.. . r 4 N»;
Emu”?

u..l..’u
3!)". .

, {ﬂatunqmww
t‘kuﬂwxlu ﬂ. 3.; hr. .

nilu. ‘

.. ‘
.1 H
Frying

ii (1.15%

L...F:Enaﬁ ».

awn:

 

: .
\Vnrﬂs.
5.4 3. .

afar;
33:1?
:2? ..
u...

.1...
5.5 .2-
h... .

3

 

 

. .c
34".qu
‘ ‘9}.
Ln! 5
“NW...”
. ﬁnal.
. bun?» .9 349’... .
‘ .. .m a». . 3.3.1.1
Q .121. 01)...”
, .51.! I. ftf.’
1).. PI}: 1.!
2...? 1:... .112
5M0)” {$2.353
' .. I...) - Indeﬁnﬂrqlio... hﬂu.
. nth... Runs... . 30%;.”th
tints; *hbrt: N” P- n:
. 3|: .1: , .
324.2%- A. 3:: _
Chili . .... v J
4U m

DUDE)

K4

. ,~ ) r.
I 4’ \:

'. t
00;;b-J

This is to certify that the
dissertation entitled

PATHWAYS FROM NEIGHBORHOOD CHARACTERISTICS
TO ADOLESCENT WELL-BEING AND ACADEMIC
ACHIEVEMENT

presented by

su MIN oH

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in

 

 

Family and Child Ecology

%J,/ Mm:

MajorMes§6fs Signature

3//f/d 6'

MSU is an Afﬁrmative Action/Equal Opportunity Institution

LIBRARY
Michigan State
University

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

 

AR 3 1 2.008

 

a 910.8.

 

JUN 04 2011
1214 10

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/CTRC/DateDue.indd-p.15

 

——

PATHWAYS FROM NEIGHBORHOOD CHARACTERISTICS TO
ADOLESCENT WELL-BEING AND ACADEMIC ACHIEVEMENT

By

Su Min Oh

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Family and Child Ecology

2005

ABSTRACT

PATHWAYS FROM NEIGHBORHOOD CHARACTERSTICS TO ADOLESCENT
WELL-BEING AND ACADEMIC ACHIEVEMENT

By
Su Min Oh

The purpose of this study was to examine the process by which neighborhood
conditions inﬂuence parents, peers, and schools, and the combined effect of
neighborhood conditions, parents, peers, and schools on adolescents’ achievement and
subjective well-being. Using data from the 2002 wave of the Panel Study of Income
Dynamics — Child Development Supplement (PSID CDS-II), the study focused on 1178
adolescents between 12 and 17 years old. A path analysis was used to test the conceptual
model for this study.

Consistent with Leventhal and Brooks-Gum’s theoretical frameworks for linking
children and youth behaviors and neighborhood inﬂuences, the study shows that
neighborhood characteristics inﬂuenced adolescents’ subjective well being through
parental characteristics, parenting behaviors, peer interactions, and school characteristics.
Similarly, parental characteristics and parenting behaviors mediated the relation between

neighborhood characteristics and adolescent academic achievement.

TABLE OF CONTENTS

LIST OF TABLE ........................................................................................................... v
LIST OF FIGURES .............................................................................. vi
CHAPTER ONE
INTRODUCTION ......................................................................................................... 1
Purpose of Study ..................................................................................................... 3
Statement of Problem .............................................................................................. 3
Theoretical Model ................................................................................................... 4
Availability of Institutional Resources ............................................................. 4
Relationships .................................................................................................... 5
Norms/Collective Efﬁcacy ............................................................................... 5
Conceptual Model ................................................................................................... 5
Conceptual and Operational Deﬁnitions ................................................................. 8
Assumptions .......................................................................................................... 12
Overview ............................................................................................................... 1 3
CHAPTER TWO
REVIEW OF THE LITERATURE ............................................................................. 13
Neighborhoods ...................................................................................................... 14
Methodological Issues Related to Neighborhood Studies ..................................... 16
A. Perception of Neighborhood ................................................................... 16
B. Theoretical Perspectives on Neighborhood Inﬂuences ........................... 17
C. Research Designs Used to Study Neighborhoods .................................. 18
D. Limitations/Challenges to Neighborhood Research ............................... 19
Processes of Neighborhood Effects on Children ................................................... 20
Relationship between Neighborhood Characteristics and Behavioral Development
............................................................................................................................... 25
Relationship between Neighborhood Characteristics and Academic Achievement.
..................................................................... 28
Summary ............................................................................................................... 30
CHAPTER THREE ..................................................................................................... 32
METHODS .................................................................................................................. 32
Research Questions ............................................................................................... 32
Research Hypotheses ............................................................................................. 33
Participants ............................................................................................................ 34
Measures ................................................................................................................ 35
Analyses ................................................................................................................ 41
CHAPTER FOUR ........................................................................................................ 43
RESULTS .................................................................................................................... 43
Relations among the Predictor Variables .............................................................. 43
Relations between the Predictor Variables and Outcome variables ...................... 47
Hierarchical Regression Analyses of Adolescent Broad Reading Scores on the
WJ-R Achievement Test ....................................................................................... 49

iii

Hierarchical Regression Analyses for WJ-R Applied Problems Achievement Test

Scores .................................................................................................................... 52
Hierarchical Regression Analyses of Adolescent Subjective Well-being ............ 54
Results of a Path Analyses .................................................................................... 56
Summary of Results .............................................................................................. 59
CHAPTER FIVE ......................................................................................................... 64
DISCUSSION .............................................................................................................. 64
Summary of the study ........................................................................................... 64
Conclusions ........................................................................................................... 68
Limitations ............................................................................................................ 70
Theoretical Implications ........................................................................................ 71
Suggestions for future research ............................................................................. 72
BIBLOGRAPHY ......................................................................................................... 75

iv

LIST OF TABLES
Table 1 Relations among the Predictor Variables .......................................... 46
Table 2 Relations between the Predictor Variables and Outcome Variables .......... 48

Table 3 Hierarchical Regression Analyses Predicting Adolescent Broad Reading
Score .................................................................................... 51

Table 4 Hierarchical Regression Analyses Predicting Adolescent Applied Problems
......................................................................................... 53

Table 5 Hierarchical Regression Analyses Predicting Adolescent Subjective Well-
Being... 55

LIST OF FIGURES

Figure 1 Conceptual model: Pathways from neighborhood characteristics to
adolescent well-being and academic achievement ............................... 6

Figure 2 Path analyses featuring neighborhood characteristics and intervening

variables (parents, peers, school) as predictors of adolescent well-being and
academic achievement .............................................................. 58

vi

CHAPTER ONE
INTRODUCTION

Although it is widely accepted that the multiple contexts in which individuals
interact inﬂuence children and families, only recently have neighborhood effects on
adolescent outcomes been studied (Meyers & Miller, 2004). Given every generation of
adults is concerned about the conditions of its children and youth (Moore, 1999), more
attention has to be given to these largely neglected contexts of development to
supplement our understanding of factors that inﬂuence the well-being of children and
youth.

Throughout the history of development of psychological theory and methods,
social factors have drawn much attention from researchers interested in individual
differences in children’s outcomes (Pollard & Davidson, 2001); however, relatively little
has been done to investigate the ways in which factors outside of the children’s and
youth’s immediate environments inﬂuence children and adolescent outcomes. Along
with peers and school factors, the role of neighborhood inﬂuences on adolescent
outcomes has only been studied extensively in recent years.

Neighborhood refers to both the subjective (i.e., perceptions of neighborhood) and
objective (e.g., community resources, economic sources) qualities of this context which
could affect adolescents’ lives. In the existing literature, neighborhoods are described in
terms of different levels of economic resources, opportunities, community resources,
social cohesion/control, and safety (Pinderhughes, Nix, Foster, Jones, & The Conduct

Problems Prevention Research Group, 2001; Sampson, 2001). However, as noted in an

earlier review of the literature by Leventhal and Brooks-Gunn (2000), it is less than clear
how exactly neighborhoods affect adolescents’ outcomes.

While there is still much to learn about the pathways of neighborhood inﬂuences,
previous studies have suggested that neighborhood characteristics were directly related to
adolescents’ psychological adjustment and school problems (Meyers & Miller, 2004),
and adolescents living in poor neighborhoods, where most residents have little education
and have difﬁculty obtaining jobs, perceived that they have little control over their lives
and a poor chance of success (Wilson, 1991). Furthermore, the body of research that
explored the direct relations between neighborhood characteristics and adolescents’
academic achievement suggested that adolescents in disadvantaged neighborhoods
showed lower level of performance on standardized tests (Halpem-Felsher et al., 1997),
lower school grades (Dombusch, 'Ritter, & Steinberg, 1991), and lower high school
graduation rates (Brooks-Gunn, Duncan, Glebanov, & Sealand, 1993; Duncan, 1994). In
addition, neighborhood characteristics (such as SES) have been related to adolescent
aggression and conduct disorders (Aneshensel & Sucoff, 1996) as well as delinquent
behavior (Kalil & Eccles, 1998).

Some of the previous studies also suggested that neighborhood characteristics
inﬂuence other immediate environments (e.g., parents) in the adolescents’ everyday lives
(Baldwin et al., 1990; Furstenberg, Cook, Eccles, Elder, & Sameroff, 1999; Oh & Luster,
1998). Notably, Oh and Luster demonstrated that mother’s psychological well-being and
parenting behavior mediated the relation between neighborhood characteristics and

children’s academic achievement.

Recently, it has been suggested that peer characteristics mediate the relation
between neighborhoods and adolescent outcomes as well. Adolescents who live in poor
neighborhoods are at greater risk than other adolescents of afﬁliating with peers who
exhibit delinquent behavior, such as damaging property, stealing, or skipping school
(Brody, Ge, Conger, Gibbons, Murry, Gerrard, & Simons, 2001). In summary, most
research on neighborhood characteristics and adolescent outcomes focused on either
academic achievement or behavior problems and included some but not all of the
potential mediating factors in the analysis. Further research is necessary to enhance our
understanding of: (1) how neighborhoods directly inﬂuence adolescent outcomes --
speciﬁcally, both subjective (i.e., dimensions of well-being) and objective outcomes (i.e.,
standardized tests), and (2) how parents, peers, and schools mediate this association.

Purpose of Study

The purpose of this study is to examine the process by which neighborhood
conditions inﬂuence parents, peers, and schools, and the combined effect of
neighborhood conditions, parents, peers, and schools on adolescent achievement and
subjective well-being. The study will be based on a secondary analysis of the data from
the Panel Study of Income Dynamics - Child Development Supplement 11 (PSID CDS-
II).

Statement of Problem

It is widely believed that neighborhood characteristics affect adolescent outcomes
both directly and indirectly. Afﬂuent parents carefully select the neighborhoods in which
to rear their children. Yet there is limited research on the processes by which

neighborhoods inﬂuence adolescent development. This study will attempt to address this

gap in the literature. Building on previous studies, this study will attempt to answer the

question: how do neighborhoods inﬂuence adolescents’ subjective well-being as well as

objective indicators of academic achievement. Speciﬁcally, the inﬂuence of
neighborhoods will be investigated in a path model which includes direct and indirect

relations to adolescent outcomes and examines the relative importance of neighborhood

 

.- characteristics, parents’ attributes, parenting behaviors, peer attributes, and schooling
experiences in predicting these adolescent outcomes.
Theoretical Model
This study will be based on a theoretical model developed byuLeventhal and
Brooks-Gum (2000), in which they identiﬁed three theoretical frameworks for linking
children and youth behaviors and neighborhood inﬂuences. The following section
provides a very brief explanation of their model.
Availability of Institutional Resources
The existence of institutional resources in the neighborhood -- learning,
recreational, and social activities; child care; schools; medical facilities; and employment
opportunities -- may inﬂuence child and adolescent outcomes (Leventhal & Brooks-Gunn,
2000). According to Leventhal and Brooks-Gunn’s framework? the institutional resource
model may be useful for studying achievement outcomes. The presence of learning
activities (e. g., libraries, family resource center, literacy programs, and museums) may
stimulate children’s development, especially school readiness and achievement outcomes.
:Economic resources .of the neighborhood are important factors inﬂuencingparenting
behaviors as well as caregivers’ psychological well-being, and they may also inﬂuence

how caregivers perceive their neighborhoods. I‘Neighborhood resources) may also

inﬂuence adolescent outcomes through characteristics of the, school environment (e.g.,
school safety, connectedness to school). In this study, adolescent connectedness to
school will be viewed as a mediating factor in the relationship between neighborhood
characteristics and adolescent achievement.
Relationships

It is suggested that parental characteristics (e. g., mental health, irritability, and
physical health), support networks available to parents, parental behavior, and the quality
and structure of the home environment may mediate the relationship between
neighborhood characteristics and adolescent well-being. Parental characteristics not only
moderate or mediate the inﬂuence of their mental and physical health on parenting and
child outcomes, but also they are important antecedents of parental coping skills and
efﬁcacy. Thus, neighborhood characteristics may inﬂuence parental mental health and
parental behavior, which in turn, affect adolescent outcomes.
Norms/Collective Efﬁcacy

Collective efﬁcacy refers to the extent of formal and informal social connections
in the neighborhood and the degree to which residents supervise or monitor the behavior
of others in accordance with community standards including shared values of mutual trust,
safety, the willingness to intervene for the common good, and supportive childrearing by
way of supervising and monitoring children and youth. In this study, sqqiaicontrol,will
be viewed as an important factor for supervising and controlling adolescent peer groups,

which in turn, affect adolescent well-being.

Conceptual Model

The conceptual model presented in Figure 1 shows the expected relationships
among the variables of interest in this study. Speciﬁcally, the relationships between the
independent variables (neighborhood characteristics, connectedness to neighborhood),
the potential mediating variables (PCG’s self-esteem, self-efﬁcacy, and psychological
distress, parental difﬁculty, quality of the home environment, parental monitoring, peer
interactions, and connectedness to school) and the dependent variables (adolescent’s

performance on WJ-R Achievement tests and subjective well-being) are presented in a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

graphic format.

Parental characteristics

Parental difﬁculty

Self—efﬁcacy

Self-esteem

Psychological distress

Parenting behaviors

Home environment

Parental monitoring

WJ-R achievemnt test;
Parent’s perceptions of neighborhood Passage Conu‘ehension,
characteristics; Calculation
Residential stability, -
Safety Subjective well-being;
Neighbors know each other Emotional, Social, Psychological
Satisfaction V
Social control
Peers
Youth’s connectedness to am to friends
neighborhoodas reported by parent; Peer Victimization _
Participation in oormnunity activities Peers’ devrant behavror .
Peers’ academic orientatlon

 

 

 

 

 

 

 

School

Connectedness to school;
Inclusivenms, Closeness,
Happiness, and Safety at school

 

 

 

Figure 1. Conceptual model: Pathways from neighborhood characteristics to
adolescent well-being and academic achievement

Primary caregivers who perceived their neighborhoods as close-knit ones are
likely to show higher levels of self-efﬁcacy, higher levels of self-esteem, and lower levels
of psychological distress. If the neighborhood is viewed as safe and a desirable place to
rear children, a primary caregiver will be likely to perceive fewer difﬁculties with rearing
an adolescent there. If the neighborhood contributes to better adjustment on the part of
parents, the primary caregiver may provide a more cognitively stimulating home
environment, which in turn, is likely to affect adolescents’ scores on the WJ-R
achievement test. In addition, the PCG’s characteristics (i.e., self-esteem, self-efﬁcacy,
and psychological distress) and parenting behaviors (i.e., home environment and parental
monitoring) are likely to inﬂuence adolescent’s peer interaction and schooling
experiences. If parents who are distressed provide a less cognitively stimulating home
environment and monitor their children’s activity less diligently, an adolescent is more
likely to afﬁliate with peers who engage in deviant behaviors. Neighborhoods are also
likely to have an effect on the pool of peers that the adolescents can draw from. The
school characteristics may also vary by neighborhood, and adolescents who attend

schools with lower levels of safety or schools with high numbers of low achieving

/

/ 1
students may feel less connected to their school. Thustarental characteristics, peer

interactions and connectedness to school are likely to mediate the relationship between
_r_1_ei_ ghborhood characteristics and adolescent academic achievement.

Adolescents who reside in disadvantaged neighborhoods may be more likely to be
exposed to peer victimization and to peers who engage in deviant behaviors, which in
turn, may affect their experiences at school and their subjective well-being. On the other

hand, an adolescent who feels close to his or her friends may also feel connectedness to

school, which in turn, affects his or her subjective well-being. Furthermore, peer

interactions and connectedness to school, along with parental characteristics and practices,

are likely to mediate the relationship between neighborhood characteristics and

adolescents’ subjective well-being.

Conceptual and Operational Deﬁnitions

The following section provides the conceptual and operational deﬁnitions for the key

concepts in this study.

Neighborhood characteristics

Conceptual deﬁnition: Overall neighborhood characteristics including residential stability,

residential satisfaction, neighbors know each other, and neighborhood safety as reported

by the primary care giver (PCG).

Operational deﬁnition: Items measuring the primary caregiver’s perceptions about the

neighborhood, including stability, satisfaction, familiarity with neighbors, and safety,

were combined into a single scale to create an overall indicator of neighborhood quality.
__S_9_cia_lncontrol

Conceptual deﬁnition: The degree to which residents intervene for the common good and

monitor the behavior of adolescents.

Operational deﬁnition: The willingness to intervene in response to the behavior of

adolescents was measured with four items. Primary caregivers were asked how likely

neighbors would do something if someone was trying to sell drugs to children in plain

sight, if children were getting into trouble, if a child was showing disrespect to an adult,

and if a child was taking something out of a neighbor’s apartment, house, garage, car or

yard.

Connectedness to neighborhood

Conceptual deﬁnition: Participation in community activities by youth, for example,
religious services, athletic team, physical exercise, library story hour, and scouting.
Operational deﬁnition: The frequency of adolescents’ participation in each community
activity within the past 12 months as reported by caregivers was computed to create a
total score on this measure.

Parental difﬁculty

Conceptual deﬁnition: The primary caregiver’s perceptions of difﬁculties with raising
children.

Operational deﬁnition: The primary caregiver’s rating on three statements (i.e., bothered
by child, gives up life to meet the child’s needs, and angry with the child) for a CDS-II
target child. A total score from these three items was computed, with higher scores
reﬂecting greater parental perceptions of difﬁculty with their child

Primary caregiver’s Self-esteem

Conceptual deﬁnition: The primary caregivers’ evaluation of their overall self worth.
Operational deﬁnition: Self-esteem was measured with the Rosenberg Self-esteem Scale.
It is a 10-item scale with responses scored on a four-point scale. Higher scores reﬂect
higher levels of self-esteem.

Primary caregiver’s Self-efﬁcacy

Conceptual deﬁnition: The level of mastery over events in their lives reported by the

PCG.

Operational deﬁnition: The Pearlin Self-efﬁcacy Scale was used to assess PCG self-
efﬁcacy. It is a 4-item scale with responses scored on a four-point scale. Higher scores
indicate higher levels of self-efﬁcacy.

Primary caregiver’s psychological distress

Conceptual deﬁnition: The primary caregiver’s psychological state of mind or level of
distress.

Operational deﬁnition: The primary caregiver’s score on the Psychological Distress
Scales (Kessler, Andrews, Colpe, Hiripi, Mroczek, Normand, Walter, & Zaslavsky, 2002)
was used in this study. It is a 6-item scale with responses scored on a ﬁve-point scale.
Home environment

Conceptual deﬁnition: The quality of the rearing environment provided by the primary
care givers for their child.

Operational deﬁnition: TheHome Observation for Measurement of the Environment
(HOME) by Caldwell and Bradley (1984) was employed to measure parenting behavior.
It is an index of the quality of cognitive stimulation and emotional support provided in
the home. The age-appropriate items for older children (10 years old and above) were
used for this study.

Parental monitoring

Conceptual deﬁnition: The adolescent’s perceptions of parental monitoring of their
behavior and activity. For example, it includes parents’ knowledge of adolescent’s

ﬁ'iends and what adolescents do with their free time.

10

Operational deﬁnition: Perceived parental monitoring by adolescents was measured with
a 3-item scale. It collects information on parents’ knowledge (i.e., knowledge about
friends, spending money, and spending free time).

Closeness to friends

Conceptual deﬁnition: Adolescents’ perceptions of their closeness with friends.
Operational deﬁnition: Closeness to friends was measured with a 4-item scale assessing
how close adolescents feel towards their ﬁ'iends, how often they talk with their friends
about friends, future plans, and problems at school.

Peer victimization

Conceptual deﬁnition: Adolescents’ experiences related to victimization by their peers.
Operational deﬁnition: The frequency of adolescents’ victimization experiences by other
adolescents at school and in the neighborhood within the last month was measured with a
four-item scale.

Peers’ deviant behaviors

Conceptual deﬁnition: The involvement of the adolescents’ peers in antisocial behaviors.
Operational deﬁnition: Adolescent’s perceptions of the degree of engagement in
negative (e.g., disobey parents, drink alcohol) behaviors by their ﬁ'iends was measured.
Peers’ academic orientation

Conceptual deﬁnition: The involvement of adolescents’ peers in academic behaviors.
Operational deﬁnition: Adolescent’s perceptions of the degree of engagement in
academic (e.g., plan to go to college, emphasis on school work) behaviors by their ﬁiends
were measured.

Connectedness to school

11

Conceptual deﬁnition: Adolescents’ perceptions of the degree of inclusiveness, closeness,
and happiness with school, as well as safety at school.

Operational deﬁnition: Adolescents’ experiences at school in the last month were
assessed by four questions related to inclusiveness, closeness, general happiness with
school, and safety at school.

Adolescents’ academic achievement

Conceptual deﬁnition: Adolescents’ ability to perform age-appropriate cognitive tasks
relative to peers.

Operational deﬁnition: The adolescents’ scores on the Woodcock-Johnson Revised Tests
of Achievement (WI-R) (Woodcock & Johnson, 1989). The standardized scores for
broad reading scores combining Letter-Word and Passage Comprehension and the
standardized scores for Applied Problems were used.

Adolescents’ subjective well-being

Conceptual deﬁnition: Adolescents’ perceived quality of life based on their personal
values, views, and assessment of the circumstances of their lives.

Operational deﬁnition: Adolescent’s score on a measure labeled Subjective Well-Being.
It measures several dimensions of subjective well-being: Emotional Well-Being; Social
Well-Being; and Psychological Well-Being. It was adapted from the MacArthur MIDUS
_ Youth measure (MIDUS, n.d.).

Assumptions

The following underlying assumptions have been made in this study.

12

1. The scales used to measure variables in the study are valid (e. g., Rosenberg Self-
esteem Scale, Pearlin Self-efﬁcacy Scale, HOME inventory, WJ-R, MIDUS
Subjective Well-being).

2. Quality of the home environment, adolescents’ academic achievement, and subjective
well-being can be measured.

3. Primary caregivers and adolescent respondents will answer the questions honestly.

Overview
This chapter included an introduction, purpose of the study, and statement of the
problem. It also included a conceptual model which is based on a theoretical model
proposed by Leventhal and Brooks-Gunn (2000). Finally, operational deﬁnitions and
assumptions were delineated.
A review of the literature relevant to this study is presented in Chapter 11. Chapter

111 contains research questions, a description of the data set, participants, and measures

for this study. The hypotheses to be tested in this study are also included.

Methodological issues related to the study including planned analysis of the data. The

results of the analysis are presented in Chapter IV. Chapter V discusses the results and

suggests implications for theory and future research.
CHAPTER TWO
REVIEW OF THE LITERATURE
The review of literature is divided into three main parts. The ﬁrst presents an
overview of neighborhoods and the role of neighborhoods in children and youth
outcomes. This section also discusses methodological issues related to neighborhood

studies, as well as several ways in which neighborhood inﬂuences on children and youth

13

have been addressed. The second part of the review summarizes prior research on the
relationship between neighborhoods and children’s behavioral development and
academic achievement. The last section of this chapter brieﬂy describes the direction of
this study.
Neighborhoods

Families with children living in stressful neighborhoods have become a national
concern in the United States. It is more problematic for metropolitan areas and for
neighborhoods with a high concentration of people with low incomes (J argowsky &
Bane, 1990). How neighborhood characteristics, particularly poverty, affects families
and children has been a critical question in urban neighborhood and child development
research (Burton & Jarrett, 2000; J encks & Mayer, 1990; Wilson, 1987) over the last 20
years. Wilson (1987) argued that neighborhoods played an important role in the well-
being of children, especially those who live in low-income neighborhoods. He suggested
that living in neighborhoods with few jobs and a high concentration of single-parent
households may produce conditions of “social isolation”; these conditions, in turn, may
produce socialization practices and family life styles that are not associated with steady
employment (Wilson, 1991). Wilson also suggested that poor planning and organization,
little sense of personal control over events, and a lack of emphasis on school or job-
related skills of parents may inﬂuence their parenting behavior, organization of the
household, and the provision of learning experiences for their children. _However, little is
__k_nownabout how neighborhood characteristics affect maternal characteristics and
adolescent behaviors, because the primary focus of research has been on how family-

level poverty affects parents and children (Klebanov, Brooks-Gunn, & Duncan, 1994).

14

Some of the linkages between family poverty and psychological distress and

depression have been examined. Studies of the association between poverty and

-..+

psychological distress of parents suggest that low income families have to deal with a

‘. h._~_...

 

greater number of daily stresses, which may lead to poor- or impaired parenting behavior
”(McLoyd, 1990). Poor families have to deal with a greater number of daily stresses
which over time may weaken their ability to control the source of the stress and their
ability to handle subsequent stress. Thus, inability to cope with stress affects
psychological functioning, which in turn, may lead to poor parenting behavior (McLoyd
& Wilson, 1991), and child abuse (Garbarino, 1976).

Several studies have examined the ways in which conditions within high-risk,
stressful communities affect parents and child development (Brooks-Gunn, Duncan, &
Aber, 1997; Dubrow & Garbarino, 1989; Garbarino, Kostelny, & Dubrow, 1991).
Brooks-Gunn and colleagues (1997) argued that the aggregate of individuals and families
within a neighborhood setting creates a context that affects child outcomes. That context
consists of the resources, role models, and safety of the neighborhood. Dubrow and
Garbarino (1989) conducted a study of safety issues for children in two Chicago
communities. Mothers in low-income public housing identiﬁed shootings as the most
serious danger for their children. Dubrow and Garbarino found that all the children in
their study had experience with shootings by age 5. Therefore, the existence of
neighborhoods with high rates of violent crime means some children and their parents
face violence as a day-to-day life experience.

Neighborhoods provide one context in which children are raised, and these

*neighborhonds affect parental behavior.__ One of the consequences of living in a violent

15

neighborhood is that the problems have a greater impact on parents, especially on
mothers, because mothers are still typically the primary caregivers of their children
(Garbarino & Kostelny, 1993). In one exemplary work, Garbarino and Kostelny (1993)
argued that living in stressful neighborhoods and communities inﬂuences the stress on
parents, especially on mothers. “If the family is in a highly stressful neighborhood, the
neighborhoodmay. inﬂuence the parent’s psychological well-being, which may in turn,
gaffecthis other. parenting behavior.

Analysis of the Infant Health and Development Program by Klebanov and her
colleagues (1994) also suggested that neighborhood poverty was associated with less
maternal warmth and responsiveness. Such maternal behaviors may be seen as adaptive
in an impoverished neighborhood where parents may want to teach their children to
adjust to a harsh environment (Klebanov, Brooks-Gunn, & Duncan, 1994).
Methodological Issues Related to Neighborhood Studies

A. Perception of Neighborhood

Most studies on the effects of neighborhood on families and children have used
census tracts, postal codes, or other administrative units as proxies for neighborhoods
(Coulton, Korbin, Chan, & Su, 2001). The census tract has been widely used because of
the large amount of information available as well as the potential to use administrative
data (e. g., crime reports, housing values) that are aggregated into a census unit.
However, there is a concern that neighborhood boundaries that consider residents’
perceptions may produce more meaningful and relevant settings that are more closely
representative of the neighborhood construct (Burton, Price-Spratlen, & Spencer, 1997).

There is a tradition within urban sociology and environmental psychology of using

16

residents’ deﬁnition of neighborhoods to study neighborhoods. Perceived neighborhood
boundaries may be inﬂuenced by several factors, including the location of neighborhoods
(i.e., urban vs. rural), race, gender, and age. Furthermore, there is concern about whether
residents living in spatial proximity to one another share similar neighborhood deﬁnitions
(Coulton et al., 2001). Coulton and her colleagues demonstrated that residents within a
single block showed variation and disagreement about neighborhood boundaries and
these disagreements were inﬂuenced by race, age, and gender. Thus, variation among
residents and how they see their neighborhood may be responsible for the weak
neighborhood effects found in many studies (Meyer & Jencks, 1989). Therefore, it is
important to note that research that relies on census deﬁnitions alone may underestimate
neighborhood effects; the study of neighborhoods requires correct speciﬁcation of the
neighborhood context in order to investigate its effects on children and families.
B. Theoretical Perspectives on Neighborhood Inﬂuences

-Jencks and Mayer (1990) identiﬁed several theories regarding the ways in which
neighborhoods may affect child outcomes: (1) peighborhood resource theories, which
consider that child outcomes are related to the level of resources (e. g., community
centers, parks, day care) available in neighborhoods; (2) collective socialization theories,
which suggest that neighborhood monitoring (monitoring by adults in the neighborhood
rather than just by parents), supervision, and role modeling affect the behavior of
neighborhood children; (3) contagion theories, which consider the inﬂuence of negative
behavior of neighborhood peers on behavior problems among children in socioeconomic
and racially homogeneous communities; (4) competition theories, which suggest that

neighborhood effects on children are a function of community residents competing for

17

scarce resources; and (5) relative deprivation theories, which suggest that neighborhood
effects are due to the evaluation of their circumstances relative to their neighbors and
peers. Although these theoretical frameworks primarily focused on deﬁcit-based child
outcomes, the results of analyses have supported contagion and socialization theories
(Brooks-Gum, Duncan, Klebanov, & Sealand, 1993; Duncan, Brooks-Gunn, &
Klebanov, 1994).
C. Research Designs Used to Study Neighborhoods

In a review of the literature, Leventhal and Brooks-Gunn (2000) identiﬁed four
prevalent research design types used in the area of neighborhood effects on children and
youth: (l)_n_ational or multisite large studies, (2) city or regional studies, (3)
neighborhood-based designs, and (4)_exp,erimental or quasi-experimental designs. The
ﬁrst strategy to study neighborhoods includes the use of national or multisite studies of
individuals and families. Some data sets have been widely used including the Panel
Study of Income Dynamics (PSID; Survey Research Center, 1984), the National
Longitudinal Study of Youth-Child Supplement (NLSY-CS; Baker & Mott, 1989), and
the Infant Health and Development Program (IHDP; Gross, Spiker, & Haynes, 1997;
Infant Health and Development Program, 1990). The weakness of this type of approach
is that most of these data sets were not originally intended to examine speciﬁc
neighborhood effects (Leventhal & Brooks-Gunn, 2000).

City or regional studies have been done within a speciﬁc city or metropolitan area.
Therefore, there is concern regarding the strength and consistency of the results of these

types of studies compared to national studies. It is often the case that neighborhood

l8

characteristics are not widely varied across participants in city or regional studies
(Leventhal & Brooks-Gunn, 2000).

The neighborhood-based designs have been used with representation of
neighborhoods in mind, which includes a wide range of neighborhoods and has a certain
number of individuals per neighborhood unit. This approach is preferable to national or
regional studies for researchers to examine the variability of outcomes both within and
across neighborhoods, which provides more reliable estimates of neighborhood effects
(Sampson, Raudenbush, & Earls, 1997)

The experimental designs provide better estimates of the effects of neighborhood
because of the beneﬁt of using random assignment. An example of this type of study is
often found in the study of housing mobility programs (e.g., Moving to Opportunity),
which allow researchers to examine how a change in neighborhood context affects
families and children. However, often times the selection of neighborhood residence is
not random; it is a choice by each family. This is one of the limitations related to
neighborhood research which will be discussed in the next section.

D. Limitations/Challenges to Neighborhood Research
When conducting neighborhood research, it is important to note that relationships
“between families and neighborhoods are bidirectional. It is important to consider the fact
that families are not randomly assigned to neighborhoods but generally select the
neighborhoods in which they live within certain economic and social conditions. _ In other
words, the likelihood of children living in better or worse neighborhoods or attending
”getter or worse schools depends on parental characteristics (e. g., parental employment,

income). Families with problems and few resources may end up in less desirable

l9

neighborhoods and these neighborhoods may further undermine family functioning.
Thus, it should be noted that it is difﬁcult to examine the interactions between families
and neighborhoods. Another problem is related to measures of neighborhood. As
mentioned in a previous section, if a study relied on census data alone, the researcher(s)
must take into consideration both structural dimensions of the neighborhood and the
dimensions the census cannot measure, such as social control and social cohesion. In
addition, selection of neighborhood residence is a choice by each family. More afﬂuent
families may choose to live in a neighborhood where better child care services and better
schools are available to them.

Although it is still difﬁcult to address the bidirectional relationship between
families and neighborhoods, several strategies can be employed to minimize these
limitations. “Using measures of neighborhoods at different levels is one way to produce
better estimates of neighborhood effects. Several studies suggested that including both

..-

measures of structural characteristics and individual and family characteristics could
improve estimates of neighborhood effects (Duncan, Connell, & Klebanov, 1997; Elliott,
“Wilson, Huizinga, Sampson, Elliott, & Rankin, 1996; Sampson, Raudenbush, & Earls,
1997)

Processes of Neighborhood Effects on Children

A. Availability of Institutional Resources
The availability, accessibility, affordability, and quality of resources available in

communities could inﬂuence children’s outcome (Leventhal & Brooks-Gunn, 2000). For

example, the presence of learning activities in communities, such as libraries, family

resource centers, literacy programs, and museums, may stimulate children’s learning, in

20

turn affecting children’s academic development. To date, however, research suggests
that learning experience inside the home was a mediator of neighborhood effects on
children’s school readiness, especially for young children Gﬂebanov, Brooks-Gunn,
McCarton, & McCormick, 1998).

In addition to educational resources, the availability and quality of child care in
the neighborhood may play an important role in young children’s outcomes. High quality
child care and early intervention programs showed positive effects on children’s
cognitive and socioemotional outcomes in numerous studies (Campbell & Ramey, 1994;
Lee, Brooks-Gunn, Schnur, & Liaw, 1990; Reynolds, 1994; Yoshikawa, 1994).
Concerns regarding the quantity and quality of child care in poor neighborhoods have
been raised (Fuller, Coonerty, Kipnis, & Choong, 1997). For many poor families, child
care is not affordable, especially for low- and middle-income working families who do
not qualify for government sponsored programs such as Head Start.

For adolescents, schools are more salient institutional resources that could be a
potential mediator of neighborhood effects. ‘ Ennett and colleagues (1997) found that
school characteristics (e. g., school safety, school attachment, and availability of
substances) mediated the relationship between neighborhood characteristics (e.g.,
neighborhood attachment and safety, census measures of residential mobility and density)
and alcohol and cigarette use among ﬁfth and sixth graders.

B. Relationships

Within a framework of parental relationships, more and more studies suggest that

parental behavior (e. g., supervision/monitoring) and the quality of the home environment

are mediators of neighborhood effects on children and adolescents (Jarrett, 1997; Bradley,

21

 

 

1995; Klebanov, Brooks-Gunn, Chase—Landsdale, & Gordon, 1997). Parenting behaviors
have been related to social (Furstenberg, Cook, Eccles, Elder, & Sameroff, 1999),
emotional (Gray & Steinberg, 1999; Lambom, Mounts, Steinberg, & Dombusch, 1991),
and cognitive (Steinberg, Lambom, Dombusch, & Darling, 1992) outcomes for children
and youth. Theory and research indicated that multiple factors inﬂuence parenting
(Belsky, 1984, 1990), and parenting is related to children’s development, particularly
cognitive development (Bradley & Caldwell, 1984; Bradley et al., 1989). Bradley and
colleagues (1989) found that when the home environment is not a favorable one, the
child’s developmental outcome, such as cognitive performance, is likely to be low. They
also found that mothers who provide highly stimulating home environments have
children who score relatively high on intelligence tests. More recent studies have looked
at the processes by which neighborhoods inﬂuence families, especially parents, and in
turn, how parents are likely to inﬂuence their children._ Klebanov and colleagues (1998)
suggested that the provision of learning experiences inside the home was a mediator of
neighborhood effects on 3-year-old children’s IQ scores.-_In addition, a study showed an
association between home environments and children’s and adolescents’ physical health.
Children who lived in low-income neighborhoods showed higher rates of injury, partly
due to unsafe play areas within the home (Durkin, Davidson, Kuhn, O’Connor, &
Barlow, 1994).

Belsky (1984) developed a conceptual model, which proposes thatparenting
developmental history and personality), contextual factors (e.g., marital relations and

work), and characteristics of the child (e. g., temperament). Of the three factors that affect

22

parenting behavior, the most inﬂuential factor is the parents’ characteristics, followed by
contextual factors and child’s characteristics. Belsky also argued that stress in one factor
may be buffered by support in other factors (Belsky, 1984).

Based on his research, Belsky noted that the parents’ characteristics could affect
parenting; at least to some extent, developmental history and the psychological well-
being of parents can modify their parenting behavior. Research on mother’s L
psychological well—being indicated that depressed mothers are more likely than other

mothers to provide a disruptive, hostile, rejecting home environment to their children

 

(Colletta, 1983). Interestingly, these earlier studies on parent-child relationships have
focused on how multiple factors inﬂuence parenting and the development of the child,
but neighborhood inﬂuences were not extensively discussed.

Bronfenbrenner’s (1979, 1986, 1989) theory of the ecology of human
development proposed that people’s characteristics are related to the environment in
which they live. In other words, studies in parenting and child development must
examine not only the parent’s social and psychological contexts but also other aspects of
the larger context (e.g., neighborhood) in which parenting occurs. Therefore, the effects
of parenting on child development may depend, to some extent, on the particular
neighborhoods in which they live.

C. Norms/Collective Efﬁcacy

Many studies have shown a link between the social organizational characteristics
of neighborhoods and child and family well-being (Sampson et al., 1997). Sampson and
his colleagues deﬁnedcollective efﬁcacy as social control and social cohesion that arise

among residents within the neighborhood. Social control can be measured by the

23

likelihood that neighbors intervene in various situation, such as children skipping school,
a ﬁght in front of their house, or a decision to close the local ﬁre station. Social cohesion
is present in a community when mutual trust is observed. Social cohesion can be
measured from a series of questions including how strongly residents agreed that people
are willing to help neighbors, the neighborhood is close-knit, and residents share common
values. According to Sampson et a1. (1997), collective efﬁcacy is negatively related with

community violence, and it also mediated the association between neighborhood factors L

(i.e., concentrated poverty and residential instability) and community violence. In

 

addition, informal social control was negatively associated with adolescent problem I
behaviors after controlling for individual characteristics (Elliott, Wilson, Huizinga,
Sampson, Elliott, & Rankin, 1996; Sampson, 1997).

For adolescents, several studies suggested that peer inﬂuences may be associated
with adolescents’ behavior problems (Sampson, 1992; Sampson & Groves, 1989). Peer
inﬂuences were considered as a risk factor for adolescents who are exposed to peer
groups that are involved in delinquency and antisocial behaviors. Dishon and colleagues
(1995) found that young adolescent males who engaged in antisocial behaviors and their
close friends lived in the same neighborhood and they had more unstructured and
unsupervised activities in the neighborhoods. In addition, peer deviance (e. g., peers’
risky behavior and antisocial peer pressure) were associated with low grade point
averages (Darling & Steinberg, 1997), and it also mediated the effect of neighborhood
distress on adolescents’ mental health (e.g., distress) as well academiciand behavioral
difﬁculties at school (Meyers & Miller, 2004). Across this research, it has been

suggested that peers are an important aspect of the lives of adolescents.

24

The following section reviews research on the relationship between neighborhood
characteristics and children’s behavioral development and academic achievement.
Relationship between Neighborhood Characteristics and Behavioral Development

The studies of behavioral outcomes for adolescents have shown some evidences
that there is relationship between neighborhood characteristics, especially for low-SES
neighborhoods, and adolescents’ externalizing (e.g., aggression, delinquency) behaviors
as well as internalizing (e. g., depression, anxiety) behaviors (Anesshensel & Sucoff,
1996; Kalil & Eccles, 1998; Meyers & Miller, 2004). For example, among male
adolescents ages 13 to 16 in a Pittsburgh sample, neighborhood-level characteristics (i.e.,
poverty, unemployment, male joblessness, female farmly headship, race/ethnicity, out-of
wedlock children, and welfare receipt) were related with their delinquent and criminal
behavior. The male adolescents who lived in a low-SES neighborhood showed higher
rates of delinquent and criminal behaviors. (Loeber & Wikstrom, 1993; Peeples & Loeber,
1994). Findings from national data also suggested that living in a low-SES neighborhood
was associated with adolescents’ criminal and delinquent behaviors (Sampson & Groves,
1989; Simons, Johnson, Bearnan, Conger, & Whitbeck, 1996). Interestingly, there was
an experimental study (i.e., the Moving to Opportunity), which examined neighborhood
effects on adolescent outcomes. The Moving to Opportunity (MTO) project randomly
assigned housing-project residents in ﬁve of the nation’s largest cities to one of three
groups: 1) a group receiving housing subsidies to move into low-poverty neighborhoods;
2) a comparison group receiving Section 8 housing assistance but not constrained in their

locations; 3) a second comparison group receiving no special assistance. Ludwing and

25

colleagues (1998) used the data from the Baltimore site to evaluate the effects of the
program on adolescents’ criminal activity as reported in the Maryland Department of
Juvenile Justice’s criminal-offender records. All of the families who volunteered for
Baltimore MTO program were African American families headed by females. The
baseline survey showed that the most important reason for participation in the MTO
programs was to escape from gangs and drugs. Ludwig and colleagues analyzed 358
adolescents who ranged in age from 13 to 17 and lived in the state of Maryland for at
least one year following baseline random assignment. For boys in the experimental
groups, there was a statistically signiﬁcant, 17% reduction in the proportion who were
arrested for violent offenses (e.g., rape, robbery, assault) and a 13 percent reduction for
other offenses (e.g., disorderly conduct, weapons, drugs) compared to boys in the control
groups. For girls, there were no signiﬁcant differences in crime rates across experimental
and control groups (Ludwig, Duncan, & Hirschﬁeld, 1998).

In addition to the effects of low-SES on adolescent outcomes, few studies have
examined the role of residential ethnic diversity and residential instability and
adolescents’ behavioral outcomes. Sampson and Groves (1989) found that living in
neighborhoods with greater ethnic heterogeneity and with high rates of residential
instability were associated with adolescents’ delinquent behaviors (e.g., personal
victimization, property victimization). These results were supported and examined in
detail by Aneshensel and Sucoff (1996); neighborhoods were divided into clusters using
neighborhood SES and racial diversity information (e.g., low-, middle, high-SE8,
European American, Latinos, African American) in their study. Among adolescents,

aggression was to be found highest in middles-SES communities with high

26

concentrations of European Americans and Latinos and lowest in low-SES
neighborhoods with high concentrations of Aﬁ'ican Americans. On the contrary, conduct
disorders were most prevalent among youth living in low-SES African American
neighborhoods and least prevalent among youth in low-SES Latino neighborhoods.

Studies presented thus far assumed that neighborhood conditions affect adolescent
directly. However, it is likely that neighborhood characteristics affect family
characteristics such as income, living arrangement, parenting, and parental mental health,
which in turn, affect adolescent outcomes (Duncan, Connell, & Klebanov, 1997; Korbin
& Coulton, 1997).

Korbin and Coulton (1997) found that single mothers who lived in a low-SES
neighborhood with high levels of crime, low levels of economic opportunities, and poor
transportation face more challenges as parents; they would like to move to a
neighborhood where they can work and provide safe activities outside the home for their
children. However, it is suggested that residential moves typically occur between similar
neighborhoods with similar characteristics (Solon, Page, & Duncan, 1997).

Although research regarding the relationship between neighborhoods and family
characteristics is important, much of this research has focused on the extent to which
parenting strategies vary by community contexts (Simons, Lin, Gordon, Brody, &
Conger, 2002). Parents who use inductive reasoning to explain rules, monitor their
child’s behavior, positively reinforce desired actions, and are consistent in their use of
discipline are likely to have children who are less involved in antisocial behavior
(Simons, Johnson, Conger, & Elder, 1998). Sirnons and colleagues (1998) also suggested

that the effect of such parenting is much greater in communities where deviant behavior

27

 

is prevalent. Children who live in conventional neighborhoods may be at low risk for
behavioral problems regardless of how they are reared, whereas parental control may be
particularly important if a child is living in a community where antisocial behavior is
often modeled and even encouraged (Simons et al., 2002). In other words, the effect of
parentalcontrol may be more critical when children live in a hi gh-risk neighborhood, as
parental control may reduce their child’s involvement in deviant behaviors even if there
are pressures and encouragements supporting deviance. It is also important to note that
parenting strategies that are effective in conventional neighborhoods may not be
sufﬁcient to prevent child antisocial behavior in high-risk areas. Thus, further research
efforts are needed to identify those parenting behaviors that are effective in such
communities.
Relationship between Neighborhood Characteristics and Academic Achievement

Most of the studies of neighborhoods, families, and children in past decades have
looked at the developmental outcomes of children and adolescents including physical
health, school readiness, educational attainment, depression, childhood aggression,
juvenile delinquency, antisocial behavior, substance abuse, child maltreatment, and
teenage pregnancy (Burton & Jarrett, 2000). In the area of children’s school
achievement, Dombusch, Ritter, and Steinberg (1991) found that neighborhood
socioeconomic contexts were associated with secondary school grades; students who
lived in neighborhoods with fewer socioeconomic resources did more poorly in school
than did those who resided in neighborhoods with more resources. A study of
associations between neighborhood socioeconomic status and academic performance

among low-income elementary school children suggested that children’s academic

28

performance in ﬁfth grade was negatively associated with an indicator of neighborhood
risk even after controlling for individual family characteristics (Shumow, Vandell, &
Posner, 1999). Others showed neighborhood effects on young children; for example,
Chase-Lansdale and Gordon (1996) found that neighborhood risk was associated with 5
and 6-year-old children’s intelligence test scores, but not with their reading readiness
skills after controlling for family factors. Prior research found that neighborhood
characteristics were related to adolescent school outcomes (Brooks-Gunn, Duncan, Kato,
& Sealand, 1991) and academic problems among 6th to 8th graders (Halpem—Felsher et al.,
1997), but not among preschoolers and ﬁrst graders (Duncan et al., 1994; Chase-Lansdale
& Gordon, 1996).-Therefore, these previous studies suggest that neighborhood effects on
‘ children’s academic achievement may become stronger during the school-age years.

Furthermore, the effects of neighborhood SES on academic achievement have
been more prevalent for male adolescents than for female adolescents (Entwisle,
Alexander, & Olson, 1994; Halpem-Felsher, Connell, Spencer, Aber, Duncan, Clifford,
Crichlow, Usinger, & Coles, 1997). Male adolescents showed higher scores on basic
skills tests and math achievement compared to their female peers. Finding from several
studies also indicated that there is relationship between neighborhood high SES and
adolescents’ educational attainment (Duncan, 1994; Halpem—Felsher et al., 1997). These
studies found that neighborhood afﬂuence was related to adolescents’ chances of
completing high school, attending college, and years of schooling completed in the PSID
sample.

Further, Kauﬁnan and Rosenbaum (1992) analyzed the Gautreaux project data, in

which a quasi-random experiment was employed. Nearly all participants were Aﬁ'ican

29

American and ﬁ'om the same low-income public-housing projects. They volunteered to
participate in a subsidized program that arranged for private housing. Some moved to
middle-income white suburbs, while others moved to white and black urban
neighborhoods. Although ﬁndings regarding neighborhood effects are limited to low-
SES families who volunteered for the program, adolescents who moved to the more
afﬂuent suburbs were more likely to stay in high school, more often in college tracks, and
showed high rates of college enrollment.
Summary

It is evident from prior studies that neighborhoods inﬂuence adolescent outcomes
directly and indirectly. Both academic achievement and behavioral development were
inﬂuenced by neighborhood characteristics by way of more immediate environments (e.g.,
parents). However, most studies examined the effects of neighborhood on adolescent
outcomes within a framework of deﬁcit-based models. Along with poverty,
neighborhood characteristics have been looked at as problems or risk factors associated
with adolescents’ behavioral and cognitive development. Further, neighborhood
information was almost always from census tract data, in which one aspect of
neighborhoods (i.e., economic and social conditions) were measured including income,
percent of families living in poverty, percent of high school graduates, and percent of
families headed by females. Thus, aspects of the individual’s subjective experiences of
the neighborhood have been overlooked with the exception of few studies (Aneshensel &
Sucoff, 1996; Meyers & Miller, 2004). Again, perceived neighborhood characteristics
were assessed as risk factors for adolescents’ school problems, adolescent distress

(Meyers & Miller, 2004) and adolescents’ mental health (Aneshensel & Sucoff, 1996).

30

 

In addition, none of the previous studies employed a more complete and complex
conceptual model in which adolescents’ immediate (e.g., parents, peers, school)
environments as well as more distal environments (e.g., neighborhoods) were included in
their studies. In other words, most studies have not examined all different levels
(individual, family, school, peer, and community) simultaneously to see if individual,
family, school, and peer groups mediated the association between neighborhood
characteristics and adolescent outcomes. While parents may be the primary mediator of
neighborhood inﬂuences for younger children, for adolescents neighborhood effects may

be more direct rather than mediated by parents, and may be mediated by other factors,

 

such as peer group behavior (Sampson, 1992; Sampson & Groves, 1989). Thus, this
study will examine the ways characteristics of neighborhoods inﬂuence parenting, peers,
and schools and subsequently adolescent outcomes._ﬁThe primary purpose of the present
study is to determine if the neighborhood environment of the families is predictive of _
academic achievement of adolescents and their subjective well-being. Moreover, this
:tudywill provide additional information regarding the processes by which the larger

context (e.g., neighborhood) inﬂuences characteristics of the parent, peers, schools, and

adolescent development.

31

CHAPTER THREE
METHODS

The purpose of this study is to examine the process by which neighborhood
conditions inﬂuence parents, peers, and schools and the combined effect of neighborhood
conditions, parents, peers, and, schools on adolescent academic achievement and well-
being. This chapter describes the methodology of this study in detail.

Research Questions

This study will address the main question of how neighborhood characteristics
inﬂuence adolescent outcomes. Moreover, ﬁve more speciﬁc questions will be
investigated. They are:

1. How do neighborhoods inﬂuence the academic achievement and subjective well-
being of adolescent?

2. Are neighborhood characteristics related to the characteristics of parents (i.e.,
self-esteem, self-efﬁcacy, psychological distress), parenting practices, peer
interactions, and connectedness to school?

3. Are the characteristics of parents (i.e., self-esteem, self-efﬁcacy, psychological
distress), parenting practices, peer interactions, and connectedness to school, in
turn, predictive of academic achievement (i.e., WJ-R Achievement test) and the
subjective well-being of adolescents?

4. Do the characteristics of parents (i.e., self-esteem, self-efﬁcacy, psychological
distress), parenting practices, peer interactions, and connectedness to school
mediate the relation between neighborhood characteristics and adolescents’

academic achievement?

32

 

5. Do the characteristics of parents (i.e., self-esteem, self-efﬁcacy, psychological
distress), parenting practices, peer interactions, and connectedness to school
mediate the relation between neighborhood characteristics and adolescents’
subjective well-being?

Research Hypotheses

The following hypotheses will be tested in this study:
HA1: Neighborhood characteristics (i.e., perceptions of neighborhood and connectedness
to neighborhood) have a positive direct effect on parental characteristics (e.g., self-esteem
and self-efﬁcacy).
HA 2: Neighborhood characteristics have a positive direct effect on quality of parenting
(e.g., HOME environment).
H A3: Neighborhood characteristics are related to the peers adolescents encounter and the
nature of the interactions that adolescents have with peers (e.g., peer deviant behavior,
victimization, and closeness to friends).
HA4: Neighborhood characteristics have a positive direct effect on adolescents’
perceptions of connectedness to school.
HA5: Neighborhood characteristics have a positive direct effect on adolescents’ academic
achievement.
HA6: Neighborhood characteristics have a positive direct effect on adolescents’
subjective well-being.
HA7: Parental characteristics, quality of parenting, peer interactions, and connectedness
to school mediate the relation between neighborhood characteristics and adolescents’

academic achievement.

33

HA8: Parental characteristics, quality of parenting, peer interactions, and connectedness
to school mediate the relation between neighborhood and adolescents’ subjective well-
being. I
Participants

The data for this study are from the Panel Study of Income Dynamics -- Child
Development Supplement H (PSID CDS-II). The PSID CDS-II is the latest survey in the
Panel Study of Income Dynamics (PSID) program. The objective of the Child
Developmental Supplement (CDS) was to collect comprehensive and nationally
representative information about children and their families to study how economic and
social differences inﬂuence child development (Survey Research Center at the University
of Michigan, 1997). As additions to the PSID, the CDS I collected reliable, age-graded
assessments of the cognitive, behavioral, and health status of 3,600 children (including
about 250 immigrant children); these data were obtained from the mother, a second
parent or parent-ﬁgure, the teacher or child care provider, and the child. From fall of
2002 through spring of 2003, the University of Michigan Survey Research Center
conducted the 2002 PSID CDS H. As PSID data were collected in 1997, children
between the age of 5 and 13 were identiﬁed for inclusion in the Child Development
Supplement. The data for this study come from the PSID CDS-II which includes 2,907
children. The sample for this study consists of youth who were 12 to 17 years of age at
the time of the 2002 assessment. The sample included 556 White (47.2%), 478 African
American (40.6%), 82 Latino origin or descent (7%), 20 Asian, Paciﬁc Islander (1.7%),

and 6 American Indian adolescents. About half (50.6%) of the adolescents were female.

34

The mean age of the parents was 41.1 years (SD. = 7.03), and the average level of
education (based on the parent with the highest level of education in the family) was 12.7
years (SD. = 2.61). In 2002, the mean family income of the sample was $69,956. In
addition, about 20% of these families had an annual family income above $100,000.

Measures

Neighborhood characteristics: The primary caregiver’s perceptions of the
neighborhood (e.g., safety, residential satisfaction, anonymity, and residential stability)
were measured. Sample items are: “How would you rate your neighborhood as a place to
raise children?” and “How safe is it to walk around alone in your neighborhood after
dark?” The items were summed to produce a composite score for the neighborhood
characteristics index. Higher scores indicate a better neighborhood environment. The
reliability coefﬁcient for these four items was .50.

Social control: The four items for social control are scored on a four-point scale
(i.e., l-very unlikely to 4-very likely). The items are: “How likely is it that a neighbor
would do something if someone was trying to sell drugs to your children in plain sight?”,
“How likely is it that a neighbor would do something if your kids were getting into
trouble?”, “How likely is it that a neighbor would do something if a child was showing
disrespect to an adult?, and “How likely is it that a neighbor would do something if a
child was taking something out of a neighbor’s apartment, house, garage, car or yard?” A
total score was computed by summing the items, with higher scores reﬂecting greater
social control among the neighbors. The reliability coefﬁcient for the social control scale

was .85 for this sample.

35

Connectedness to neighborhood: The frequency of participation in various
community activities (e.g., religious activity, physical exercise, and going to a
community center) by the youth within the past 12 months was measured with seven
items. Sample items are: “How often has your child participated in church (or other
religious) club or activity — not religious service or mass?”, “How often has your child
participated in a physical exercise, such as aerobics, running or lifting weights?”, and
“How often has your child participated in going to a community center like a YMCA?”
These questions were answered by the primary caregiver. The possible rating ranged
from 1 (never in the past 12 months) to 7 (several times a week). Scores from each item
were summed to compute a total score. Higher scores indicate greater participation in
community activities by the youth as reported by the primary caregiver. The reliability
coefﬁcient for these seven items was .51 for the present study sample.

Parental difﬁculty: Perception of parenting difﬁculty was assessed with three
questions: “There are some things about my child that really bother me a lot.” “I ﬁnd
myself giving up more of my life to meet my child’s needs than I ever expected.” “I
often feel angry with my child.” The possible ratings ranged from l-not at all true to 5-
completely true, and scores from the items were summed to compute a total score.
Higher scores indicate greater parental perceptions of difﬁculty with their child. The
reliability coefﬁcient for this scale was .70 for this sample.

Primary caregiver ’s Self-efﬁcacy: The Pearlin Self-efﬁcacy Scale (Pearlin,
Lieberman, Menaghan, & Mullan, 1981) was used to assess the primary caregiver’s self-
efﬁcacy. The CDS-H used a shortened version of this measure based on factor analysis

and reliability analysis of the measure used in wave one of the CDS in 1997. The items

36

are: “There is really no way I can solve some of the problems I have”, “Sometimes I feel
that I’m being pushed around in life”, “I have little control over the things that happen to
me”, and “I often feel helpless in dealing with the problems of life.” A total score was
computed by summing the items, with higher scores reﬂecting higher levels of self-
efﬁcacy. Coefﬁcient alpha for the present study sample was .79.

Primary caregiver 's Self-esteem: The primary caregiver’s self-esteem was
measured with Rosenberg Self-esteem Scale (Rosenberg, 1986) that assesses the way
people feel about themselves. The Rosenberg Self-esteem Scale is a lO-item scale with
responses scored on a four-point scale, ranging from 1 (Strongly disagree) to 4 (Strongly
agree). Sample items include: “I feel that I have a number of good qualities” and “I take a
positive attitude toward myself.” The scale also includes some negatively stated items
such as: “I feel I do not have much to be proud of ’, and “I certainly feel useless at times.”
Scores from these negatively stated items were recoded so that high scores on this
measure reﬂect higher levels of self-esteem. The reliability coefﬁcient for this scale
was .86 for this sample.

Primary caregiver ’s psychological distress: The primary caregiver’s
psychological state of mind (e.g., distress) was measured with the Psychological Distress
Scale (Kessler, et al., 2002). The possible rating ranged from 1(all of the time) to 5 (none
of the time). The scores on each item were recoded and summed to compute a total score.
Higher scores indicate higher levels of psychological distress. The reliability coefﬁcient
for these 6 items was .81. The items are: “During the past 30 days, how often did you feel

nervous?”, “How often did you feel hopeless?”, “How often did you feel restless or

37

ﬁdgety?”, “How often did you feel that everything was an effort?”, “How often did you
feel so sad nothing could cheer you up?”, and “How often did you feel worthless?”

Home environment: The Home Observation for Measurement of the Environment
(HOME) inventory by Caldwell and Bradley (1984) was used as a measure of the
cognitive stimulation and emotional support that parents provided for their children. This
scale consists of 81 items. Of those, 33 items which are age-appropriate questions for
older adolescent were used. Ten out of these 33 items were assessed with observations
by the interviewer, while the rest of the items were based on answered provided by the
primary caregivers. The total score was computed from the 33 items. The reliability
coefﬁcient for the HOME scale was .69 for the present sample.

Parental monitoring: The adolescents were asked three questions related to
parental monitoring of their activities and friends. The items asked adolescents about
their parents’ knowledge such as: “Do your parents know what you do during your free
time?” “Do your parents know which friends you hang out with during your free time?”
and “Do your parents know what you spend your money on?” Each item is scored on a
5-point rating scale. A total score was computed by summing these items. The reliability
coefﬁcient for these 3 items was .86.

Closeness to friends: The frequency with which an adolescent converses with
friends about school, future plans, and friends was assessed. In addition, there was a
question about the adolescent’s closeness with his or her friends. These four items were
selected from the Closeness to Others scale (Institute for Social Research, 2004). Scores

on the ﬁrst three items ranged from 1-never to 6-everyday. Scores on the closeness with

38

 

friends item ranged from l-not very close to 4-extremly close. A total score was
computed by summing these items. The reliability coefﬁcient for this scale was .72.

Peer victimization: Adolescents’ experiences with victimization at school and in
the neighborhood were measured with a set of four items. The peer victimization scale,
developed by Kochenderfer and Ladd (1996), asks adolescents questions including: “In
the last month, how often have kids in your school or neighborhood picked on you or said
mean things to you?” and “How often have kids in your school or neighborhood
purposely left you out of your friends’ activities?” The possible ratings ranged from 1
(not in the past month) to 6 (everyday). A total score was computed by summing the
items, with higher scores reﬂecting more peer victimization experiences by the
adolescent. The reliability coefﬁcient for this scale was .57 for the present study sample.

Peers ’ deviant behaviors: Adolescents’ ratings of their peers’ deviant behaviors
were based on how many of their friends have engaged in negative behaviors, including
disobeying parents, involvement in gangs, ﬁghting with other kids, encouraging others to
do dangerous things, drug and alcohol use, and getting in trouble. The possible responses
ranged from l-none to 5-ahnost all or all. A total score was computed by summing these
items. The reliability coefﬁcient for this scale was .66.

Peers’ academic orientation: Adolescents were also asked to indicate how many
of their friends have planned to go to college and how many of their friends think
schoolwork is important. The possible ratings ranged from l-none to 5-almost all or all.
A total score was computed by summing the two items. The reliability coefﬁcient for

these two items was .62.

39

 

Connectedness to school: Adolescents were asked four questions to assess the
degree of inclusiveness (i.e., “How often did you feel like you were part of your
school?”), closeness (i.e., “How often did you feel close to people at your school?”),
happiness with (i.e., “How often did you feel happy to be at your school?”), as well as
safety at school (i.e., “How often did you feel safe at your school?”). Each item is scored
on a 6-point scale with responses ranging from l-never to 6-everyday. A total score was
computed by summing the items, with higher scores indicating that the adolescents feel
more connected to their school. The reliability coefﬁcient for this scale was .74 for this
sample.

Adolescents ’ academic achievement: The adolescent’s academic achievement
was assessed with Woodcock—Johnson Revised Tests of Achievement for Reading and
Math (WJ-R). The WJ-R test consists of nine subtests measuring different aspects of
academic achievement (Woodcock & Johnson, 1989). Three subsets of the WJ-R -- the
Letter-Word, Passage Comprehension, and Applied Problems tests -- were administered
through the CDS-II. The standardized scores for the Broad Reading test, combining
Letter-Word and Passage Comprehension, and the standardized scores of the Applied
Problems test were used for the analysis.

Adolescents ’ subjective well-being: A set of items adapted from the MacArthur
MIDUS ‘Youth measure were used to measure subjective well-being. It measures several
dimensions of the subjective well-being of the adolescents including: emotional well-
being, social well-being, and psychological well-being. There are three questions on
emotional well-being such as “In the last month, how often did you feel happy?” and

“How often did you feel satisﬁed?” Five questions were asked to assess social well-

40

being: “In the last month, how often did you feel that you had something important to
contribute to society?” “How often did you feel that people are basically good?” There
are four questions in the psychological well-being domain, such as “good at managing the
responsibilities of your daily life?” and “conﬁdent to think or express your own ideas and
opinions?” The possible ratings ranged from l-never to 6-every day. Scores on these
three scales were summed to produce the youth subjective well-being index. The
reliability coefﬁcient for this scale was .91 for this sample.

Creating composite path variables: To create a parsimonious model, standardized
scores for measures assessing the constructs of interest were computed after the necessary
reverse coding. This resulted in each measure having a mean of zero and a standard
deviation of 1. The standardized scores were saved as variables and were used to create
an overall index for each domain. For instance, the scales measuring perceptions of
neighborhood, social control, and connectedness to neighborhood were combined into an
overall index of neighborhood characteristics by adding the standardized scores for the
various indicators of neighborhood quality, with higher scores indicating more positive
neighborhood characteristics. Similarly, parental difﬁculty, self-efﬁcacy, self-esteem, and
psychological distress were combined to create an overall index of parental
characteristics, with higher scores indicating positive psychological well-being. Also,
scales measuring the home environments that the primary caregivers provided for their
children and parental monitoring as reported by youth were combined into an overall
index of parenting behaviors; higher scores were indicative of more supportive parenting

practices. The overall index of peer interactions was created by combining closeness to

41

friends, peer victimization, peer deviant behavior, and peer academic orientation; higher
scores indicated positive peer relationships.
Analyses

The analyses involved four main parts: descriptive statistics, reliability analysis,
correlations between the variables, and a path analysis. Descriptive statistics (e.g., means
and standard deviation) were used to describe the distributional characteristics of each of
the variables. A reliability analysis was conducted on multi-item scales to assess the
internal consistency of the items. The relationship between variables was assessed with
zero-order correlations. To test the hypotheses in this study, separate regression analyses
were done for adolescents’ academic achievement and for adolescents’ subjective well-
being. In addition, path analysis was employed to assess the direct effects of
neighborhood characteristics on adolescents’ academic achievement and their subjective
well-being. Indirect effects of neighborhoods by way of parental characteristics and
behaviors, peers, and connectedness to school on adolescents’ academic achievement as
measured by WJ-R tests and adolescents’ subjective well-being were also assessed. In
the path analysis model, neighborhood characteristics were viewed as having a direct
effect on adolescent outcomes and indirect effects by means of parental characteristics

and behaviors, peers, and connectedness to school.

42

CHAPTER FOUR
RESULTS
This chapter ﬁrst presents the results of the relations among the predictor
variables. Second, the zero-order correlations between the predictor variables and the
dependent variables are shown. Third, the results of the hierarchical regression analyses
for adolescents’ academic achievement and subjective well-being are discussed. Finally,
the results of a path analysis testing the conceptual model are discussed.
Relations among the Predictor Variables

Correlational analyses were performed to determine the extent of associations
among the predictor variables. Table 1 presents the zero-order correlations among the
predictor variables. Several signiﬁcant correlations among the predictor variables were
found. Most of the correlations were in the expected directions, and were small to
moderate in magnitude. Signiﬁcant correlations for the sample ranged from r = .06 to
.58.

Primary caregivers who perceived their neighborhood positively tended to
perceive more social control among their neighbors, and their children participated more
in community activities. In addition, the primary caregivers who had positive perceptions
of neighborhood tended to report less parental difﬁculty, to feel less distressed, to have
higher self-efﬁcacy scores, and to have higher self-esteem scores. Primary caregivers
who perceived their neighborhood as a relatively positive one tended to provide better
home environments for their children and monitored their youth closely; because of the
large sample size, the relation between perceptions of neighborhood characteristics and

parental monitoring was statistically signiﬁcant but small in magnitude. As expected,

43

“7“" ‘_"'"’ __
i .

 

 

when primary caregivers reported better neighborhood environments, their adolescents
tended to feel closer to friends and more connected to school.

Primary caregivers who perceived more parental difﬁculties tended to provide
less stimulating home environments and to monitor their children less closely than other
parents. When primary caregivers reported more difﬁculty with rearing their adolescents,
the adolescents tended to have more victimization experiences, and to have peers who
engaged more in deviant behaviors and who were less academically oriented. These
adolescents also reported feeling less connected to their schools.

Primary caregivers’ self-efﬁcacy and self-esteem were moderately related (r =
.58). Primary caregivers who had higher self-efﬁcacy and self-esteem scores tended to
provide better home environments, and their adolescents reported more closeness to
school and their peers tended to be more academically oriented.

Primary caregivers who were more psychologically distressed tended to monitor
their adolescents less and provided a less stimulating home environment for their
children. More psychologically distressed primary caregivers tended to have adolescents
who reported more victimization experiences and who had peers who engaged in more
deviant behaviors.

Primary caregivers who monitored their adolescents closely tended to provide
better home environments, and their adolescents reported feeling close to ﬁiends and
connected to school. Closely monitored adolescents tended to have peers who were less
likely to engage in deviant behaviors and more likely to be academically oriented.

Primary caregivers who provided better home environments tended to have

adolescents who reported having peers who were less engaged in deviant behaviors and

more interested in academics. These adolescents also tended to feel connected to their
schools.

Adolescents who reported more victimization experiences tended to have peers
who engaged in more deviant behaviors and to feel less connected to school. Consistent
with expectations, adolescents who reported that their peers engaged more in deviant
behaviors tended to feel less connected to school and to have peers who were less

interested in academics.

45

S. v a ...... .3. v a ... .822 258223 P688 AQUA: EonoSo ESE-E a

 

 

Begum
:3. 2..- :8. :2.- :2. :2. eaoéaooéooi
notﬂﬁoﬂo

:E. 8.- ...S. :8.- :2. ...8. 2588“ sum .2
:8; .8. 8.- :2. .8.- 3.- .8323 2538 .82 .2
:8.- :2. 8.- :2. 8.- 8.- . 82328382.:
:2. 8.- 8. 5.- 3. :8. aqua 2 $8320 .2
:2. :2.- :~N. :8- :3. :R. squamoeaomzoma
82 .8; .8. :2.- 8. .8. 2255835an

$0me
2: :8.- :§. :2.- :2.- 323203 whom .5
:mm. :8.- :2. :2. eases-20m room e
8.2 ...-R.- :2. :2. sees-20m room .m
82 :2.- :2.- baa-ca 35am .4

.eooﬁonnwmoc 3
82 .....a _. 352828 98> .m
:2. L268 28m .m

«momentowoahﬁo
8.2 82822022

w s w v m _

 

 

moEmtw> 88:65 on“ mace-8 mcouﬁom

ﬂ 2an

46

Relations between the Predictor Variables and Outcome variables

Table 2 presents the zero-order correlations between the predictor variables and
the adolescents’ academic achievement and subjective well-being. Small to moderate
correlations were obtained between the predictor variables and the three outcome
variables.

Consistent with expectations, there were positive relationships between
adolescents’ WJ-R scores and neighborhood characteristics. However, there was not a
signiﬁcant relationship between adolescents’ subjective well-being and primary
caregivers’ perceptions about their neighborhood. There were positive relationships
between youth connectedness to neighborhood and adolescents’ academic achievement
scores, and between youth connectedness to neighborhood and subjective well-being. All
of these correlations were small (< .20).

Parental difﬁculty and primary caregivers’ psychological distress were negatively
related to the adolescents’ outcomes. On the other hand, primary caregivers’ self-esteem
and self-efﬁcacy were positively related to adolescents’ WJ-R scores and subjective well-
being as expected. Adolescents’ subjective well-being was signiﬁcantly related to
parental monitoring, while adolescents’ academic achievement scores were signiﬁcantly
related to the home environments that primary caregivers provided for their adolescents.

As expected, the result of correlational analyses among the outcome variables
indicated that adolescents’ broad reading scores were positively related to their applied
problem scores (r = .67, p < .01). There were also positive relationships between

adolescents’ broad reading scores and adolescents’ subjective well-being, and between

~—

47

adolescents’ applied problem scores and adolescents’ subjective well-being. These
correlations were small (r =.13 & .12 respectively).

Table 2

Relations between the Predictor Variables and Outcome Variables

 

 

Broad Reading Applied Subjective
Problems well-being

Neighborhood .17** .17** .04
Characteristicsal
Social controla .07* .04 .01
Youth connectedness to .14" .16** .11**
neighborhood“
Parental difﬁculty -.13** -.20** -.07**
PCG’s Self-efﬁcacy .14** .16M .06*
PCG’s Self-esteem .08** .08** {.07’“
PCS’s Psychological -.l7** -.14** {4.05
distress ‘ ‘ ‘
Parental monitoringb .18" .17** .38**
HOME environment .3 1 ** .35" .09**
Closeness to ﬁiends .09" .02 .24"
Peer victimization -.08** -.04 -.08**
Peer deviant behavior -.13** -.16** -.15*"‘
Peer academic orientation .08** .07* .35"
Connectedness to school .17" .16** .39**

 

’ Primary caregiver (PCG) rated, b Adolescents rated. * p < .05, ** p < .01

The result indicated that peer characteristics were related to the adolescents’
overall outcomes. Closeness to ﬁiends was signiﬁcantly related to adolescents’ broad
reading scores and subjective well-being. There was a negative relationship between
peer deviant behavior and adolescents’ academic achievement scores, and between peer
deviant behavior and adolescents’ subjective well-being. As expected, peer victimization

experiences were negatively related to adolescents’ broad reading scores and subjective

48

well-being. Having peers who were academically oriented was signiﬁcantly, albeit
modestly, related to adolescents’ academic achievement scores; a somewhat stronger
relation was found between having peers with an academic orientation and adolescents’
subjective well-being.

As expected, feeling connected to school was signiﬁcantly related to adolescents’
academic achievement scores. Again however, this predictor variable is more strongly

related to adolescents’ subjective well-being.

Hierarchical Regression Analyses of Adolescent Broad Reading Scores on the WJ-R
Achievement Test

In this section, the combined inﬂuence of the various factors identiﬁed in the
conceptual model on adolescents’ broad reading is assessed when family SES and other
demographic variables (parents’ age, parents’ highest level of education, family income,
number of children in family, and number of persons in the family) are controlled. Table
3 presents the results of the hierarchical regression analyses in which neighborhood
characteristics, parental characteristics, parenting behaviors, peers, and school variables
were entered in separate blocks to predict adolescents’ broad reading scores.

Among the neighborhood predictors, parents’ perceptions of neighborhood and
connectedness to neighborhood uniquely contributed to the prediction of adolescents’
broad reading scores. The primary caregiver’s psychological distress, parental
monitoring, and home environment variables were also found to be predictive of
adolescents’ broad reading scores. Although the primary caregivers’ self-esteem was

positively related to their adolescents’ broad reading score in the bivariate correlational

49

analysis, primary caregivers’ self-esteem was negatively related to broad reading scores
in the ﬁnal regression model when other predictor variables were controlled. This may
be due to the fact that parental self-esteem is correlated with other parental characteristics
(e. g., self-efﬁcacy) and parental behavior which were also entered as predictor variables.

Among the peer predictors, peer academic orientation was a signiﬁcant predictor
of the adolescents’ broad reading scores. Although closeness to ﬁiends and peer deviant
behaviors were not signiﬁcant at the .05 level, those variables were related at the .10
level, to the adolescents’ broad reading scores. Closeness to school was also signiﬁcantly
related to the adolescents’ broad reading scores.

Each system (neighborhood, parents, school), except for peers, generally made a
statistically signiﬁcant contribution to the adolescents’ broad reading scores as evidenced
by the signiﬁcant change in R2 after each step. Overall, the predictor variables accounted
for 20% of the variance in the broad reading scores of the sample.

The F (19, 1075) = 14.77 for the model was found to be signiﬁcant (p < .001). When
family SES and other demographic variables (e.g., parents’ age, parents’ level of
education, family income, number of children in family, number of persons in family)
were included in the ﬁnal model, none of the neighborhood variables had a direct effect
on adolescents’ broad reading scores. The effects of neighborhood characteristics may be

indirect, and mediated by parental characteristics, peers, or schools. This possibility is

examined later in the chapter.

50

Table 3

Hierarchical Regression Anialjses PredictingLAdolescent Broad—Reading Score

 

Broad Reading Score

Step 1 Step 2 Step 3 Step 4 Step 5

l3 l3 B l3 BAR’

 

Predictor Variables

Step 1. .05**
Neighborhood Characteristics

Perceptions of neighborhood .17** .10“ .10** .10" .05
Perceptions of social control .01 -.02 -.01 -.01 -.01
Connectedness to neighborhood .1 1** .02 .02 .01 -.03

Step 2. .09**
Parental Characteristics

Parental difﬁculty -.03 -.03 -.03 -.O4
Self-efﬁcacy .04 .05 .05 .05
Self-esteem -.O4 -.04 -.04 -.09*
Psychological distress -.11** -.10** -.10** -.09**

Home environment .22" .21“ .21 ** .14**
Parental Monitoring .1 l** .10** 10" .12"

Step 3. .01
Peers "
Closeness to ﬁiends .05 .04 .05

Peer victimization -.03 -.02 -.01

Peer deviant behaviors -.05 -.04 -.06

Peer academic orientation -.O4 -.O6 -.07*

Step 4. .01*"‘
School

Connectedness to school .08** .07 *

Step 5. .06**
Family SES

Parents’ level of education .23M
Parents’ age .11**

Family income .05

Number of persons in family -.03

Number of children in family .01

 

*p < .05, **p < .01

Note: R2 change represents the explained variance contributed by each step

51

Hierarchical Regression Analyses for WJ-R Applied Problems Achievement Test Scores

This section discusses the results of the hierarchical regression analyses that were
computed to determine which of predictor variables in each system are related to
adolescents’ applied problems scores on the WJ-R achievement test. Primary caregivers’
perceptions of neighborhood characteristics and youths’ connectedness to neighborhood
were found to be predictive of the adolescents’ applied problems scores. Except for the
primary caregiver’s psychological distress, all of the parental characteristics were
signiﬁcantly related to adolescents’ applied problems scores. Peer deviant behaviors and
closeness to school were also signiﬁcantly related to adolescents’ applied problems
scores (see Table 4).

Each system (neighborhood, parents, peers, school) made a statistically
signiﬁcant contribution to the adolescents’ applied problems scores as evidenced by the
signiﬁcant change in R2 at each step. However, none of the neighborhood variables
uniquely predicted the outcome in the ﬁnal model. The predictor variables accounted for
22% of the variance in the applied problems scores of the overall sample. The F value
for the model was found to be signiﬁcant, F (19, 1075) = 17.30, p < .001. Similar to the
results for adolescents’ broad reading scores, family SES and other demographic

variables accounted for 6% of the variance in the applied problems scores when entered

on the last step.

52

Table 4

Hieraichical Regression Analyses PredictingAdolescent Applied Problems

 

Applied Problems

Step 1 Step 2 Step 3 Step 4 Step 5

BB B B BAR’

 

Predictor Variables

Step 1. ' .05**
Neighborhood Characteristics

Perceptions of neighborhood .17** .10** .10** .09" .05
Perceptions of social control -.02 -.05 -.05 -.05 -.04
Connectedness to neighborhood .13** .02 .02 .01 -.03

Step 2. .1 1**
Parental Characteristics

Parental difﬁculty -.13** -.12** -.12** -.13**
Self-efﬁcacy .07 .08* .08* .08*
Self-esteem -.04 -.04 -.04 -.09**
Psychological distress -.05 -.05 -.05 -.04

Home environment .25" .24M .24” .17"
Parental Monitoring .09“ .08** .08* .09"

Step 3. .01*
Peers

Closeness to friends -.01 -.02 -.01

Peer victimization .02 .02 .03

Peer deviant behaviors -.09** -.08** -.10**

Peer academic orientation -.03 -.05 -.06

Step 4. .01*
School

Connectedness to school .07* .06*

Step 5. .06**
Family SES

Parents’ level of education .24"
Parents’ age .10“

Family income .05

Number of persons in family -.07

Number of children in family .10*

 

*p < .05, **p < .01

Note: R2 change represents the explained variance contributed by each step

53

Hierarchical Regression Analyses of Adolescent Subjective Well-being

Table 5 presents the results of the regression analyses that examined the relation
between the predictor variables and the adolescents’ subjective well-being. Among the
neighborhood characteristics, only youths’ connectedness to neighborhood was a
signiﬁcant predictor of the adolescents’ subjective well-being when other variables in the
model were controlled.

Parental monitoring and connectedness to school were the strongest predictor
variables that were signiﬁcantly related to adolescents’ subjective well-being (B5 = .32
and .26 respectively), followed by peers’ academic orientation. Closeness to friends was
also signiﬁcantly related to adolescents’ subjective well-being.

Each system (neighborhood, parents, peers, school) made a statistically signiﬁcant
contribution to the adolescent subjective well-being as evidenced by the signiﬁcant
change in R2 at each step. The predictor variables accounted for 32% of the variance in
the adolescents’ subjective well-being. The F (19, 1075) = 26.50 for the model was
signiﬁcant (p < .001). Interestingly, the family background characteristics accounted for
only an additional one percent of the variance when added on the ﬁnal step of the
hierarchical regression analysis: Only parents’ age was signiﬁcantly related to
adolescents’ subjective well-being. Adolescents with older parents reported higher levels

_ of subjective well-being.

54

Table 5

Hierarchﬁl Regression Analyses Predicting Adolescent Subﬁctive Well-being

 

Subjective Well-being

Step 1 Step 2 Step 3 Step 4 Step 5

B B B B BAR’

 

Predictor Variables

Step 1. .02**
Neighborhood Characteristics

Perceptions of neighborhood .03 .01 .00 -.01 -.02
Perceptions of social control -.02 -.02 -.03 -.03 -.03
Connectedness to neighborhood .12** .12** .08" .06* .07*

Step 2. .15**
Parental Characteristics

Parental difﬁculty -.03 -.02 .00 .00
Self-efﬁcacy -.04 -.03 -.03 -.02
Self-esteem .05 .03 .02 .03
Psychological distress -.01 -.02 -.03 -.02

Home environment -.04 -.07* -.07* -.06*

Parental Monitoring .38** .31" .30** .32"

Step 3. .09**
Peers

Closeness to friends » .14“ .10** .10**

Peer victimization -.02 .01 .02

Peer deviant behaviors -.03 .01 .01

Peer academic orientation .25" .19** .19**

Step 4. .06"
School

Connectedness to school .26" .26”

Step 5. .01**
Family SES

Parents’ level of education -.03

Parents’ age .10"

Family income :.05
Number of persons in family .02

Number of children in family -.03

 

*p < .05, **p < .01

Note: R2 change represents the explained variance contributed by each step

55

 

Results of a Path Analyses

To test the conceptual model for this study as a whole, a path analysis was
conducted to assess both the direct and mediated relations between neighborhood
characteristics and adolescents’ well-being and academic achievement, controlling for
family SES. The results are presented in Figure 2.

Thecomposite index of neighborhood characteristics was not directly related to
ladolescents’ subjective well-being when family SES, parental characteristics, parenting
behaviors, peers, and connectedness to school were controlled. However, the parental
characteristics composite, the parenting behaviors composite, the peers composite, and
.. the school composite mediated the relationship between neighborhood characteristics and
subjective well-being. More speciﬁcally, neighborhood characteristics were signiﬁcantly

and positively related to parental characteristics and parenting behaviors, which in turn
were signiﬁcantly related to peers characteristics, which in turn were positively related to
connectedness with school; both peer group and school characteristics were positively
related to adolescents’ subjective well-being. In addition, parenting behavior had a direct
effect on subjective well-being, with positive parenting associated with greater well-
being among adolescents.

Parental characteristics and parenting behaviors were signiﬁcantly related to both

broad reading scores and applied problems scores; parents with higher levels of
psychological well-being and parents who provided better home environments had

adolescents who had higher achievement test scores-iThe peer group composite index

was not related to adolescents’ academic achievement-2" Although not signiﬁcant at the

56

.05 level, school characteristics were related to the adolescents’ broad reading scores and
applied problems scores at the .10 level.

Neighborhood characteristics did not have a direct effect on the achievement
outcomes but had an indirect effect on both achievement indicators that was mediated by
parental characteristics and parenting behaviors. Thus, as shown in Figure 2,
neighborhood characteristics had indirect effects on adolescent’s subjective well-being by
way of parental characteristics, parenting behaviors, peer group, and school
characteristics, and indirect effects on the two achievement test scores via parental

characteristics and parenting behaviors.

57

 

Eon—2633 3:838 e5 memos-=03 Ecumﬂoca
no 308682“ mm Bonav- éoom .2538 832.5, weEoEoHE can Sam-288.85 coon-523mb: magma-w women—new 5mm .N 0.5me

5. v a ...... .3. v a ... .eoaﬂmmme one @3353 Emu—.236 bee 332% 08 $3 mucosa-Loco “~232on coﬁEmwaﬂm $32

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w .Alll rho.
S no» «on
b e m i/
......om.
~00£om ea~~.
. soo- -
:2 a 5323
_ wet—88m
mace—noun ......NN. 3N.
vow—anw myoca— * an w .
- «*NN.
sho- 8558820
13:88
enaN.
......NN. 8558830

 

 

mien-=95 econ-89362

o>uoo Bum

 

 

 

 

58

 

Summary of Results
In this section, the results of the study are smnmarized. The summary is presented

in terms of the research hypotheses addressed in the study.

HA1: Neighborhood characteristics (i.e., perceptions of neighborhood and connectedness
to neighborhood) have a positive direct effect on parental characteristics (e.g., self-esteem

and self-efﬁcacy).

The results presented in an earlier section are consistent with this hypothesis.
Parents’ perceptions of neighborhood characteristics, social control, and youth
connectedness to neighborhood were positively related to the primary caregivers’ self-
efﬁcacy and self-esteem (See Table 1). The results also indicated that neighborhood
characteristics were inversely related to the primary caregivers’ parental difﬁculty and
psychological distress. The overall index of neighborhood characteristics was also

signiﬁcantly and positively related to the parental characteristics composite index (See

Figure 2).

HA 2: Neighborhood characteristics have a positive direct effect on the quality of

parenting (e.g., HOME environment).

The hypothesis is supported by the results presented above. Positive perceptions

of neighborhood characteristics were positively related to the home environments that

primary caregivers provided and the level of parental monitoring reported by adolescents

59

(See Table 1). Similarly, primary caregivers’ perceptions of social control and youth’s
connectedness to neighborhood were related to the quality of the home environment. The
overall index of neighborhood characteristics was signiﬁcantly related to the parenting

behaviors composite (See Figure 2).

HA3: Neighborhood characteristics are related to the peers adolescents encounter and the
nature of the interactions that adolescents have with peers (e.g., peer deviant behavior,

victimization, and closeness to ﬁiends).

This hypothesis is partially supported by the results presented earlier. All of
neighborhood characteristics variables were signiﬁcantly related to the measure of
adolescents’ peers’ academic orientation. The perceptions of neighborhood were also
signiﬁcantly related to adolescents’ perceptions of closeness to friends (See Table 1).
Adolescents’ connectedness to neighborhood was signiﬁcantly and inversely related to
peers’ deviant behaviors. The overall index of neighborhood characteristics, however,

was not directly related to the peers composite in the path analysis (See Figure 2).

HA4: Neighborhood characteristics have a positive direct effect on adolescents’

perceptions of connectedness to school.

The results of the zero-order correlations indicated that there was a positive
relationship between all of the neighborhood characteristics and adolescents’ closeness to
school (See Table 1). Also, the path analysis indicated that the overall index of

neighborhood characteristics was signiﬁcantly related to the youths’ connectedness to

60

school when other variables were statistically controlled including family SES (See

Figure 2).

HA5: Neighborhood characteristics have a positive direct effect on adolescents’ academic

achievement.

This hypothesis is partially supported by the results presented in the earlier
section. One of the neighborhood characteristics variables, the primary caregivers’
perceptions of neighborhood, was related to both adolescents’ broad reading scores and
applied problem scores of the WJ-R achievement test when family SES variables were
controlled (See Table 3 and 4). However, the path analysis indicated that the composite
index of neighborhood was not directly related to both broad reading scores and applied

problems scores when other variables were statistically controlled (See Figure 2).

HA6: Neighborhood characteristics have a positive direct effect on adolescents’

subjective well-being.

This hypothesis is partially supported. There was a positive relationship between
youths’ connectedness to neighborhood and adolescents’ subjective well-being (See
Table 5). However, there was no direct relationship between the overall index of
neighborhood characteristics and adolescents’ subjective well-being in the path analysis

(See Figure 2).

61

HA7: Parental characteristics, quality of parenting, peer interactions, and connectedness
to school mediate the relation between neighborhood characteristics and adolescents’

academic achievement.

The results of the path analysis indicated that the relations between neighborhood
characteristics and adolescents’ academic achievement were mediated by parental
characteristics and parenting behaviors. More speciﬁcally, the overall index of
neighborhood characteristics was related to parental characteristics and parenting
behaviors. The parental characteristics composite index, in turn, was associated with
adolescents’ broad reading scores and applied problems scores. Also the results of the
path analysis indicated that the parenting behaviors were subsequently related to
adolescents’ academic achievement. Although the adolescents’ peer interactions and
connectedness to school signiﬁcantly related to adolescents’ academic achievement in the
hierarchical regression analyses, the results of the path analysis indicated that the overall
index of peer interactions and adolescents’ connectedness to school were not

subsequently related to adolescents’ academic achievement.

HA8: Parental characteristics, quality of parenting, peer interactions, and connectedness
to school mediate the relation between neighborhood and adolescents’ subjective well-

being.

The results of the path analysis indicated that parenting behaviors mediated the
association between neighborhood characteristics and adolescents’ subjective well-being.

In addition, parental characteristics, peer interactions, and adolescents’ connectedness to

62

school together mediated the association between neighborhood characteristics and
adolescents’ subjective well-being. More speciﬁcally, parental characteristics were
related to both peer interactions and connectedness to school, which in turn were

signiﬁcantly associated with adolescents’ subjective well-being.

63

CHAPTER FIVE
DISCUSSION

In this chapter, a summary of the results pertaining to each of the research
questions, conclusions, limitations, theoretical implications, and suggestions for future
research are presented.
Summary of the study

The purpose of this study was to examine the process by which neighborhood
conditions inﬂuence parents, peers, and schools, and the combined effect of
neighborhood conditions, parents, peers, and schools on the academic achievement and
subjective well-being of 12 to 17 year-old adolescents using the PSID -- CDS H data. In
this section, the ﬁndings of this study pertaining to the research questions of the study are

summarized.

Research question 1
How do neighborhoods inﬂuence the academic achievement and subjective well-being of

adolescent?

Results of the study show that among the neighborhood characteristics, only
youth’s connectedness to neighborhood had a direct effect on the subjective well—being of
adolescents. The result of the path analysis, however, showed that there is no direct
relationship between the overall index of neighborhood characteristics and adolescents’

subjective well-being.

When other predictor variables (i.e., parents, peers, school) including family SES
are statistically controlled, the primary caregiver’s perceptions of neighborhood and
social control were not predictive of academic achievement. The results of the path
analysis also showed that there was no direct relationship between the overall index of

neighborhood characteristics and adolescents’ academic achievement.

Reseagh Question 2
Are neighborhood characteristics related to the characteristics of parents (i.e., self—esteem,
self-efﬁcacy, psychological distress), parenting practices, peer interactions, and

connectedness to school?

The results of the correlational analysis indicated that all of the neighborhood
characteristics were signiﬁcantly related to parental characteristics such as parental
difﬁculty, self-efﬁcacy, self-esteem, and psychological distress. In terms of parenting
practices, there was a signiﬁcant relationship between perceptions of neighborhood and
parental monitoring. However, parental monitoring was not related to perceptions of
social control and youths’ connectedness to neighborhood. There were strong positive
relationships between perceptions of neighborhood and home environment, and between
youths’ connectedness to neighborhood and home environment.

There were relatively weak relationships between all of the neighborhood
predictors and peer interactions. In addition, all of the neighborhood predictor variables

were signiﬁcantly related to adolescents’ perception of connectedness to school.

65

According to the results of the path analysis, the overall index of neighborhood
characteristics had a direct effect on parental characteristics, parenting behaviors, and
connectedness to school. However, this direct relationship was not found between the

overall index of neighborhood characteristics and peer interactions.

Research question 3

Are the characteristics of parents (i.e., self-esteem, self-efﬁcacy, psychological distress),
parenting practices, peer interactions, and connectedness to school, in turn, predictive of
academic achievement (i.e., WJ-R Achievement test scores) and the subjective well-

being of adolescents?

For the adolescents’ broad reading scores, primary caregivers’ psychological
distress, parental monitoring, home environment, and connectedness to school were
signiﬁcant predictors. In other words, adolescents who had less psychologically
distressed caregivers, who were closely monitored by caregivers, and who had better
home environments obtained higher broad reading scores. Also, adolescents who were
more connected with school had higher broad reading scores. Contrary to expectations,
the primary caregiver’s self-esteem and peer academic orientation were inversely related
to the adolescents’ broad reading scores in the ﬁnal model of the regression analysis.

The results were similar for the adolescents’ applied problems scores. Parental
difﬁculty, self—efﬁcacy, self-esteem, parental monitoring, peer deviant behaviors, peer
academic orientation, and connectedness to school were predictive of the adolescents’

applied problems scores when family SES variables were statistically controlled. Again,

66

the primary caregivers’ self-esteem and peer academic orientation were negatively related
to the adolescents’ applied problem scores.

For adolescents’ subjective well-being, parental monitoring, home environments,
closeness to friends, peer academic orientation, and closeness to schools were signiﬁcant
predictors. The quality of the home environments appeared to be inversely related to
adolescents’ subjective well-being when other factors were controlled with this sample.

The results of the path analysis showed that parental characteristics and parenting
behaviors had direct effects on both broad reading scores and applied problems scores.
Although there was a direct relationship between parenting behaviors and peers and
school, the overall indices of peers and school characteristics were not related to the
adolescents’ academic achievement scores.

Although the overall index of parental characteristics was not directly related to
the adolescents’ subjective well-being, there was an indirect relationship between
parental characteristics and adolescent’s subjective well-being by way of parenting

behaviors, peers, and school.

Research question 4
Do the characteristics of the parents (i.e., self-esteem, self-efﬁcacy, psychological
distress), parenting practices, peer interactions, and connectedness to school mediate the

relation between neighborhood characteristics and adolescent academic achievement?

Parental characteristics mediated the relationship between neighborhood

characteristics and adolescents’ academic achievement. Also, parenting behaviors

67

mediated the relationship between neighborhood characteristics and adolescents’
academic achievement. Although the composite index of neighborhood characteristics
was directly related to connectedness to school, the overall indices of peer interactions
and connectedness to school did not mediate the relationship between the composite

index of neighborhood characteristics and adolescents’ academic achievement.

Reseapch guestion 5
Do the characteristics of the parents (i.e., self-esteem, self—efﬁcacy, psychological
distress), parenting practices, peer interactions, and connectedness to school mediate the

relation between neighborhood characteristics and adolescents’ subjective well-being?

For adolescents’ subjective well-being, the overall index of parental
characteristics had no direct effects on subjective well-being, but there were indirect
relations between parental characteristics and adolescents’ subjective well-being by way
of parenting behaviors, peer interactions and connectedness to school. Thus, parental
characteristics mediated the relationship between neighborhood characteristics and
adolescents’ subjective well-being through parenting behavior, peer interactions and
closeness to school. Parenting behavior mediated the relationship between the composite
index of neighborhood quality and adolescents’ subjective well-being; parenting behavior
also had an indirect effect on adolescent’s subjective well-being via peer group and
school characteristics.

Conclusions

68

Findings from this study are consistent with a model of the effects of
neighborhoods on adolescent outcomes by Leventhal and Brooks-Gunn (2000).
Leventhal and Brooks-Gunn emphasized three pathways (i.e., availability of institutional
resources, relationships, norms/collective efﬁcacy) through which neighborhoods may
inﬂuence adolescent outcomes. For example, ﬁndings from this study conﬁrm that the
institutional resources, especially adolescents’ schooling experiences, inﬂuence
adolescents’ subjective well-being. The results are consistent with the ﬁndings of
previous research about school characteristics. That is, in a prior study, school
characteristics (e.g., school safety, school attachment, and availability of substances)
mediated the relationship between neighborhood characteristics and substance use by
early adolescents (Ennett, et al., 1997).

It is important to note that family demographic variables are still important
factors predicting adolescent outcomes. In this study, parents’ level of education and
parents’ age were signiﬁcantly related to the adolescents’ academic achievement.

The ﬁndings of this study also conﬁrm the relationship between parental
characteristics and parenting behaviors. The results of the path analysis indicated that the
overall parental characteristics composite was positively related to the overall index of
parenting behaviors. In addition, the overall index of neighborhood characteristics
positively inﬂuenced the parental characteristics and parenting behaviors composites,
which in turn, were related to adolescents’ peer interactions, and connectedness to school;
both peer group characteristics and connectedness to school predicted adolescents’

subjective well-being.

69

The ﬁndings of this study extend previous research on the relationships among
parental characteristics, parenting behaviors, and the development of adolescents by
including peer interactions and schooling experiences as potential mediators between
parental characteristics and adolescent outcomes. The results of this study showed that
parental behaviors had direct effects on adolescents’ academic achievement and
subjective well-being. Also parental characteristics had a direct effect on adolescents’
academic achievement and had an indirect effect on adolescents’ subjective well-being
via peer interactions and connectedness with school.

Although perceptions of social control among neighbors and perceptions of
neighborhood were not signiﬁcant predictors of adolescents’ academic achievement
when other predictor variables were statistically controlled, the present study showed the
direct effect of youths’ connectedness to neighborhood on the adolescents’ subjective
well-being. However, the results of the path analysis indicated that the overall
neighborhood characteristics composite had no direct effect on adolescents’ subjective
well-being. Peer interactions and school experiences, along with parental characteristics
and parenting behaviors, mediated the effect of overall neighborhood characteristics on
adolescents’ subjective well-being. Moreover, it was found that neighborhood

characteristics had a direct effect on adolescents’ school experiences. Therefore, it is

.....7
Fri-‘— ,

H-ﬂ'

important to note that educational resources are important factors not only for young
children’s outcomes (Campbell & Ramey, 1994; Lee, et al., 1990; Reynolds, 1994;
Yoshikawa, 1994), but also for adolescent outcomes.

Limitations

70

 

Although the proposed model adequately represented the subjective measures of
neighborhood characteristics, several other variables would have been useful to include in
the analysis in order to test a more complete model. For example, objective
neighborhood characteristics variables (i.e., census tract variables) were not included in
this study due to the limited ﬁnancial resources available to conduct this study. In
addition, information regarding school characteristics was limited. Additional measures
of school environments and the quality of the schools would have strengthened the study
further.

In this study, 88% “of the» primary caregivers were mothers; thus, it was not
possible to focus on the relationship between paternal characteristics and the
developmental outcomes of adolescents even with national-level data such as the PSID
CDS H.

Theoretical Implications

This study was based on the framework of Leventhal and Brooks-Gunn, in which
they identiﬁed three theoretical models for linking neighborhood inﬂuences and
adolescent outcomes. The results of the research provide support for their theoretical
models.

Leventhal and Brooks-Gunn proposed that the quality and availability of
institutional resources in the neighborhood may inﬂuence adolescent outcomes. Data
from this study provided evidence that family economic resources and characteristics of
the school environment inﬂuenced adolescents’ subjective well-being. Adolescents’

connectedness to school functioned as Leventhal and Brooks-Gunn had proposed.

71

The results of this study also conﬁrmed that parental characteristics and
parenting behaviors mediated the relationship between neighborhood characteristics and
adolescent outcomes; a wide range of parental characteristics and parenting behaviors
was included in the analysis. The results are consistent with the view that parental
characteristics and parenting behaviors are still powerful mediators of neighborhood
inﬂuences on adolescence, while other contexts such as peers and schools become more
salient inﬂuences presumably as adolescents’ contact and exposure to these environments
increases.

Although previous research found that norms/collective efﬁcacy (e. g., social
control) were related to adolescents’ problem behaviors (Elliott, et al., 1996; Sampson,
1997), the results of this study indicated that social control had no effect on adolescents’
academic achievement and subjective well-being. However, one should not disregard the
effects of other aspects of the neighborhood, such as subjective quality of the
neighborhood and community participation, on adolescent outcomes. The result of this
study indicated that the youths’ connectedness to neighborhood is associated with
adolescents’ subjective well-being.

Finally, the outcomes of this study provide further support for an ecological
perspective, showing that researchers need to examine the multiple contexts that
inﬂuence adolescents and their families. The ﬁndings imply that it is important to
examine the different contexts where adolescents spend their time (home, school, peer
group, community) in order to understand neighborhood inﬂuences on adolescent
outcomes.

Suggestions for future research

72

Much of the previous research on the effects of neighborhood on adolescent
outcomes has focused on identifying the negative consequences of neighborhood
income/SES on adolescents’ developmental outcome. Most studies, therefore, have
relied on census-based measures of neighborhoods. Researchers have tended to overlook
the social organizational dimensions of neighborhoods, which are not based on census
tract level data, and these aspects of neighborhoods may have an effect on youth
outcomes. However, recently some sociologists and urban scholars have examined
neighborhood characteristics that are associated with neighborhood social organization
and their impact on adolescents’ outcomes (Coulton et al., 1995; Elliott et al., 1996;
Sampson & Groves, 1989). In addition, scholars with an ecological perspective should
also examine the links between neighborhoods and other contexts where adolescents
spend their time, including peers and schools.

Ideally, the effects of both objective and subjective indicators of neighborhood
quality on adolescents’ outcome should be examined together; however, the present study
could not investigate these objective variables because of the lack of the funding and time
constraints. In the future, I would like to extend the ﬁndings from this study by adding
objective neighborhood indicators to the analysis. It is important to explore the relation
between these objective indicators of neighborhoods and the key contexts where
adolescents spend much of their time, such as at home, with peers, and in school. Future
studies may also focus on the variables used in this study with children in different age
groups (i.e., early childhood, late childhood, early adolescence, and late adolescence). It
may be meaningful to study the unique contributions of each system (e.g., parents, peers,

school) on the various age groups because each age group will experience one or more

73

major transitions over time. For instance, school entrance and biological maturation are
some of the developmental challenges in a child’s life. Thus, the salience of mediators

linking neighborhood characteristics and developmental outcomes may vary as a function

of the age of the children.

74

BIBLOGRAPHY

Aneshensel, C. S., & Sucoff, C. A. (1996). The neighborhood context and adolescent
mental health. Journal of Health and Social Behavior, 3 7, 293-310.

Baker, P. C., & Mott, F. L. (1989). NLS Y child handbook 1989: A guide and resource
documents for the National Longitudinal Survey of Youth 1986 Child Data.
Columbus, OH: Center for Human Resources Research, Ohio State University.

Baldwin, A. L., Baldwin, C., & Cole, R. E. (1990). Stress-resistant families and
stress-resistant children. In J. Rolf, A. S. Masten, D. Cicchetti, K. H.
Nuechterlein, & S. Weintraub (Eds.), Risk and protective factors in the
development of psychopathology (pp. 257-280). Cambridge, England:
Cambridge University Press.

Belsky, J. (1984). The determinants of parenting: A process model. Child
Development,_55, 83-96.

Belsky, J. (1990). Parental and nonparental child care and children’s socioemotional
development: A decade in review. Journal of Marriage and Family, 52, 885-903.

Bradley, R. H. (1995). Environment and parenting. In M. Bomstein (Ed.), Handbook of
parenting (pp. 235-261), Hillsdale, NJ: Erlbaum.

Bradley, R. H., & Caldwell, B. M. (1984). The relation of infants’ home environments to
achievement test performance in ﬁrst grade: A follow-up study. Child
Development, 55, 803-809.

Bradley, R. H., Caldwell, B. M., Rock, 8. L., Barnard, K. B, Gray, C., Hammond, M. A.,
Mitchell, 8., Siegel, L., Ramey, C. T., Gottfried, A. W., & Johnson, D. L. (1989).
Home environment and cognitive development in the ﬁrst 3 years of life: A
collaborative study involving six sites and three ethnic groups in North America.
Developmental Psychology, 25, 217-235.

Bronfenbrenner, U. (1979). The ecology of human development. Cambridge: Harvard
University Press.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development
research perspectives. Developmental Psychology, 22, 723-742.

Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Annals of child
development-Six theories of child development: Revised formulation and current
issues (pp. 187-250). Greenwich, CT: JAI Press.

Brooks-Gunn, J ., Duncan, G. J ., & Aber, J. L. (1997). Neighborhood poverty: Vol I.
Context and consequences for children. New York: Russell Sage Foundation.

75

Brooks-Gunn. J ., Duncan, G. J ., Klebanov, P. K, & Sealand, N. (1993). Do
neighborhoods inﬂuence child and adolescent development? American Journal of
Sociology, 99, 353-395.

Brooks—Gunn. J ., Duncan, G. J ., Kato, P., & Sealand, N. (1991). Do neighborhoods
inﬂuence child and adolescent development? Paper presented at the meeting of

the Society for Research in Child Development. Seattle, WA.

Burton, L. M., Price-Spratlen, T., & Spencer, M. B. (1997). On ways of thinking about
measuring neighborhoods: Implication for studying context and developmental
outcomes for children. In J. Brooks-Gunn, G. J. Duncan, & J. L. Aber (Eds.),
Neighborhood poverty: Contexts and consequences for children (Vol. I) (pp. 132-
144). New York: Russell Sage Foundation.

Burton, L. M., & Jarrett, R. L. (2000). In the mix, yet on the margins: The place of
families in urban neighborhood and child development research. Journal of
Marriage and the Family, 62, 1 114-1135.

Caldwell, B. M., & Bradley, R. H. (1984). Home observation for measurement of the
environment. Little Rock, AR.

Campbell, F. A., & Ramey, C. T. (1994). Effects of early intervention on intellectual and
academic achievement: A follow-up study of children from low-income families.
Child Development, 65, 684-698.

Chase-Lansdale, P. L., Gordon, R. (1996). Economic hardship and development of ﬁve
and six-year olds: Neighborhood and regional perspectives. Child Development,
67, 3338-3385.

Colletta, N. D. (1983). At risk for depression: A study of young mothers. Journal of
Genetic Psychology, 142L 301 -3 10.

Coulton, C. J ., Korbin, J., Chan, T., & Su, M. (2001). Mapping residents’ perception of
neighborhood boundaries: A methodological note. American Journal of
Community Psychology, 29, 371-383.

Dishon, T. J ., Andrews, D. W., & Crosby, L. (1995). Antisocial boys and their friends in
early adolescence: Relationship characteristics, quality, and interactional process.
Child Development, 66, 139-151.

Dombusch, S., Ritter, P., & Steinberg, L. (1991). Community inﬂuences on the relation

of family statuses to adolescent school performance among Aﬁican-American and
non-Hispanic Whites. American Journal of Education, 38, 543-567.

76

Dubrow, N., & Garbarino, J. (1989). Living in the war zone: Mothers and children in
public housing development. Child Welfare, 68, 3-20.

Duncan, G. J. (1994). Families and neighbors as sources of disadvantage in the school
decisions of White and Black adolescents. American Journal of Education, 103,
20-53.

Duncan, G. J ., Brooks-Gunn, J. P., & Klebanov, P. K. (1994). Economic deprivation and
early-childhood development. Child Development, 65, 296-318.

Duncan, G. J ., Connell, J. P., & Klebanov, P. K. (1997). Conceptual and methodological
issues in estimating causal effects of neighborhoods and family conditions on
individual development. In J. Brooks-Gunn, G. J. Duncan, 8L J. L. Aber (Eds.),
Neighborhood poverty: Contexts and consequences for children (Vol. I) (pp. 219-
250). New York: Russell Sage Foundation.

Durkin, M. 8., Davidson, L. L., Kuhn, L., O’Connor, P., 8; Barlow, B. (1994). Low-
income neighborhoods and the risk of severe pediatric injury: A small-area
analysis in Northern Manhattan. American Journal of Public Health, 84, 587-592.

Elliott, D., Wilson, W. J ., Huizinga, D., Sampson, R. J ., Elliott, A., & Rankin, B. (1996).
Effects of neighborhood disadvantages on adolescent development. Journal of
Research in Crime and Delinquency, 33, 389-426.

Ennett, S. T., F lewelling, R. L. Lindrooth, R. C., & Norton, E. C. (1997). School and
neighborhood characteristics associated with school rates of alcohol, cigarette,
and marijuana use. Journal of Health and Social Behavior, 38, 55-71.

Entwisle, D. R., Alexander, K. L., & Olson, L. S. (1994). The gender gap in math: Its
possible origins in neighborhood effects. American Sociological Review, 59, 822-
838.

Fuller, 8., Coonerty, C., Kipnis, F ., & Choong, Y. (1997). An unfair head start:
California families face gaps in preschool and child care availability. Berkeley,
CA: Berkeley-Stanford PACE Center, Yale University, and the Califomia Child
Care Resource and Referral Network: Growing up in Poverty Project.

Furstenberg, F. F., Jr., Coo, T. D., Eccles, J ., Elder, G. H., Jr., & Sameroff, A. (1999).
Managing to make it: Urban families and adolescent success. Chicago:
University of Chicago Press.

Garbarino, J. (1976). A preliminary study of some ecological correlates of child abuse:
The impact of socioeconomic stress on mothers. Child Development, 47, 178-185.

Garbarino, J ., Kostelny, K., & Dubrow, N. (1991). No place to be a child: Growing up in
a war zone. New York, NY: Lexington Books.

77

Garbarino, J ., & Kostelny K. (1993). Neighborhood and community inﬂuences on
parenting. In T. Luster, & L. Okagaki (Eds.), Parenting: An ecological
perspective (pp. 203-226). Hillsdale, NJ: Erlbaum.

Gross, R. T., Spiker, D., & Haynes, C. W. (Eds.). (1997). Helping low birth weight,
premature babies: The infant health and development program. Stanford, CA:
Stanford University Press.

Halpem-Felsher, B., Connell, J ., Spencer, M., Aber, L., Duncan, G., Clifford, E.,
Crichlow. W., Usinger, P., Cole, 8., Allen, L., & Seidman, E. (1997).
Neighborhood and family factors predicting educational risk and attainment in
African American and White Children and adolescents. In J. Brooks-Gunn, G. J.
Duncan, & J. L. Aber (Eds.), Neighborhood poverty: Contexts and consequences
for children (Vol. I) (pp. 146-173). New York: Russell Sage Foundation.

Infant Health and Development Program. (1990). Enhancing the outcomes of low
birthweight, premature infants: A multisite randomized trial. Journal of the
American Medical Association, 263, 3035-3042.

Institute for Social Research. (2004). The Child Development Supplement to The
Panel Study of Income Dynamics. Ann Arbor. MI: University of Michigan

J argowsky, P. A., & Bane, M. J. (1990). Ghetto-poverty: Basic questions. In L. E.
Lynn, Jr., & G.H. McGeary (Eds.), Inner-city poverty in the United States (pp.
235-280). Washington, DC: National Academy Press.

Jarrett, R. L. (1997). Bringing families back in: Neighborhoods’ effects on child
development. In J. Brooks-Gunn, G. J. Duncan, & J. L. Aber (Eds.),

Neighborhood poverty: Policy implications in studying neighborhood (Vol. 2)
(pp. 48-64). New York: Russell Sage Foundation.

J encks, C., & Mayer, S. E. (1990). The social consequences of growing up in a poor
neighborhood. In L. E. Lynn, Jr., & G.H. McGeary (Eds.), Inner-city poverty
in the United States (pp. 111-186). Washington, DC: National Academy Press.

Kalil, A., & Eccles, J. S. (1998). Does welfare affect family processes and adolescent
adjustment? Child Development, 69, 1597-1613.

Kaufman, J ., & Rosenbaum, J. (1992). The education and employment of low-income
black youth in white suburbs. Educational Evaluation and Policy Analysis,14,
229-240.

Klebanov, P.K. Brooks-Gunn, J ., Chase-Lansdale, P. L., & Gordon, R. (1997). Are

neighborhood effects on young children mediated by features of the home
environment? In J. Brooks-Gunn, G. J. Duncan, & J. L. Aber (Eds.),

78

Neighborhood poverty: Contexts and consequences for children (Vol. I) (pp.
119-145). New York: Russell Sage Foundation.

Klebanov, P. K., Brooks-Gunn, J ., & Duncan, G. J. (1994). Does neighborhood and
family poverty affect mothers’ parenting, mental health, and social support?
Journal of Marriage and the Family, 5 6, 441-455.

Klebanov, P. K., Brooks-Gunn, J ., McCarton, C. M., & McCormick, M. C. (1998).
The contribution of neighborhood and family income upon developmental test
scores over the ﬁrst three years of life. Child Development, 69, 1420-1436.

Kochenderfer, B.J., & Ladd, G. W. (1996). Peer victimization: Cause of consequence
of school maladjustment? Child Development, 67, 1305-1317.

Korbin, J. & Coulton, C. (1997). Understanding the neighborhood context for children
and families: Combining epidemiological and ethnographic approaches. In J.
Brooks-Gunn, G. J. Duncan, & J. L. Aber (Eds.), Neighborhood poverty:
Policy implications in studying neighborhood (Vol. 2) (pp. 65 - 79). New
York: Russell Sage Foundation.

Lee, V., Brooks-Gunn, J ., Schnur, E., & Liaw, F. (1990). Are Head Start effects
sustained? A longitudinal comparison of disadvantaged children attending

Head Start, no preschool, and other preschool programs. Child Development,
61, 495-507.

Leventhal, T., & Brooks-Gunn, J. (2000). The neighborhoods they live in: The
effects of neighborhood residence on child and adolescent outcomes.
Psychological Bulletin, 126, 309-337.

Loeber, R., & Wikstrom, P. H. (1993). Individual pathways to crime in different types
of neighborhoods. In D. P. Farrington, R.J. Sampson, & P.H. Wikstrom (Eds.),
Integrating individual and ecological aspects of crime (pp. 169-204).
Stokholrn: National Council for Crime Prevention.

Ludwig, J ., Duncan, G., & Hirschﬁeld, P. (1998). Urban poverty and juvenile crime:
Evidence from a randomized housing-mobility experiment. Unpublished
manuscript. Northwestern University.

McLoyd, V. C. (1990). The impact of economic hardship on black families and
children: Psychological distress, parenting, and socioemotional development.
Child Development, 61 , 311-346.

McLoyd, V. C., & Wilson, L. (1991) The strain of living poor: Parenting, social

support, child mental health. In A. C. Huston (Ed.), Children in poverty (pp.
105-135). Canada: Cambridge University Press.

79

Meyer, S. E., & J encks, C. (1989). Growing up in poor neighborhoods: How much
does it matter? Science, 1 7, 1441-1445.

Meyers, S. A., & Miller, C. (2004). Direct, mediated, moderated, and cumulative

relations between neighborhood characteristics and adolescent outcomes.
Adolescence, 39, 121-144.

MIDUS (n.d.) (October 15, 2004) http://midmac/med/harvard/edu/

Moore, K. A. (1999). Indicators of Child and Family Well-Being: The Good, the Bad,

and the Ugly. Bethesda, MD: National Institutes of Health, Ofﬁce of
Behavioral and Social Sciences, 1999 Seminar Series.

Oh, S., & Luster, T. (1998). The inﬂuence of neighborhood characteristics on home
environment and children’s achievement. Paper presented at the National
Council on Family Relations Annual Meeting, Milwaukee, WI.

Pearlin, L. 1., Lieberman, M. A., Menaghan, E. G., & Mullan, J. T. (1981). The stress
process. Journal of Health and Social Behavior, 22, 337-356.

Peeples, F., & Loeber, R. (1994). Do individual factors and neighborhood context explain
ethnic differences in juvenile delinquency. Journal of Quantitative Criminology,
10, 141-157.

Pinderhughes, E. E., Nix, R., Foster, E. M., Jones, D., & The Conduct Problems
Prevention Research Group (2001). Parenting in context: Impact of neighborhood
poverty, public services, social networks, and danger on parental behaviors.
Journal of Marriage and the Family, 63, 941-953.

Pollard, E. L., & Davidson, L. (2001). Foundations of Child Well-being. UNESCO
Education Sector Monograph, No 18/2001.

Reynolds, A. J. (1994). Effects of a preschool plus follow-on intervention for children at
risk. Developmental Psychology, 30, 787-804.

Rosenberg, M. (1986). Conceiving the Self. New York: Basic Books.

Sampson, R. J. (2001). How do communities undergird or undermine human
development? Relevant contexts and social mechanisms. In A. Booth & A. C.
Crouter (Eds.), Does it take a village? Community eﬂects on children,
adolescents, and families (pp. 3-30). Mahwah, NJ: Erlbaum

Sampson, R. J. (1992). Family management and child development: Insights from
social disorganization theory. In J. McCord (Ed.), Advances in criminological
theory (Vol. 3) (pp. 63-93). New Brunswick, NJ: Transaction Books.

80

\
\

{ __... .

Sampson, R. J. (1997). Collective regulation of adolescent misbehavior: Validation
results from eighty Chicago neighborhoods. Journal of Adolescent Research,
12, 227-244.

Sampson, R. J ., & Groves, W. B. (1989). Community structure and crime: Testing
social-disorganizational theory. American Journal of Sociologi, 94, 774-780.

Sampson, R. J ., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent
crime: A multilevel study of collective efﬁcacy. Science, 277, 918-924.

Shumow, L., Vandell, D. L., Posner, J. (1999). Risk and resilience in the urban
neighborhood: Predictors of academic performance among low-income
elementary school children. Merrill-Palmer Quarterly, 45, 309-331.

Simons, R.L. Johnson, C., Beaman, J. J ., Conger, R. D., & Whitbeck, L. B. (1996).
Parents and peer group as mediators of the effect of community structure on
adolescent behavior. American Journal of Community Psychology, 24, 145-
171.

Simons, R. L., Johnson, C., Conger, R. D., & Elder, G. R, Jr. (1998). A test of latent
trait versus life course perspectives on the stability of adolescent antisocial
behavior. Criminology, 36, 217-244.

Simons, R. L., Lin, K, Gordon, L. C., Brody, G. H., & Conger, R. D. (2002).
Community differences in the association between parenting practices and
child conduct problems. Journal of Marriage and Family, 64, 331-345.

Solon, G., Page, M., & Duncan, G. (1997). Correlations between neighboring children
in their socioeconomic status as adults. Unpublished manuscript.
University of Michigan.

Survey Research Center. (1984). User guide to the Panel Study of Income Dynamics.
Ann Arbor, MI: Inter-University Consortium for Political and Social Research.

Survey Research Center. (1997). User guide to the Child Development Supplement to
the Panel Study of Income Dynamics. Ann Arbor, M1: the University of
Michigan, Institute for Social Research.

Wilson, W. J. (1987). The truly disadvantaged: The inner city, the underclass, and
public policy. Chicago: University of Chicago Press.

Wilson, W. J. (1991). Public policy research and the truly disadvantages. In C. J encks, &
P. E. Peterson (Eds.), The urban underclass (pp. 460-481). Washington, DC:
Brookings Institution.

Woodcock, R. W., & Johnson, M. B. (1989). Tests of Achievement, Standard Battery

81

[Form B]. Chicago, IL: Riverside Publishing (Houghton Mifﬂin).

Yoshikawa, H. (1994). Prevention as cumulative protection: Effects of early family
support and education on chronic delinquency and its risks. Psychological
Bulletin, 115, 28-54.

82

llllllllllllllllllll