administrator Surveys

 

Teacher Survey

Administrator Survey

 

Teacher Working Conditions

- Available time for teachers (Bl)

- Available resources for teachers (BZ)

- Empowerment (B3)

- Leadership (B4)

- Collegial support (BS)

- Professional development (B6)

- Perceptions about school problems (sources of
problems/teacher belief) (B7 and B8)

Teacher Practice and Beliefs

- Teaching practice (C 1)

- General beliefs about student ability (C2)

- General beliefs about accountability (C3)

- Participation in professional development (C4)
- Time commitment (C5 and C6)

- .lob satisfaction (C7)

School Choice

- Acquiring student enrollment information (D1)
- Changes in enrollment and awareness of them
(D2 and D3)

- Information about choice students (D4)

- Impact of enrollment changes on teachers’ job
security, salary, and beneﬁts (D5)

- Reasons for parents’ school decisions (D6)

- Evaluation of their own schools in terms of
competitiveness (D7)

- Endeavors at the school to attract students (D8)
- Evaluation of the endeavors (D9)

- Impact of school choice programs on teachers
(D10) '

- Impact of school choice programs on school
(D11)

Demographics

- Working years
- Tenure

- Education

- Certiﬁcation

- Ethnicity

- School level

Teacher Working Conditions

- Available time for teachers (Bl)

- Available resources for teachers (BZ)

- Empowerment (B3)

- Leadership (B4)

- Collegial support (B5)

- Professional development (B6)

- Barriers to the dismissal of poor-performing
teachers (B7)

- Perceptions about school problems (sources of
problems/teacher belief) (B8 and B9)

Satisfaction and Commitment
- Job satisfaction (B10)
- Time commitment (Bl l and 812)

School Choice

- Acquiring student enrollment information (C 1)
- Changes in enrollment and awareness of them
(C2 and C3)

- Information about choice students (C4)

- Impact of enrollment changes on his/her own
job security, salary, and beneﬁts (C5)

- Impact of enrollment changes on staff
members’ job security, salary, and beneﬁts (C6)
- Reasons for parents’ school decisions (C 7)

- Evaluation of their own schools in terms of
competitiveness (C8)

- Endeavors at the school to attract students (C9)
- Evaluation of the endeavors (C10)

- Impact of school choice programs on teachers
(Cl 1)

- Impact of school choice programs on school
(C12)

Demographics

- Working years
- Education

- Certiﬁcation

- Ethnicity

- School level

The survey items for Teacher Working Conditions, Teacher Practices and Beliefs,

and Demographics were drawn from the 2007-08 Schools and Staffing Survey (SASS) of

the National Center for Education Statistics (NCES) and the Arizona Teacher Working

37

Conditions Survey, conducted by the Center for Teaching Quality (Berry, Fuller, &
Williams, 2007). The items for Perceptions of and Responses to School Choice Programs
were created for this study based on the interviews with school administrators and
teachers. The validity of the survey items was tested with four teachers and two school
administrators from gaining and losing school districts that were not participating in the
dissertation study.

Analysis Scheme
Statistical Models

First, exploratory factor analyses were conducted to see if making one composite
score for each construct of teachers’ working conditions could be justiﬁed. The factor
analyses showed that sub items of each construct were well correlated and could be
composited into one score. In addition, in order to check the internal consistency among
the sub items for each construct, Cronbach’s alpha coefﬁcients were calculated. The
Cronbach’s alpha coefﬁcients ranged from .733 to .926, showing very good reliability
(see Table A3 in Appendix C).

Second, to see overall trends, the mean scores of all responses, those for losing
school districts, those for gaining school districts, those for teachers, and those for
administrators for the variables of interest were calculated, ignoring clustering of I
educators within schools. Then, because of the multilevel nature of the survey data (i.e.,
teachers were nested in schools), hierarchical (generalized) linear modeling (H(G)LM)
analyses were employed to examine the statistical differences between gaining and losing

school districts, across school levels, between administrators and teachers, between

38

female and male educators, and between tenured and non-tenured educators with respect

to the following:

awareness of enrollment changes from 2004-05 to 2008-09 at their schools
awareness of the percentage of choice students at their schools in 2008-09
perceptions of the impact of changes in enrollment on educators’ job security,
salaries, and beneﬁts

perceptions of teachers’ working conditions

perceptions of school activities for attracting more students

perceptions of school problems

perceptions of school choice reasons

perceptions of the impact of school choice programs on educators’ practices

perceptions of the impact of school choice programs on schools

Then, I conducted hierarchical (generalized) linear modeling (H(G)LM) to test the

following associations:

between educators’ perceptions of job security, salaries, and beneﬁts and their
awareness of student enrollment changes from 2004-05 to 2008-09 at their
schools

between educators’ perceptions of job security, salaries, and beneﬁts and the
percentage of choice students in 2008—09 at their schools

between competition among schools and teachers’ working conditions
between school competition and school activities for attracting more students
between school problems and school activities for attracting more students

between choice reasons and school activities for attracting more students

39

- between school competition and educators’ perceptions of the impact of

school choice programs on their practices

- between school competition and educators’ perceptions of the impact of
school choice programs on schools and education

The detailed models were illustrated in the related parts. Position (administrators or

teachers) at the individual level was included as control variables. At the school level, the

following variables were included as explanatory or control variables depending on

research questions: the gaining or losing status of school districts, student enrollment

changes over the past five years, the percentage of choice students, the student/teacher

ratio, school size, local funding, the percentage of students eligible for free and reduced

lunch, the percentage of minority students, and school level. See Table 3-6 for school-

level variable descriptive statistics.

Table 3-6: School-Level Variable Descriptive Statistics

 

N Min. Max. Mean SD.
Enrollment Changes from 2004-05 to 2008-09 30 -45 .9 66.0 -3.9 17.8
% of Choice Students in 2008-09 30 .0 41.0 14.1 1 1.9
Student/1' eacher Ratio in 2008-09 30 15.5 23.9 20.2 2.2
School Size (100 students) in 2008-09 30 1.3 15.5 5.3 3.4
Local Funding ($100) in 2006-07 30 13.9 52.3 29.5 12.7
% of Students Eligible for Free or Reduced Lunch in 2008-09 30 6.4 82.9 38.6 23.6
% of Minority Students in 2008-09 30 .0 69.1 23.0 21.6

 

In this study, statistical signiﬁcance was determined at the P value equal to or less

than 0.10. Each coefﬁcient in the results has information about statistical signiﬁcance

denoted by *"‘* P5001, ** P3005, or * P5010.

Regarding the outcome variables for teachers’ working conditions which are in

continuous scale, intra-class correlation coefﬁcients (ICCs) were calculated to determine

40

whether there was enough variance among schools to employ multilevel modeling. As
shown in Table A4 in Appendix D, the ICCs ranged from .20 to 54, which meant that
school-level variance accounted for 30% to 54% of the total variance in educators’
perceptions of teachers’ working conditions. We could say that there is a non-negligible
nesting effect.

Missing Data

The missing rate for the administrator survey was less than 5%, and that for the
teacher survey ranged from 5% to 16% depending on the items. There were more missing
data for the items on the last section of the survey. Overall, the survey data had a
monotone missing pattern where in the event that a variable YJ- was missing for the
individual i, the individual i was missing all subsequent variables Yk, k>j (Horton &
Kleinman, 2007).

Relatively more teachers at losing schools than those at gaining schools were in
the missing group whose members did not respond to most items on the last parts of the
survey. The differences between the missing group and total sample seemed to be small.
There were almost no differences between the two groups with respect to their
perceptions of teachers’ working conditions which were addressed in the ﬁrst part of the
surveys (see Table A5 in Appendix E). In addition, little differences were found between
the two groups regarding the impact of student enrollment increases/decreases on
teachers’ job security, beneﬁts, and salary which were addressed in the last part of the
survey. In sum, the data sets seemed to be missing data at random (MAR).

There could be many methods for the analysis of incomplete data regression

models such as the complete case model, ad-hoc methods, multiple imputation,

4l

likelihood-based approaches, weighting methods, and Bayesian approaches (Horton &
Kleinman, 2007). Among them, I used the simplest method: the complete case method,
which involved the analysis of the set of observations with no missing values because the
data sets in this study were close to missing at random and had monotone missing
patterns. When missing data is missing completely at random (MCAR), then the
complete case (CC) estimator is unbiased. The main drawback of the CC estimator is that
if there are many different variables with missing values, then a large fraction of the
observations may be dropped. However, since the data sets in this study had monotone
missing patterns, dropping missing values did not substantially increase the proportion of

the observations that were excluded from analyses due to this method.

42

CHAPTER 4

AWARENESS OF ENROLLMENT CHANGES

In order to improve school quality through school choice programs, school
administrators and teachers should know about student enrollment changes in their
schools, and then they should employ appropriate actions to reach the optimal number of
students in their schools. It might be assumed that educators would know about student
enrollment changes once school choice programs are put in place. That might be one
reason why there have been few research studies about educators’ awareness of student
enrollment changes. Paying attention to social institution, human beliefs, and humans’
limited information processing capacity, however, the institutional and behavioral
economics (IBE) theory leads us to attend to educators’ awareness and beliefs with
regard to school choice policies. Whether teachers know about student enrollment
changes is a question not to be assumed but to be examined. To address that, this chapter
addresses the following Research Question 1: How much are teachers and school
administrators aware of changes in enrollment? Do schools have incentive systems
aligned with decreases or increases in student enrollment?

With respect to factors affecting educators’ awareness of enrollment changes, the
neoclassical economics (NCE) perspective would posit that educators, as rational actors,
would know about student enrollment changes when the awareness of the changes and
doing something, if needed, affect their beneﬁts. For example, when there are rewards or

sanctions for student enrollment changes, educators would be aware of enrollment

43

changes. In addition, educators would be more likely to know about student enrollment
changes when they are affected by rewards or sanctions more seriously.

Both the NCE perspective and the IBE perspective agree that educators are more
likely to know about student enrollment changes when they are signiﬁcantly affected by
rewards or sanctions more seriously. However, the IBE perspective implies that educators
attach different meanings to the rewards or sanctions depending on their values or belief
systems. As a result, the same rewards or sanctions could make some educators more
sensitive than others to enrollment changes.

Based on the NCE and IBE perspectives, this study generates the following two
research hypotheses and tests them.

Hypothesis 1a (NCE): When there are rewards or sanctions for student enrollment
changes, educators would be aware of enrollment changes. In addition, the degree of
educators’ awareness of student enrollment changes (or the percentage of choice students)
at their schools would be positively associated with the degree of the impact of student
enrollment changes on them.

Hypothesis 1b (IBE): Educators attach different meanings to the rewards and
sanctions depending on their values or belief systems. Thus, the impact of the rewards
and sanctions would be different among educators in a school. As a consequence,the
same rewards or sanctions could make some educators more sensitive than others to
enrollment changes.

In the following sections, I ﬁrst address whether educators knew about student
enrollment changes from 2004 to 2008 and whether they knew about the percentage of

choice students in their schools. Second, I analyze how educators’ job security, salaries,

44

and beneﬁts were affected by student enrollment changes. Finally, I examine the
associations between rewards or sanctions and teachers’ awareness of enrollment changes.
Educators’ Awareness of Enrollment Changes and the Percentage of Choice
Students

To examine whether administrators and teachers knew about student enrollment
changes from 2004 to 2008 at their schools, they were asked to mark one among the
following choices: no change (in enrollment), less than 10% increase, 10% to 20%
increase, 20% to 3 0% increase, more than 3 0% increase, less than I 0% decrease, 10%
to 20% decrease, 20% to 3 0% decrease, more than 3 0% decrease, and don ’t know.

In addition, to explore whether administrators and teachers correctly knew about
the percentage of choice students in 2008-09 at their schools, they were asked to mark
one among the following choices: none, less than 10%, 10% to 20%, 20% to 30%, more
than 3 0%, and don ’t know.

To check whether administrators and teachers have accurate information, I
rescored their answers compared to the actual changes at their schools. Correct answers
were coded as 1, otherwise 0. Those who marked do not know were coded as 0. To
provide some margin for errors, I considered choices within +/- 5% as correct. For
example, if student enrollment increased by 7% during the past ﬁve years, both less than
10% increase and 10 to 20% increase were treated as correct answers.

As shown in Table 4-1, many educators did not know about recent student
enrollment changes. About 73% of the teachers and 45% of the administrators did not

provided correct answers about student enrollment changes at their schools.

45

Administrators were more likely than teachers to know about student enrollment changes.
Note that the sample size for the administrator survey is small (N=31).

Table 4-1: Educators’ Awareness of Student Enrollment Change

 

 

Wrong Correct Total
Teachers (N=315) 72.7% 27.3% 100.0%
Administrators (N=3 1) 45.2% 54.8% 100.0%

 

In addition, educators generally did not know correctly about the percentage of
choice students at their schools. As indicated in Table 4-2, about 56% of the teachers and
26% of the administrators did not have correct perceptions of the percentage of choice
students at their schools. Similar to student enrollment changes, administrators were more
likely to know about the percentage of choice students in their schools. These differences
between administrators and teachers seem to result from their different positions. Since
one of the main jobs administrators are responsible for is school management, they would
be in a better position to know about enrollment changes. Even though higher
percentages of administrators than teachers knew about both student enrollment changes
and the percentage of choice students, the proportion of administrators who did not
provide accurate information about those changes appears to be rather high, considering
their supposed role. The ﬁndings about educators’ awareness imply that changes in
student enrollment due to school choice policies fail to signal to educators that they
should respond to decreasing or increasing student enrollment.

Table 4-2: Awareness of the Percentage of Choice Students

 

 

Wrong Correct Total
Teachers (N=313) 56.2% 43.8% 100.0%
Administrators (N=31) 25.8% 74.2% 100.0%

 

46

Differences between Subgroups

In order to check whether educators’ awareness of enrollment changes and the
percentage of choice students at their schools were statistically different among
subgroups related to position (administrator or teacher), gender, tenure, the gaining or
losing status of their school districts, or school level, I conducted two-level hierarchical
generalized linear models. Two outcome variables were included in the analyses. One
was whether educators provided accurate information about student enrollment changes
at their schools from 2004 to 2008. The other was whether educators provided accurate
information about the percentage of choice students at their schools. Both of them are
binary variables. Each regression had one outcome variable and a common set of
explanatory variables. Regarding explanatory variables, the following variables at the
school level were included: the gaining or losing status of school districts, middle school,
and high school. Position, tenure and gender at the individual level were included. The
models are as follows.
Level-1 Model

Prob(Y=l |B) = P
log[P/(1-P)] = BO + B1*(ADMINIST) + B2*(MALE) + B3*(NONTENUR)

Y1 = Awareness of Student Enrollment Change over the Past Five Years ~
Y2 = Awareness of Percentage of Choice Students

Level-2 Model

BO = GOO + G01*(LOSING) + G02*(MIDDLE) + G03*(HIGH) + no

B1=G10+Ul
B2=G20+U2
B3 =G30+U3

As shown in Table 4-3, there was no statistically signiﬁcant difference between

gaining and losing school districts with regard to educators’ awareness of student

47

enrollment changes. Educators at middle schools were more likely than those at
elementary schools to know about student enrollment changes at their schools. No
signiﬁcant difference was found between elementary and high school educators.

Educators at high schools were less likely than those at middle schools to know about

student enrollment changes at their schools.3 At the individual level, administrators were
more likely than teachers to know about enrollment changes at their schools. Male
educators were more likely than female educators to know about them. There was no
signiﬁcant difference between tenured and non-tenured educators.

Table 4-3: Differences between Subgroups in Awareness of Student Enrollment Changes

 

 

Coefﬁcient Odds Ratio S.E. P.Value

Intercept, BO -1 239*" 0.248 0.345 0.001
Level 2

Losing Districts 0.103 1.109 0.301 0.735
Middle School 0.732M 2.079 0.352 0.047
High School 0046 0.955 0.393 0.909
Level 1

Administrators 0.903 * 2.466 0.510 0.087
Male 0679* 1.972 0.347 0.060
Non-tenured -0.353 0.703 0.43 1 0.420
N 329

 

Note. *** P5001; ** P5005; * P5010

As shown in Table 4-4, there is signiﬁcant unexplained variance in the intercept
for the three level-2 predictors. The slope parameters for administrator (vs. teacher), male
(vs. female), and non-tenured educator (vs. tenured educator) do not have any error

variances signiﬁcantly different from zero.

 

To examine the statistical differences between middle and high school educators, another analysis was

conducted in which middle school was set as a reference group. The resulting coefﬁcient for high school in
the analysis was -0.777 with the P-value of 0.049.

48

Table 4-4: Variance Components in the Differences between Subgroups in Awareness of
Student Enrollment Changes

 

S.D. Variance Component df Chi-square P-value
INTRCPTl, U0 0658*" 0.433 8 24.030 0.003
ADMINIST slope, U1 1.280 1.637 1 1 14.433 ' 0.209
MALE slope, U2 0.784 0.615 11 16.026 0.140
NONTENUR slope, U3 0.772 0.596 11 12.289 0.342

 

Note. **"‘ P5001; ** P5005; * P5010

As Table 45 indicated, no signiﬁcant differences were found between gaining

and losing school districts, between elementary and middle schools, between elementary

and high schools, or between middle and high schools4 with regard to educators’
awareness of the percentage of choice students at their schools. There was a signiﬁcant
difference between administrators and teachers. Administrators were more likely than
teachers to know about the percentage of choice students at their schools. I did not ﬁnd
signiﬁcant differences between female and male educators and between tenured and non-
tenured educators.

Table 4-5: Differences between Subgroups in Awareness of the Percentage of Choice
Students

 

 

Coefﬁcient Odds Ratio S.E. P.Value

Intercept, BO 0.028 1.029 0.343 0.934
Level 2

Losing Districts -0.350 0.705 0.340 0.314
Middle School -0242 0.785 0.394 0.544
High School -0.096 0.909 0.430 0.826
Level 1

Administrators 1.145" 3.141 0.523 0.037
Male 0.428 1.535 0.296 0.159
Non-tenured -0.198 0.820 0.340 0.564
N 327

 

Note. *""" P5001; ** P5005; * P5010

 

4 . . . . . .
To examine the statistical differences between middle and high school educators, another analysrs was

conducted in which middle school was set as a reference group. The resulting coefﬁcient for high school in
the analysis was 0146 with the P-value of 0.753.

49

As shown in Table 4-6, there is signiﬁcant unexplained variance in the intercept
for the three level-2 predictors. The slope parameters for administrator (vs. teacher), male
(vs. female), and non-tenured educator (vs. tenured educator) do not have any error
variances signiﬁcantly different from zero.

Table 4-6: Variance Components in the Differences between Subgroups in Awareness of
the Percentage of Choice Students

 

S.D. Variance Component df Chi-square P-value
INTRCPT 1, U0 0.635 0.404 8 15.451 0.050
ADMINIST slope, U1 1.152 1.327 1 1 12.841 0.303
MALE slope, U2 0.178 0.032 11 9.873 >.500
NONTENUR slope, U3 0.453 0.206 1 1 11.496 0.403

 

Educators’ Job Security, Salaries, and Benefits

In order for educators to respond to student enrollment changes due to school
choice programs, educators’ job security, salaries, and beneﬁts would need to be affected
by student enrollment changes. To examine this, administrators and teachers were asked
to indicate the degree to which changes in student enrollment affected them in terms of
job security, salary, and beneﬁts respectively. They could indicate one of the following
ﬁve choices: 1 =not at all, 2=to some extent, 3=to a moderate extent, 4 =to a great extent,
and 5 =not sure. For the analyses, the responses with 5 =not sure were excluded. The
results suggest that public school choice policies did not prompt school districts to create
market incentives for educators.

Teachers

In terms of job security, salaries, and beneﬁts, more than two-thirds of teachers
responded that they were not affected by changes in student enrollment at all or they were
affected only to some extent (see Table 4-7). In addition, teachers appeared to be more

affected in terms of job security than salary or beneﬁts by student enrollment changes. It

50

is also notable that teachers in a school were affected differently by student enrollment

changes. For example, in terms of job security, at school 1315, while 62% of teachers
responded that they were not affected by student enrollment at all, 23% of teachers
responded that they were affected to a great extent. With regard to salary, at school 1 11,
44.4%, 22.2%, 11.1%, and 22.2% of the teachers responded that they were affected by
student enrollment changes not at all, to some extent, to a moderate extent, and to a great
extent respectively. Regarding beneﬁts, at school 132, 30.8%, 15.4%, 30.8%, and 23.1%
of the teachers responded that they were affected by student enrollment changes not at all,
to some extent, to a moderate extent, and to a great extent respectively (see Tables A6-1,
A6-2, and A6-3 in Appendix F).

Table 4-7: Impact of Student Enrollment Changes on Teachers (%)

 

 

1 Not at all 2 Some extent 3 Moderate extent 4 Great extent Total
Job security (n=305) 50.2 21.0 10.8 18.0 100.0
Salary (n=295) 44.7 21.0 1 1.5 22.7 100.0
Beneﬁts 9:298) 40.9 22.1 14.4 22.5 100.0

 

Administrators

More than 60% of administrators responded that changes in student enrollment
affected their job security, salaries, and beneﬁts either not at all or only to some extent
(see Table 4-8). Administrators appeared to be more affected in terms of salaries and
benefits than job security.

Table 4-8: Impact of Student Enrollment Changes on Administrators (%)
1 Not at all 2 Some extent 3 Moderate extent 4 Great extent Total

 

 

Job security (n=32) 31.3 31.3 18.8 18.8 100.0
Salary (n=3l) 41.9 29.0 16.1 12.9 100.0
Beneﬁts (n=3l) 35.5 32.3 19.4 12.9 100.0

 

 

For conﬁdentiality, a series of numbers were assigned to schools in the surveys

51

Diﬂerences between Subgroups

To examine whether educators’ perceptions of job security, salaries, and beneﬁts
were statistically different among subgroups related to position (administrator or teacher),
gender, tenure, the gaining or losing status of their school districts, or school level, I
conducted two-level hierarchical generalized linear models. Three outcome variables
were included in the analyses: Job Security, Salary, and Beneﬁts. The outcome variables
were in ordinal scale. Each regression had one outcome variable and a common set of
explanatory variables. Regarding explanatory variables, the following variables at the
school level were included: the gaining or losing status of school districts, middle school,
and high school. Position, tenure and gender at the individual level were included. The
models are as follows.
Level-1 Model

Prob[R = 11B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(1)= 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

Y1 = Job Security

Y2 = Salary

Y3 = Beneﬁts

log[P'(1)/(1 - P'(l))] = B0 + Bl *(ADMINIST) + B2*(MALE) +

B3*(NONTENUR)

log[P'(2)/(l - P'(2))] = B0 + Bl *(ADMINIST) + B2*(MALE) +

B3*(NONTENUR) + d(2)

log[P'(3)/(1 - P'(3))] = BO + Bl*(ADMINIST) + BZ*(MALE) +

B3"‘(NONTENUR) + d(3)

Level-2 Model

52

B0 = G00 + G01*(LOSING) + G02*(MIDDLE) + G03*(HIGH) + U0

B1=G10+Ul
BZ=GZO+U2
B3 =G30+U3

In terms of educators’ perceptions of effects on job security, educators in losing
school districts were more likely than those in gaining school districts to be affected by
student enrollment changes (see Table 4-9). There were no signiﬁcant differences
between elementary and middle schools and between elementary and high schools. At the
individual level, administrators were more likely than teachers to be affected. No
signiﬁcant differences were observed between female and male educators. Non-tenured
educators were more likely than tenured educators to be affected.

Table 4-9: Differences between Subgroups in Educators’ Perceptions of Job Security

 

Coefﬁcient Odd Ratio S.E. P.Value

Level 2

Losing Districts 0900* 0.407 0.502 0.084
Middle School 0.307 0.736 0.593 0.609
High School 0.097 0.908 0.629 0.880
Level I

Administrators 1.201 *** 0.301 0.381 0.004
Male 0.157 0.855 0.298 0.601
Non-tenured 1.980*** 0.138 0.379 0.000
N

 

Note. *** P5001; ** P5005; * P5010. Each multilevel model has an unreported
intercept as well as unreported error variances for the intercept and the slope(s).

With regard to educators’ perceptions of effects on salaries, no signiﬁcant
differences were observed among subgroups of interest in this study (see Table 4-10). In
particular, no signiﬁcant differences were observed between administrators and teachers.

This might be related to similar schedules across different teacher union contracts.

53

Table 4-10: Differences between Subgroups in Educators’ Perceptions of Salary

 

 

Coefﬁcient Odd Ratio S.E. P.Value

Level 2

Losing Districts 0.696 0.499 0.464 0.145
Middle School 0.259 0.772 0.532 0.630
High School 0.101 0.904 0.589 0.866
Level 1

Administrators -0. 107 1.1 13 0.389 0.785
Male 0.277 0.758 0.281 0.332
Non-tenured -0.624 1.866 0.378 0.109
N

 

Note. *** P5001; ** P5005; * P5010. Each multilevel model has an unreported
intercept as well as unreported error variances for the intercept and the slope(s).

Regarding educators’ perceptions of beneﬁts, there were no signiﬁcant
differences between gaining and losing school districts, between elementary and high
schools, between administrators and teachers, or between female and male educators.
However, statistically signiﬁcant differences were observed between elementary and
middle schools and between tenured and non-tenured educators. Educators at middle
school were more likely than those at elementary schools to be affected. Non-tenured
educators were less likely than tenured educators to be affected (see Table 4-1 1).

Table 4-11: Differences between Subgroups in Educators’ Perceptions of Beneﬁts

 

 

Coefﬁcient Odd Ratio S.E. P.Value

Level 2

Losing Districts 0.608 0.545 0.386 0.127
Middle School 0.781 * 0.458 0.458 0.100
High School 0.074 0.928 0.492 0.881
Level 1

Administrators -0.388 1.475 0.431 0.375
Male 0.284 0.753 0.286 0.329
Non-tenured -0.875 ** 2.399 0.413 0.043
N

 

Note. *** P5001; ** P5005; * P5010. Each multilevel model has an unreported
intercept as well as unreported error variances for the intercept and the slope(s).

54

Association between Educators ’ Perceptions of Job Security, Salary and Beneﬁts and
their Awareness of Student Enrollment Changes and the Percentage of Choice Students.

Two-level hierarchical generalized liner models (HGLM) were conducted to
explore whether the impact of student enrollment changes on administrators and teachers
in terms of job security, salary, and beneﬁts was associated with educators’ awareness of
a) student enrollment changes or b) the percentage of choice students in their schools. To
this end, two outcome variables were included in the analyses. One outcome variable was
whether teachers knew about student enrollment changes at their schools; another was
whether teachers knew about the percentage of choice students at their schools. These
were binary variables (Y=l if a teacher knew correctly about enrollment changes or the
percentage of choice students at his/her school, Y=0 if not). Maj or explanatory variables
were the degree of teachers’ perceptions of the impact of student enrollment changes on
their job security, salary, and beneﬁts, respectively.

There were two models for each outcome variable. Model 1 for each outcome
variable examined the relationship between the outcome variable and the explanatory
variable with only two control variables added to the model. These two variables were
the gaining or losing status of school districts at the school level and position at the
individual level. The following additional variables at the school level were added in
Model 2 for each outcome variable: student enrollment changes over the past ﬁve years,
the percentage of choice students, student/teacher ratio, school size, local funding, the
percentage of students eligible for free and reduced lunch, and the percentage of minority

students. These additional variables in Model 2 allow us to examine the impact of

55

competition among schools and socioeconomic characteristics of schools on educators’
awareness of student enrollment changes and the percentage of choice students.
Level-1 Model

Prob(Y=l |B) = P
log[P/(l-P)] = B0 + Bl*(ADMINIST) + BZ*(D5)

Y1 = Awareness of Student Enrollment Change over the Past Five Years
Y2 = Awareness of Percentage of Choice Students

D5-1 = Job Security
D5-2 = Salary
D5-3 = Beneﬁts

Level-2 Model

Model 1

BO = G00 + G01 *(LOSING) + U0
Bl = G10 + U1

BZ = 620 + U2

Model 2
B0 = G00 + G01*(LOSING) + G02*(ENROLCHG) + G03*(CHOICE) +
G04*(TSRATIO) + G05*(SCHSIZE) + G06*(LOFUND) + G07*(FRL) +

G08*(MINORITY) + U0
El = 010 + U1
BZ = G20 + U2

Noticeably, as indicated in Table 4- 12, the degree of impact of student enrollment
changes on educators with regard to job security, salary, and beneﬁts was not
signiﬁcantly associated with their awareness of student enrollment changes. In addition,
the gaining or losing status of school districts, the degree of student enrollment changes,
and the percentage of choice students in a school did not seem to make educators aware
of student enrollment changes. There was no signiﬁcant association between educators’
awareness of student enrollment changes and the factors. Finally, socioeconomic

characteristics of schools (such as the amount of local funding, the percentage of students

56

eligible for free and reduced lunch, and the percentage of minority students) were not
associated with educators’ awareness of student enrollment changes.

As indicated in Table 4-13, the degree of impact of student enrollment changes on
educators with regard to job security, salary, and beneﬁts was not signiﬁcantly
associated with their awareness of the percentage of choice students in their schools. In
addition, the gaining or losing status of school districts, the degree of student enrollment
changes, and the percentage of choice students in a school did not seem to make teachers
aware of the percentage of choice students. There was no signiﬁcant association between
educators ’ awareness of the percentage of choice students in their schools and the factors.
Finally, socioeconomic characteristics of schools (such as the amount of local funding,
the percentage of students eligible for free and reduced lunch, and the percentage of
minority students) were not associated with educators’ awareness of the percentage of

choice students at their schools.

57

.328? 2: one 385:: 05
coo $281? 8.5 @3583: mm :95 we 50885 wotoooas no mm: .085 65:22: scam .2.on * ”3.on .1. ”5.on .13.. .632.

 

 

 

wmm nNm mmm mmm mmm mmm Z
w _ o.o mood foo wooo w _ o.o o_o.o $535 3:62;: 00 .x.
o_o.o Nooo- o_o.o vooo- o_o.o ooo.o :83 863.com Ea 8E do o\e
mmoo mmoo- mmoo $9? 366 Red. wEE—sm Woo-H
omoo omoo omoo mmoo omoo mvoo cum _oonom
owoo oooo- owoo o_o.o- owoo vmod comm conoeogcooam
wmoo mmoo omoo wmoo omoo _Noo ﬂavoam 86:0 oo o\..
m o co 18.? m _oo mmoo- v _ o.o o _ o.o- mowﬁﬁu 2:08:83. o0 .x.
homo _woo oomo oooo homo mmoo wwvo omoo End 2 _.o- wwvo mood- $2.5me maﬁa-—
ammo :Nmm._ Emo 1&3 4 o _ mo in _ N; Emo INN _ ._ movo 1&8.— Vnmo aroma.— EoﬁmemE—é
om _ .o mm _ .o- 2 _.o wmoo- mowoeom
m~_.o $5.? 2 _.o _~__.o- cCan—em
m: .o 33.. X _ .o 38. €83. p2

m.m E80 m.m E80 m.m $80 mm E80 mom ucooo m.m E80

N .252 _ .082 N 682 _ _oooE N _oooZ _ 682

 

mowamzu Boa—_oam mo $053344 :05 one muuocom EB saw—mm .braoom n2. .mcoowosom 2853 mcooﬂooaxx UNIV 2an

58

.325? o5 98 30885 05
com 82393 echo cocoa—2:: mm :03 mm 30885 @2583: an mm: 688 .3222: comm .o _ .owm * ”moowm .1. :oowm .13.. .262

 

 

 

mm». 9% _mm mm 8m .2 z
486 Rod- 23. Rod- _Nod 83. accuse. .9852 co .x.
m 55 53 ~23 88 :8 Sod- :23 e883. 25 8; .8 as
awed _ :3. £3 :8. 83 m 8d- Masai .83
Eod ~35 ES 9.3 88 Sod new 8an
i _.o :36. 2: .o Rod- 2: .o %od. 25. 5:085:85
was mad 3:. 23 Rod 28 $525 866 co as
a Sc 23 o :3 23 m :5 god 855 205.65 .0 .x.
owed Rod m _ no _ 3o- :2 a; 53 $3- 33 ~86 memo 83- 3055 was:
89o tow: $2 _..-.82 $2 8: 83 _..-82 $2 :2; was _..... _ N3 2982553
mm _ .o Sod mm; 39o- 38:5
83 Rod _ _ 3 £3. 33
em; 53. : _.o Sod- bessm .5
mm £89 mm E80 mm e08 mm e08 mm E80 mm .88
N _082 _ 382 N 382 _ .082 N .082 _ .082
353%

8:50 we omega—8.5m 2: oo wmocoBB< :2: ES mEocom one spa—mm 550% 22. .Eoﬁosom c8382 accommoommx‘ ”2% 2an

59

Discussion

First, it is notable that many educators did not know about recent student
enrollment changes or the portion of choice students at their schools. More than 72% of
the teachers and more than 45% of the administrators did not know about student
enrollment changes in their schools. Regarding educators’ awareness of the percentage of
choice students, about 56% of the teachers and 26% of the administrators did not provide
accurate information about these percentages. Even though administrators were more
likely than teachers to know about student enrollment changes and the percentage of
students at their schools, we could say that a large portion of administrators still did not
know about them considering their position where they were in charge of counting the
number of students and maintaining school budget. It is important for educators to know
about student enrollment changes in that their awareness of the changes could be a signal
that they should take appropriate actions to address them.

Second, public schools seem to fail to build market incentives for educators. In
terms of job security, salaries, and beneﬁts, more than two-thirds of teachers and
administrators responded that they were not affected by changes in student enrollment at
all or they were affected only to some extent. This implies that the current incentive
systems for educators may have limitations in motivating educators to be aware about
student enrollment changes and to improve school quality in order to attract more
students.

Third, as the IBE hypothesis expected, in a school where it can be assumed that
the same incentive systems are applied to teachers in the school, teachers perceived the

degree of effects of student enrollment changes on their job security, salaries, and

60

beneﬁts differently. This can be explained from the IBE perspective, which takes into
account emotions and beliefs. Educators’ responses to job security, salaries, and beneﬁts
are mediated by emotions like pleasure or pain. However, the key issue is here whether
the effects were large enough for an emotional stimulus to motivate educators to take
action. In addition, emotions are triggered by beliefs, which provide a meaning and sense
of direction to educators. In a given policy context, it is important that schools build
incentive systems that impact teachers’ job security, salaries, and beneﬁts sufﬁciently to
motivate educators to take action in order to attract more students. In addition, attending
to educators’ beliefs which provide a meaning and sense of direction to them,
policymakers need to inform educators that student enrollment changes are inﬂuenced by
educators’ practices.

Finally, the ﬁnding does not support either the NCE or the IBE perspective about
the relationship between the degree of impact of student enrollment changes on teachers’
job security, salary, and beneﬁts and the degree of their awareness of enrollment changes
and the percentage of choice students in their schools. The degree of impact of student
enrollment changes on teachers with regard to job security, salary, and beneﬁts was not
signiﬁcantly associated with their awareness of student enrollment changes and the
percentage of choice students in their schools. In addition, the degree of competition
among schools was not associated with educators’ awareness of student enrollment
changes and the percentage of choice students. The gaining or losing status of school
districts, the degree of student enrollment changes, and the percentage of choice students

in a school, which are indicators of the degree of competition among schools, did not

61

seem to make teachers aware of student enrollment changes and of the percentage of

choice students at their schools.

62

CHAPTER 5

TEACHERS’ WORKING CONDITIONS

This chapter addresses the second guiding research question: How are
schools/districts that lose students as part of school choice policies (i. e., losing
schools/districts) different from schools/districts that gain students (i. e., gaining
schools/districts) in terms of teachers ’ working conditions such as available time and
resources for teachers, leadership, and opportunities for professional development? Are
teachers’ working conditions associated with the intensiveness of school competition?
Many research studies have shown that teachers are the most important factor for student
achievement. Therefore, it is important for schools to maintain good working conditions
that will help in recruiting, retaining, and motivating quality teachers(Berry, Smylie, &
Fuller, 2008). If school choice programs can contribute to providing better teacher
working conditions, that could be one justiﬁcation of introducing such programs.
Regarding this issue, the neoclassical economics (NCE) perspective and the institutional
and behavioral economics (IBE) perspective have different expectations.

From the NCE perspective, school administrators would be expected to make an
effort to provide better working conditions in order to attract more students to their
schools if attracting more students in their schools produces proﬁts for them. In addition,
it is assumed that school administrators know the relationship between teachers’ working
conditions and their performance or commitment to schools and how to provide better
working conditions. Parents also would be expected to choose schools which provide

better working conditions, because teachers at those schools would serve their children

63

better. Similar to school administrators, it is assumed that parents know the differences
among schools with regard to teachers’ working conditions. Finally, as economic actors,
educators would balance their effort with resulting proﬁts. This implies that as
competition among schools becomes severe, educators would be more likely to make an
effort to provide better working conditions, if other conditions are equal.

In contrast to the NCE perspective, the IBE perspective questions whether
educators know about the relationship between teachers’ working conditions and their
performance and how to provide better working conditions. In addition, they question
whether educators can calculate the utility of certain measures for providing better
working conditions. This implies that educators might overestimate or underestimate the
effects of improving teachers’ working conditions. When they underestimate such effects,
they might not pay attention to teachers’ working conditions. For parents, the IBE
perspective does not assume that parents have the capacity to appreciate the differences
among schools in terms of teachers working conditions. Teachers’ working conditions
could be categorized as high information goods.

The two competing perspectives lead to different hypotheses as follows with
regard to the second research question.

Hypothesis 2a (NCE): Schools in gaining school districts would have better
working conditions because rational parents would choose schools with better working
conditions and educators would make an effort to provide better working conditions. The
degree of the differences in working conditions between gaining and losing schools

would be associated with the degree of school competition (i.e., the degree of student

64

enrollment change). As competition among schools becomes more intensive, schools will
make more effort to improve working conditions.

Hypothesis 2b (IBE): There might be no signiﬁcant difference in teacher working
conditions between schools in gaining and losing school districts when educators do not
know the importance of teachers’ working conditions or when they underestimate the
effects of improving teachers’ working conditions. In addition, parents might not
consider teachers’ working conditions when they choose schools for their children. In this
case, parental choice would not lead educators to provide better working conditions.

Before I test the two competing hypotheses, I examine the differences between
gaining and losing school districts with regard to teachers’ working conditions. Then, I
use two-level hierarchical linear models to examine whether the degree of school
competition is associated with teachers’ working conditions.

Educators’ Perceptions of Working Conditions

Teachers’ Working Conditions were measured with the following six constructs:
Use of Time, Facilities and Resources, Empowerment, Leadership, Collegial Support,
and Professional Development. Each construct had four to eight sub items. Teachers and
administrators were asked to answer the extent to which they agree with statements about
teachers’ working conditions. Each item had a ﬁve-point scale: 1=Strongly agree,
2=Disagree, 3=Neither disagree nor agree, 4=Agree, and 5=Strongly agree. Higher values
are interpreted as better teachers’ working conditions. For the analyses, sub items in each
construct were composited into one mean score.

Teachers

65

As shown in Table 5-1, in all areas of teacher working conditions, teachers in
gaining school districts rated their working conditions higher than those in losing school
districts. This ﬁnding implies that schools in the gaining districts have better teacher
working conditions. Differences in teacher empowerment, leadership, and facilities and
resources were relatively larger than differences regarding other working conditions.

Table 5-1: Teachers' Perceptions of Working Conditions

 

 

Both Gaining Losing Diff.
Use ofTime 2.87 2.98 2.78 0.20
Facilities and Resources 3.93 4.28 3.62 0.65
Empowerment 3.39 3.82 3.03 0.79
Leadership 3.38 3.77 3.04 0.73
Collegial Support 3.81 3.99 3.65 0.34
Professional Development 3.1 5 3.24 3.08 0.1 7

 

Note. l=strongly disagree, 2=disagree, 3=neither disagree nor agree, 4=agree, and
5=strongly agree. Clustering of educators within schools was ignored.

Administrators

Administrators had different perceptions of teachers’ working conditions as
compared to teachers (see Table 5-2). While teachers in gaining school districts valued
their working conditions more favorably in all areas than those in losing school districts,
administrators in losing school districts perceived teachers’ working conditions more
favorably than those in gaining school districts with respect to the following two items:
use of time and professional development.

Table 5-2: Administrators’ Perceptions of Working Conditions

 

 

Both Gaining Losing Diff.
Use of Time 3.73 3.65 3.81 -0. 17
Facilities and Resources 4.44 4.55 4.31 0.24
Empowerment 4.21 4.39 4.00 0.39
Leadership 4.24 4.43 4.03 0.40
Collegial Support 3.95 4.12 3 .76 0.36
Professional Development 3 .96 3 .91 4.02 -0. 10

 

Note. l=strongly disagree, 2=disagree, 3=neither disagree nor agree, 4=agree, and
5=strongly agree. Clustering of educators within schools was ignored.

66

It is also interesting is that there was a big gap between teachers’ and
administrators’ perceptions of teacher working conditions. As shown in the Table 5-3,
administrators in both gaining and losing districts perceived teachers’ working conditions
more favorably than teachers. Most administrators tended to think that their schools had
better working conditions than teachers themselves.

Table 5-3: Differences between Teachers’ and Administrators’ Perceptions of Teacher
Working Conditions

 

 

Gaining (A) Losing (B)

Teachers Admi. Diff. Teachers Admi. Diff.
Use of Time 2.98 3.65 -0.67 2.78 3.81 -l .03
Facilities and Resources 4.28 4.55 -0.27 3.62 4.31 -0.69
Empowerment 3.82 4.39 -0.57 3.03 4.00 -0.97
Leadership 3.77 4.43 -0.65 3.04 4.03 -0.99
Collegial Support 3.99 4.12 -013 3.65 3.76 -0.1 1
Professional Development 3.24 3.91 -0.67 3.08 4.02 -0.94

 

Note. 1=strongly disagree, 2=disagree, 3=neither disagree nor agree, 4=agree, and
5=strongly agree. Clustering of educators within schools was ignored.

Differences in Subgroups

To examine whether educators’ perceptions of working conditions were
signiﬁcantly different between subgroups related to position (administrator or teacher),
gender, tenure, the gaining or losing status of their school districts, or school level, I
conducted two-level hierarchical linear models. The outcome variables were the six key
constructs of teachers’ working conditions. The outcome variables were treated as
continuous variables. Regarding explanatory variables, the following variables at the
school level were included: the gaining or losing status of school districts, middle school,
and high school. In addition, Position, tenure and gender at the individual level were
included. The models are as follows.
Level-l Model

Y = B0 + Bl*(ADMINIST) + B2*(MALE) + B3*(NONTENUR) + R
Y1 = Use of Time

67

Y2 = Facilities and Resources
Y3 = Empowerment

Y4 = Leadership

Y5 = Collegial Support

Y6 = Professional Development

Level-2 Model

BO = GOO + G01*(LOSING) + G02*(MIDDLE) + GO3*(HIGH) + U0

Bl =GlO+Ul
B2=GZO+U2
B3=G30+U3

As indicated in Table 5-4, signiﬁcant differences were found between gaining and
losing school districts in terms of the following four key constructs: facilities and
resources, teacher empowerment, leadership, and collegial support. Educators in gaining
school districts rated these working conditions more favorably than those in losing school
districts. Second, while there were no signiﬁcant differences between elementary and
middle school educators with regard to the six key constructs, educators at high schools
rated teachers’ working conditions less favorably than those at elementary schools. Third,
at the individual level, signiﬁcant differences were observed between administrators and
teachers (with regard to each aspect of working conditions) except for collegial support.
In general, administrators rated teachers’ working conditions more favorably than
teachers. Fourth, there were no statistically signiﬁcant differences between female and
male teachers. Finally, non-tenured educators rated the following key constructs more
favorably than tenured educators: use of time, collegial support, and professional

development.

68

.Amooooﬁ 2: one 8888. 05
8o 828.8.» 8bo 3:308: mm :03 we E888. outage: Sm mm: .288 88.5.2: comm. .2.on .- ”modwn. _..... ..o.own. 12.. $82

 

vmm vmm Nmm Nmm omm vmm 2

o. .d 1&de nwod *2. . .o hood .Vm . .o mm . .o S . .o mm . .o m. ..o oo . .o :smwvd dodged-.82
wood whod- mwod wood- wood wnod- wood .vod mwod oo. .o- oood vmod o.3).
wm.d Icmmwd ...d wad mN.d :LPd om.d ***ooo.o om. .o unawwwd om . .o ***mood muo.mbm.:.Eo<
— .954

.o. .o gmd- on .d ***Vwm.o- mwmd mgd- whmd .ovd- oo.d .VNd- mo.d .ood- .oosow FEE
hid .mmd- mm .d mid- wmmd homd- mwmd mm .d- mm .d 25.? oCd oood- .oonom o.35).
mm .d otd- o. .d InnovMo- mmmd aaoood- o .md _..:Ebd- Nm.d _..:Nvmd- om .d oo.d- 39:35 9:84
N 88-—

 

.m..m .tooo .m.m E80 .m.m E80 .m..m deco .m.m deco .m.m E80
..o>oQ p.85. tooosm 39:00 9:80.89. 8080338.”. 88:08.». a. due... 05.... ..o 8..
283680 $5.83 no £8.88qu .mcoomosom. 3 Pawn: 5.3 31:20 8932. 8058....5 ”v-m 03m...

0338., «8.0880

 

69

Associations between Competition among Schools and Teachers’ Working
Conditions

To examine whether teachers’ working conditions were associated with the
degree of competition among schools, two-level hierarchical linear regressions were
employed. The following six composite scores regarding teachers’ working conditions
were used as outcome variables: use of time, facilities and resources, empowerment,
leadership, collegial support, and professional development. The degree of competition
was measured by school location (gaining district or losing district), student enrollment
changes over the past ﬁve years, and the percentage of choice students. In this study, it
was assumed that educators in losing school districts and at schools with decreasing
student enrollment and smaller percentages of choice students would feel competition to
a greater extent than their counterparts.

There were two models for each outcome variable. Model 1 for each outcome
variable examined the relationship between the outcome variable and the explanatory
variables with only one control variable included in the model. The explanatory variables
were school location (gaining district or losing district), student enrollment changes over
the past five years, and the percentage of choice students. As a common control variable
for both Model 1 and Model 2, the position variable (administrator or teacher) at, the
individual level was included. For Model 2, the following control variables at the school
level were added in Model 2 for each outcome variable: student/teacher ratio, school size,
local funding, the percentage of students eligible for free and reduced lunch, and the
percentage of minority students.

Level-1 Model

70

Y = BO + Bl*(ADMINIST) + R
Y1 = Use of Time
Y2 = Facilities and Resources
Y3 = Empowerment
Y4 = Leadership
Y5 = Collegial Support
Y6 = Professional Development

Level-2 Model

Model 1

B0 = GOO + G01*(LOSING) + G02*(ENROLCHG) + GO3*(CHOICE)

B1 = G10 + Ul

Model 2

B0 = GOO + G01*(LOSING) + 602*(ENROLCHG) + GO3*(CHOICE) +
G04*(TSRATIO) + G05*(SCHSIZE) + G06*(LOFUND) + GO7*(FRL) +
GO8*(MINORITY) + U0

B1 = 010 + U1

The results indicate that no aspects of teachers’ working conditions except for
facilities and resources were associated with the location of the school, student
enrollment changes, or the percentage of choice students. This implies that teachers’
working conditions are not associated with the intensiveness of competition among
schools. Rather, teachers’ working conditions seemed to be associated with other control
variables (see Table 5-5). First, regarding use of time, educators at schools with higher
percentage of free and reduced lunch eligible students tended to agree that they had
available time to meet the educational needs of all students, collaborate with their
colleagues, and so on. Second, in terms of facilities and resources, educators in losing
school districts were less likely than those at gaining school districts to agree that their
schools had sufﬁcient facilities and resources for education. In addition, educators at

schools with higher percentage of choice students were less likely to agree. While there

were positive associations between facilities and resources and the amount of local

71

funding and between facilities and resources and the percentage of students eligible for
free and reduced lunch, there was a negative association between facilities and resources
and the percentage of minority students.

Third, when controlling in Model 1 for associations between school competition
and teachers’ working conditions without considering schools’ socioeconomic factors,
negative associations between the outcome variables (empowerment and leadership) and
losing school district and between the outcome variables and the percentage of choice
students were observed. However, the negative associations disappeared when control
variables on school characteristics were also taken into account. Fourth, with respect to
collegial support and professional development, no signiﬁcant associations between the
outcome variables and explanatory variables were observed. However, there were
negative associations between collegial support and school size and between professional

development and the amount of local funding.

72

.Amo0o0_m 05 0:0 8008:: 05
8o 0000009» 800 000-8008: mm :03 mm 8020:: 0008008: :0 mm: _0008 08.22: nomm .o_ .owo ... ”modwo _..... :odwo 1.... .082

 

 

 

 

73

 

wwm wwm wwm wwm mwm mwm Z
wood Noo.o- wood mood- _ _od *NNod- broom—2 «0 o\e
mood _ood wood mood- ooo.o wood- AME o0 °\c
_ _od *oNod- o_o.o mood m _ o.o b so wEnEE 7000-—
wNod m_o.o- hNod ..:.:..Nwo.o- owod oNod- 0N5 _oosom
wmod owod- omod wmod- _mod mmod- 030M u0£000§0003m
m—od m_o.o- o_o.o mood- m_o.o ooo.o- o_o.o ooo.o o.o.o :od- N_ o.o _....Nmod- 300035 020:0 o0 o\o
ooo.o ooo.o mood _ood- ooo.o mood- wood ooo.o wood Nood ooo.o _ _od 0wac0 E0E=OEm o0 o\o
$2 32. :3 50¢- 82 $3- 20¢ 22. 82. $2. 33 .........0.% o 00.500 0580
mm: .o «aawmod mm _ .o :‘nhwbd wo _ .o E _.o 02 .o wood _N _ .o *aeNood MN _ .o tit-0d whoombmmﬁauxx
.m.m E000 .m.m “$000 .m.m “$000 .m.m E000 .m.m E000 .m.m E000

N _0002 — 0002 N .0002 _ _0002 N _0002 — _0002
E08020>0Q 802000-85 tooozm _Ew0=00 ﬂow—00004 0Eatw> E00083—

wwm wwm hwm owm mwm wwm Z
m _od o _ o.o- ooo.o **m~o.o- wood wood- hum-50:2 o0 o\0
ooo.o mood- mood *Oood wood _...L —o.o AME m0 ..\0
:od wood wood fun—o.o :od ooo.o- $2055 300-—
owod wNod- o _ o.o w _od- oNod wNod 0N5 _oonom
omod Nmod- oNod _Nod- omod Nood 050d 8503:00035
NNod NNod- N _od .3.“ _ mod- ooo.o *sNNod- ooo.o ***o_o.o- w_o.o mood- ooo.o ooo.o £00035 020:0 «0 o\o
ooo.o wood wood o_o.o wood Nood wood _ood wood wood wood ooo.o 098:0 “005:0.Em o0 o\c
_wmd 55.? thd ***o\-o.o- _de 3.050.? on _ .o .Iimwhd- hwmd _ _ _.o- _NNd w—od- $2.5me woo-.00:—
mm—o tin—o.o wm _ .o Iioooo _N _ .o “I"...mmwo w. _.o sagammwo Nw _ .o ***o_w.o mm _ .o ***wNw.o Eoumbmmcmeﬂuxx

.m.m E000 .m.m E000 .md H000 .m.w “$000 Maw 5000 .m.m .0000
N .0002 _ _0002 N _0002 _ .0002 N _0002 ~ $002
0000003008m 0008000d .80 00.5500; 05:- oo 00: 03093 80000000

 

$83980 @0283 .8283. 000 80:00:80 _oosom 0002500 mcooﬂoommxx ”Wm 050,—-

Discussion

Simple comparison of teachers’ working conditions indicated that schools in
gaining school districts had better working conditions than schools in losing school
districts. However, most differences between gaining and losing school districts (except
for facilities and resources) disappeared when school characteristics were controlled for.
Student enrollment changes from 2004-2005 to 2008-2009 and the percentage of choice
students, which are the indicators of the degree of competition among schools, were not
signiﬁcantly associated with teachers’ working conditions except for facilities and
resources. Educators in losing school districts were less likely than those in gaining
school districts to agree that their schools had good facilities and resources. In addition,
educators at schools with less choice students were less likely to agree that their schools
had good facilities and resources. These negative associations seemed to be related to the
location of the school districts and schools. Losing school districts and schools with less
choice students were usually located in less afﬂuent communities than their competing
school districts and schools (Ni & Arsen, 2008). In this respect, it is plausible to speculate
that school facilities and resources as part of teachers’ working conditions were related to
the location of school districts and schools and not simply to competition among schools.
As a result, it is difﬁcult to retain the NCE hypothesis that as competition among schools
becomes more intensive, schools will make more efforts to improve working conditions.
Instead, teachers’ working conditions appeared to be associated with teacher
characteristics (such as the status of tenure) and school environment (such as local

funding, student socioeconomic composition, and school level). In addition, the

74

likelihood that other latent factors not included in this study might affect teachers’
working conditions cannot be excluded.

It is notable that the rejection of the NCE hypothesis does not necessarily imply
that educators’ responses were not rational. This is because there was not clear evidence
that improving teachers’ working conditions could contribute to attracting more students
in their schools. In addition, as seen in Table 5-3, administrators might think that their
schools already have good working conditions for teachers, thus they might not see the
need to improve working conditions. As rational actors, educators might have chosen
other strategies to attract more students. This issue is beyond the scope of this study. The
clearer point here is that we cannot expect that school choice programs will necessarily
lead schools to provide better working conditions conducive to student achievement and
school improvement.

Regarding the IBE hypothesis, this study has limitations in conﬁrming that
administrators and teachers do not know the importance of working conditions or how to
improve them. However, a large gap between teachers’ and administrators’ perceptions
regarding teachers’ working conditions suggests the possibility that administrators might
not have clear knowledge about teachers’ working conditions.

The IBE perspective also suggests that even when administrators know about the
importance of improving teachers’ working conditions and how to improve them, they
might not be interested in doing so. When we look at the impact of student enrollment
changes on administrators’ and teachers’ job security, salaries, and beneﬁts, we cannot
exclude this possibility. As shown in Chapter 4, more than two—thirds of administrators

and teachers responded that changes in student enrollment affected their job security,

75

salaries, and beneﬁts either not at all or only to some extent. Student enrollment changes
due to school choice programs might fail to motivate administrators to do something to

attract more students.

76

CHAPTER 6

EFFORTS FOR ATTRACTING STUDENTS

This chapter focuses on Research Question 3: What are schools doing to attract
students? Is the degree of organizing activities for attracting more students associated
with school competition? Are the activities associated with school problems or parents’
school choice reasons? To answer these questions, 10 possible activities for attracting
students were listed on the surveys, and administrators and teachers were asked to rate
the extent to which their schools or districts had done each activity. This study also
explores whether the degree of organizing these school activities is commensurate with
the intensity of school competition measured by the degree of student enrollment changes
and the percentage of choice students at the school level. In addition, I explore whether
the degree of organizing school activities for attracting students is consistent with the
degree of school problems perceived by teachers. Finally, I examine school activities for
attracting students address what parents consider when they choose schools for their
children.

In regard to this research question, the neoclassical economics (NCE) perspective
and the institutional and behavioral economics (IBE) perspective would have different
expectations. First, the NCE perspective would expect schools under more severe
competition to make greater effort to attract students. Regarding school problems,
schools would keenly attend to their problems, if any, because parents, as rational actors,
would not choose schools with problems. In terms of the associations between the school

activities and the reasons for choosing schools, the NCE perspective would assume that

77

schools will try to identify what factors parents consider important when they choose
schools for their children. As a result, the NCE perspective would draw a conclusion that
school choice policies can contribute to addressing school problems and what parents
want.

However, the IBE perspective would question the central logic of the NCE
perspective. With regard to the associations between school problems and school
activities for attracting students, the IBE perspective does not assume that administrators
will know about their school problems. In addition, even though administrators perceive
the school problems, it cannot be assumed that they will know about appropriate
strategies to solve the problems. Finally, we cannot know whether or not administrators
try to address the problems even when they know the problems and ways to address them.
Similarly, in terms of the associations between school activities for attracting students
and parents’ reasons for choosing schools, the IBE perspective would not assume that
administrators know about the reasons for choosing schools and that they know ways to
address what parents consider when they choose schools. Lastly, the IBE perspective
would not assume that administrators will try to address parents’ needs even when they
are aware of parents’ needs and the ways to address them.

Based on the two competing perspectives, this study generates the following two
hypotheses.

Hypothesis 3a (N CE): The schools would examine who choice students/families
are and what types of schooling options they are seeking. As competition among schools
becomes severe, schools would make more efforts to attract students. In addition, as a

given school problem becomes more serious, schools would make more efforts to address

78

that problem. Finally, schools’ strategies for attracting students would be closely
associated with the reasons for parents’ school choice decisions.

Hypothesis 3b (IBE): There might be no difference between schools when schools
fail to analyze problems, do not have enough capacity to come up with appropriate
strategies, or just adopt other schools’ strategies. Schools might rely only on their
perceptions of the reasons for parents’ school choice decisions without doing systematic
research about parents’ actual decisions.

Activities for Attracting Students
Teachers

The degree of organizing various activities for attracting students under school
choice programs was measured by educators’ perceptions. Administrators and teachers
were asked to indicate the degree to which their schools were engaged in each of 10
possible school activities for attracting students. They could indicate one of the following
choices: I =not at all, 2 =t0 some extent, 3 =t0 a moderate extent, 4 =to a great extent, and
5 =not sure. Responses with not sure were excluded from the analyses.

The ﬁnding on teachers’ perceptions of schools’ and districts’ activities to
increase student enrollment revealed that their schools and districts did not make much
effort in order to increase student enrollment (see Table 6-1). Of 10 possible activities for
attracting students, teachers responded that their schools had done each activity to “some
extent” or to “a moderate extent.” Among the 10 possible activities, teachers responded
that schools and districts had made more efforts to improve student achievement,
improve school curriculum, maintain a safe environment, invest in school facilities and

resources, and improve relationships with parents than the others. There were some

79

differences between gaining and losing school districts with regard to the extent to which
they were engaged in these activities. It is notable that among the 10 possible activities,
both gaining and losing school districts made less effort to research who choice
students/families were or to counsel out low-performing teachers. Another noticeable
point is that gaining school districts made more efforts than losing school districts except
with regard to improving school curriculum and advertising their schools.

Table 6-1: Teachers’ Perceptions of Schools’ Activities for Increasing Student
Enrollment

 

 

Activities Both Gaining Losing Diff.
Improve student achievement 2.92 3.03 2.83 .20
Improve school curriculum 2.81 2.72 2.88 -. l 6
Maintain a safe envrronment (school order and 2.78 2.98 2.62 .36
drscrplrne)

Invest in school facilities and resources 2.69 2.76 2.63 .13
Improve relationships with parents 2.59 2.68 2.52 .16
Advertise school 2.44 2.23 2.60 -.37
Improve extra-curricular activities 2.27 2.33 2.22 .l l
Recruit more highly qualiﬁed teachers 2.26 2.41 2.13 .28

Research who choice students/families are and what

types of schooling options they are seeking 2'07 1'99 2'12 " l 3
Counsel out low-performing teachers 1.49 1.48 1 .50 -.01
Means 2.43 2.46 2.41 .06

 

Note. l=not at all, 2=some extent, 3=moderate extent, and 4=great extent. Clustering of
educators within schools was ignored.

Administrators

Administrators’ responses showed that they made more efforts to maintain a safe
environment, improve student achievement, improve school curriculum, and invest in
school facilities and resources than the other activities (see Table 6-2). It is noticeable
that schools and districts made less effort to counsel out low-performing teachers or to
research who choice students/families were than the other activities.
Teachers vs. Administrators

Compared with teachers, administrators seemed to think that schools and districts

had made more efforts to attract students than teachers did (see Table 63). Except for

80

researching who choice students/families were, administrators rated the degree of
organizing school activities for attracting more students more highly than teachers did.

Table 6-2: Administrators’ Perceptions of Schools’ Activities for Increasing Student
Enrollment i

 

Both Gainirg Losing Diff.
3.40 3.40 3.40 .00

 

Maintain a safe environment (school order and

discipline)

Improve student achievement 3.37 3.13 3.60 —.47
Improve school curriculum 3.30 3.27 3.33 -.07
Invest in school facilities and resources 3.03 3.07 3.00 .07
Improve relationships with parents 2.77 2.87 2.67 .20
Recruit more highly qualiﬁed teachers 2.62 2.80 2.43 .37
Improve extra-curricular activities 2.60 2.40 2.80 -.40
Advertise school 2.58 2.38 2.80 -.43
Counsel out low-performing teachers 2.00 2.43 1.57 .86

Research who chorce students/families are and what 1.85 1.38 2.29 _.90
types of schooling options they are seeking
Means 2.75 2.71 2.79 -.08

Note. l=not at all, 2=some extent, 3=moderate extent, and 4=great extent. Clustering of
educators within schools was ignored.

 

Table 6-3: Differences in Perceptions of Schools’ Activities for Increasing Student
Enrollment Between Teachers and Administrators

 

 

Teacher Adm. Diff.
Advertise school 2.44 2.58 -0. l4
Invest in school facilities and resources 2.69 3.03 -0.34
Recruit more highly qualiﬁed teachers 2.26 2.62 -0.36
Improve school cuniculum 2.81 3.30 -0.49
Counsel out low-performing teachers 1.49 2.00 -0.51
Improve relationships with parents 2.59 2.77 -O.18
Maintain a safe environment (school order and discipline) 2.78 3.40 -0.62
Improve student achievement 2.92 3.37 -0.44
Improve extra-curricular activities 2.27 2.60 033
Research who choice students/families are and what types 2 07 l 85 0 21
of schooling options they are seeking ' ' '
Means 2.43 2.75 -0.32

 

Note. l=not at all, 2=some extent, 3=moderate extent, and 4=great extent. Clustering of
educators within schools was ignored.

As Table 6-4 shows, for gaining school districts, there were relatively larger
differences between teachers and administrators with regard to the following items than
the others: counsel out low-performing teachers, research who choice students/families
are, and improve school curriculum. With respect to losing school districts, there were
larger differences between teachers and administrators with regard to the following items

81

than the others: maintain a safe environment, improve student achievement, and improve
extra-curricular activities.

Table 6-4: Differences in Perceptions of Schools’ Activities for Increasing Student
Enrollment Between Teacher and Administrators in Gaining and Losing School Districts

 

 

Gaining Losing
Teacher Adm. Diff. Teacher Adm. Diff.
Advertise school 2.23 2.38 -0.14 2.60 2.80 -0.20
Invest in school facilities and resources 2.76 3.07 -0.30 2.63 3.00 -0.37
Recruit more highly qualiﬁed teachers 2.41 2.80 -0.39 2.13 2.43 —0.29
Improve school curriculum 2.72 3.27 -0.55 2.88 3.33 -0.45
Counsel out low-performing teachers 1.48 2.43 -0.94 1.50 1.57 -0.08
Improve relationships with parents 2.68 2.87 -0.19 2.52 2.67 -0.15
Maintain a safe envrronment (school order and 2.98 3.40 _0'42 2. 62 3.40 -0.78
discrpline)
Improve student achievement 3.03 3.13 -0.10 2.83 3.60 -0.77
Improve extra-cunicular activities 2.33 2.40 -0.07 2.22 2.80 058
Research who choice students/families are and
what types of schooling options they are seeking 1'99 1'38 0'60 2'12 2'29 ’0' I 7
Means 2.46 2.71 -0.25 2.41 2.79 -0.38

 

Note. l=not at all, 2=some extent, 3=moderate extent, and 4=great extent. Clustering of
educators within schools was ignored.

Differences between Subgroups

To examine whether educators’ perceptions of school activities for attracting
students were signiﬁcantly different between subgroups related to position (administrator
or teacher), gender, tenure, the gaining or losing status of their school districts, or school
level, I conducted two-level hierarchical non-linear models. Outcome variables were the
10 possible school activities for attracting students listed in Table 6-5. The outcome
variables were ordinal. Regarding explanatory variables, the following variables at the
school level were included: the gaining or losing status of school districts, middle school,
and high school. Position, tenure and gender at the individual level were included. The
models are as follows.
Level-l Model

Prob[R =1|B] = P'(l) = P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)
Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

82

Prob[R <= 4|B] = 1.0

where

P(l) = Prob[Y(1)= 1|B]
P(2) = Prob[Y(2) = 1|B]
P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(l - P'(l))] = BO + Bl *(ADMINIST) + B2"‘(MALE) +
B3*(NONTENUR)

log[P'(2)/(l - P'(2))] = BO + 131 *(ADMINIST) + B2*(MALE) +
B3*(NONTENUR) + d(2)

log[P'(3)/(1 — P'(3))] = BO + B1*(ADMINIST) + B2*(MALE) +
B3*(NONTENUR) + d(3)

Level-2 Model

B0 = GOO + G01*(LOSING) + GOZ*(MIDDLE) + G03*(HIGH) + U0
B1 = 010 + U1
B2 = G20 + U2
B3 = G30 + U3

As shown in Table 6-5, no signiﬁcant differences between gaining and losing

school districts were observed with regard to the degree of organizing school activities

for attracting students except for maintaining safe environment. Losing school districts

were less likely than gaining school districts to make an effort to maintain a safe

environment. Second, middle schools were less likely than elementary schools to make

an effort to maintain a safe environment. There were no signiﬁcant differences between

elementary and middle schools with regard to the other activities. High schools were less

likely than elementary schools to make an effort to maintain a safe environment, improve

student achievement, or do research about choice students and families. No signiﬁcant

differences between elementary and high schools were found with regard to the other

activities.

Third, administrators were more likely than teachers to think that schools made

effort to recruit more highly qualiﬁed teachers, improve school curriculum, counsel out

83

low-performing teachers, maintain a safe environment, and improve student environment.
Administrators and teachers did not differ in the other activities. Fourth, male educators
were less likely than female educators to think that schools made effort to improve the
school curriculum. There were no signiﬁcant differences between female and male
educators with regard to their perceptions of the other activities. Finally, non-tenured
educators were more likely than tenured educators to think that schools made effort to
maintain a safe environment and improve extra-curricular activities. Tenured and non-

tenured educators did not differ in their perceptions of the other activities.

84

0.08.0 0... .000 2.00.0.0. 0...
8. 000:0..0> 8.8 00.8008: 00 :03 00 2.00.0.0. 00.8008: 00 00.. .000... .0>0....:E 80m. .o. .own. ... ”mod-vs. I :odwn. .12.. .082

 

 

 

Ewd wNNd .- . wd **o.-o.. nwmd mnmd mmmd ..:..mw..d ommd owmd 0088.002
owmd No . d moNd mwod woNd MNod moNd m0..o- RNd o. .d 0.0.).
wowd howd- momd wmod . .wd «so .od ond 3.3.0.... Nwmd Nmmd 8.0..0....E.0<
.wmd tom...- nNod .ood- Nowd _..:wmw. .- o.m.o 0mg..- wowd MNod- .008m aw...
wowd owmd- ohmd nde- mowd mood- wwwd robwd- mwwd ow . .o- .008m 0.0.0:).
wNwd womd wowd ww. .o- oomd NNod- wowd 00.0w...- ommd ode- 0.0.00.0 980-.
.m..m E000 .m..m E000 .m..m E000 .m.m E000 .m..m E000
0.0 02.80.080.030 00...>..00 0.00080 80:80.80 80:80.38 0.8.0..

00.08 055 8.00000.

0.00 08.9...

2.0.5.0 088....

0.00 0 0.0.0.0.).

5.3 00.88.08. 08.0....

 

 

 

 

wowd .ood- ommd mwmd Fond coo. mnmd onNd o.w.o omod 0088.002
.mmd NoNd NbNd smowd- moNd Nw..o- .de .w..o wde ode- 0.0.).
wwwd _..:wmw. wowd t..w.o.. .omd **me.o wowd .wwd wwwd MNNd .o.0..0.=.E.0<
.mmd wmod ohmd Sumo- .mwd owed- momd «Ed- mond mw..o- .008m aw...
w. md wmod wwmd w.o.o- .owd wwod- mwwd h. .d mohd wood .008m 0.00.2
owwd ooo.o- hoNd wo..o .wmd mwwd- womd mod. oNod mend 0.0.00.9 9:00...
.m.m E000 .m..m E000 .m..m E000 .m.m E000 .m..m E000
0.080%..000E00 05.00an 0.0800. .00....0so 0008000. .80
-30. .30 80.800 .0080 08.9... raw... 08:. .883. 00.....00. .0080 0. 00>... .0080 00.8>.0<

 

080.5% we..00...< 8.. 00...>..0< .0.008m ..0 08:80.00 0... s. 008.05% 803.00 0008.0...5 0.0 0.00...

85

Associations between School Competition and School Activities for Attracting Students.
In order to examine whether the degree of organizing school activities for
attracting students is associated with the degree of competition among schools, two-level

non-linear regressions were employed. The degree of competition was measured by
school location (gaining district or losing district), student enrollment changes over the
past five years, and the percentage of choice students. In this study, it was assumed that
educators in losing school districts and at schools with decreasing student enrollment and
smaller percentages of choice students would feel competition to a greater extent than
their counterparts.

There were two models for each outcome variable. Model 1 for each outcome
variable examined the relationship between the outcome variable and the explanatory
variables with only one control variable included in the model. The explanatory variables
were school location (gaining district or losing district), student enrollment changes over
the past five years, and the percentage of choice students. As a common control variable
for both Model 1 and Model 2, the position variable (administrator or teacher) at the
individual level was included. For Model 2, the following control variables at the school
level were added in Model 2 for each outcome variable: student/teacher ratio, school size,
local funding, the percentage of students eligible for free and reduced lunch, and the
percentage of minority students.

Level-1 Model

Prob[R = 1|B] = P'(l) = P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P( 1) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

86

P(1)= Prob[Y(1)= 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(l - P’(l))] = B0 + B1*(ADMINIST)

log[P'(2)/(1 - P'(2))] = BO + B1*(ADMINIST)+ d(2)

log[P'(3)/(1 - P'(3))] = BO + Bl*(ADMINIST) + d(3)
Level-2 Model

Model 1

B0 = GOO + G01 *(LOSIN G) + GOZ*(ENROLCHG) + G03*(CHOICE) + U0

B1 = G10 + U1

Model 2

BO = GOO + 601 *(LOSING) + 602*(ENROLCHG) + G03*(CHOICE) +
GO4*(TSRATIO) + GOS*(SCHSIZE) + G06*(LOFUND) + GO7"‘(FRL) +
GOS*(MINORITY) + U0

Bl = 610 + U1

As shown in Table 6-6, according to educators’ perceptions, losing school
districts were more likely than gaining school districts to make an effort to advertise and
do research who choice students/families were. However, the differences disappeared
when controlling for school socioeconomic environment such as student/teacher ratio,
school size, local funding, and the percentages of students eligible for free and reduced
lunch and minority students (see Table 6-7). In addition, losing school districts were less
likely than gaining school districts to make an effort to maintain a safe environment.
There were no statistically signiﬁcant differences between gaining and losing school
districts with regard to the other school activities for attracting students. Second, the
degree of organizing activities was not statistically associated with the degree of student
enrollment changes during the past ﬁve years except for counseling out low-performing

teachers. Schools with decreasing student enrollment were less likely than schools with

increasing student enrollment to make an effort to counsel out low-performing teachers.

87

Finally, schools with higher percentages of choice students were more likely than
schools with smaller percentages to make an effort to advertise schools and counsel out
low-performing teachers. However, the differences disappeared when controlling for
school-level socioeconomic factors such as student/teacher ratio, school size, local
funding, and percentages of students eligible for free and reduced lunch and minority
students (see Table 6-7). In addition, schools with higher percentages of choice students
were less likely than schools with smaller percentages of choice students to make an

effort to maintain a safe environment.

88

0.0020 00. 000 .000.0.0. 0...
.0. 00000..0> .000 000000.00 00 :03 00 .000.0.... 000000.00 00 00.. .0000. .0>0....0.0 000m. .0. .oW0 * .modwﬁ. .1. ”5...”. 1.... .0002

 

 

 

 

 

mmod So... wmod hood- emod wmod- 08.0 1.03...- .mod 08...- 0.0000.m 00.000 .0 ..\0
m5... 0 . o.o m3... o.o... m . o.o N 5.0 m .od .3... N56 to... 00w00..0 .000...0.0m. ..0 .x.
00m... 00.0.... 000.0 .00...- @000 N000- wwmd 1.0.3..- m .md @00- 0.0.0.0.n. w0.00-.
mow... 0mm..- womd N20... 80.0 180.0 2...... from... 0.0.... mo . .o 0.0.0..0.0..00<
.m..m ...-.000 .m..m 0.000 .m..m .....000 .m..m ...-.000 .m..w .-.-.000
0.0 000.800.850.00 00...>..00 0.00.000 .000.0>0...00 00.008300 0.0000
00.000 003 00.0003. 0.0.0 0>0.0.... .0000.0 0>0.00.. 0.00 0 0.0.0.0.). .03 002000020. 03.0.0.
mmod _..mvod wad omod- N8... wood 08... wood- 03... 0000.0 0.0000.m 00.000 ..0 o\°
0...... .08...- o.o.o wood NS... Nood- v.o.o 806- OS... 80.0- 00w0000 .008..0.0m. ..0 a\o
mum... 08.0 2...... .mod- man... .000..- 000... 5.....- Ex... Imam. 0.0.00.0 w0.00-.
. .0... 03.00%. 30... 0.2.0.0 00m... and No.0... .00... v . 0.0 mm . .0 0.0.0.00.0..00<
.m..m H.000 .m.m H.000 .m..m H.000 .m..m .....000 .m..m 0.000
0.00000. w0..0.0...00 80.00....00 0.00000. 00......000 000.0000.
-30. .00 .000000 .00000 0>0.0.... 00%... 0.00. .0003. 000 00.....00. .00000 000.0>0<

0.0000.m w0..00...< .0.. 000.>..0< .0.0000m 000 000000.000 .0000m 0003.00 000..0.0000< ”0-0 0.00 .-

.00000 0. .0030

9
00

00.0020 00. 000 .000.0.0. 00.

.0. 00000..0> .0..0 000000.00 00 :03 00 .000.0.0. 000000.00 00 000 .000... .0>0.0.00. 000m. .0 . 0W0 _.. .000W0 .1. ..0.0W0 *1. 0.02

 

 

9O

 

 

 

 

0N00 0N00 0N0.0 000.0 0N00 000.0 mN00 000.0- NN00 m .00- b00002 .0 o\°
0.0.0 0.0.0- 0.0.0 0 . 00 m .00 m000 N.0.0 000.0 N.0.0 000.0- 0y... .0 .0..
0m00 .000- N80 0000- 0000 00.0.0- 0N0.0 0000- 080 000.0- w0.000n. .0000
000.0 00. .0- 000.0 ... :- . .0- 0000 «00 . .0- 2.00 N. .0- 0.0.0 00. .0- 0~.m .0000m
0. .0 0N . .0 N. .0 0.. . .0 0N..0 000.0- 00.0 0000- .0. .0 .N . .0 0000 00000-500000
N000 0N00- 0000 1.0000- 0000 *0000- 0.00 3.0000- 0m00 0m00- 0.503m 00.000 .0 00
0.00 0.0.0 0 .00 mN00 0.0.0 0.0.0 0.0.0 .080 0 .00 .000 0»0000 E00030... .0 c\o
00.0.. 0.00 0.0.. .m0. .- 0N... 0000- 0000 3.000..- 0000 00N0- 0.0.00.0. w0.00-.
0 .00 Nmm0- 000.0 0000 N000 2.0000 0N00 1.000.. 0mm .0 0000 0.0.000.0.E0<

.00 ...000 .m.m H.000 .00 0.000 .m..m ...000 .00 ...000

0.0 000.50.300.00... 000.>000 .0.00...00 .00E0>0.000 .00E00..>00 0.00.00 003
00.000 003 00.00000 -0..x0 0>0.0E. 0.0030 0>0.00.. 0.00 0 0.0.0.02 00.000000.0. 0>0.00..

08.0 0000- 0N00 000.0- NN00 0000- 0N00 5.....N000- 500 Nm00 00.00.00 .0 .0.
0.0.0 N000- . .00 .000- N . 00 .N00 0.0.0 .3080 .N00 0N00 1.00 .0 ..\c
0.00 0000- 0N00 0N00- 080 m .00- mN00 _...I. . 00.0 000.0 mwo.0- w0.0000 .0000
000.0 0000- 0000 0000- 0000 0000- 0000 1mm . .0 0N .0 0000- 00.m .0000m
mm . .0 mm . .0 N000 N000 .0. .0 000.0 000.0 0000- 00 . .0 mNNO 000.0 .00000000003m
000.0 0N00 .m00 08.0- 20.0 000.0- 0m00 .80 000.0 0N00- 0.00000 00.000 .0 o\o
.N00 _.. .000- m .00 .000 m . 0.0 0.0.0 N .00 000.0- 0N00 0N00 000000 .00—000.00 .0 £0
.0.... 00.0- 200 mm . .0- 80.0 0mN..- 000.0 100.. 000.. wNNO 0.0.00.0. 00.000
0.0.0 6.3.0 .0.. w0m0 .030 N20 00 m .0 0000 0000 .000 90.0 0.0.0..0.0.0.0<
.m..m ...000 .m.m H.000 .m..m 0.000 .m.m H.000 .m..m H.000
0.00000. 0000.0..00 0.200.000 0.00000. 000000.. 000.0000. 000

-30. .00 .000000 .00000 0>0.00.. 0.00.0 0.00. 00.03. 0000.00. .8000 0. .00>0. .00000 000.0>0<

00..0..0.00.000

.0000m .0. 00500000 .000 0.0000.m 000000.... .0. 00...>..0< .0000m 000 000000.000 .0000m 0003.00 00000.0000< ”0-0 0.00...

School Problems

School problems were measured by educators’ perceptions. Administrators and
teachers were asked to indicate the degree to which each of the listed school problems
was a problem in their schools. They could indicate one of the following choices: l=not a
problem, 2=minor problem, 3=moderate problem, and 4=serious problem.
Teachers

In the views of teachers, the following problems seemed more serious than the
others: student tardiness, student absenteeism, student apathy (lack of engagement),
poverty, lack of parental involvement, student unpreparedness, and decrease in student
enrollment (see Table 6-8). However, teachers considered the following as “not a
problem” or “a minor problem”: student class cutting, student dropping out, quality of
instruction, school curriculum, teacher absenteeism, and increase in student enrollment. It
is noticeable that teachers in losing school districts were more likely than those in gaining

school districts to consider each problem serious.

91

Table 6-8: Teachers’ Perceptions of School Problems

 

Both Gaining LosinL Diff.
Student tardiness 2.56 2.27 2.81 -0.54
Student absenteeism 2.57 2.20 2.88 -O.68
Students class cutting 1.70 1.46 1.92 -O.46
Students dropping out 1.67 1.39 1.92 -O.53
Student apathy (lack of engagement) 2.63 2.32 2.90 -0.58
Lack of parental involvement 2.48 1.93 2.95 -1.02
Poverty 2.74 2.26 3.15 -0.89
Students come to school unprepared 2.76 2.35 3.12 -0.77
Quality of instruction is poor 1.88 1.61 2.1 l -0.50
Lack of school resources 1.86 1.47 2.20 -0.73
Poor teaching quality 1.32 1.16 1.45 —0.29
Inappropriate school curriculum 1.36 1.24 1.46 022
School order and discipline policies 1.92 1.51 2.28 -0.77
Teacher absenteeism 1.35 1.18 1.51 -0.33
Decrease in student enrollment 2.54 2.12 2.90 -0.78
Increase in student enrollment 1.14 1.08 1.19 -O.1 1
Meeting Adequate Yearly Progress (AYP) 1.84 1.27 2.35 -1.08

 

Note. l=not a problem, 2=minor problem, 3=moderate problem, and 4=serious problem.
Clustering of educators within schools was ignored.

Administrators

Administrators in both gaining and losing school districts considered the
following problems more serious than the others: poverty, student unpreparedness,
student apathy (lack of engagement), decrease in student enrollment, student tardiness,
and student absenteeism (see Table 6-9). In contrast, administrators viewed the following
problems as “not a problem” or as “minor problems”: school curriculum, increase in
student enrollment, school order and discipline, and student dropping out. In addition,
overall, administrators in losing school districts considered each problem more serious

than those in gaining school districts.

92

Table 6-9: Administrators’ Perceptions of School Problems

 

 

Both Gaining Losing Diff.
Student tardiness 2.19 2.00 2.40 -0.40
Student absenteeism 2.19 1.88 2.53 -O.65
Students class cutting 1.41 1.41 1.40 0.0]
Students dropping out 1.28 1.29 1.27 0.02
Student apathy (lack of engagement) 2.28 2.18 2.40 -0.22
0.00
Lack of parental involvement 2.19 1.88 2.53 -0.65
Poverty 2.66 2.24 3.13 -0.89
Students come to school unprepared 2.47 2.06 2.93 -0.87
0.00
Quality of instruction is poor 1.75 1.47 2.07 -0.60
Lack of school resources 1.53 1.53 1.53 0.00
Poor teaching quality 1.38 1.12 1.67 -O.55
Inappropriate school curriculum 1.19 1.00 1.40 -0.40
School order and discipline policies 1.22 1.06 1.40 -0.34
Teacher absenteeism 1.69 1.35 2.07 -0.72
0.00
Decrease in student enrollment 2.22 2.06 2.40 -0.34
Increase in student enrollment 1.03 1.06 1.00 0.06
Meetiﬂ Adequate Yearly Progﬁs (AYP) 1.72 1.47 2.00 -O.53

 

Note. l=not a problem, 2=minor problem, 3=moderate problem, and 4=serious problem.
Clustering of educators within schools was ignored.

Administrators vs. Teachers

Teachers were more likely than administrators to consider most school-related
problems serious except poor teaching quality and teacher absenteeism. Noticeably, both
teachers and administrators did not appear to view teaching quality and the school
curriculum as serious problems. The largest difference was observed in perceptions of
school order and discipline policies between teachers and administrators. Teachers tended
to consider the problem of school order and discipline policies more serious than

administrators (see Table 6-10).

93

Table 6-10: Differences in Perceptions of School Problems between Teachers and
Administrators

 

 

Teach. Admi. Diff.
Student tardiness 2.56 2.19 .37
Student absenteeism 2.5 7 2.1 9 .3 8
Students class cutting 1.70 1.41 .30
Students dropping out 1.67 1.28 .39
Student apathy (lack of engagement) 2.63 2.28 .35
Lack of parental involvement 2.48 2.19 .29
Poverty p 2.74 2.66 .08
Students come to school unprepared 2.76 2.47 .29
Quality of instruction is poor 1.88 1.75 .13
Lack of school resources 1.86 1.53 .33
Poor teaching quality 1.32 1.38 -.06
Inappropriate school curriculum 1.36 1.19 .17
School order and discipline policies 1.92 1.22 .70
Teacher absenteeism 1.35 1.69 -.33
Decrease in student enrollment 2.54 2.22 .32
Increase in student enrollment 1.14 1.03 .1 1
Meeting Adequate Yearly Progress (AYP) 1.84 1.72 .13

 

Note. l=not a problem, 2=minor problem, 3=moderate problem, and 4=serious problem.
Clustering of educators within schools was ignored.

As indicated in Table 6-1 1, the magnitude of the difference in most items between
teachers and administrators was larger in losing school districts than in gaining school
districts. While the differences in gaining school districts were less than .45 for all school
problems, the differences in losing school districts were more than .50 with regard to the
following problems: student class cutting, student dropping out, student apathy (lack of
engagement), lack of school resources, school order and discipline policies, teacher

absenteeism, and decrease in student enrollment.

94

Table 6-11: Differences in Perceptions of School Problems between Teachers and
Administrators in Gaining and Losing School Districts

 

Gaining School Districts Losing School Districts

 

Teach. Admi. Diff. Teach. Admi. Diff.

 

Student tardiness 2.27 2.00 .27 2.81 2.40 .41
Student absenteeism 2.20 1.88 .32 2.88 2.53 .35
Students class cutting 1.46 1.41 .04 1.92 1.40 .52
Students dropping out 1.39 1.29 .09 1.92 1.27 .66
Student apathy (lack of engagement) 2.32 2.18 .14 2.90 2.40 .50
Lack of parental involvement 1.93 1.88 .05 2.95 2.53 .42
Poverty 2.26 2.24 .02 3.15 3.13 .01
Students come to school unprepared 2.35 2.06 .29 3.12 2.93 .19
Quality of instruction is poor 1.61 1.47 .14 2.1 1 2.07 .05
Lack of school resources 1.47 1.53 -.06 2.20 1.53 .66
Poor teaching quality 1.16 1.12 .04 1.45 1.67 —.21
Inappropriate school curriculum 1.24 1.00 .24 1.46 1.40 .06
School order and discipline policies 1.51 1.06 .45 2.28 1.40 .88
Teacher absenteeism 1.18 1.35 -. l 8 1.51 2.07 -.56
Decrease in student enrollment 2.12 2.06 .06 2.90 2.40 .50
Increase in student enrollment 1.08 1.06 .03 1.19 1.00 .19
Meeting Adequate Yearly Progress (AYP) 1.27 1.47 -.20 2.35 2.00 .35

 

Note. l=not a problem, 2=minor problem, 3=moderate problem, and 4=serious problem.
Clustering of educators within schools was ignored.

Diﬂerences between Subgroups

To examine whether educators’ perceptions of school problems were signiﬁcantly
different between subgroups related to position (administrator or teacher), gender, tenure,
the gaining or losing status of their school districts, or school level, I conducted two-level
hierarchical non-linear models. Seventeen outcome variables can be seen in Table 6- 12.
The outcome variables were ordinal. Regarding explanatory variables, the following

variables at the school level were included: the gaining or losing status of school districts,

95

middle school, and high school. Position, tenure and gender at the individual level were
included. The models are as follows.
Level-1 Model

Prob[R =1|B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)
Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)
Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(l) = 1|B]
P(2) = Prob[Y(2) = 1|B]
P(3) = Prob[Y(3) = 1|B]

log[P'(1)/(l - P'(l))] = B0 + Bl*(ADMINIST) + B2*(MALE) +
B3*(NONTENUR)

log[P'(2)/(1 - P'(2))] = BO + B1*(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR) + d(2)

log[P'(3)/(l - P'(3))] = BO + B1*(ADMINIST) + B2*(MALE) +
B3*(NONTENUR) + d(3)

Level-2 Model
B0 = G00 + G01*(LOSING) + 002*(MIDDLE) + GO3*(HIGH) + U0

Bl =G10+U1
BZ=GZO+U2
B3 =G30+U3

As shown in Table 6-12, it is first notable that educators in losing school districts
were more likely than those in gaining school districts to view all the school problems
seriously. Second, middle school educators tended to view the following problems more
seriously than elementary school educators: student class cutting, students dropping out,
student apathy, students come to school unprepared, lack of school resources, poor
teaching quality, school order and discipline policies, school order and discipline
policies, teacher absenteeism, and decrease in student enrollment. There were no
signiﬁcant differences between elementary and middle school educators with regard to

the other problems. High school educators tended to view all the school problems more

96

seriously than elementary school educators except with regard to the following three
problems: poverty, quality of instruction is poor, and decrease in student enrollment.
There were no statistically signiﬁcant differences between elementary and high school
educators with regard to those problems.

Third, administrators were less likely than teachers to view the following school
problems seriously: student absenteeism, student apathy, students come to school
unprepared, and school order and discipline policies. In contrast, they were more likely
than teachers to view teacher absenteeism seriously. There were no signiﬁcant
differences between administrators and teachers with regard to the other school problems.
Fourth, male educators tended to view the following problems more seriously than
female educators: student apathy, inappropriate school curriculum, and meeting
adequate yearly progress. No signiﬁcant differences between female and male educators
were observed with regard to the other school problems. Finally, non-tenured educators
tended to view lack of parental involvement and poor teaching quality less seriously than
tenured educators. There were no signiﬁcant differences between tenured and non-

tenured educators with regard to the other school problems.

97

A308? 05 95 E0885 2:
Sm 828:? Ste 8:032: mm :03 mm 588:: venomous: cm mm: 338 35:22: comm d_dwn_ _.. ﬁdde I :ddwm .12.. «~82

 

 

 

 

 

 

 

 

wnvd mood- nomd nmmd- mmvd dood- chmd mmmd- m _ vd dwmd Beans-52
vad _..mE-d Rm d mm _ d wcmd N _ md cdmd m. _d- _omd ...m$d 2&2
mam d cmmd 3 wd dand- mdwd aimed.— thd :33. _- Smd _wwd- LougﬂEEdax
dad _.....Kmd owed Sued mX-d :36.— ZKd _..:movN Nde ammo; _oozom :wf
Rwd 3N— Emd 15w.— Ecd imam.— Ndcd 1.50.— ﬁnd on _d 32% 2252
VON-d *aacﬁd fwd ***~om.m dmod 23?; mddd 1...:m.~ omvd iwm _.. SPEED wEmoq
.md .toou .m.m “Coco .m.m .toou .m.m .tcoou .m.m .toou
mmohwoi 3.80% EBB—Eco 86:8 8:986 8:33.55 3058
Buscou< wezoo—z EodBm E 83800 Emmooucomnn 358,—. use 52¢ 305% Bargains:—
mdmd 1.3—.7 dwvd nmdd $md wood nwmd mid- Nwmd Q _d- mmmd aid—wd- 35:8-ch
N _ vd cdvd Ndmd 2. _ d- vmmd 2 _d- domd _m _ d- bdvd omvd- _dmd cnmd 222
com d _ddd mwvd owed- _mvd C _ d- wmvd sownd- vvvd o _ md _mvd vood- uoﬁbmmEEc<
acmd 1.38:6. wdwd them.— Nved Nddd wood Edam.— omnd NNdd- vad 3mm»; .09—om :wi
ndmd _..Vddd hmvd *ddwd Smd ~wood mmod _.....wwm._ m Ed Nd. d- _:d mdwd _oozom 23:2
Z-vd 3:53.— mdvd _..:won; End “.102“.— wde 1...».va wmcd 1.3va weed «tom—N 3365 $53
.m.m .boou .m.m .toou .m.m E80 .m.m .tooU .m.m E80 .m.m “Coco
bzasc moo-582 coon m_ :288me 839895 _oosom 2880235
war—03. Loon— _oo:om .3 x23 .8 bzmso 9 088 Scodam Eugen. 3:83 «o x03
_de wmmd Svd m: _d odvd mm _ d nmmd dmdd- N. _ md mm _ d- 355752
dwmd ammmd vovd Nwod- wmmd Fwd- mwmd vwmd mdmd Kmd 232
w :d L Ed. N _ ed mdwd- 0mm d 3 _ d- and _..wRd- dmvd howd- SSbﬂEEd<
_Nod _..:omdd mm: :KEd 5d; 5:. 50.5 and ..:.:..mmm.m dmnd :aommd 33% :3:
End _..:mNNN. 05.— ...t..$w.m mmdd *aawndé med mmwd coed did 32% 22:2
_dvd :13; mad _..:ow_.v vowd 1.3—Wm chad I...ooo.~ mwmd 1&3.— ﬂoEmE wEmS
.m.m “coco .m.m .toou .m.m .boou .md “coco .m.m .booU
>595 So weaned maﬁa mum—o 8202:0QO 30:63...
“Scam ﬂaodam mucodam Eva—yam EocEm

 

Eco—poi _oonom do mcocmoobm .Eoumondm madam»: mascumnzm smegma oozes-$5 ”N70 oBm-w

98

Educators’ Perceptions of the Reasons for Parents’ School Choice Decisions

In order to measure the importance of various reasons for choosing schools,
administrators and teachers were asked to indicate the extent to which parents in their
school districts consider each of 12 possible reasons when they choose schools for their
children. Administrators and teachers could indicate one of the following choices: l=not
at all, 2 =to some extent, 3 =to a moderate extent, 4 =to a great extent, and 5 =not sure.
Responses with not sure were excluded from the analyses.
Teachers

Among 12 possible reasons for parents’ choices regarding schools for their
children, teachers responded that parents would place the most importance on the
following three reasons: school reputation and history, safe environment (school order
and discipline), and academic programs (see Table 6-13). Teachers in gaining school
districts were more likely than those in losing school districts to consider all possible
parental reasons except the following two as important: location or convenience and
student racial composition. Teachers in losing school districts were more likely than
those in gaining school districts to perceive that parents consider those reasons as

important factors.

99

Table 6-13: Teachers’ Perspectives on Parents’ School Choice Reasons

 

 

Both Gaining Losing Diff.
School reputation and history 3.62 3.87 3.40 .47
Safe environment (school order and discipline) 3.57 3.66 3.48 .18
Academic programs 3.38 3.63 3.15 .48
Teacher quality 3.36 3.61 3.13 .49
School facilities 3.35 3.51 3.20 .30
Academic performance/test scores 3.28 3.54 3.05 .48
School resources 3.25 3.47 3.05 .42
Extra-curricular activities 3.20 3.33 3.09 .24
Location or convenience 3.03 2.87 3.17 -.31
Teaching practices 2.93 3.09 2.78 .31
Socio-economic status of the community 2.81 2.96 2.68 .28
Student racial composition 2.37 2.04 2.64 -.60
Mean 3.18 3.30 3.07 .23

 

Note. 1=Not at all, 2=Some extent, 3=Moderate extent, and 4=Great extent. Clustering of
educators within schools was ignored.

Administrators

As Table 6-14 shows, among 12 possible reasons for parents’ choices regarding
schools for their children, administrators responded that parents would consider the
following reasons more seriously than the others: safe environment (school order and
discipline), school reputation and history, academic programs, and academic
performance/test scores. Generally speaking, administrators’ responses were similar to
teachers’. Administrators in gaining school districts were more likely than those in losing
school districts to consider all the reasons except the following two as more important:
location or convenience and student racial composition. Administrators in losing school
districts shared the same pattern of responses as their teachers regarding their perceptions
of school location or convenience and student racial composition. It means that
administrators in losing school districts were more likely than those in gaining school
districts to perceive that parents consider school location and student racial composition

as more important factors.

100

Table 6-14: Administrators’ Perspectives on Parents’ School Choice Reasons

 

 

Both Gaining Losing Diff.

Safe environment (school order and discipline) 3.75 3.88 3.60 .28
School reputation and history 3.72 4.00 3.40 .60
Academic programs 3.50 3.76 3.20 .56
Academic performance/test scores 3.34 3.65 3.00 .65
Teacher quality 3.31 3.82 2.73 1.09
Extra-curricular activities 3.29 3.56 3.00 .56
School facilities 3.22 3.35 3.07 .29
School resources 3.00 3.56 2.36 l .21
Socio-economic status of the community 2.93 3.13 2.73 .40
Location or convenience 2.87 2.56 3.20 -.64
Teaching practices 2.87 3.18 2.50 .68
Student racial composition 2.20 1.88 2.57 -.70
Mean 3.17 3.36 2.95 .42

 

Note. 1=Not at all, 2=Some extent, 3=Moderate extent, and 4=Great extent. Clustering of
educators within schools was ignored.

Teachers vs. Administrators

As shown in Table 6-15, teachers were more likely than administrators to
consider the following as more important factors for parents: location or convenience,
school facilities, teacher quality, school resources, student racial composition, and
teaching practice.

Table 6-15: Differences between Teachers’ and Administrators’ Perspectives on Parents’
School Choice Reasons

 

 

Teach. Admi. Diff.
Location or convenience 3.03 2.87 .16
Safe environment (school order and discipline) 3.57 3.75 -.18
Socio-economic status of the community 2.81 2.93 -. 12
School facilities 3.35 3.22 . l 3
Academic performance/test scores 3.28 3.34 -.06
Academic programs 3.38 3.50 -. 12
Extra-curricular activities 3.20 3.29 -.09
Teacher quality 3.36 3.31 .05
School resources 3.25 3 .00 .25
School reputation and history 3.62 3.72 -. l 0
Student racial composition 2.37 2.20 .17
Teaching practices 2.93 2.87 .06

 

Note. 1=Not at all, 2=Some extent, 3=Moderate extent, and 4=Great extent. Clustering of
educators within schools was ignored.

101

As indicated in Table 6-16, in gaining school districts, teachers were more likely
than administrators to consider the following as more important factors for parents:
location, school facilities, and student racial composition. However, in losing school
districts, teachers were more likely than administrators to consider the following as more
important factors for parents: school facilities, teacher quality, school resources, and
teaching practices.

Table 6-16: Differences between Teachers’ and Administrators’ Perspectives on Parents’
School Choice Reasons in Gaining and LosingSchool Districts

 

 

 

Gargggﬁsccirool Losing School Districts

Teach. Admi. Diff. Teach. Admi. Diff.

Location or convenience 2.87 2.56 .31 3.17 3.20 -.03
Safe environment (school order and discipline) 3.66 3.88 -.22 3.48 3.60 -.12
Socio-economic status of the community 2.96 3.13 -.17 2.68 2.73 -.05
School facilities 3.51 3.35 .15 3.20 3.07 . 14
Academic performance/test scores 3.54 3.65 -.l l 3.05 3.00 .05
Academic programs 3.63 3.76 -.14 3.15 3.20 -.05
Extra—curricular activities 3.33 3.56 -.24 3.09 3.00 .09
Teacher quality 3.61 3.82 -.21 3.13 2.73 .39
School resources 3.47 3.56 -.09 3.05 2.36 .70
School reputation and history 3.87 4.00 -. l 3 3.40 3.40 .00
Student racial composition 2.04 1.88 .17 2.64 2.57 .07
Teaching practices 3.09 3.18 -.09 2.78 2.50 .28

 

Note. 1=Not at all, 2=Some extent, 3=Moderate extent, and 4=Great extent. Clustering of
educators within schools was ignored.

Diﬂ'erences between Subgroups

To examine whether educators’ perceptions of school choice reasons were
signiﬁcantly different between subgroups related to position (administrator or teacher),
gender, tenure, the gaining or losing status of their school districts, or school level, I
conducted two-level hierarchical non-linear models. Outcome variables were educators’
perceptions of the importance of 12 possible school choice reasons for parents. The
outcome variables were ordinal. Regarding explanatory variables, the following variables

at the school level were included: the gaining or losing status of school districts, middle

102

school, and high school. Position, tenure and gender at the individual level were included.
The models are as follows.
Level-1 Model

Prob[R =1|B] = P'(l) = P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)
Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)
Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(l) = 1|B]
P(2) = Prob[Y(2) = 1|B]
P(3) = Prob[Y(3) = 1|B]

log[P'(1)/(1 - P'(l))] = BO + B1*(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR)

log[P'(2)/(1 - P'(2))] = BO + B1 *(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR) + d(2)

log[P’(3)/(1 - P'(3))] = B0 + Bl*(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR) + d(3)

Level-2 Model

B0 = GOO + GOI*(LOSING) + G02*(MIDDLE) + GO3*(HIGH) + U0

B1=G10+U1
BZ=GZO+U2
B3=G30+U3

As shown in Table 6-17, it is ﬁrst notable that educators in losing school districts
were less likely than those in gaining schools to think that parents considered each school
choice reason seriously except with regard to location or convenience, socioeconomic
status of the community, extra-curricular activities, and student racial composition.
There were no statistically signiﬁcant differences between gaining and losing school
districts regarding socioeconomic status of the community and extra-curricular activities.

However, educators in losing school districts were more likely than those in gaining

103

school districts to think that parents considered location or convenience and student
racial composition.

Second, educators in middle schools were more likely than those in elementary
schools to think that parents considered extra-curricular activities and school resources
seriously. No statistically signiﬁcant differences were found with regard to the other
choice reasons. Educators in high schools were more likely than those in elementary
schools to think that parents considered school facilities, extra-curricular activities, and
school resources seriously. However, they were less likely than educators in elementary
schools to think that parents considered teaching practice seriously. Educators in
elementary and high schools did not differ with regard to the other choice reasons. Third,
no statistically signiﬁcant differences between teachers and administrators were observed.
Fourth, male educators were less likely than female educators to think that parents
considered academic performance/test scores, school resources, and teaching practices
seriously. Male and female educators did not differ with regard to the other school choice
reasons. Finally, non-tenured educators were more likely than tenured educators to think
that parents considered teacher quality and teaching practices seriously. Tenured and

non-tenured educators did not statistically differ with regard to the other choice reasons.

104

.303? 05 ﬁg Eoouodﬁ on:
:8 30:23., :oto 3:033: mm :25 mm EoBBE octaves: :m we: 68.: 65:22: :omm d—dwm ... ”modwm ...... xodwm 1.... $82

 

 

 

 

 

 

 

 

105

_vvd a... _ m _ ._ o _ md momd- REES-:02
thd 32.5.? womd oomd- 2:2
wvvd onmd movd Rmd- 8852583..
$3 3.8:? 3...... .33- .858 gm...
ommd ohvd- Nomd v2 .o- 30:5 23:2
Nomd ............mm~. _- Nmmd .1253.— mdoEmE mime-—

.md .tooo .md .booU

5:509:00
8288a @223..- . .
.308 Eovﬁm

_mvd omod whmd o _ md omvd _..mowd vomd ovmd mnmd womd- 8:38:02
mmvd oo _ .o Rmd _.....hwod- mmmd o _ vd- womd oomd- n _ md Nomd- 2:2
mmwd oomd movd vad- vovd 31d oovd 3 _ .o onvd mmmd :oﬁbﬂﬁﬁox‘
Nmmd ommd- mvmd sumod vovd ......w~o._- 53d Immmg nmwd :Nd- _oonom :wE
owvd _mod- oomd *vmwd momd Iod- movd _....imvmg bond onmd _oocom 22:2
onvd .31—Nd- Numd 3.5mm;- nmmd 3.3%;- wvmd Rond- omvd 333.. _- $2.55 $63

.m.m E80 .m.m .tooU .m.m E000 .m.m “coco .m.m .b000

.0052 88:82 _oosom .0530 8:08... .8238: 2:8on o_Eoowo<

gm 558:3: _oonom SBoEU-mbxm

h :d mu _ .o med mood- _omd mmmd- demo 3 _ .o- E md ~wood 3:38-82
mmmd added. 3 md oovd- mwmd ooNd 2 md onod- momd mwod- 222
vad E- . .o Rvd mhmd- owed momd Sand _ _wd vomd B: .o- cougmmﬁﬁoaw
End _omd- wmvd *mvnd 2nd «Ed- dwmd vMod de mnmd _oonom awe
novd mm _.o owmd bond wovd vomd womd wwvd nwmd 52d- _oonom 22:2
omvd .........wmm. T ommd *omod- oovd god. 2 md .Lmod- wvmd .........N~o.o 82:85 wEmS

.md .booo .m.m “Coco .m.m .toou .m.m .tooU .m.m H300

8:08 “moubocmgoﬁoq moczmoﬂ _oosom 35888 05 oo EoE:o.:>:o comm 8:35:00 :o 5:80-—
o_Eoowo< mBSm o_Eo:ooo-o_oom

 

28%-m 36:0 .0055 .«o £8398:ko .muoumosom E asepwosm 59.30: 80:08th K To 038.

Associations between School Problems and Activities for Attracting Students

To examine whether schools’ activities for attracting students were commensurate
with the degree of school problems perceived by teachers, two—level hierarchical non-
linear regressions were conducted. Outcome variables were the degree of organizing
various school activities for attracting students as indicated in the following model. The
outcome variables were ordinal in scale. Main explanatory variables were the extent to
which teachers viewed each problem as serious. The explanatory variables were treated
as continuous to examine the direction of the associations.

Each regression has one outcome variable and one matched explanatory variable.
In addition, each regression has two common control variables. One is the gaining or
losing status of school district at the school level. The other is position at the individual
level.
Level-1 Model

Prob[R =1|B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

P(l) = Prob[Y(1)= 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(l - P'(l))] = BO + B1*(ADMINIST) + B2*(SPOO)

log[P'(2)/(l - P'(2))] = BO + Bl *(ADMINIST) + B2*(SPOO) + d(2)

log[P'(3)/(l - P'(3))] = BO + B1*(ADMINIST) + BZ*(SPOO) + d(3)

Y1 = invest in school facilities and resources

Y2 = improve school curriculum

Y3 = counsel out low-performing teachers

Y4 = improve relationships with parents
Y5 = maintain a safe environment

106

Y6 = improve student achievement

School Problems SPOO

SPOl = lack of school resources

SPOZ = inappropriate school curriculum

SPO3 = quality of instruction is poor

SPO4 = lack of parental involvement

SPOS = school order and discipline policies

SPO6 = meeting AYP
Level-2 Model

B0 = GOO + 601 *(LOSING) + U0

B1 = 610 + U1

82 = 620 + U2

Schools’ activities for attracting students did not appear to address the school
problems that teachers viewed seriously. As shown in Table 6-18, it is notable that
negative associations were observed between the degree to which the following school
problems were perceived as serious and their effort to address those problems:
inappropriate school curriculum, lack of parental involvement, school order and

discipline policies, and meeting A YP. There were no statistically signiﬁcant associations

with respect to the other school problems.

107

.328? 05 28 385:: 2:

8m meant? coho Banana: 3 :03 mm 5095:: 3:885. cm mm: 3.58 ESE—=8 comm .3 .owm _.. ”3.on ...... :odwm 1.... 682

 

aomd _.....mood-

owmd mm _ .c-

hm _ .o tibovd-

wovd owed- mcmd 2 _.o- NNYO
2 m6 Smd w _ md N3 .o- 796 _wmd mbmd _omd- momd wmmd-

AVG—d ***cmh.os

3.... _ow. _- movd _..mtd- mwmd 09.6-

30520 wEmoq

a: @582
axiom 05—385
Ea EEO _oosom
235335
.853; .«o :03
coon— mm 5:252:
do 3:20
EgsoEsu _oonom
Bar—mega:—
much—08m

620m .8 V.2...

N36 _.....nmmd-

3N6 med

hafo ***m_©.o..

mac—d N25-

 

.m.m .toou
E25523.
«co—cam 389:.

.m.m .tooU .m.m .tooU .m.m E80 .m.m .tooU .m.m .tooU

$535— Eosowoh 88:83— 033%;

ESBEEE—u/x W

aGOEGOhm>CQ

£8 a 53532 :23 8.55533—

389:—

wig—eta

:33 «so _DMGZOU

823530

_oozom 3895 can ﬁzz—Ban

.025 E 60>:—

“uneconom—

 

mEovBm wczombﬁx com 853:3. can mace—no.5 _oonom 5233 mcocﬂoame. H26 2me

Associations between Choice Reasons and Activities for Attracting Students

To examine whether schools’ activities for attracting students were commensurate
with teachers’ perceptions of the importance of various reasons for choosing a school,
two-level multiple regressions were conducted while controlling for school
characteristics. Outcome variables were the degree of organizing various school activities
for attracting students as indicated in the following model. Main explanatory variables
were the extent to which teachers viewed various reasons for choosing schools as serious.
Each regression has one outcome variable and one matched explanatory variable. In
addition, each regression has two common control variables. One is the gaining or losing
status of school district at the school level. The other is position at the individual level.
Level-l Model

Prob[R =1|B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(l) = 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(1)/(1 - P'(l))] = BO + Bl*(ADMINIST) + B2*(CROO)

log[P'(2)/(1 - P'(2))] = BO + Bl *(ADMINIST) + BZ*(CROO) + d(2)

log[P'(3)/(l - P'(3))] = B0 + Bl*(ADMINIST) + B2*(CROO) + d(3)

Y1 = invest in school facilities and resources

Y2 = recruit more highly qualiﬁed teachers

Y3 = improve school curriculum

Y4 = counsel out low-performing teachers

Y5 = maintain a safe environment

Y6 = improve student achievement

Y7 = improve extra-curricular activities

Choice Reasons CROO

109

CROl = school facilities

CR02 = teacher quality

CRO3 = academic programs

CR04 = teacher quality

CROS = safe environment

CRO6 = academic performance/test scores

CRO7 = extra-curricular activities
Level-2 Model

B0 = GOO + G01*(LOSING) + UO

B1 = 610 + U1

82 = (320 + U2

As shown in Table 6-19, in the views of teachers, many school activities for
attracting students seemed to be commensurate with teachers’ perceptions of the
importance of various reasons for choosing schools. Of seven possible school activities
included in the analyses, the following ﬁve school activities were signiﬁcantly and
positively associated with the degree of the importance perceived by teachers regarding a
given reason for choosing a school: recruit more highly qualiﬁed teachers, improve
school curriculum, maintain a safe environment, improve student achievement, and

improve extra-curricular activities. The other two activities were positively related but

were not statistically signiﬁcant.

110

.3286 05 van £0285 05

.8“ meager? hobo Btomoan mm :03 mm E852: Bahama: cm was $on 65:23: :8m .2 .on ... Hedwm .2. :odwm 1.... .302

 

 

 

 

 

wwmd emvd o _ to 33.56 wvvd 1&2. _ ﬁoﬁbmmEEEw
3:3 03.? vad oﬂd- owed oovd- 33.55 @501.
cc _ .o _..:med mo£>co< uﬂsogohbxm
_m _ .o 35on 880m “mgoocactotom 2:598"
g _ .o 1.1306 EoEcohém ﬂaw
.m.m .tcoou .m.m .boou .m.m .tooo
853:3. 8_:oE:u-abxm 3895 EoEo>oEo< 2.25% 3895 825.835 £8 a ESEmE
mmvd ***m~m._ oovd soond 250 med 936 $32. ﬂoabmmﬁﬁcxx
vmvd fwd- 30o wmmd cemd 30o- Svd mm _ .o- 8255 maﬁa»—
8 _ .o ego £20 22E.
2 _ .o 2536 meewoa 2.532
$3 _..:Svo Ego 28¢
on _ .o Rmd magmas”. _oozom
.m.m E80 .m.m .toou .m.m “coco .m.m “coco
E238... wﬁctotoaéoq 52:39.0 _oosom £05on 35:30 88:83. can
So _omcsou "SPEE— bnwf 232 23.83. mochmm .025 5 not:

 

3:255 MEBEHE ecu mo£>uo< EB 3083— 86:0 5.02502 323683. ”3-0 oEmh

111

Discussion

With regard to school activities for increasing student enrollment, it is ﬁrst
notable that both gaining and losing school districts made only a limited amount of effort
to research who choice students/families were and what types of schooling options they
were seeking. We may suppose that in order to develop appropriate strategies for
attracting more students, districts or schools would study what factors parents consider
important when they choose schools for their children, who exercises school choice
options, and who does not. We need to examine the reasons why districts or schools do
not make serious effort to research these issues. There could be many reasons.
Administrators might assume that they already know what parents need and who choice
students/families are. Thus, they might not feel the necessity for such research. However,
administrators’ and teachers’ knowledge about choice students does not seem to be
empirically supported. For example, in the surveys, about 22% of administrators and 30%
of teachers responded that they did not know whether choice students were highly
motivated or whether they were high achievers (see Appendix G). Another alternative
explanation is that administrators did not feel that the research was necessary or they did
not have enough capacity to conduct the research. This issue goes beyond the boundary
of this study. Future studies may need to address this concern.

Another notable point is that while losing school districts have made more efforts
in advertising than gaining school districts, gaining school districts had made more
efforts to improve student achievement, maintain a safe environment, and recruit more
highly qualiﬁed teachers. In order for school choice programs to be a mechanism for

improving school quality, schools which are losing students to neighboring schools need

112

to address issues related to school quality. Advertising schools might be an effective way
to attract students but that is not directly related to school improvement.

Finally, in the views of educators, the degree of organizing various activities for
attracting students did not appear to be associated with the degree of competition among
schools. Only a few activities were signiﬁcantly associated with the degree of
competition among schools but the directions of the associations were not consistent.
This could imply that student enrollment changes caused by school choice programs do
not strongly inﬂuence schools’ responses. This could be partly explained by incentive
systems which were not aligned with decreases or increases in student enrollment. From
the NCE perspective, the current incentive systems might fail to motivate educators to
respond to decreasing student enrollment. This could be also explained from the IBE
perspective. Even though they know about enrollment changes, they might be interested
in other personal issues instead of attracting more students or believe that attracting more
students is not important.

In terms of school problems, both teachers and administrators appeared to think
that the problems related to students or parents themselves were more serious than the
problems related to educators themselves, school curriculum, or school policies. It would
be hard to expect schools to make an effort to improve teaching quality or school
curriculum when they do not view these as serious problems. It is also notable that
administrators tended to consider school problems to be less serious than teachers. This
could imply that administrators might not be active in addressing school problems even

when teachers ask them to do so.

113

Regarding educators’ perceptions of the reasons for parents’ school choice
decisions, we need to pay attention to teachers and administrators in losing school
districts. First, educators in losing school districts were more likely than those in gaining
school districts to consider the following two reasons as important factors: location or
convenience and student racial composition. We could conclude that when educators in
losing school districts believe that parents consider location or convenience and student
racial composition when they choose schools, they might think that they have nothing to
do in order to attract more students. In this context, it would be hard to expect them to
make an effort to improve school quality through upgrading school curriculum or
introducing extra-curricular activities. Another notable point is that educators in losing
school districts seem to believe that parents in their school districts are not interested in
choosing schools for their children. In this context, educators in losing school districts
might not be active in doing things in order to attract more students.

School activities for attracting students did not appear to be commensurate with
the degree of school problems perceived by educators. The results do not seem to support
the NCE hypothesis: as a given school problem becomes more serious, schools would
make greater efforts to address that problem. It is interesting that negative associations
were observed between the degree to which the following were perceived as school
problems and their effort to address those problems: inappropriate school curriculum,
lack of parental involvement, school order and discipline policies, and meeting A YP. One
could explain this by drawing on the notion that educators have different reference points
(beliefs). For example, educators whose schools have serious problems with curriculum

might think that their schools do not make an effort even though their schools make much

114

effort. In contrast, educators whose schools do not have any serious problem with
curriculum might think that their schools make much effort even when their schools do
not. There were no statistically signiﬁcant associations with respect to the other school
problems. To explain why school activities are not commensurate with school problems,
the IBE perspective might be of help. At ﬁrst, administrators might not know about the
problems because they were not interested in school problems or they did not consider
the issues as problems to be addressed. Second, administrators might not think that they
should address school problems in order to attract more students. Finally, administrators
might not know how to address the problems. This study does not address these issues
directly.

In regard to the association between school activities for attracting students and
perceptions of parents’ reasons for choosing a school, the results seem to partially support
the NCE hypothesis. Many school activities for attracting students seemed to be
commensurate with teachers’ perceptions of the importance of various reasons for
choosing schools. Of seven possible school activities included in the analyses, ﬁve school
activities were signiﬁcantly and positively associated with the degree of the importance
perceived by teachers of a given reason for choosing a school. The other two activities
were positively related but were not statistically signiﬁcant. Yet, there is a causality iSsue
here. The NCE perspective assumes that parental reasons for choosing schools for their
children have a casual effect on educators’ actions. This ﬁnding, however, does not
necessarily conﬁrm this assumption. For example, school practices might also affect
educators’ perceptions, as expected from the IBE perspective. Educators at schools in

afﬂuent districts that have made greater efforts to recruit high quality teachers might

115

perceive that parents consider teacher quality as a more important factor when they

choose schools for their children.

116’

CHAPTER 7

IMPACT OF SCHOOL CHOICE POLICES ON EDUCATORS AND SCHOOLS

This chapter aims to address the last research question: How do teachers and
administrators perceive the impact of school choice policies on teachers and
administrators themselves and their schools? Are their perceptions of the impact
associated with the degree of competition among schools? One of the various purposes of
school choice policies is to improve school quality with the help of competition among
schools. Teaching and leadership practices (Barnett & McCormick, 2004; Demoss, 2002;
Leithwood & Jantzi, 2006; Marks & Printy, 2003), collaboration with colleagues
(Goddard, Goddard, & Tschannen-Moran, 2007), cuniculum(Newmann, Smith,
Allensworth, & Bryk, 2001), extra-curricular activities , and relationships with parents
(Henderson & Mapp, 2002) are important factors which contribute to improving school
quality including student achievement. Assuming that school competition due to school
choice policies would affect these factors conducive to student achievement, previous
studies focused on the associations between school choice and student achievement
(growth). However, previous studies failed to show a clear causal link between school
choice programs and student achievement growth (Holmes et al., 2003; Hoxby, 2004’;
Shanker & Rosenberg, 1992). To explain this, we need to examine how school choice
programs affect teaching (leadership) practices and school organization such as
collaboration with colleagues, curriculum, extra-curricular activities, and relationships

with parents, which are important factors in improving student achievement. Based on

117

 

[CE

W

it

teachers’ and administrators’ perceptions of the impact of school choice programs on
educators themselves and schools, this study aims to address that issue.

According to the neoclassical economics (NCE) perspective, parents would
choose schools for their children based on school quality such as school curriculum,
teaching quality and methods, and student achievement. Responding to parents, schools
would make an effort to improve school quality. Therefore, we can expect educators to
believe that school choice programs could contribute to improving school quality by
leading to improvement in their teaching (leadership) practices, programs, extra-
curricular activities, and orderly environment. In addition, as competition among schools
becomes intensive, educators would believe that school choice programs could contribute
more to improving school quality.

However, the institutional and behavioral economics (IBE) perspective draws our
attention to educators’ beliefs and limited information processing capacity. Many
educators might not be interested in improving school quality when they are more
interested in other issues. In addition, the IBE perspective does not posit that all educators
would know about student enrollment changes, suggesting the possibility that enrollment
changes might fail to prompt educators to do something to attract more students. As a
result, there might be no associations between the degree of competition and the degree
of the impact of school choice programs on educators themselves and/or schools.

Based on the two competing perspectives, I generated two research hypotheses as
follows and tested them with the surveys of administrators and teachers.

Hypothesis 4a (NCE): Educators would believe that school choice programs could

contribute to improving school quality through improving their teaching (leadership)

118

practices, academic programs, extra-curricular activities, collaboration with colleagues,
and relationships with parents. In addition, educators in schools under more intensive
competition would be more likely than those in schools under less intensive competition
to have these beliefs.

Hypothesis 4b (IBE): Educators might not believe that school choice programs
could contribute to improving school quality through improving their teaching
(leadership) practices, academic programs, extra-curricular activities, collaboration with
colleagues, or relationships with parents. In addition, there might be no associations
between the intensiveness of competition and their perceptions of the degree of the
impact of school choice programs on educators themselves and/or schools.

Educators’ Perceptions of Impact of School Choice Policies on Their Practices

To examine educators’ perceptions of the impact of school choice policies on
their practices, teachers and administrators were asked to indicate the extent to which
they agreed that school choice policies were likely to affect each of several areas of
practice. The areas of practice are seen in Table 7-1 for teachers and in Table 7-3 for
administrators.

Teachers

According to teachers’ perceptions, school choice policies appeared to fail to
motivate teachers to improve their teaching practices, collaboration with colleagues,
curriculum, extra-curricular activities, or relationships with parents. As indicated in Table
7-1, more than two-thirds of the teachers responded negatively when asked about the
effects of school choice policies on efforts to improve their own teaching practice and

curriculum, collaborate with colleagues, or build relationships with parents.

119

Table 7-1: Impact of School Choice Programs on Teachers (%)

 

 

itrongly Disagree Agree Strongly Total
1sagree agree
5:38;? to improve my own teaching practice 18.4 55.9 23-3 2.4 1000
Lead me to more collaborate with colleagues
to attract more students. (n=287) 18'8 58'2 206 2'4 1000
Lead me to rethink my curriculum (n=288) 18.8 54.9 24.7 1.7 100.0
1““?le ”fem“ my ema'cumcma’ 22.6 59.7 16.3 1.4 100.0
activ1ties(n-283)
Lead me tp improve my relationships With 19.6 44.8 31.8 3.8 100.0
parents(n—286)

Mean 19.6 54.7 23.3 2.4 100.0

 

As shown in Table 7-2, school choice policies were more likely to affect teachers

in losing school districts rather than those in gaining school districts.

Table 7-2: Impact of School Choice Programs on Teachers in Gaining and Losing School

 

 

Districts
Both Gaining Losing Diff.
Lead me to improve my own teaching 2 10 2 01 2 18 - l7
practice. ' ' ' '
Lead me to more collaborate with colleagues
to attract more students. 2'07 1'92 2'21 "29
Lead me to rethink my curriculum. 2.09 1.96 2.21 -.25
Lead me to rethink my extra-curricular l 97 1 86 2 07 _ 21
activities. ' ' ' '
Lead me to improve my relationships with 2 20 2 01 2 36 _ 36
parents. ' ' ' '
Mean 2.09 1.95 2.21 -.25

 

Note. l=strongly disagree, 2=disagree, 3=agree, 4=strongly agree

Administrators

Administrators tended to evaluate the impact of school choice policies on

themselves more positively than teachers. As Table 7-3 indicates, while 70% of the

administrators responded negatively to whether school choice programs led them to

improve their leadership, more than half of the administrators reported that school choice

programs led them to collaborate with colleagues, rethink school curriculum, and

improve relationships with parents.

120

Table 7-3: Impact of School Choice Programs on Administrators

 

 

Strongly . Strongly
disagree Disagree Agree agree Total
Lead me to improve my leadership (n=30) 6.7 63.3 26.7 3.3 100.0
Lead me to more collaborate with colleagues to
attract more students (n=30) 10.0 36.7 46.7 6.7 100.0
Lead me to rethink my school’s curriculum (n=30) 6.7 36.7 50.0 6.7 100.0
Lead' me to rethink my school 5 extra-curricular 10.0 433 43.3 3.3 100.0
actiVities (n—30)
Lead me to improve my school 5 relationships With 6.7 23.3 60.0 100 100.0
parents (n—30)
Mean 8.0 40.7 45.3 6.0 100.0

 

As shown in Table 7-4, administrators in losing school districts were more likely
than those in gaining school districts to be affected by school choice programs.

Table 7-4: Impact of School Choice Programs on Administrators in Gaining and Losing
School Districts

 

 

Both Gaining Losing Diff.
Lead me to improve my leadership 2.27 2.27 2.27 0.00
Lead me to more collaborate With colleagues to 2.50 2.47 2.53 -0.0 6
attract more students
Lead me to rethink my school’s curriculum 2.57 2.40 2.73 -0.33
Leadme to rethink my school 8 extra-curricular 2.40 2.20 2' 60 _040
activmes
Lead me to improve my school 5 relationships With 2.73 2.53 2.93 -0. 40
parents

 

Note. 1=strongly disagree, 2=disagree, 3=agree, 4=strongly agree
Differences between Subgroups

To examine whether educators’ perceptions of the impact of school choice
programs on their practices were signiﬁcantly different between subgroups related to
position (administrator or teacher), gender, tenure, the gaining or losing status of their
school districts, or school level, I conducted two-level hierarchical non-linear models.
Outcome variables are seen in Table 7-4. The outcome variables were ordinal. Regarding
explanatory variables, the following variables at the school level were included: the
gaining or losing status of school districts, middle school, and high school. Position,

tenure and gender at the individual level were included. The models are as follows.

121

Level-1 Model

Prob[R =1|B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)
Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)
Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(1)=1|B]
P(2) = Prob[Y(2) = 1|B]
P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(l - P'(l))] = BO + Bl*(ADMINIST) + B2*(MALE) +
B3*(NONTENUR)

log[P'(2)/(1 - P'(2))] = BO + Bl *(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR) + d(2)

log[P'(3)/(1 - P'(3))] = BO + B] *(ADMINIST) + BZ*(MALE) +
B3*(NONTENUR) + d(3)

Level-2 Model

BO = G00 + GOI*(LOSING) + G02*(MIDDLE) + GO3*(HIGH) + U0

Bl =GlO+Ul
B2=GZO+U2
B3=GBO+U3

As indicated in Table 7-5, educators in losing school districts were more likely
than those in gaining school districts to agree that school choice programs led them to
collaborate more with colleagues, rethink their curriculum and extra-curricular activities,
and improve their relationships with parents. However, there were no signiﬁcant
difference between gaining and losing school districts in terms of whether school choice
programs led them to improve their own teaching or leadership practices. Second, no
statistically signiﬁcant differences were observed between elementary and middle
schools and between elementary and high schools. Third, administrators were more likely
than teachers to agree that school choice programs led them to collaborate more with

colleagues, rethink their curriculum and extra-curricular activities, and improve their

122

relationships with parents. However, there was no signiﬁcant difference between
administrators and teachers with regard to the impact of school choice policies on
teaching (leadership) practices. Finally, there were no signiﬁcant differences between

male and female educators or between tenured and non-tenured educators.

123

.330? 2: can E0885 2:
.8.“ 82893 Hobo cor—once: mm :95 3 E6885 3:682: an mm: 638 35:22: comm .o _ .owm ... $9ch I. :odwm *1. .682

 

 

mwmd m . Nd m 36 w _ Nd momd wwod- h _ vd god and R: .o- 8:52-52
wwmd no _ .o- wmmd m _ od- vomd wovd- _omd v5.0- _amd onmd- 232
aovd _..:Ncog ohvd ***cmm._ movd 1.1—om.— thd :32.— o _ v.9 39o hoagmmﬁﬁgx
3nd vmmd- Nmmd oomd- mmmd wmod- wand owed- owmd mmmd- _oonom nwmm
med 80o- wmmd v3.0- 3 md mvod- oomd Sad. and mood- 75:5 2352
mwmd Imago Nomd *owmd wcmd cited wmmd sonwd aomd .36 32.5.5 mime;
.m.m E000 .m.m .&BU .m.m .tooU .m.m “coco .m.m $80
3:28 85558 9 m8 no
55, nimcoua—ox B_:ow.:.=n.rgxm EBaoEsv =o_§onm=ou A w.”_:omo._._v

 

83895 :05 :o mEEon 86:0 _oonom mo 8me— 2: mo mcocmoocom .muoumoswm E anewnsm 503.63 moococoeca ”9.5 05.2.

124

Association between School Competition and Educators ’ Perceptions of the Impact of
School Choice Programs on their Practices

In order to examine whether educators’ perceptions of the impact of school choice
programs on their practices were associated with the degree of competition among
schools, two-level non-linear regressions were employed. The degree of competition was
measured by school location (gaining district or losing district), student enrollment
changes over the past ﬁve years, and the percentage of choice students. In this study, it
was assumed that educators in losing school districts and at schools with decreasing
student enrollment and smaller percentages of choice students would feel competition to
a greater extent than their counterparts.

There were two models for each outcome variable. Model 1 for each outcome
variable examined the relationship between the outcome variable and the explanatory
variables with only one control variable included in the model. The explanatory variables
were school location (gaining district or losing district), student enrollment changes over
the past ﬁve years, and the percentage of choice students. As a common control variable
for both Model 1 and Model 2, the position variable (administrator or teacher) at the
individual level was included. For Model 2, the following control variables at the school
level were added in Model 2 for each outcome variable: student/teacher ratio, school size,
local funding, the percentage of students eligible for free and reduced lunch, and the
percentage of minority students.

Level-l Model
Prob[R =1|B] = P'(1)= P(l)
Prob[R <= 218] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)
Prob[R <= 4|B] = 1.0

125

where

P(1)= Prob[Y(1)= 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(l - P'(l))] = BO + Bl*(ADMINIST)

log[P'(2)/(1 - P'(2))] = BO + Bl*(ADMINIST) + d(2)

log[P'(3)/(1 - P'(3))] = BO + B1*(ADMINIST) + d(3)
Level-2 Model

hdodell

B0 = GOO + (301 *(LOSING) + GOZ*(ENROLCHG) + GO3*(CHOICE) + U0

B1 = (310 + U1

Model 2

B0 = G00 + G01 *(LOSING) + GOZ*(ENROLCHG) + GOB*(CHOICE) +
GO4*(TSRATIO) + GOS*(SCHSIZE) + GO6*(LOFUND) + GO7*(FRL) + U0

B1 = 610 + U1

Educators’ perceptions of the impact of school choice policies on their practices
did not tend to be associated with the intensiveness of competition among schools. First,
in Model 1, educators in losing school districts were more likely than those in gaining
school districts to agree that school choice policies led them to improve their teaching or
leadership practices, rethink their curriculum and extra-activities, collaborate with their
colleagues, and improve relationships with parents (see Table 7-6). However, the
differences between gaining and losing school districts disappeared after controlling for
school socioeconomic factors (see Table 7-7). Second, no signiﬁcant associations
between educators’ perceptions and percentages of enrollment changes were observed.
Finally, in Model 1, there were positive associations between educators’ perceptions and

the percentages of choice students. However, these associations disappeared after

controlling for school socioeconomic factors.

126

A303? 05 0:0 E08005 0.:

:8 80:05; :80 00:32:: mm :03 00 E08003 00:88:: :0 was 0:9: .05—23:: comm .3.on ... ”3.on .1. :9on .12.. .082

 

 

Cod Bed 0 _ ed *wmod Sod _Nod o _ o.o god 0 :3 m _ o.o 5:052 .3 ..\o
ooo.o tamed ooo.o *wcod ooo.o o _ :6 o _ o.o wood wood m _ o.o Adm 00 .x.
_Nod wood- a _ o.o awed- omod Rod- mmod omod- o.o.o cmod- 04:65:— :0qu
N36 pod- $56 1.0— ..o- owed mood- wwcd .253 .o- wvod Load- 05m _oosom
owed mN_.o owed wocd 2.06 03.0 god m _ _.o 086 936 0:3. 05000::003m
owed a _ o.o- 3:6 0 _ o.o- 30.: N. 56. owed 29o- god 39o- 350:5 005:0 .00 .x.
Sod Cod _ “o.o Eod _ :3 wood Sod wood _ 56 Bed 0wSEU E0E=oEm 0: .x.
$06 mmmd wood mtd- chd 53d- 3:. wood. mmcd ammd- £0505 wEmoA
vmvd 31.2 W. and iiwnm; a _ vd .1305».— mmvd 3.30m.— Vwmd wood E28§EE0<

.m.m E000 .m.m .0000 .m.m E000 .m.m E000 .m.m E000

3:03: 00:38: 82:02:50 8:880:00 3205000:
5:5 E55323“ :0_=0E=0-E§m . . 32000:.

 

AN .000—20 80:85
:05 :o 0:8:on 008:0 33% mo 83:: 05 00 0:20:80: .m:800::m :5 5:53:50 _oonom :0050: :o:0_00mm< ”W: 030:.

A308? 05 ::0 30800:: 05
:8 805:? 8:0 00:38:: mm :03 mm 3080:: 00:38:: :0 m0: .009: $5.22: :00m .o_.own_ * ”3.on 1. :odwm 1.... .032

 

 

m :3 .58: 23 2.036 M 5o :23 0:3 :36 m :3 _..:Qod 9585 866:0 as
ooo.o Sod- ooo.o :55 wood wood- 23. 83. 83 83. 805.5 0:05:05 .8 as
:3 2:2 2 $3 _..:N. 2 X2 :38: 820 :52 83 _..:23 9052: 05.0.3
ﬁg _..:22 83 2.3.32 33 1.: :2 was 582 £3 83 assuage?
.m.m :80 0.0 :80 new .080 .m.m .080 .m.m :80
3:08: 002280 E=_=0E=0 5:830:00 320.5000:
55 :Emgzﬂox 830E=9gxm . . @2800.

 

: 3821888:
:05 :o mega—moi 008:0 _oonom mo 88:: 05 mo m:o:a0o:0m .Eoﬁoswm :5 56:09:80 _oonom 503:0: :ouﬁoomm/x 6.5 033.

127

Educators’ Perceptions of the Impact of School Choice Policies on Their Schools

To examine educators’ perceptions of the impact of school choice policies on
their schools and education in general, teachers and administrators were asked to indicate
the extent to which they agreed with each of possible contributions which school choice
policies could make. The possible contributions listed on the surveys are seen in Table 7-
8.

Teachers

According to teachers’ responses, school choice programs did not appear to
contribute much to improving school quality and education in general. As indicated in
Table 7-8, most teachers agreed that school choice programs brought about competition
among schools. However, competition among schools did not seem to cause school
improvement in terms of school programs, student achievement, reducing socioeconomic
segregation in education, serving the interests of schools for low-income or minority
students, or teacher working conditions.

Noticeably, teachers in losing school districts were more likely than teachers in
gaining school districts to disagree that school choice programs can reduce
socioeconomic segregation in education, serve the interests of schools serving low-
income or minority students, or lead schools to provide better working conditions for

teachers.

128

 

Table 7-8: Teachers’ Perceptions of the Impact of School Choice Programs on Schools
Both Gaining Losing Diff.

 

Contribute to improving school programs. 2.37 2.41 2.33 0.08
Contribute to improving student achievement. 2.28 2.32 2.23 0.09
Reduce socioeconomic segregation in education. 2.03 2.23 1.86 0.37
Bring about competition among schools. 2.93 2.94 2.93 0.01
Serve the interests of schools servmg low-income or minority 2.1 l 2.27 1.98 029
students.

Lead schools to provide better working conditions for teachers. 1.93 2.07 1.82 0.25

 

Note. l=strongly disagree, 2=disagree, 3=agree, and 4=strongly agree. Clustering of
educators within schools was ignored.

Administrators

Most administrators seemed to agree that school choice programs brought about
competition among schools (see Table 7-9). In addition, administrators appeared to agree
that school choice programs contribute to improving school programs and enhancing
student achievement. However, administrators did not seem to agree that school choice
programs reduced socioeconomic segregation in education, served the interests of schools
serving low-income or minority students, or led schools to provide better working
conditions for teachers. Administrators in gaining school districts were more likely than
those in losing school districts to agree that school choice programs reduced
socioeconomic segregation in education and served the interests of schools serving low-

income or minority students.

129

Table 7-9: Administrators’ Perceptions of the Impact of School Choice Programs on
Schools

 

Both Gaining Losing Diff.

 

School chorce programs contnbute to improvmg school 2. 67 2.67 2.67 000

programs.

School chorce programs contribute to improvmg student 2. 60 2. 40 2.80 _040
achievement.

School. chorce programs reduce socroeconomic segregation in 2.10 2.33 1.8 6 0.4.7
education.

School choice programs bring about competition among schools. 3.20 3.07 3.33 026

School choice programs serve the interests of schools serving
low-income or minority students.

School choice programs lead schools to provide better working
conditions for teachers.

1.97 2.20 1.73 0.47

2.20 2.33 2.07 0.26

 

Note. l=strongly disagree, 2=disagree, 3=agree, and 4=strongly agree. Clustering of
educators within schools was ignored.

Teachers vs. Administrators

Administrators were more likely than teachers to think that school choice
programs could contribute to school improvement. As shown in Table 7—10,
administrators were more likely than teachers to agree that school choice programs
improved school programs and student achievement, brought about competition among
schools, and led schools to provide better working conditions for teachers. However,
teachers were more likely than administrators to agree that school choice programs

served the interests of schools serving low-income or minority students.

130

Table 7-10: Differences in the Perceptions of the Impact of School Choice Programs on
Schools between Teachers and Administrators

 

 

Teach. Admi. Diff.

School choice programs contribute to improving 2 37 2 67 30
school programs. ' ' "
School choice programs contribute to improving
student achievement. 2'28 2'60 "33
School choice programs reduce socioeconomic
segregation in education. 2'03 2'10 "07
School choice programs bring about competition
among schools. 2.93 3.20 -.27
School choice programs serve the interests of

. . . . 2.1 l 1.97 .14
schools servmg low-income or minority students.
School choice programs lead schools to provide 1 93 2 20 _ 27

better working conditions for teachers.

 

Note. l=strongly disagree, 2=disagree, 3=agree, and 4=strongly agree. Clustering of
educators within schools was ignored.

As Table 7-11 indicates, the differences between administrators’ and teachers’
perceptions of the impact of school choice programs on schools were larger in losing
school districts than in gaining school districts.

Table 7-11: Differences in the Perceptions of the Impact of School Choice Programs on
Schools between Teachers and Administrators in Gainingand Losing School Districts

 

Gaining School Districts Losing School Districts

 

Teach. Admi. Diff. Teach. Admi. Diff.

 

School choice programs contribute to improving

2.41 2.67 -.25 2.33 2.67 -.34
school programs.

School choice programs contribute to improving

student achievement. 2.32 2.40 -.08 2.23 2.80 -.57

School choice programs reduce socioeconomic

. . . 2.23 2.33 -.10 1.86 1.86 .00
segregation in education.

School choice programs bring about competition

2.94 3.07 -.13 2.93 3.33 -.41
among schools.

School choice programs serve the interests of

schools serving low-income or minority students. 2'27 2‘20 '07 1°98 [-73 '25

School choice programs lead schools to provide

better working conditions for teachers. 2‘07 2'33 "26 1°82 2-07 '25

 

Note. l=strongly disagree, 2=disagree, 3=agree, and 4=strongly agree. Clustering of
educators within schools was ignored.

131

Differences between Subgroups

To examine whether educators’ perceptions of the impact of school choice
programs on their schools and education were signiﬁcantly different between subgroups
related to position (administrator or teacher), gender, tenure, the gaining or losing status
of their school districts, or school level, I conducted two-level hierarchical non-linear
models. Outcome variables were seen in Table 7-12. The outcome variables were ordinal.
Regarding explanatory variables, the following variables at the school level were
included: the gaining or losing status of school districts, middle school, and high school.
Position, tenure and gender at the individual level were included. The models are as
follows.
Level-1 Model

Prob[R =1|B] = P'(l) = P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(l) = 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(1 - P'(l))] = BO + B1*(ADMINIST) + BZ*(MALE) +

B3*(NONTENUR)

log[P'(2)/(l - P'(2))] = BO + B1*(ADMINIST) + B2*(MALE) +

B3*(NONTENUR) + d(2)

log[P'(3)/(1 - P'(3))] = BO + Bl*(ADMINIST) + B2*(MALE) +

B3*(NONTENUR) + d(3)
Level-2 Model

B0 = GOO + 601 *(LOSING) + GOZ*(MIDDLE) + GO3*(HIGH) + U0

B1 = 610 + U1

32 = G20+U2
B3 = G30 + U3

132

As shown in Table 7-12, educators in losing schools were less likely than those in
gaining schools to agree that school choice programs reduced socioeconomic segregation
in education, served the interests of schools accommodating low-income or minority
students, or led schools to provide better working conditions. There were no signiﬁcant
differences between gaining and losing school districts with regard to the other possible
contributions of school choice programs. Second, educators in middle schools were less
likely than those in elementary schools to agree that school choice programs brought
about competition among schools and led schools to provide better working conditions.
Educators in high schools were less likely to agree that school choice programs led
schools to provide better working conditions. Third, administrators were more likely than
teachers to agree that school choice programs contributed to improving school programs
and student achievement, brought about competition among schools, and led schools to
provide better working conditions. Fourth, there were no signiﬁcant differences between
male and female educators. Finally, non-tenured educators were more likely than tenured
educators to agree that school choice programs brought about competition among schools

and led schools to provide better working conditions.

133

.3090? 05 :5 0:00:85 05
:8 80555 8:0 00:38:: 3 :03 mm 0:00:85 00:03:: :5 mm: .002: .05—22: comm .o _ .cwm * Hedwm .1. 23%: I... .082

 

 

 

134

 

 

wnmd :vwwd _mmd 0mm.o _vmd .3306 85:00:02
wmmd aw _ .o- bend :bd :36 3:6 .032
_mvd :25.— wovd cmmd- wovd *o _ wd :38EEE:<
Rmd 5.52.7 8nd 032? wind omvd- _oonom :3:
End .1. :56. mhmd 2 md- Nomd ... _ mmd- 60:5 0:022
Smd «1.9—.7 3N6 siﬂwd- mwmd avmd 30.935 $504
.m.m E000 .m.m .tooU .m.m E000
M03000: :0: €25.50 3:095 3:055 .5 05005.32 ﬂoor—om w:oEw 5:50:88 Son: w:cm
:25...» 8:0: 0:305 wEEom ﬂoozom :0 30885 05 020m . . .
3 m6 mmmd ommd wvvd Nwmd mead 35:07:02
mmmd wwmd End acmd and wood- 032
_ med E _.o oavd *Nood vovd *Ngd :memmﬁﬁgw
hmmd 5:6. :36 «cad- wovd *XKd- _oosom :35
:md ocmd- mmmd vocd wwmd *wood- _oozom 0:022
womd ***mN_ . T :Nd End- vmmd Q36- $2.35 wEmoq
.m.m E000 .m.m E000 .m.m E000

8:830 :_ :ozawoaom

E25530: 309% @589:— mEﬁwoa _ooc0m w:_>oa::
288000068 833m . . .

 

33:00 E 8:00:3—
EE 0.02—om :05 :o mEEwo£ 008:0 32% :0 .009:— 05 mo m:o_E00:0m .Eoﬁosum E musewnsm 50300: 80:80:05 ”~75 2an

Associations between School Competition and Perceptions of the Impact of School
Choice Programs on Schools and Education

In order to examine whether educators’ perceptions of the impact of school choice
programs on their schools and education are associated with the intensiveness of
competition among schools, two-level non-linear regressions were employed. The degree
of competition was measured by school location (gaining district or losing district),
student enrollment changes over the past ﬁve years, and the percentage of choice students.
In this study, it was assumed that educators in losing school districts and at schools with
decreasing student enrollment and smaller percentages of choice students would feel
competition to a greater extent than their counterparts.

There were two models for each outcome variable. Model 1 for each outcome
variable examined the relationship between the outcome variable and the explanatory
variables with only one control variable included in the model. The explanatory variables
were school location (gaining district or losing district), student enrollment changes over
the past ﬁve years, and the percentage of choice students. As a common control variable
for both Model 1 and Model 2, the position variable (administrator or teacher) at the
individual level was included. For Model 2, the following control variables at the school
level were added in Model 2 for each outcome variable: student/teacher ratio, school size,
local funding, the percentage of students eligible for free and reduced lunch, and the
percentage of minority students.

Level-1 Model
Prob[R =1|B] = P'(1)= P(l)
Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)
Prob[R <= 4|B] = 1.0

135

where

P(l) = Prob[Y(1)= 1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(l)/(1 - P'(l))] = B0 + B1*(ADMINIST)

log[P'(2)/(1 - P'(2))] = B0 + Bl*(ADMINIST) + d(2)

log[P'(3)/(l - P'(3))] = BO + B1*(ADMINIST) + d(3)
Level-2 Model

Model 1

B0 = GOO + G01*(LOSING) + GOZ*(ENROLCHG) + GO3*(CHOICE) + U0

B1 == 610 + U1

Model 2

B0 = GOO + 001 *(LOSING) + G02*(ENROLCHG) + GO3*(CHOICE) +
GO4*(TSRATIO) + G05*(SCHSIZE) + GO6*(LOFUND) + GO7*(FRL) + U0

B1 = 610 + U1

Overall, educators’ perceptions of contributions of school choice programs did
not appear to be associated with the intensiveness of school competition. As shown in
Table 7-13 educators in losing school districts were less likely than those in gaining
school districts to agree that school choice programs contributed to improving student
achievement. In addition, they were more likely than those in gaining school districts to
agree that school choice programs reduced socioeconomic segregation in education and
brought about competition among schools. However, after controlling for school
socioeconomic factors, no signiﬁcant differences were observed between gaining and
losing school districts (see Table 7-14).

Second, there were no signiﬁcant associations between educators’ perceptions and

the percentage of enrollment changes in Model 1. However, after controlling for school

socioeconomic factors, educators in schools where student enrollment increased during

136

the past ﬁve years were more likely to agree that school choice programs served the
interests of schools serving low-income or minority students. Finally, educators in
schools with higher percentages of choice students were more likely than educators in
schools with smaller percentages of choice students to agree that school choice programs
contributed to improving school programs, reduced socioeconomic segregation in
education, brought about competition among schools, and provided better working
conditions for teachers. However, these differences disappeared after controlling for

school socioeconomic factors.

137

050020 0:: 0:0 00088:: 0::
:0: 80:0_:0> :88 08.5002: 00 :03 00 0:00:83 00:88:: :0 00: 000:: 0853:: :00”: .5: .5w: 0. $5.5M: .1. ”55$: 1.... 0:02

 

 

w 5.5 1.0.05.5 m 5.5 55.5 05.5 155.5 0:825 00:0:0 :0 .x.
_ 5.5 ~55 55.5 005.5 055.5 55.5 890:0 ::0::=0::m :0 .x.
20.5 5mm.5 mmmd _ 5.5- wmmd _..—5.5 000.505 0500:
0.0.0.5 2.55.5 52.5 mind- 50.5 1.03.5 E0:E:0_:_E:<
.m.m .::000 .m.m .::000 .m.w .::000
00:08: :0: 0:00:30 5:058 :0 08005.32 0:00:00

0:056:00 52:03 8:0: 0030::

w::>:00 0:00:00 :0 00880:: 0:: 0>:0m

w:0E0 5:009:00 30:0 w:::m

 

 

 

:55 :~55 m 5.5 w5.5.. 55.5 3.555 0::05_:m 00:0:0 :0 °\°

555.5 55.5 555.5 N 5.5 _ 5.5 55.5 8w:0:0 ::0::=0::m :0 o\o
nmm5 * $5.5 mmmd 3.1:..- vad 08.5 000505 5:00:
_Nvd 1555.5 55.5 55.5 «~05 Imwwd 0:0:0::0_E:::<
.m.m .5000 .m.m .::000 .m.m .::000

8:00:00 E 8:08:80
2505000800 00:53:

::0::0>0_:00 0:00:00 5:30:08:

05000:: 3080 530:9:—

 

5 08:2: 8:830

5:0 0:00:0m :0 08000:: 00:0:0 _00:0m :0 000:8: 05 :0 0:00:88: 0:0 :0203800 _00:0m 803:0: 0:0::0_0000< ”0.7:. 030:.

00

3

1

.A000020 0:: 0:0 :000:0::_ 0::
:0: 80:00:? :0::0 08308:: 00 :03 00 :000:0::: 00::000::: :0 00: _0000: 080:2: :00m .5: .5-v-0 * 50.0-v.0 .2. ”5.0%— ..cI. 0:02

 

 

05.5 55.5 m 5.5 555- w 5.5 50.5- broEZ :0 00
55.5 55.5 055.0 *m _ 5.5 050.5 50.5 AM"— :0 o\o
mN05 5055- w _ 5.5 v _ 5.5- NN05 055.0- w:_0::.: .000:
005.5 05.5- 0055 55.5 005.5 55.0. 03m 30:0m
:005 005.0- 35.5 m055- $05 :55 2:0”: :0:000:\::00::m
55.5 55.5 mm55 05.5- mm55 05.5 030005 003:0 :0 c\o
0 _ 5.5 :- 55 _ 5.5 3.055 m 5.5 55.5- 0w:0:0 ::0E_3::m_ :0 o\o
000.5 «.905- wm05 000.5- 30.5 Nmmd 0:00:05 00:00:
50.5 0 _ 00.5 0005 02.5- Nor-0.5 1&005 0:0:0::0::_E0<
.m.m .::000 .m.m .::000 .m.m .::000

0:0:000: :0: 0:003:00

0::0030 :0:-5:3: :0 0800:730—

 

 

 

139

 

0::-4:03 :0::0: 0030:: 0538 28:8 :0 0:8:0::_ 0:: 0>:0m 0:050 3:009:00 30:0 w:_:m

05.5 :- 5.0- w 5.5 55.5 05.5 50.5- :0:-5:52 :0 Ax.
055.5 _ 5.0 55.5 55.5 55.5 m 5.5 AM“: :0 o\o
55.0 55.5 mm55 95.5- 05.5 :55- 050:0: .000..—
0055 w 5.0 $55 05.5- 0055 005.5- 005 30:0m
:55 3.5.5- 355 35.5- 005.5 05.5 0:04: :0:000.C::00Em
05.5 05.5- 05.5 055.0- 555 55.5- 030005 003:0 :0 00
N 5.0 w 5.5 m 5.0 005.5 m 5.5 N55 00:0:0 ::0E=0::m :0 o\o
000.5 m0w5- 50.0 wwm0- :0:-.5 $1.5- 08.505 5:00.:
0wm5 m0 _ .5 N95 1.50.0 0.20.5 1.05.0 0:0:0::0_:_::0<
.m.m .::000 .m.m .::000 .m.m .::000

:0::00000 :_ :0::000:w00 ::0::0>0::00 ::0030 0:30:05: 0:00:08: 30:00 w::>0:0:::

0383000300 00000”: . . .

a 0:52: 508%:

0:0 030:0m :0 0E0:w0:: 003:0 30:0m :0 :0008: 0:: :0 0:00:88: 0:0 :0::000800 30:0m :003:0: 0:0::0:0000< ”3.0 030:-

Associations between Educators ’ Perceptions of School Choice Reasons and their
Perceptions of the Impact of School Choice Programs on Schools

Two-level multiple regressions were conducted to explore whether educators’
perceptions of various reasons for choosing schools were associated with their
perceptions of the impact of school choice programs on schools. Outcome variables were
the degree, perceived by teachers, of the impact of school choice programs on schools.
Major explanatory variables were the degree of importance of various reasons for
choosing a school.

Each model has one outcome variable and one explanatory variable matched with
it. In addition, each regression has two common control variables. One is the gaining or
losing status of school district at the school level. The other is position at the individual
level.

Level-1 Model

Prob[R =1|B] = P'(1)= P(l)

Prob[R <= 2|B] = P'(2) = P(l) + P(2)

Prob[R <= 3|B] = P'(3) = P(l) + P(2) + P(3)

Prob[R <= 4|B] = 1.0

where

P(1)= Prob[Y(1)=1|B]

P(2) = Prob[Y(2) = 1|B]

P(3) = Prob[Y(3) = 1|B]

log[P'(1)/(1 - P'(l))] = BO + Bl*(ADMINIST) + B2*(Choice Reasons)

log[P'(2)/(1 - P'(2))] = B0 + B1*(ADMINIST) + BZ*(Choice Reasons) + d(2)

log[P'(3)/(l - P'(3))] = BO + B1*(ADMINIST) + B2*(Choice Reasons) + d(3)

Y1 = improve school programs

Y2 = improve student achievement

Y3 = reduce socioeconomic segregation in education
Y4 = provide better working conditions

140

Choice Reasons CROO

CROl = academic programs

CR02 = extra-curricular activities

CR03 = academic performance/test scores

CR04 = socio-economic status of the community

CROS = student racial composition

CR06 = teacher quality

CRO? = teaching practices
Level-2 Model

BO = G00 + 001 *(LOSING) + UO

B1 = 610 + U1

BZ = 620 + U2

Even though educators agreed that parents chose schools based on some reasons
related to school improvement or education development, they did not appear to agree
that school choice programs led to school improvement or educational development. As
shown in Table 7-15, educators’ perceptions of the importance of each school choice
reason were not statistically associated with their expectations of their impact on schools
except for teacher quality and teaching practices. Educators who considered teacher
quality and teaching practices as more important factors were more likely to agree that
school choice programs led schools to provide better working conditions. No statistically
signiﬁcant associations between teachers’ perceptions of school choice reasons and their

perceptions of the impact of school choice programs on schools were observed with

respect to the other items.

141

000020 0:: 0:0 2:00:02: 0::

:0: 00000::0> :0::0 0000002: 00 :03 00 :000:0::: 00:88:: :0 00: _0002 385:2: :00m. 5:.0w: _.. H55W: I. ”5.5M: .12.. 0:02

 

 

090.5 1.05.— 595 0000.5 0~v5 055 3:05 035 0:0:0::0_::E0<
m0m5 3.00:5- :mm5 *0005- m0~5 1302 . T 50N5 21.0—N:- :0_::0_Q 3:00,:
00 _ .5 1.0005 80:00:: w:::000:.
5N5 .0005 3:000 8:000:-
wm _ .5 5— .5- 32000200 50:00”: ::00::m
m3 .5 05:5. 5:25:00 0:: :0 030:0 03:0:000-200m
.m.m .::000 .m.m .::000 .m.m .::000
0000000.":

0:003:00 59:03 :0::0m 0030::

E :0:0w0:w0m 0:50:000030m 00:00:

 

 

m0v5 000:5 03.0 000.5 050.5 _..0m05 0:0:0::0_::E0<
00.015 N2 .0- 05m5 mm _ .0- $005 02.5- :0505 w:_00-:
hm _ .5 0555- 8:000 :00:\00:0:::0::0: 0_E0000<
:0 _ .0 m3 .5- 00::_>_:0< :0_:0_::00-0::xm
00—5 N55- 020:w0:: 320000.00
.m.m .::000 .m.m .::000 .m.m .::000
::0E0>0_:0<

::00::m 035::—

0E0:w0:: 30:0m 03:2:—

030:0m :0 0E0:w0:.: 003:0

30:0m :0 609:: 0:: :0 0:00:38: :0:: 0:0 0:0000M 003:0 30:0m :0 0:00:88: .0:0:00:0m 803:0: 0:0_:0_0000< ”2-0 030:.

142

Discussion

In sum, the ﬁndings in this chapter did not seem to support the neoclassical
economics (N CE) perspective that educators would believe that school choice programs
could contribute to improving school quality through improving their teaching
(leadership) practices, academic programs, extra-curricular activities, collaboration with
colleagues, and relationships with parents. In addition, the ﬁndings did not support the
NCE perspective’s assumption that educators in schools under more intensive
competition would be more likely than those in schools under less intensive competition
to have these beliefs.

First, we need to note that school choice policies did not tend to motivate
educators to improve their teaching or leadership practices as the NCE perspective
expected. More than two-thirds of the teachers responded negatively when asked about
the effects of school choice policies on efforts to improve their own teaching practice and
curriculum, collaborate with colleagues, or build relationships with parents. In addition,
even though administrators tended to evaluate the impact of school choice policies on
their practices more positively than teachers, still a higher portion of the administrators
had a negative view of this impact. Seventy percent of the administrators responded
negatively to whether school choice programs led them to improve their leadership, and
only half of the administrators reported that school choice programs led them to
collaborate with colleagues or rethink school extra-curriculum activities.

Second, the intensiveness of competition among schools did not seem to inﬂuence
educators’ practices related to school improvement. Educators’ perceptions of the impact

of school choice policies on their practices did not tend to be associated with the

143

intensiveness of competition among schools. Aﬁer school socioeconomic factors were
controlled for, the gaining or losing status of school districts, the percentage of
enrollment changes, and the percentage of choice students were found to not be
associated with educators’ perceptions.

Third, school choice programs did not appear to contribute much to improving
school quality and education in general. Most teachers agreed that school choice
programs brought about competition among schools. However, competition among
schools did not seem to cause school improvement in terms of improving school
programs, increasing student achievement, reducing socioeconomic segregation in
education, serving the interests of schools for low-income or minority students, or
improving teacher working conditions.

Finally, educators’ perceptions of the contributions of school choice programs did
not appear to be associated with the intensiveness of school competition. After school
socioeconomic factors were controlled for, the gaining or losing status of school districts,
the percentage of enrollment changes, and the percentage of choice students were found
to not be associated with educators’ perceptions. In addition, even though educators
agreed that parents chose schools based on some reasons related to school improvement
or education development, they did not appear to agree that school choice programs led
to school improvement or educational development.

As I discussed in chapter 4, many educators were not aware of student enrollment
changes; thus, it would make sense that the intensiveness of competition among schools
was not associated with their perceptions of the impact of school choice programs on

their practices and schools. In addition, incentive systems were not aligned with changes

144

in student enrollment, and this also could inﬂuence educators’ perceptions of the impact
of school choice programs on their practices and schools. Educators might not be

interested in attracting more students by improving school quality.

145

CHAPTER 8.

IMPLICATIONS FOR POLICY AND FUTURE RESEARCH

In this chapter, I ﬁrst brieﬂy summarize the background of this research,
theoretical framework, and research questions. The ﬁndings regarding the research
questions are followed. I then use the ﬁndings to draw some implications for policy.
Finally, I provide some thoughts on the limitations of this study and some implications
for future research on school choice policies.

Background of This Research, Theoretical Framework, and Research Questions

Previous studies on school choice policies focused primarily on the impact of
school choice programs on students’ academic achievement, assuming that school
competition due to school choice policies would affect educators’ practices and school
programs conducive to student achievement. However, previous studies generally failed
to show a clear causal link between school choice programs and student achievement
growth (Hoxby, 2004; Shanker & Rosenberg, 1992; Witte, 1996). To address this, we
need to examine whether this popular assumption is valid in practice. To this end, this
study examined how administrators and teachers responded to school choice policies
based on interviews and surveys. Two competing economics perspectives provided useful
insights into examining the issue: the neoclassical economics (NCE) perspective
(Friedman, 1962; Samuelson, 1973) and the institutional and behavioral economics (IBE)
perspective (Hirschman, 1970; Schmid, 2004). These theories have different assumptions

with regard to human actions, the roles of a market and government (or institutions), and

146

the performance of a market. Regarding educators’ responses to school choice policies,
the NCE and IBE perspectives have different expectations.

According to the NCE perspective, parents would make rational choice based on
school quality with perfect information. Educators’ incentive systems would be aligned
with decreases or increases in student enrollment. To attract more students to their
schools, educators would do research on what parents want as rational consumers of
educational services, evaluate their own and competitors’ (competing schools’) strengths
or weaknesses, and develop effective responses. As a result, school choice programs are
believed to contribute to improving school quality.

In contrast, the IBE perspective posits that school competition could be limited
due to parents’ limited information, limited capacity with regard to information
processing, and avoidance of competition. In addition, parents might choose schools
based on status ideology or convenient location rather than factors that constitute school
quality. In terms of educators, they also have limited information, limited capacity with
regard to information processing, or beliefs inconsistent with expectations from the NC E
perspective. Consequently, they might fail to be aware of student enrollment changes,
might not be interested in increasing student enrollment, might not know how to attract
students, or show socially inappropriate market behaviors such as cost creaming (Arsen
et al., 1999; Ni & Arsen, 2008), watering down curriculum, or cream skimming
(Lacireno—Paquet et al., 2002). Therefore, school choice programs might not contribute to
improving school quality. The IBE perspective allows us to pay attention to parents’ and

educators’ beliefs, capacity, and actual behaviors.

147

Base on literature review and the NCE and the IBE perspectives, this study posed
the following four research questions.

1. How much are teachers and school administrators aware of changes in

enrollment? D0 schools have incentive systems aligned with decreases or

increases in student enrollment?

2. How are schools/districts that lose students as part of school choice policies

(i.e., losing schools/districts) different from schools/districts that gain students

(i.e., gaining schools/districts) in terms of teachers’ working conditions such as

available time and resources for teachers, leadership, and opportunities for

professional development? Are teachers’ working conditions associated with the

intensiveness of school competition?

3. What are schools doing to attract students? Is the degree of organizing activities

for attracting more students associated with school competition? Are the activities

associated with school problems or parents’ school choice reasons?

4. How do teachers and administrators perceive the impact of school choice

policies on teachers and administrators themselves and their schools? Are their

perceptions of the impact associated with the degree of competition among

schools?

Summary of the Findings and Discussion

In Chapter 4, I addressed Research Question 1. First, it is notable that many
educators did not know about recent student enrollment changes or the portion of choice
students at their schools. Even though administrators were more likely than teachers to

know about student enrollment changes and the percentage of students at their schools,

148

we could say that a large portion of administrators still did not know about them
considering their position. It is important for educators to know about student enrollment
changes in that their awareness of the changes could be a signal that they should take
appropriate actions to address them.

Second, public schools seem to fail to build market incentives for educators. In
terms of job security, salaries, and beneﬁts, more than two-thirds of teachers and
administrators responded that they were not affected by changes in student enrollment at
all or they were affected only to some extent. This implies that the current incentive
systems for educators may have limitations in motivating educators to be aware of
student enrollment changes or to improve school quality in order to attract more students.

Third, as expected from the IBE perspective, in a school where it can be assumed
that the same incentive systems are applied to all teachers, teachers perceived the degree
of the impact of student enrollment changes on their job security, salary, and beneﬁts
differently. In a given policy context, it is important that schools build incentive systems
that impact teachers’ job security, salaries, and beneﬁts enough to motivate educators to
do some actions in order to attract more students.

Finally, the ﬁndings did not support either the NCE or the IBE perspective about
the relationship between the degree of impact of student enrollment changes on teachers’
job security, salary, and benefits and the degree of their awareness of enrollment changes
and the percentage of choice students in their schools. The degree of impact of student
enrollment changes on teachers with regard to job security, salary, and beneﬁts was not
signiﬁcantly associated with their awareness of student enrollment changes and the

percentage of choice students in their schools. In addition, the degree of competition

149

among schools was not associated with educators’ awareness of student enrollment
changes and the percentage of choice students.

In Chapter 5, I addressed Research Question 2. Simple comparison of teachers’
working conditions indicated that schools in gaining school districts had better working
conditions than schools in losing school districts. However, most differences between
gaining and losing school districts (except for facilities and resources) disappeared when
school characteristics were controlled for. The intensiveness of school competition was
not signiﬁcantly associated with teachers’ working conditions except for facilities and
resources. As a result, it is difﬁcult to retain the NCE hypothesis that as competition
among schools becomes more intensive, schools will make more efforts to improve
working conditions. Instead, teachers’ working conditions appeared to be associated with
individual characteristics such as the status of tenure and school environment such as
local funding, student socioeconomic composition, and school level. In addition, the
likelihood that other latent factors not included in this study might affect teachers’
working conditions cannot be excluded.

It is notable that the rejection of the NCE hypothesis does not necessarily imply
that educators’ responses were not rational. When we look at the limited impact of
student enrollment changes on administrators’ and teachers’ job security, salaries, and
beneﬁts, it makes sense that teachers’ working conditions were not associated with the
degree of school competition. In addition, administrators might think that their schools
already have adequate working conditions for teachers; thus, they might not see the need
to improve working conditions. As rational actors, educators might have chosen other

strategies to attract more students. This issue is beyond the scope of this study. The

150

clearer point here is that we cannot expect that school choice programs will lead schools
to provide better working conditions conducive to student achievement and school
improvement.

Regarding the IBE hypothesis, this study has limitations in conﬁrming that
administrators and teachers do not know the importance of working conditions or how to
improve them. However, the large gap between teachers’ and administrators’ perceptions
regarding teachers’ working conditions suggests the possibility that administrators might
not have clear knowledge about teachers’ working conditions. The IBE perspective also
suggests that even when administrators know about the importance of improving teachers’
working conditions and how to improve them, they might not be interested in doing so.

In Chapter 6, I addressed Research Question 3. With regard to school activities
for increasing student enrollment, it is ﬁrst notable that both gaining and losing school
districts made only a limited amount of effort to research who choice students/ families
are and what types of schooling options they are seeking. We may suppose that in order
to develop appropriate strategies for attracting more students, districts or schools would
study about the factors their parents consider important when they choose schools for
their children, who exercises school choice options, and who does not. We need to
examine the reasons why districts or schools do not make serious efforts to research these
issues. There could be many reasons. Administrators might assume that they already
know what parents need and who choice students/families are. Thus, they might not feel
the necessity for such research. However, administrators’ and teachers’ knowledge about
choice students does not seem to be empirically supported in the surveys. Another

alternative explanation is that administrators did not feel that the research was necessary

151

or they did not have enough capacity to conduct the research. This issue goes beyond the
scope of this study. Future studies may need to address this concern.

Another notable point is that while losing school districts made more efforts in
advertising than gaining school districts, gaining school districts made more efforts in
improving student achievement, maintaining safe environment, and recruiting more
highly qualiﬁed teachers. In order for school choice programs to be a mechanism for
improving school quality, schools which are losing students to neighboring schools need
to address issues related to school quality. Advertising schools might be an effective way
to attract students, but that is not directly related to school improvement.

Finally, in the views of educators, the degree of organizing various activities for
attracting students did not appear to be associated with the degree of competition among
schools. Only a few activities were signiﬁcantly associated with the degree of
competition among schools but the directions of the associations were not consistent.
This could imply that student enrollment changes caused by school choice programs do
not strongly inﬂuence schools’ responses. This could be partly explained by incentive
systems which were not aligned with changes in student enrollment. From the NCE
perspective, the current incentive systems might fail to motivate educators to respond to
decreasing student enrollment. This could be also explained from the IBE perspective.
Even though they know about enrollment changes, they might be interested in other
issues instead of attracting more students, or they might believe that attracting more
students is not important.

In terms of school problems, both teachers and administrators appeared to think

that the problems related to students or parents themselves were more serious than the

152

problems related to educators themselves, school curriculum, or school policies. It would
be hard to expect that schools would make an effort to improve teaching quality or school
curriculum when they did not view these as serious problems. It is also notable that
administrators tended to consider school problems to be less serious than teachers. This
could imply that administrators might not be active in addressing schools problems even
when teachers ask them to do so.

Regarding educators’ perceptions of the reasons for parents’ school choice
decisions, we need to pay attention to teachers and administrators in losing school
districts. First, educators in losing school districts were more likely than those in gaining
school districts to consider the following two reasons as important factors: location or
convenience and student racial composition. We could conclude that when educators in
losing school districts believe that parents consider location or convenience and student
racial composition when they choose schools, they might think that they have nothing to
do in order to attract more students. In this context, it would be hard to expect them to
make an effort to improve school quality through upgrading school curriculum or
introducing new extra-activities. Another notable point is that educators in losing school
districts seem to believe that parents in their school districts are not interested in choosing
schools for their children. In this context, educators in losing school districts might not be
active in taking some actions in order to attract more students.

School activities for attracting students did not appear to be commensurate with
the degree of school problems perceived by educators. The results do not seem to support
the NCE hypothesis: as a given school problem becomes more serious, schools would

make greater efforts to address that problem. It is interesting that negative associations

153

were observed between the degree of the following school problems and their effort to
address those problems: inappropriate school curriculum, lack of parental involvement,
school order and discipline policies, and meeting A YP. There were no statistically
signiﬁcant associations with respect to the other school problems.

To explain why school activities are not commensurate with school problems, the
IBE perspective might be of help. First, administrators might not know about the
problems because they were not interested in them or they did not consider the issues as
problems to be addressed. Second, administrators might not think that they should
address school problems in order to attract more students. Finally, administrators might
not know how to address the problems. This study does not address these issues directly.

In regard to the association between school activities for attracting students and
the reasons for choosing a school, the results seem to partially support the NCE
hypothesis. Many school activities for attracting students seemed to be commensurate
with teachers’ perceptions of the importance of various reasons for choosing schools. Of
seven possible school activities included in the analyses, ﬁve school activities were
signiﬁcantly and positively associated with the degree of the importance perceived by
teachers of a given reason for choosing a school. The other two activities were positively
related but were not statistically signiﬁcant.

In Chapter 7, I addressed Research Question 4. In sum, the ﬁndings in this chapter
did not seem to support the neoclassical economics (NCE) perspective that educators
would believe that school choice programs could contribute to improving school quality
through improving their teaching (leadership) practices, academic programs, extra-

curricular activities, collaboration with colleagues, or relationships with parents. In

154

addition, the ﬁndings did not support the NCE perspective that educators in schools under
more intensive competition would be more likely than those in schools under less
intensive competition to believe that.

First, we need to note that school choice policies did not tend to motivate
educators to improve their teaching or leadership practices as the NCE perspective
expected. More than two-thirds of the teachers responded negatively when asked about
the effects of school choice policies on efforts to improve their own teaching practice and
curriculum, collaborate with colleagues, and build relationships with parents. In addition,
even though administrators tended to evaluate the impact of school choice policies on
their practices more positively than teachers, still a higher portion of the administrators
had a negative view of this impact. Seventy percent of the administrators responded
negatively to whether school choice programs led them to improve their leadership, and
only half of the administrators reported that school choice programs led them to
collaborate with colleagues and rethink school extra-curriculum activities.

Second, the intensiveness of competition among schools did not seem to inﬂuence
educators’ practices related to school improvement. Educators’ perceptions of the impact
of school choice policies on their practices did not tend to be associated with the
intensiveness of competition among schools. After school socioeconomic factors were
controlled for, the gaining or losing status of school districts, the percentage of
enrollment changes, and the percentage of choice students were found not to be
associated with educators’ perceptions.

Third, school choice programs did not appear to contribute much to improving

school quality and education in general. Most teachers agreed that school choice

155

programs brought about competition among schools. However, competition among
schools did not seem to cause school improvement in terms of school programs, student
achievement, reducing socioeconomic segregation in education, serving the interests of
schools for low-income or minority students, or teacher working conditions.

Finally, educators’ perceptions of the contributions of school choice programs did
not appear to be associated with the intensiveness of school competition. After school
socioeconomic factors were controlled for, the gaining or losing status of school districts,
the percentage of enrollment changes, and the percentage of choice students were found
to not be associated with educators’ perceptions. In addition, even though educators
agreed that parents chose schools based on some reasons related to school improvement
or education development, they did not appear to agree that school choice programs led
to school improvement or educational development.

As I discussed in chapter 4, many educators were not aware of student enrollment
changes; thus, it would make sense that the intensiveness of competition among schools
was not associated with their perceptions of the impact of school choice programs on
their practices and schools. In addition, incentive systems were not aligned with changing
student enrollment, and this also could inﬂuence educators’ perceptions of the impact of
school choice programs on their practices and schools. Educators might not be interested
in attracting more students by improving school quality.

Implications for Policy and Practice

Based on the ﬁndings, I propose three general implications for policy and

practices. First, policymakers need to be aware that school choice programs under current

contexts have limitations in improving school quality. Teachers’ perceptions of the

156

degree of organizing various activities for attracting students under school choice
programs revealed that schools and districts did not make much effort in order to increase
student enrollment. Educators tended to think that the problems related to students or
parents were more serious than the problems related to educators themselves. It would be
hard to expect schools to make an effort to improve teaching quality or the school
curriculum when they did not view these as serious problems. In addition, school
activities for attracting students did not appear to be commensurate with the degree of
school problems perceived by educators.

Second, policymakers should make more effort to research who choice students
and families are and what types of schooling options they are seeking. For example, in
order to develop appropriate strategies for attracting more students, districts or schools
need to know what factors parents consider important when they choose schools for their
children, who exercises school choice options, and who does not. This does not imply
that school systems try to meet parents’ needs uncritically because parents’ needs might
be in conﬂict with the goals of public education. For instance, when parents choose
schools based on status ideology, educators could limit parental choice or encourage
parents to choose based on school quality.

Third, policymakers need to pay close attention to individual differences among
educators. Educators do not attach the same meaning to the same incentive or sanctions.
For example, in a school where it can be assumed that the same incentive systems are
applied to teachers in the school, teachers perceived the degree of impact of student
enrollment changes on their job security, salaries, and beneﬁts differently. Therefore, in a

given policy context, it is important that schools build incentive systems that impact

157

teachers’ job security, salaries, and beneﬁts enough to motivate educators to take some
actions in order to attract more students. It is also notable that there might be a difference
between administrators and teachers with regard to making sense of school problems or
issues. For instance, administrators might think that their schools already have adequate
working conditions for teachers. However, teachers might not think so.

Finally, policymakers need to pay attention to educators’ capacity to analyze and
develop strategies along with physical resources. This study did not focus on whether
educators have enough capacity to analyze and develop strategies, but the ﬁnding that
educators did not make much effort to research who choice students/families were
implies that educators might not have the capacity.

Limitations of the Study and Implications for Future Research

There are several limitations of this study which can be addressed by future
research. The research ﬁndings in this study show the limitations of this study and
suggest a need for further studies. First, this study showed that educators did not know a
lot about student enrollment changes and the percentage of choice students within their
schools. To explain this, this study examined the association between incentive systems
for educators and their awareness. However, the association was found to be statistically
insigniﬁcant. This study did not focus on examining the reasons why educators were not
aware of enrollment changes or the percentage of choice students or why their awareness
was not associated with the incentive systems. Further studies need to address these
issues.

Second, this study found that schools and districts did not make much effort to

attract more students and that there were no signiﬁcant associations between activities for

158

attracting more students and the intensiveness of competition among schools. However,
the reasons for these ﬁndings were not fully examined in this study. Incentive systems for
educators which were not aligned with changes in student enrollment could explain this,
but there might be other reasons. For example, educators might not have enough capacity
to develop appropriate strategies for attracting more students or schools might not have
enough resources to do these. Future studies need to explore these reasons.

Third, educators tended to consider the problems related to students or parents
more serious than the problems related to educators themselves. In addition, school
activities for attracting students did not seem to be commensurate with the seriousness of
school problems perceived by educators. Many reasons could explain these ﬁndings. For
example, educators might not know about the problems or they might place more
emphasis on other issues. Future studies need to address these reasons in order to give
practical implications to policy and practices.

Finally, this study found that educators tended to evaluate the effects of school
choice programs on their practices and schools negatively. However, this study did not
address the reasons why educators viewed school choice programs this way. Research
that examines these reasons would be beneﬁcial for policymakers and others.

There were also some technical limitations with this study, particularly
concerning the generalizability of the ﬁndings. This study tried to represent public
schools which were affected by school choice programs by intentionally recruiting public
schools which varied in the portion of choice students and the student enrollment changes.
In addition, the response rates were adequate to address the research questions, with a

65.3% for the administrator survey and a 60.1% for the teacher survey from 30 schools in

159

the state of Michigan. However, this study was limited geographically since it was
conducted in one state (Michigan). Schools and districts in other states might have
different policy contexts, which might contribute to different responses by educators to
changes in enrollment and school choice policies. Therefore, it is important to consider
the policy contexts when we interpret the research ﬁndings. Additionally, this study was
based on educators’ perceptions, but their perceptions might not represent reality. For
example, we cannot exclude the likelihood that educators’ perceptions of the degree of
organizing activities for attracting more students might be different from reality. Another
source of problems in the surveys is social desirability bias (Hays, Hayashi, & Stewart,
1989). Educators in this study might have answered in a socially desirable, rather than in
a truthful manner. For example, educators tended to consider school problems related to
students or parents as more serious than those related to educators themselves. This could
be either true representation or the result of social desirability bias. I cannot fully exclude
the possibility of social desirability bias.

Although there are limitations with this study, the ﬁndings from this research do
add to previous research on school choice policies. First, this study builds on prior
research by focusing on school competition in traditional public schools. While there are
many studies which compared public schools with private schools (Benveniste et al.,
2003; Chubb & Moe, 1990) or public schools with charter schools (Miron & Nelson,
2002) to examine the effects of school competition, there are few research studies which
have focused on traditional public schools (Arsen et al., 1999; Ni & Arsen, 2008). Second,
this study challenges the basic assumptions of proponents of school choice policies by

addressing educators’ responses to school choice policies. According to school choice

160

proponents, in a free education market where parents choose schools for their kids and
schools make efforts to attract more students, parents will choose the most appropriate
schools for their children with perfect information about available schools in order to
maximize their interests or preferences. In addition, school systems will create market
incentives under which educators get rewards such as job security, a bonus, or a raise
when they attract more students and get sanctions or penalties such as a wage cut or loss
of their jobs when they lose students. In order to get rewards by attracting more students,
educators will make efforts to meet parents’ needs or preferences, evaluate and compare
their own programs with those of competing schools, and reduce costs (Friedman, 1962;
Hoxby, 2003). This study found that public school systems did not create market
incentive systems aligned with changes in student enrollment and that educators in public
schools did not seem to make much effort to research what parents desired for their
children’s schools.

Finally, by introducing the institutional and behavioral perspective to research on
school choice policies, this study encourages scholars of school choice policies to pay
attention to educators’ limited information, their capacity for information processing, and
their beliefs. There are many research studies on parents’ limited information, their
capacity for information processing and their beliefs with respect to school choice
policies (Holme, 2002; Martinez et al., 1996; Schneider & Buckley, 2002), but there have
been few studies on educators’ access to information, capacity for processing information,
or beliefs. Although there are many studies in other areas of education which share the
fundamental ideas of the IBE perspective (even though they did not use the term), the

implication from those ﬁndings did not apply to school choice policy studies. For

161

example, curriculum reform studies have shown educators’ limited capacity for
processing new information (Cohen & Ball, 1990; Spillane, 2004). This study addressed
educators’ information and capacity issues by showing that educators had limited
information about student enrollment changes and that they might not have appropriate

capacity to cope with changes in student enrollment.

162

APPENDICES

163

APPENDIX A: ADMINISTRATOR SURVEY INSTRUMENT

Study of How School Administrators and Teachers
Respond to School Choice Policies in Michigan
Administrator Survey

This survey is part of a research project that examines how school administrators and
teachers respond to school choice policies in several Michigan school districts.

We want you to know that:

1. We are asking you these questions to better understand how school choice policies
impact school administrators and teachers in Michigan.

2. Your name, the name of your school, the name of your district, and your responses to the
questions in this survey will be kept strictly conﬁdential among members of the research
team at Michigan State University.

3. The survey should take about 20 to 30 minutes to complete. You may skip any questions
you do not wish to answer; however, we hope that you will answer as many questions as
you can.

If you desire further information about this research study, you may contact Peter Youngs,
School Choice Study, 364 Erickson Hall, Dept. of Teacher Education, Michigan State
University, E. Lansing, MI 48824. You may also contact Dr. Youngs by email at
pyoungs@msu.edu or by phone at (517) 353-4348. Thank you very much!

Questions A1-A4 ask about your school, your role in your school, and the teachers in
your school.

Al. Your school is an . . .

Mark one
El school

Middle school
school

 

A2. Please indicate your position:

Mark one

Assistant ' ' O
Other e '

 

A3. How many students were enrolled in your school as of fall 2008?

students

 

A4. How many full-time classroom teachers were teaching your school as of fall 2008?

teachers

 

I64

Questions 81-39 ask about the teacher working conditions in your school.

Bl. To what extent do you agree with each of the following statements about the use of time

in your school?

 

 

 

 

 

 

routine administrative paperwork I am

 

 

 

 

 

 

Neither
(Strongly Disagree disagree Agree Strongly
rsagree nor agree
Mark (X) one box for each row agree
Teachers have reasonable class sizes,
affording them time to meet the educational O O O O 0
Teachers have time available to collaborate O O O O O
with their colleagues.
The non-instructional time provided for
teachers in my school is sufﬁcient. O O O O 0
Teachers are protected from duties that
interfere with their essential role of educating O O O O 0
Efforts are made to minimize the amount of O O O O O

 

BZ. To what extent do you agree with each of the following statements about your school

facilities and resources?

 

 

 

 

 

 

 

environment that is safe.

 

 

 

 

 

 

Neither
itrongly Disagree disagree Agree Strongly
lsagree nor agree
Mark (X) one box for each row agree
Teachers have sufﬁcient access to appropriate 0 O O O O
instructional materials and resources.
Teachers have sufﬁcient access to
communications technology, including 0 O O O 0
Teachers have sufﬁcient access to ofﬁce
equipment and supplies such as copy 0 O O O 0
Teachers have sufﬁcient access to O O O O O
instructional technology.
Teachers have adequate professional space to O O O O 0
work productively.
Teachers and staff work in a school 0 O O O O

 

165

 

 

B3. To what extent do you agree with each of the following statements about empowerment

in your school?

 

 

 

 

 

 

making group decisions and solving problems.

 

 

 

 

 

 

Neither
(Strongly Disagree disagree Agree Strongly
isagree nor agree
Mark (X) one box for each row agree
Teachers are centrally involved in decision- 0 O O O 0
making about educational issues.
Teachers are trusted to make sound
professional decisions about instruction. 0 O O O 0
Teachers are recognized as educational O O O O O
professionals.
In this school, we take steps to solve O O O O 0
problems.
The faculty has an effective process for O O O O O

 

B4. To what extent do you agree with the following statements about your leadership as an

administrator?

 

Mark (X) one box for each row

Strongly
disagree

Disagree

Neither
disagree
nor

Agree

Strongly
agree

 

There is an atmosphere of trust and mutual
respect within the school.

0

O

O

 

Teachers feel comfortable raising issues and
concerns that are important to them.

 

The school leadership communicates clear
expectations to students and parents.

 

The school leadership consistently enforces
rules for student conduct.

 

School leaders effectively communicate

 

Teachers are held to high professional
standards for delivering instruction.

00000

000000

00000

000000

000 O

 

Teacher performance evaluations are fair in
my school.

0

O

O

O

O

 

Teachers receive feedback that can help them
improve teaching.

 

 

 

 

 

 

 

166

 

 

B5. To what extent do you agree with the following statements about collegial support in

your school?

 

 

 

 

 

 

 

Neither
itrongly Disagree disagree Agree Strongly
isagree nor agree
Mark (X) one box for each row agree
All of the faculty are committed to helping O O O O 0
every student learn.
Most of the teachers at this school share my
beliefs and values about what the central 0 O O O 0
mission of the school should be.
Rules for student behavior are consistently
enforced by teachers in this school, even for O O O O 0
students who are not in their classes.
Teachers do favors for each other. 0 O O O 0
Teachers believe that their students have the O O O O O

ability to achieve academically.

 

 

 

 

 

 

B6. To what extent do you agree with each of the following statements about professional

development in your school?

 

 

 

 

 

 

Neither

Strongly Disagree disagree Agree Strongly

disagree nor agree
Mark (X) one box for each row agree
Sufﬁcient resources are available to allow
teachers to take advantage of professional 0 O O O 0
Professional development provides teachers
with the knowledge and skills most needed to O O O O 0
Teachers are provided opportunities to learn 0 O O O 0
from one another.
Adequate time is provided for professional 0 O O O 0

development.

 

 

 

 

 

 

167

 

 

B7. In your opinion, to what extent is each of the following considered a barrier to
the dismissal of poor-performing or incompetent teachers in this school?

 

Not at Some Moderate Great

Not sure

Mark (X) one box for each row

all

extent

extent

extent

 

Personnel policies

0

O

 

Termination decisions not upheld

O
O

O
O

 

Length of time required for termination
process

 

Effort required for documentation

 

Tight deadlines for completing
documentation

 

Tenure

 

Teacher associations or unions

 

Dismissal is too stressful and/or
uncomfortable for you

 

Difﬁculty in obtaining suitable
replacements

0 000000

0 000000

 

 

Resistance from parents

 

00 00000000

 

O

 

O

00 000000

 

00 000000

 

 

B8. To what extent is each of the following a problem in your school?

 

Mark (X) one box for each row

Not a
problem

Minor
problem

Moderate
problem

Seﬁous
Problem

 

Student tardiness

 

Student absenteeism

 

Students class cutting

 

Students dropping out

 

Student apathy (lack of engagement)

 

 

00000

00000

 

 

OOOOO

 

OOOOO

 

B9. To what extent is each of the following a problem in your school?

 

Mark (X) one box for each row

Not a
problem

Minor
problem

Moderate
problem

Senous
Problem

 

Lack of parental involvement

0

 

Poverty

 

Students come to school unprepared

 

Poor student health

 

Lack of school resources

 

Quality of instruction is poor

 

Inappropriate school curriculum

 

School order and discipline policies

 

Teacher absenteeism

 

Decrease in student enrollment

 

Increase in student enrollment

 

 

Meeting Adequate Yearly Progress (AYP)

 

OOOOOOOOOOOO

 

OOOOOOOOOOOO

 

 

OOOOOOOOOOO

OOOOOOOOOOOO

 

168

 

 

 

Questions BI 0-312 ask about your commitment to your school and your work effort as a
principal.

810. To what extent do you agree with each of the following statements?

Strongly . Strongly
. D A

Mark (X) one box for each row disagree isagree gree agree

The stress and disappointment involved in

serving as an administrator at this school 0 O O O

aren’t really worth it.

The faculty and staff at this school like
being here; I would describe them as a
satisﬁed group.

 

 

 

 

I like the way things are run at this district.

If I could get a higher paying job, I’d leave
education as soon as possible.

 

 

000 0
000 0
000 0
000 O

I think about transferring to another school.

 

I don’t seem to have as much enthusiasm
now as I did when I began my career as an
administrator.

O
O
O
O

 

I think about staying home from school 0 O O 0
because I’m just too tired to go.

 

 

 

 

 

 

 

B] 1. Including hours spent during the school day, before and after school, and on the
weekends, how many hours do you spend on ALL school-related activities during a typical full
week at this school?

hours
812. How many days per year are you required to work under your current contract?
days

Questions C1-C12 ask about school choice policies.

C 1. During this school year (2008-09), how many times have you talked with your
staff about enrollment changes in your school?

 

 

 

 

 

Mark (X) only one

None 0 6 to 8 times 0
Less than 3 times 0 More than 8 times 0
4 to 5 times 0 Not sure 0

 

 

 

 

 

169

C2. Please indicate how much enrollment has changed in your school over the past

five years (2004-05 to 2008-09).

 

 

 

 

 

 

 

 

 

Mark (X) only one

No change 0 Less than 10% decrease 0
Less than 10% increase 0 10 % to 20% decrease O
10% to 20% increase 0 20% to 30% decrease O
20% to 30% increase 0 More than 30% decrease 0
More than 30% increase 0 Don’t know 0

 

 

 

C3. What percentage of the students in your school in 2008-09 are students from

other school districts?

 

 

 

 

 

 

 

Mark(X) only one

None 0 20% to 30% 0
Less than 10% 0 More than 30% O
10% to 20% 0 Don’t know 0

 

 

 

C4. To what extent do you agree with each of the following statements about choice

students?

 

Mark (X) one box for each row

Strongly
disagree

Disagree

Agree

Strongly
agree

Not
Sure

 

Choice students are highly motivated.

 

Choice students are high achievers.

 

Choice students come from higher socio-
economic baggrounds.

 

There are no differences between choice
students and non-choice students.

 

I try to identify the choice students who are
in my school.

 

 

OOOOO

 

OOOOO

 

OOOOO

 

00000

00000

 

 

C5. To what extent have changes in student enrollment affected you with regard to

 

 

 

 

the following?
Not at Some Moderate Great
Not sure
Mark (A? one box for each row all extent extent extent
Job security 0 O O O 0
Salary 0 O O O 0
Beneﬁts 0 O O O O

 

 

 

 

 

 

 

C6. To what extent have changes in student enrollment affected your staff with

regard to the following?

 

 

 

 

 

 

 

 

 

 

Not at Some Moderate Great N
ot sure
Mark (X) one box for each row all extent extent extent
Job security 0 O O O 0
Salary 0 O O O 0
Beneﬁts 0 O O O O

 

170

 

 

 

C7. When parents in your school district choose schools for their children, to what

extent do you think they consider each of the following?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not at Some Moderate Great Not sure

Mark (X) one box for each row all extent extent extent

Location or convenience O O O O O
CSiafe. environment (school order and O O O O O

rscrpline)

Socio-economic status of the community 0 O O O 0
School facilities 0 O O O 0
Academic performance/test scores 0 O O O 0
Academic programs 0 O O O O
Extra-curricular activities 0 O O O 0
Teacher quality 0 O O O Q
School resources 0 O O O 0
School reputation and history 0 O O O 0
Student racial composition 0 O O O 0
Teaching practices 0 O O O O

 

 

 

 

 

 

C8. When your school has attracted students from or lost students to neighboring
schools, to what extent have there been differences between your school and the
neighboring schools with regard to each of the following?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not at Some Moderate Great Not sure

Mark (X) one box for each row all extent extent extent

Location or convenience O O O O 0
Safe environment (school order and O O O O O
Socio-economic status of the community 0 O O O 0
School facilities 0 O O O 0
Academic performance/test scores 0 O O O 0
Academic Eograms O O O O O
Extra-curricular activities 0 O O O 0
Teacher quality 0 O O O Q
School resources 0 O O O 0
School reputation and history 0 O O O 0
Student racial composition 0 O O O 0
Teaching practices 0 O O O O

 

171

 

 

C9. To what extent has your school or school district done each of the following in

order to increase student enrollment?

 

Mark (X) one box for each row

Not at

Some
extent

Moderate
extent

Great
extent

Not sure

 

Advertise school

 

Invest in school facilities and resources

 

Recruit more highly qualiﬁed teachers

 

Improve school curriculum

 

Counsel out low-performing teachers

 

Improve relationships with parents

 

Maintain a safe environment (school order

 

Improve student achievement

 

Improve extra-curricular activities

 

Research who choice students/ families are
and what types of schooling options they are

 

 

O OOOOOOOOO-‘P;

 

0 000000000

 

0 000000000

 

0 000000000

 

0 000000000

 

C10. To what extent do you think the following activities can contribute to
increasing student enrollment in your school?

 

Mark (X) one box for each row

Not at
all

Some
extent

Moderate
extent

Great
extent

Not sure

 

Advertise school

 

Invest in school facilities and resources

 

Recruit more highly qualiﬁed teachers

 

Improve school curriculum

 

Counsel out low-performing teachers

 

Improve relationships with parents

 

Maintain a safe environment (school order

 

Improve student achievement

 

Improve extra-curricular activities

 

Research who choice students/families are
and what types of schooling options they are

 

 

0 000000000

 

0 000000000

 

0 000000000

 

0 000000000

0 000000000

 

 

C11. To what extent do you agree with each of the following statements about how

school choice affects you?

 

 

 

 

 

 

 

my school’s relationships with parents.

 

 

 

 

Mark (X) one box for each row 3:22:12: Disagree Agree 5:23: 1y
School choice programs lead teachers to O 0
improve my leadership.

School choice programs lead me to more

collaborate with colleagues to attract more 0 0
students.

School choice programs lead me to rethink

m school’s curriculum.

School choice programs lead me to rethink

my school’s extra-curricular activities.

School choice programs lead me to improve 0

 

172

 

 

 

C12. To what extent do you agree with each of the following statements about how
school choice affects your school?

 

 

 

 

 

 

 

 

 

 

 

 

 

Mark (X) one box for each row 3:232: Disagree Agree 8:223),
School choice programs contribute to improving school 0 O O O
proggms. ‘

School choice programs contribute to improving student 0 O

achievement.

School choice programs reduce socioeconomic segregation O O O O
in education.

School choice programs bring about competition among 0 O O 0
schools.

School choice programs serve the interests of schools 0 O O O
servinglow-income or minority students.

School choice programs lead schools to provide better 0 O O 0
working conditions for teachers.

My school has enough capacity to readily cope with O O O 0
student enrollment decreases/increases.

 

 

Questions D1-D8 ask about your background information.

D1. How many years have you been working in this school, including 2008-09?

Mark one
1-3

4-8
9-12
More than 13

 

D2. How many years have you worked as an administrator in K—12 schools,
including 2008-09?

Mark one
1-3

4-8

9- 12

More than 13

 

D3. How many years did you work as a full-time teacher in K-12 schools?

Mark one
1-3

4-8
9-12
More than 13

 

173

D4. Were you transferred to this school in the past three years (i.e., since the
beginning of the 2006—07 school year)?

 

Mark (X) only one
Yes 0
No (Go to D5 below) 0

 

 

 

 

 

 

D4-1. If so, were you transferred to this school because of declining enrollments in
your previous school?

 

Mark (X) only one
Yes 0
No (Go to D5 below) 0

 

 

 

 

 

 

D4-2. If so, were the enrollment decreases in your previous school due to school
choice policies?

 

Mark (X) only one
Yes
No

 

 

 

OO

 

 

 

D5. What is the highest degree you have earned?

Mark one
Bachelor’s

Master’s
Doctoral
Other ease

 

D6. Please indicate your gender:

 

Mark (X) only one
Female
Male

 

 

OO

 

 

 

 

D7. Are you Hispanic?

 

Mark (X) only one
Yes

No

 

 

 

OO

 

 

 

D8. Please indicate your race/ethnicity:

 

Mark (X) only one
Caucasian

African American
Asian/Paciﬁc Islander
Native American

Other (please specrjy)

 

 

 

 

OOOOO

 

 

 

 

 

 

174

El. So that we can record your participation in this research study, please provide
your full name and the name of your school below. Your name, the name of your
school, and the name of your district will not be used in any data analysis, reports
or publications.

Your full name: Your school name:

 

 

If you would like to make any comments in response to any of the survey questions,
please do so here:

 

 

 

 

T hank you for your time!

Please return the completed survey by in the enclosed envelope to:

 

Peter Youngs, School Choice Study

364 Erickson Hall

Michigan State University, College of Education
E. Lansing, MI 48824

If you have any questions about this study, please contact:

Peter Youngs, Assistant Professor Wang Jun Kim, PhD Student
Michigan State University Michigan State University
(517)353-4348 (517)505-6762
pyoungs@msu.edu kimwangl @msu .edu

175

APPENDIX B: TEACHER SURVEY INSTRUMENT

Study of How School Administrators and Teachers
Respond to School Choice Policies in Michigan
Teacher Survey

This survey is part of a research project that examines how school administrators and
teachers respond to school choice policies in several Michigan school districts.

We want you to know that:

I. We are asking you these questions to better understand how school choice policies
impact teachers and school administrators in Michigan.

2. Your name, the name of your school, the name of your district, and your responses to
the questions in this survey will be kept strictly conﬁdential among members of the
research team at Michigan State University.

3. The survey should take about 20 to 30 minutes to complete. You may skip any questions
you do not wish to answer; however, we hope that you will answer as many questions as
you can.

If you desire further information about this research study, you may contact Peter Youngs,
School Choice Study, 364 Erickson Hall, Dept. of Teacher Education, Michigan State
University, E. Lansing, MI 48824. You may also contact Dr. Youngs by email at
pyoungs@msu.edu or by phone at (517) 353-4348. Thank you very much!

Questions A1-A4 ask about your school and your roles in your school.

Al. Which statement best describes your primary role at your school?

 

Mark (X) only one

 

I teach the same class of students all or most of the day in multiple subjects

 

I teach one class of students in one subject

 

I teach several classes of students in one subject

 

I teach several classes of students in one grade

 

I serve as a classroom aide or assistant teacher

 

 

000000

Other (please specify):

 

 

 

A2. Which of the following non-teaching duties do you have at your school?

 

Mark (X) all that apply

 

Administrator

 

Guidance counselor

 

School reform/ improvement coach or facilitator

 

Program coordinator

 

Master/mentor teacher or teacher consultant

 

I do not have any non-teaching duties

 

 

OOOOOOO

 

 

Other (please specify):

 

A3. During your most recent full week of teaching at this school, what was the total number
of students enrolled in the classes you taught?

176

students

 

A4. During your most recent full week of teaching at this school, what was the average

number of students you taught at any one time?

students

 

Questions BI-B8 ask about the teacher working conditions in your school.

Bl. To what extent do you agree with each of the following statements about the use of time

in your school?

 

 

 

 

 

 

required to do.

 

 

 

 

 

 

Stron 1 Neither Stron l
. g y Disagree disagree Agree g y
disagree nor agree agree
Mark (X) one box for each row
Teachers have reasonable class sizes,
affording them time to meet the educational O O O O 0
needs of all students.
Teachers have time available to collaborate O O O O O
with their colleagues.
The non-instructional time provided for O O O O 0
teachers in my school is sufﬁcient.
Teachers are protected from duties that
interfere with their essential role of educating O O O O 0
Efforts are made to minimize the amount of
routine administrative paperwork I am 0 O O O O

 

82. To what extent do you agree with each of the following statements about your school

facilities and resources?

 

 

 

 

 

 

 

 

environment that is safe.

 

 

 

 

 

Strongly Neither Strongly
. Disagree disagree Agree

disagree nor agree agree
Mark (X) one box for each row
Teachers have sufﬁcient access to appropriate 0 O O O O
instructional materials and resources.
Teachers have sufﬁcient access to
communications technology, including 0 O O O 0
phones, faxes and email.
Teachers have sufﬁcient access to ofﬁce
equipment and supplies such as copy 0 O O O 0
machines, paper, chalk, etc.
Teachers have sufﬁcient access to O O O O O
instructional technology.
Teachers have adequate professional space to O O O O 0
work productively.
Teachers and staff work in a school 0 O O O O

 

177

 

 

B3. To what extent do you agree with each of the following statements about empowerment

in your school?

 

 

 

 

 

 

 

making group decisions and solving problems.

 

 

 

 

 

Neither
itrongly Disagree disagree Agree Strongly
isagree nor agree
Mark (X) one box for each row agree
Teachers are centrally involved in decision— 0 O O O 0
making about educational issues.
Teachers are trusted to make sound
professional decisions about instruction. 0 O O O 0
Teachers are recognized as educational O O O O O
professionals.
In this school, we take steps to solve O O O O 0
problems.
The faculty has an effective process for O O O O O

 

B4. To what extent do you agree with the following statements about leadership in your

school?

 

Mark (X) one box for each row

Strongly
disagree

Disagree

Neither
disagree
nor
agree

Agree

Strongly
agree

 

There is an atmosphere of trust and mutual
respect within the school.

0

 

Teachers feel comfortable raising issues and
concerns that are important to them.

 

The school leadership communicates clear
expectations to students and parents.

 

The school leadership consistently enforces
rules for student conduct.

OOOO

 

School leaders effectively communicate
policies.

00000

O

 

Teachers are held to high professional
standards for delivering instruction.

0

O

 

Teacher performance evaluations are fair in
my school.

OOOOOOO

0000000

000000

0

 

Teachers receive feedback that can help them
improve teaching.

 

O

 

O

 

O

 

 

 

 

178

 

 

B5. To what extent do you agree with the following statements about collegial support in

your school?

 

 

 

 

 

 

 

 

ability to achieve academically.

 

 

 

 

Neither
gtrongly Disagree disagree Agree Strongly
isagree nor agree
Mark (X) one box for each row agree
There is an atmosphere of trust and mutual O O
respect amorg teachers.
All of the faculty are committed to helping O 0
every student learn.
Most of my colleagues share my beliefs and
values about what the central mission of the O O O O 0
school should be.
Rules for student behavior are consistently
enforced by teachers in this school, even for O O O 0
students who are not in their classes.
Teachers do favors for each other. 0 O O 0
Teachers believe that their students have the O O O O O

 

 

B6. To what extent do you agree with each of the following statements about professional

development in your school?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Neither
ztrongly Disagree disagree Agree Strongly

isagree nor agree
Mark (X) one box for each row agree
Sufﬁcient resources are available to allow
teachers to take advantage of professional 0 O O O 0
Professional development provides teachers
with the knowledge and skills most needed to O O O O 0
Teachers are provided opportunities to learn 0 O O O 0
from one another.
Adequate time is provided for professional 0 O O O 0
development.

B7. To what extent is each of the following a problem in your school?
Not a Minor Moderate Serious

Mark (X) one box for each row problem problem problem Problem
Student tardiness O O O 0
Student absenteeism O O O 0
Students class cutting O O O 0
Students dropping out 0 O O 0
Student apathy (lack of engagement) 0 O O O

 

 

 

 

 

 

179

 

 

 

BS. To what extent is each of the following a problem in your school?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not a Minor Moderate Serious
Mark (A? one box for each row problem problem problem Problem
Lack of parental involvement 0 O O 0
Poverty 0 0
Students come to school unprepared O O O 0
Quality of instruction is poor 0 O O 0
Lack of school resources 0 O O 0
Poor teaching quality 0 O O O
Inappropriate school curriculum 0 O O 0
School order and discipline policies 0 O O 0
Teacher absenteeism O O O O
Decrease in student enrollment O O O 0
Increase in student enrollment O O O 0
Meeting Adequate Yearly Progress (AYP) O O O O

 

Questions CI-C 7 ask about your teaching practices and your perceptions of your
school.

C1. Please indicate the extent to which each statement describes you as a teacher in relation
to students:

 

 

 

 

 

 

 

 

 

 

 

 

 

N“. at A little 30m?“ A lot Not
all like like me at hke like me sure
Mark (X) one box for each row me me
I solve all problems that students raise. O O O O O
I respect students’ choices and values. 0 O O O O
I effectively use instructional time to teach
students according to their needs. 0 O O O O
I know what each student is doing. 0 O O O 0
Order and discrpline come ﬁrst in my 0 O O O O
classroom.
I set high standards for student performance. 0 O O O O

 

 

C2. To what extent do you agree with each of the following statements about student ability
and learning?

 

 

 

 

 

 

 

 

 

 

 

Not at Some Moderate Great Not sure
Mark (X) one box [or each row all extent extent extent
Student ability is more fixed than variable. 0 O O O 0
Success in learning is directly related to the
amount of effort a student is willing to O O O O O
expend.
I believe I can inﬂuence a student’s learning. 0 O O O O

 

180

C3. To what extent do you agree with each of the following statements about accountability
In education?

 

 

 

 

 

 

 

 

 

 

 

Strongly . Strongly
. D

Mark (X) one box for each row disagree isagree Agree agree
I am responsible for student achievement. 0 O O 0
Parents are responsrble for student 0 O O O
achievement.
The whole school, including all teachers,
principals, and other staff, should be 0 O O O
repponsible for student achievement.
I should change my teaching if students don’t

erform well on standardized tests. 0 O O O

 

C4. In the past 12 months, how many hours have you participated in each of the following
activities?

 

8 hours 9-16 l7-32 33 hours

 

Mark (X) one box for each row None or less hours hours or more
Professional development activities speciﬁc to

and concentrating on the content of the subject 0 O O O 0
you teach

 

Professional development activities that
focused on student discipline and management 0 O O O O
in the classroom

 

Professronal development activrties that O O O O O
focused on school improvement

 

 

 

 

 

 

 

 

C5. Including hours spent during the school day, before and after school, and on the
weekends, how many hours do you spend on ALL teaching and other school-related
activities during a typical FULL WEEK at THIS school?

hours

C6. How many hours are you required to work to receive BASE PAY during a typical
FULL WEEK at THIS school?

hours

181

C7. To what extent do you agree with each of the following statements?

 

Strongly . Strongly
. D
Mark (X) one box for each row disagree isagree Agree agree

 

The stress and disappointment involved in O O O 0
teaching at this school aren’t really worth it.

 

The teachers at this school like being here; I
would describe us a satisﬁed group.

 

I like the way things are run at this school.

 

If I could get a higher paying job, I’d leave
teaching as soon as possible.

 

I think about transferringﬁto another school.

 

I don’t seem to have as much enthusiasm as I
did when I began teaching.

 

I think about staying home from school
because I’m just too tired to go.

0 00000
0 00000
0 00000
0 00000

 

 

 

 

 

 

 

Questions Dl-Dll ask about school choice policies.

D1. During this school year (2008-09), how many times did you hear about
enrollment changes in your school from the principal or central office?

 

 

 

 

Mark (X) onb’ one

None 0 6 to 8 times 0
Less than 3 times 0 More than 8 times 0
4 to 5 times 0 Not sure 0

 

 

 

 

 

 

D2. Please indicate how much enrollment has changed in your school over the past
five years (2004-05 to 2008-09).

 

 

 

 

 

 

Mark (X) onb’ one

No change 0 Less than 10% decrease 0
Less than 10% increase 0 10% to 20% decrease O
10% to 20% increase 0 20% to 30% decrease O
20% to 30% increase 0 More than 30% decrease C
More than 30% increase 0 Don’t know 0

 

 

 

 

 

 

D3. What percentage of the students in your school in 2008-09 are choice students
from other school districts?

 

 

 

 

 

 

Mark (X) only one

None 0 20% to 30% 0
Less than 10% 0 More than 30% O
10% to 20% 0 Don’t know 0

 

 

 

 

182

D4. To what extent do you agree with each of the following statements about choice

 

 

 

 

 

 

students?
Strongly . Strongly Not

Mark (X) one box for each row disagree Disagree Agree agree sure
Choice students are highly motivated. O O O O 0
Choice students are high achievers. O O O O 0
Choice students come from higher socio- O O O O 0
economic backgrounds. -
There are no differences between choice
students and non-choice students. 0 O O O O
I try to identify the choice students who are in O O O O O
my classes.

 

 

 

 

 

 

 

D5. To what extent have changes in student enrollment affected you with regard to

 

 

 

 

the following?
Not at Some Moderate Great Not sure
Mark (X) one box for each row all extent extent extent
Job security 0 O O O 0
Salary 0 O O O ' 0
Beneﬁts 0 O O O O

 

 

 

 

 

 

 

D6. When parents in your school district choose schools for their children, to what
extent do you think they consider each of the following?

 

Mark (X) one box for each row

Nmmw

Some
extent

Moderate
extent

Great
extent

Not sure

 

Location 01' convenience

O

O

O

O

O

 

Safe environment (school order and
discipline)

 

Socio-economic status of the

 

School facilities

 

Academic performance/test scores

 

Academic programs

 

Extra-curricular activities

 

Teacher quality

 

School resources

 

School reputation and history

 

Student racial conmosition

 

 

 

Teaching practices

0000000000 0

 

0000000000 0

 

 

0000000000 0

 

0000000000 0

0000000000 0

 

183

 

 

 

D7. When your school has attracted students from or lost students to neighboring
schools, to what extent have there been differences between your school and the

neighboring schools?

 

Mark (X) one box for each row

Not at all

Some
extent

Moderate
extent

Great
extent

Not sure

 

Location or convenience

O

O

O

O

O

 

Safe environment (school order and
discipline)

 

Socio-economic status of the

 

School facilities

 

Academic performance/test scores

 

Academic programs

 

Extra-curricular activities

 

Teacher quality

 

School resources

 

School reputation and history

 

Student racial composition

 

 

Teaching practices

 

OOOOOOOOOO

 

0000000000 0

 

0000000000 0

 

0000000000 0

 

0000000000 0

 

D8. To what extent has your school or school district done each of the following in
order to increase student enrollment?

 

Mark (X) one box for each row

Not at all

Some
extent

Moderate
extent

Great
extent

Not sure

 

Advertise school

0

 

Invest in school facilities and

00

 

Recruit more highly qualiﬁed teachers

 

Improve school curriculum

 

Counsel out low-performingteachers

 

Improve relationships with parents

 

Maintain a safe environment (school
order and discipline)

 

Improve student achievement

 

Improve extra-curricular activities

 

Research who choice students/families
are and what types of schooling
options they are seeking

 

 

O OO 0 000000

 

O OO 0 000000

 

O OO O 0000

 

O OO 0 000000

 

O OO 0 00000

 

184

 

 

D9. To what extent do you think the following activities can contribute to increasing

student enrollment in your school?

 

Mark (X) one box for each row

Nmmml

Some
extent

Moderate
extent

Great
extent

Not sure

 

Advertise school

 

Invest in school facilities and

 

Recruit more highly qualiﬁed teachers

 

Improve school curriculum

 

Counsel out low-performing teachers

 

Improve relationships with parents

 

Maintain a safe environment (school
order and discipline)

 

Improve student achievement

 

Improve extra-curricular activities

00 0 000000

00 0 000000

00 0 000000

 

Research who choice students/ families
are and what types of schooling
options they are seeking

 

 

O

 

O

 

O OO 0 000000

0

 

 

O OO 0 000000

 

D10. To what extent do you agree with each of the following statements about how

school choice affects you?

 

 

 

 

 

 

 

parents.

 

 

 

 

Mark (X) one box for each row 3:225:21 Disagree Agree Sggeiily
School choice programs lead me to O O O 0
improve my own teaching practice.

School choice programs lead me to

more collaborate with colleagues to O O O O
attract more students.

School choice programs lead me to O

rethink my curriculum.

School choice programs lead me to O O

rethink my extra-curricular activities.

School choice programs lead me to

improve my relationships with O O O O

 

185

 

 

D11. To what extent do you agree with each of the following statements about how
school choice affects your school?

 

 

 

 

 

Mark (X) one box for each row 3:223: Disagree Agree 8:23:61),
School choice programs contribute to O O O O
improving school programs.

School .choice programs contribute to O O O O
rmprovrng student achievement.

School choice programs reduce

socioeconomic segregation in O O O 0
education.

School choice programs bring about 0 O O 0

competition among schools.

 

School choice programs serve the
interests of schools serving low- 0 O O 0
income or minority students.

 

School choice programs lead schools
to provide better working conditions 0 O O O
for teachers.

 

My school has enough capacity to
readily cope with student enrollment O O O O
decreases/increases.

 

 

 

 

 

 

 

Questions E I -E I 1 ask about your background information.

E1. How many years have you been working in this school, including 2008-09?

Mark one
1
2-3

4-8
9-12
More than 13

 

E2. How many years have you worked as a full-time teacher in K-12 schools,
including 2008-09?

Mark one
1-3

4-8
9-12
More than 13

 

E3. Were you transferred to this school in the past three years (i.e., since the
beginning of the 2006-07
school year)?

 

Mark (X) only one
Yes
No (Go to E4 below)

 

 

 

OO

 

 

 

186

E3-1. If so, were you transferred to this school because of declining enrollments in
your previous school?

 

Mark (X) only one
Yes 0
No (Go to E4 below) 0

 

 

 

 

 

 

E3-2. If so, were the enrollment decreases in your previous school due to school
choice policies?

 

Mark (X) only one
Yes
No

 

 

 

OO

 

 

 

E4. Are you a tenured teacher?

 

Mark (X) only one
Yes 0
No 0

 

 

 

 

 

 

E5. What is the highest level of formal education you have completed?

Mark one
Bachelor’s

Master’s
Doctoral
Other ease

 

E6. Which of the following describes the teaching certiﬁcate you currently hold in
Michigan?

 

Mark (X)only one

 

Regular or standard state certiﬁcate or advanced professional certiﬁcate

 

Certiﬁcate issued after satisfying all requirements except the completion of a
probationary period

 

Certiﬁcate that requires some additional coursework, student teaching, or passage
of a test before regular certiﬁcation can be obtained

 

 

OO 00

 

I do not hold any of the above certiﬁcations in Michigan

 

 

187

E7. This school year, are you a highly Qualified Teacher (HQT) according to
Michigan’s requirements?

(Generally, to be Highly Qualiﬁed, teachers must meet requirements related to 1)
having a bachelor’s degree, 2) earning full state certiﬁcation, and 3) demonstrating
competency in the subject area(s) taught. The HQT requirement is a provision under No
Child Left Behind (NCLB))

 

Mark (X) only one
Yes
No

 

 

 

 

OO

 

 

E8. Please indicate your gender:

 

Mark (X) only one
Female
Male

 

OO

 

 

 

 

 

E9. Are you Hispanic?

 

Mark (X) only one
Yes
No

 

 

 

OO

 

 

 

E10. Please indicate your race/ethnicity:

 

Mark (X) only one
Caucasian

African American
Asian/Paciﬁc Islander
Native American
Other (please specify)

 

 

 

 

 

OOOOO

 

 

 

 

E11. Your school is an . . .

Mark one
school

Middle school
school

 

188

Fl. So that we can record your participation in this research study, please provide
your full name and the name of your school below. Your name, the name of your
school, and the name of your district will not be used in any data analysis, reports
or publications.

Your full name: Your school name:

 

 

If you would like to make any comments in response to any of the survey questions,
please do so here:

 

 

 

 

Thank you for your time!

Please return the completed survey by in the enclosed envelope to:

 

Peter Youngs, School Choice Study

364 Erickson Hall

Michigan State University, College of Education
E. Lansing, MI 48824

If you have any questions about this study, please contact:

Peter Youngs, Assistant Professor Wang Jun Kim, PhD Student
Michigan State University Michigan State University
(517)353-4348 (517)505-6762
pyoungs@msu.edu kimwangl @msu.edu

189

APPENDIX C: FACTOR ANALYSIS AND RELIABILITY CHECK FOR
TEACHERS’ WORKING CONDITIONS: TEACHER SURVEY

1. Use of Time

A3-l: Component Matrix

 

 

Component
1
B l TimeUse4 .765
B l TimeUse2 .733
B l TimeUse3 .731
B l TimeUseS .661
BlTimeUsel .582

 

A3-2: Reliability Statistics
Cronbach's Alpha N of Items
.733 5

 

 

 

2. Facilities and Resources

A3-3: Component Matrix

 

 

Component
1
B28chFacility3 .781
BZSchFacility4 .760
BZSchFacilityl .760
B2SchFacility5 .755
BZSchFacilityZ .754
BZSchFacility6 .695

 

A3-4: Reliability Statistics
Cronbach's Alpha N of Items
.841 6

 

 

 

3. Empowerment

A3-5: Component Matrix

 

 

Component
1
B3 Empower3 .883
B3 Empower4 .882
B3Empowerl .863
B3 Empower2 .852
B3Empower5 .824

 

A3-6: Reliability Statistics
Cronbach's Alpha N of Items
.912 5

 

 

 

190

4. Leadership

A3-7: Component Matrix

 

 

Component
1
B4Leadership3 .894
B4Leadership5 .89 1
B4Leadershipl .879
B4Leadership2 .846
B4Leadership4 .82 l
B4Leadership8 .789
B4Leadership7 .775
B4Leadership6 .565

 

A3-8: Reliability Statistics
Cronbach's Alpha N of Items
.926 8

 

 

 

5. Collegial Support

A3-9: Component Matrix

 

 

Component
1
BSColleSupport3 .825
BSColleSupportZ .782
BSColleSupportl .777
BSColleSupport6 .760
BSColleSupport4 .731
BSColleSupportS .666

 

A3-10: Reliability Statistics

Cronbach's Alpha N of Items
.84 l 6

 

 

 

6. Professional Development

A3-1 l: ComLonent Matrix

 

 

Component
1
B6PD1 .788
B6PD4 .783
B6PD3 .755
B6PD2 .702

 

A3-12: Reliability Statistics
Cronbach's Alpha N of Items

 

 

.751 4

 

191

APPENDIX D: INTRA-CLASS CORRELATION COEFFICIENTS (ICCS) FOR
TEACHERS’ WORKING CONDITIONS

Model

Level-1 Model

 

 

Y = BO + R
Level-2 Model
B0 = G00 + U0
Table A4: Intra-class Correlation Coefﬁcients
ICC Sigma2 Tau Total Variance
Use of Time 0.20 0.55 0.12 0.63
Facilities and Resources 0.32 0.35 0.15 0.50
Empowerment 0.49 0.46 0.44 0.90
Leadership 0.54 0.41 0.45 0.84
Collegial Support 0.39 0.29 0.17 0.47
Professional Development 0.21 0.53 0.1 3 0.63

 

192

APPENDIX E: MISSING DATA ANALYSIS FOR TEACHER SURVEY

l. Relatively more teachers at losing schools than those at gaining schools were in the
missing group whose members did not respond the most of last part of the survey (see
Tables A5-l and A5-2).

Table A5-l: Missing Group

 

 

 

 

 

School Level Total
Elementary Middle High
Gaining l Gaining 5 6 4 15
2 Losing l3 8 6 27
Total 18 14 10 42
Table A5-2: Total Sample
School Level Total
Elementary Middle High
Gaining l Gaining 42 47 59 148
2 Losing 93 51 37 181
Total 135 98 96 329

 

Note: Three teachers are missing because they did not provide school information.

2. The differences between the missing group and total sample seemed to be little. There were
almost no differences between the two groups with respect to their perceptions of teachers’
working conditions which were located in the ﬁrst part of the surveys (see Tables A5-3 and
A5-4). In addition, little differences were found between two groups regarding the impact of
student enrollment increase/decrease on teachers’ job security, beneﬁts, and salary which
were located in the last part of the survey (see Tables A5-5 and A5-6).

193

Table A5—3: Teachers’ Perceptions of Teacher Working Conditions in the Missing Group

 

 

95% Cl.
N Mean SD. SE. Lower Upper
BlTimeUse the use of time in l Gaining 13 2.85 0.78 0.22 2.38 3.31
your school 2 Losing 18 2.74 0.73 0.17 2.38 3.11
Total 31 2.79 0.74 0.13 2.52 3.06
BZSchFacility your school I Gaining 14 4.33 0.55 0.15 4.02 4.65
facilities and resources 2 Losing 18 3.35 0.67 0.16 3.02 3.68
Total 32 3.78 0.78 0.14 3.50 4.06
B3Empower empowerment in l Gaining 13 3.91 0.71 0.20 3.48 4.34
your school 2 Losing 19 2.72 0.89 0.20 2.29 3.14
Total 32 3.20 1.00 0.18 2.84 3.56
B4Leadership leadership in I Gaining 14 3.79 0.61 0.16 3.43 4.14
your SChOOI 2 Losing 18 2.84 1.07 0.25 2.31 3.37
Total 32 3.25 1.01 0.18 2.89 3.62
BSColleSupport collegial 1 Gaining 14 4.01 0.44 0.12 3.76 4.26
support in your school 2 Losing 16 3.49 0.84 0.21 3.04 3.94
Total 30 3.73 0.72 0.13 3.46 4.00
B6PD professional 1 Gaining 13 3.25 0.82 0.23 2.76 3.74
development in your suhooi 2 Losing 16 2.92 0.76 0.19 2.52 3.33
Total 29 3.07 0.79 0.15 2.77 3.37

 

Table A5-4. Teachers’ Perceptions of Teacher Working Conditions in the Total Sample

 

 

95% CI.
N Mean SD. SE. Lower Upper
BlTimeUse the use of time in 1 Gaining 145 2.98 0.79 0.07 2.85 3.1 1
your school 2 Losing 169 2.78 0.79 0.06 2.66 2.90
Total 314 2.87 0.79 0.04 2.79 2.96
B2SchFacility your school 1 Gaining 146 4.28 0.60 0.05 4.18 4.37
facilities and resources 2 Losing 170 3.62 0.69 0.05 3.52 3.73
Total 316 3.93 0.73 0.04 3.84 4.01
B3Empower empowerment in l Gaining 143 3.82 0.84 0.07 3.68 3.96
your school 2 Losing 171 3.03 0.97 0.07 2.89 - 3.18
Total 314 3.39 0.99 0.06 3.28 3.50
B4Leadership leadership in l Gaining 146 3.77 0.82 0.07 3.64 3.91
your school 2 Losing 166 3.04 0.91 0.07 2.90 3.18
Total 312 3.38 0.95 0.05 3.28 3.49
BSColleSupport collegial l Gaining 146 3.99 0.56 0.05 3.90 4.08
support in your school 2 Losing 167 3.65 0.72 0.06 3.54 3.76
Total 313 3.81 0.67 0.04 3.73 3.88
B6PD professional 1 Gaining 144 3.24 0.85 0.07 3.10 3.38
development in your school 2 Losing 169 3.08 0.73 0.06 2.97 3.19
Total 313 3.15 0.79 0.04 3.07 3.24

 

194

Table A5-5: Impact of Student Enrollment Increase/decrease on Teachers’ Job Security,
Salary, and Beneﬁts in the Missing Group

 

 

95% Cl.
N Mean SD. SE. Lower Upper
DSIMPofEnCHGl Job 1 Gaining 12 2.33 1.16 0.33 1.60 3.07
security 2 Losing 13 2.15 1.41 0.39 1.30 3.00
Total 25 2.24 1.27 0.25 1.72 2.76
DSIMPofEnCHGZ Salary 1 Gaining 12 1.67 0.99 0.28 1.04 2.29
2 Losing 11 2.27 1.27 0.38 1.42 3.13
Total 23 1.96 1.15 0.24 1.46 2.45

D51MPofEnCHG3 Beneﬁts 1 Gaining 12 1.75 0.97 0.28 1.14 2.36
2 Losing 11 2.45 1.21 0.37 1.64 3.27
Total 23 2.09 1.13 0.24 1.60 2.57

 

Table A5-6: Impact of Student Enrollment Increase/decrease on Teachers’ Job Security,
Salary, and Beneﬁts in the Total Sample

 

 

95% CI.
N Mean SD. SE. Lower Upper
D51MPofEnCHG1 Job 1 Gaining 137 1.72 0.98 0.08 1.56 1.89
security 2 Losing 165 2.16 1.25 0.10 1.97 2.35
Total 302 1.96 1.15 0.07 1.83 2.09
D51MPOfEnCHG2 Salary 1 Gaining 132 1.83 1.04 0.09 1.66 2.01
2 Losing 160 2.36 1.29 0.10 2.16 2.56
Total 292 2.12 1.21 0.07 1.98 2.26

DSIMPofEnCHG3 Beneﬁts 1 Gaining 135 1.93 1.07 0.09 1.75 2.12
2 Losing 160 2.40 1.25 0.10 2.21 2.59
Total 295 2.19 1.19 0.07 2.05 2.32

 

195

APPENDIX F: IMPACT OF ENROLLMENT CHANGES ON TEACHERS’ JOB

SECURITY, SALARIES, AND BENEFITS BY SCHOOL

Table A6-l: Impact of Enrollment Changes on Teachers’ Job Security by School

 

 

School ID N 1 Not at all 2 Some extent 3 Moderate extent 4 Great extent
101 7 42.9% 42.9% 14.3% .0%
102 6 66.7% .0% 16.7% 16.7%
103 7 57.1% 42.9% .0% .0%
104 16 37.5% 25.0% 18.8% 18.8%
105 8 50.0% .0% 25.0% 25.0%
106 12 41 .7% 25.0% 33.3% .0%
107 12 41.7% 50.0% 8.3% .0%
108 24 70.8% 8.3% 16.7% 4.2%
111 9 66.7% 11.1% 11.1% 11.1%
1 12 8 50.0% 12.5% .0% 37.5%
1 l3 13 76.9% 23.1% .0% .0%

l 14 15 66.7% 26.7% 6.7% .0%
120 5 40.0% 40.0% .0% 20.0%
121 3 33.3% 66.7% .0% .0%
122 5 40.0% 40.0% 20.0% .0%
123 3 100.0% .0% .0% .0%
124 6 83.3% 16.7% .0% .0%
125 6 83.3% .0% 16.7% .0%
126 10 70.0% .0% 20.0% 10.0%
127 11 81.8% 18.2% .0% .0%
128 13 84.6% ' 15.4% .0% .0%
129 7 57.1% 14.3% .0% 28.6%
130 11 36.4% 27.3% 18.2% 18.2%
131 13 61.5% 7.7% 7.7% 23.1 °/o
132 13 23.1% 46.2% 15.4% 15.4%
133 13 53.8% 23.1% 7.7% 15.4%
134 9 .0% 22.2% 11.1% 66.7%
135 9 11.1% 11.1% 11.1% 66.7%
136 16 12.5% 25.0% .0% 62.5%
137 12 .O% 16.7% 16.7% 66.7%

 

196

Table A6-2: Impact of Enrollment Changes on Teachers’ Salary by School

 

 

SCI?“ N 1 Not at all 2 Some extent 3 Moderate extent 4 Great extent
101 6 50.0% 16.7% 33.3% .00/0
102 6 50.0% 33.3% .0% 16.7%
103 7 57.1% 14.3% 14.3% 14.3%
104 15 73.3% 20.0% .0% 6.7%
105 8 75.0% 12.5% 12.5% .0%
106 12 41.7% 25.0% 16.7% 16.7%
107 12 58.3% 25.0% 8.3% 8.3%
108 23 43.5% 34.8% 17.4% 4.3%
l 1 l 9 44.4% 22.2% 1 1.1% 22.2%
112 6 33.3% 33.3% .0% 33.3%
113 13 46.2% 38.5% 7.7% 7.7%
114 15 46.7% 13.3% 20.0% 20.0%
120 4 50.0% 25.0% 25.0% .0%
121 2 50.0% 50.0% .0% .0%
122 4 50.0% 25.0% .0% 25.0%
123 3 100.0% .0% .0% .0%
124 6 50.0% 50.0% .0% .00/0
125 6 100.0% .0% .0% .0%
126 10 70.0% .0% 20.0% 10.0%
127 1 1 72.7% 18.2% 9.1% .0%
128 13 53.8% 30.8% 7.7% 7.7%
129 7 28.6% 28.6% .0% 42.9%
130 1 1 36.4% 18.2% 9.1% 36.4%
131 13 38.5% 23.1% 15.4% 23.1%
132 13 30.8% 15.4% 23.1% 30.8%
133 13 46.2% 30.8% .0% 23.1%
134 8 25.0% .0% 12.5% 62.5%
135 8 .0% 25.0% .0% 75.0%
136 16 .0% 6.3% 18.8% 75.0%
137 12 .0% 8.3% 25.0% 66.7%

 

197

Table A6-3: Impact of Enrollment Changes on Teachers’ Beneﬁts by School

 

School

 

ID N 1 Not at all 2 Some extent 3 Moderate extent 4 Great extent
101 6 50.0% 16.7% 33.3% .0%
102 6 50.0% 16.7% 16.7% 16.7%
103 7 42.9% 28.6% 14.3% 14.3%
104 16 75.0% 12.5% .0% 12.5%
105 8 62.5% 25.0% 12.5% .0%
106 12 33.3% 33.3% 25.0% 8.3%
107 12 41 .7% 25.0% 16.7% 16.7%
108 24 29.2% 37.5% 25.0% 8.3%
1 1 1 10 60.0% 10.0% 10.0% 20.0%
112 6 33.3% 33.3% .0% 33.3%
113 13 53.8% 30.8% 7.7% 7.7%
1 14 15 46.7% 20.0% 6.7% 26.7%
120 4 50.0% .0% 25.0% 25.0%
121 2 50.0% 50.0% .0% .0%
122 4 25.0% 50.0% .0% 25.0%
123 3 66.7% 33.3% .0% .0%
124 6 50.0% 50.0% .0% .0%
125 6 100.0% .0% .0% .0%
126 10 60.0% 10.0% 30.0% .0%
127 1 1 72.7% 18.2% 9.1% .0%
128 13 38.5% 38.5% 15.4% 7.7%
129 7 42.9% 28.6% .0% 28.6%
130 11 27.3% 18.2% 36.4% 18.2%
131 13 38.5% 23.1% 7.7% 30.8%
132 13 30.8% 15.4% 30.8% 23.1%
133 13 38.5% 46.2% .0% 15.4%
134 8 12.5% 12.5% 12.5% 62.5%
135 8 12.5% .0% 12.5% 75.0%
136 16 .0% .0% 18.8% 81.3%
137 12 0%) 83%) 250%) 66T%

 

198

APPENDIX G: EDUCATORS’ PERCEPTION OF CHOICE STUDENTS

Table A7-1: Teachers’ Perceptions of Choice Students

 

 

 

Strongly Disagree Agree Strongly NOI Missing Total

isagree agree sure
3‘0”" ““96““ are 5.7% 34.3% 19.3% 2.4% 32.2% 6.0% ”0'0

1gth motivated. %

Chi?“ Students are “31’ 6.0% 43.1% 12.3% 2.4% 30.4% 5.7% ”3,00
achievers. /o
Choice students come 100 0
from higher socio- 12.3% 44.3% 9.3% 3.0% 25.9% 5.1% 0/'
economic backgrounds. 0
There are no differences 100 0
between choice students 4.8% 24.1% 39.8% 6.9% 19.3% 5.1% 0/'
and non-choice students. 0
1 try to identify the 100 0
choice students who are 48.8% 32.5% 3.6% 1.8% 7.5% 5.7% 0/'
in my classes. 0

Note. N=322

Table A7-2: Administrators’ Perceptions of Choice Students

 

 

Sltrongly Disagree Agree Strongl NOt Missing Total
isagree y agree sure

Sigi‘iieﬁﬁiﬁfifm 0% 40.6% 28.1% 6.3% 21 .9% 3.1% 19,20
Ecﬁgcveefgudems are h‘gh 0% 46.9% 21.9% 6.3% 21.9% 3.1% 19,20
Choice students come '00 0
from higher socio- 21.9% 53.1% 15.6% .0% 6.3% 3.1% 0/'
economic backgrounds. 0
There are no differences '00 0
between choice students 12.5% 28.1% 50.0% 6.3% 3.1% 0% %'

and non-choice students.
I try to identify the 100 0
choice students who are 40.6% 25.0% 28.1% 3.1% 3.1% 0% %'

in my classes.

Note. N=32

199

REFERENCES

200

REFERENCES

Abelson, R. (1979). Differences between belief systems and knowledge systems.
Cognitive Science, 3, 355-366.

Abernathy, S. F. (2008). School choice and American democracy. Ann Arbor, MI: The
University of Michigan Press.

Armor, D. J., & Peiser, B. M. (1998). Interdistrict choice in Massachusetts. In P. E.
Peterson & B. C. Hassel (Eds), Learning from school choice. Washington, DC:
Brookings Institution Press.

Arsen, D., Plank, D. N., & Sykes, G. (1999). School choice policies in Michigan: The
rules matter: Michigan State University.

Arsen, D., Plank, D. N., & Sykes, G. (2002). School choice policies: How have they
aﬂected Michigan ’3 education system: Education Policy Center, Michigan State
University.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
New York: Longman.

Barnett, K., & McCormick, J. (2004). Leadership and individual principal-teacher
relationships in schools. Educational Administration Quarterly, 40(3), 406-434.

Benveniste, L., Carnoy, M., & Rothstein, R. (2003). All else equal: Are public and
private schools diﬂerent? New York: RoutledgeFalmer.

Berry, 8., Fuller, E., & Williams, A. (2007). Stemming the tide of teacher attrition: how
working conditions inﬂuence teacher career intentions and other key outcomes in
Arizona: Center for Teaching Quality.

Berry, 8., Smylie, M. A., & Fuller, E. (2008). Understanding teacher working conditions:
A review and look to the future. Hillsborough, North Carolina: Center for
Teaching Quality.

Brown, C. A., & Cooney, T. J. (1982). Research on teacher education: A philosophical
orientation. Journal of Research and Development in Education, 15(4), 13-18.

201

Carter, R. (1998). Mapping the mind. Berkeley: University of California Press.

Chubb, J. E., & Moe, T. M. (1988). Politics, markets, and the organization of schools.
American Political Science Review(82), 1065-1087.

Chubb, J. E., & Moe, T. M. (1990). Politics, markets, and America ’3 schools.
Washington , DC: The Brookings Institution.

Clark, C. M., & Peterson, P. L. (1986). Teachers' thought processes. In M. C. Wittrock
(Ed.), Handbook of research on teaching (pp. 255-296). New York: Macmillan.

Cohen, D. K., & Ball, D. L. (1990). Policy and Practice: An Overview. Educational
Evaluation and Policy Analysis, 12(3), 233-239.

Demoss, K. (2002). Leadership styles and high-stakes testing. Education and Urban
Society, 35(1), 111-132.

Dillman, D. A. (2007). Mail and internet surveys: The tailored design method. Hoboken,
NJ : John Wiley & Sons, Inc.

Earl, P. (1983). The economic imagination: Towards a behavioral analysis of choice.
Armonk, New York: ME. Sharpe.

Eisenhart, M. A., Shrum, J. L., Harding, J. R., & Cuthbert, A. M. (1988). Teacher beliefs:
Deﬁnitions, ﬁndings, and directions. Educational Policy, 2(1), 51-70.

Elmore, R. (1986). Choice in public education. Santa Monica, California: The Rand
Corporation.

Elster, J. (1998). Emotions and economic theory. Journal of Economic Literature,
36(March), 47-74.

Friedman, M. (1962). Capitalism and freedom. Chicago: University of Chicago Press.

F rijda, N. H. (1999). Emotions and hedonic experience. In D. Kahneman, E. Deiner & N.
Schwartz (Eds), Well-being: The foundations of hedonic psychology (pp. 190-
212). New York: Russell Sage Foundation.

202

Fuller, 8., Elmore, R., & Orﬁeld, G. (1996). Who chooses? Who loses?: Culture,
institutions, and the unequal effects of school choice. New York: Teachers
College Press.

Goddard, Y. L., Goddard, R. D., & Tschannen-Moran, M. (2007). A theoretical and
empirical investigation of teacher collaboration for school improvement and

student achievement in public emementary schools. Teachers College Record,
109(4), 877-896.

Greene, J. P., Howell, W. G., & Peterson, P. E. (1998). Lessons from the Cleveland
Scholarship Program. In P. E. Peterson & B. C. Hassel (Eds), Learning from
school choice. Washington, DC: Brookings Institution Press.

Hays, R. D., Hayashi, T., & Stewart, A. L. (1989). A ﬁve-item measure of socially
desirable response set. Educational and Psychological Measurement, 49, 629-636.

Heckathorn, D. D. (2002). Respondent-driven sampling II: Deriving valid population
estimates from chain-referral samples of hidden populations. Social Problems,
49(1), 11-34.

Henderson, A. T., & Mapp, K. L. (2002). A new wave of evidence: The impact of school,
family, and community connections on student achievement. Austin, Texas:
National Center for Family & Community Connections with Schools.

Henig, J. (1996). The local dynamics of choice: Ethnic preferences and institutional
responses. In B. Fuller, R. F. Elmore & G. Orﬁeld (Eds.), Who choose? Who
loses? .' Culture, institutions, and the unequal eﬂects of school choice New York:
Teachers College Press.

Hirschman, A. (1970). Exit, voice and loyalty. Cambridge: Harvard University Press.

Holme, J. J. (2002). Buying homes, buying schools: School choice and the social
construction of school quality. Harvard Educational Review, 72(2), 177-205.

Holmes, G. M., DeSimone, J., & Rupp, N. G. (2003). Does school choice increase school
quality? NBER Working Paper(9683).

203

Horton, N. J., & Kleinman, K. P. (2007). Much ado about nothing: A comparison of
missing data methods and software to ﬁt incomplete data regression models. The
American Statistician, 61(1), 12.

Hoxby, C. M. (2003). School choice and school productivity: could school choice be the
tide that lifts all boats? In C. M. Hoxby (Ed), The economics of school choice (pp.
287-341). Chicago: University of Chicago Press.

Hoxby, C. M. (2004). Achievement in charter schools and regular public schools in the
United States: Understanding the diﬂerences: National Institute for Child Health
and Human Development.

Kahneman, D. (1999). Objective happiness. In D. Kahneman, E. Deiner & G. Schwartz
(Eds.), Well-being: T he foundations of hedonic psychology (pp. 3-25). New York:
Russell Sage Foundation.

Kleitz, B., Weiher, G. R., Tedin, K., & Matland, R. (2000). Choices, charter schools, and
household preferences. Social Science Quarterly, 81(3), 846-854.

Lacireno-Paquet, Holyoke, T. T., Moser, M., & Henig, J. R. (2002). Creaming versus
cropping: Charter school enrollment practices in response to market incentives.
Educational Evaluation and Policy Analysis, 24(2), 145-158.

Lee, V., Croninger, R., & Smith, J. (1996). Equity and choice in Detroit. In B. Fuller, R.
Elmore & G. Orﬁeld (Eds.), Who chooses? Who loses?: Culture, institutions, and
the unequal eﬂects of school choice. New York: Teachers College Press.

Leithwood, K., & Jantzi, D. (2006). Transformational school leadership for large-scale
reform: Effects on students, teachers, and their classroom practices. School
Eﬂectiveness and School Improvement, 1 7(2), 201-227.

Leithwood, K., Louis, K. S., Anderson, S., & Wahlstrom, K. (2004). Review of research:
How leadership influences stuent learning: The Wallace Foundation.

Lubienski, C., Crane, C., & Lubienski, S. T. (2008). What do we know about school
effectiveness? Academic gains in public and private schools. Phi Delta Kappan,
89(9), 689-695.

204

Mantzavinos, C. (2001). Individuals, institutions, and markets. Cambridge, UK:
Cambridge University Press.

Marks, H. M., & Printy, S. M. (2003). Principal leadership and school performance: An
integration of transformational and instructional leadership. Educational
Administration Quarterly, 39(3), 370-397.

Martinez, V., Godwin, K., & Kemerer. (1996). Public school choice in San Antonio:
Who chooses and with what effects? . In B. Fuller, R. F. Elmore & G. Orﬁeld

(Eds.), Who chooses? Who loses?: Culture, institutions, and the unequal eﬂects of
school choice. New York: Teachers College Press.

Miron, G., & Nelson, C. (2002). What's public about charter schools? Lessons learned
about choice and accountability. Thousand Oaks, CA: Corwin Press.

Nespor, J. (1987). The role of beliefs in the practice of teaching. Journal of Curriculum
Studies, 19, 317-328.

Newmann, F. M., Smith, B., Allensworth, E., & Bryk, A. S. (2001). Instructional

program coherence: What it is and why it should guide school improvement
policy. Educational Evaluation and Policy Analysis, 23(4), 297-321.

Ni, Y., & Arsen, D. (2008, October 31, 2008). Impact of school choice competition on
traditional public schools: Which districts are pressured, how do they respond? .

Paper presented at the University Council for Educational Administration,
Orlando, Florida.

Pajares, M. F. (1992). Teachers' beliefs and educational research: cleaning up a messy
construct. Review of Educational Research, 62(3), 307-332.

Pearson, J. (1993). Myths of educational choice. Westport, Connecticut: Praeger.

Public Policy Forum. (1998). School choice accountability. Milwaukee, WI: Public
Policy Forum.

Rokeach, M. (1968). Belieﬁs, attitudes, and values: A theory of organization and change

San Francisco: Jossey-Bass.

205

Samuelson, P. A. (1973). Economics. New York: McGraw-Hill.

Schmid, A. A. (2004). Conﬂict and cooperation: Institutional & behavioral economics.
Oxford: Blackwell.

Schneider, M., & Buckley, J. (2002). What do parents want from schools? Evidence from
the internet. Educational Evaluation and Policy Analysis, 24(2), 133-144.

Schneider, M., Marschall, M., Teske, P., & Roch, C. (1998). School choice and culture
wars in the classroom: What different parents seek from education. Social Science
Quarterly, 79(3), 489-501.

Shanker, A., & Rosenberg, B. (1992). Do private schools outperform public schools? In P.
W. Cookson, Jr. (Ed.), The choice controversy. Newbury Park, California: Corwin
Press, Inc.

Simon, H. A. (1979). Rational decision making in business organizations. American
Economic Review, 69(4), 493-513.

Simon, H. A. (1982). Models of bounded rationality. Economic Analysis and Public
Policy, 1.

Simon, H. A. (1991). Organizations and Markets. Journal of Economic Perspectives, 5,
25-44.

Skinner, B. F. (1971). Beyond freedom and dignity. New York: Alfred A. Knopf.
Spillane, J. P. (2004). Standards deviation. Cambridge: Harvard University Press.

Thaler, R. H. (1985). Mental accounting and consumer choice. Marketing Science, 4(3),
199-214.

Vanourek, G., Manno, B. V., & F inn, C. E. (1998). Charter schools as seen by students,
teachers, and parents. In P. E. Peterson & B. C. Hassel (Eds.), Learning from
school choice. Washington, DC: Brookings Institution Press.

Wells, A. S. (1993). Time to choose: America at the crossroads of school choice policy
(1st ed.). New York: Hill and Wang.

206

Wells, A. S. (1996). African-American students' view of school choice. In B. Fuller, R.
Elmore & G. Orﬁeld (Eds.), Who chooses? Who loses?: Culture, institutions, and
the unequal. New York: Teachers College Press.

Witte, J. F. (1996). Who beneﬁts from the Milwaukee Choice Program? In B. Fuller, R. F.
Elmore & G. Orﬁeld (Eds.), Who chooses? Who loses?: Culture, institutions, and
the unequal effects of school choice. New York: Teachers College Press.

Young, M. D., Peterson, G. J ., & Short, P. M. (2002). The complexity of substantive
reform: A call for interdependence among key stakeholders. Educational
Administration Quarterly, 38(2), 137-175.

207