From conversation to conversion: children's efforts to translate their immigrant families' social networks into community connections

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

FROM CONVERSATION TO CONVERSION: CHILDREN’S EFFORTS
TO TRANSLATE THEIR IMMIGRANT FAMILIES’ SOCIAL
NETWORKS INTO COMMUNITY CONNECTIONS

by
Vikki Sara Katz
______________________________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(COMMUNICATION)

August 2007

Copyright 2007 Vikki Sara Katz
ii

DEDICATION
In memory of
My grandfather, Simon Katz (1911-2003), my family’s embodiment of the sacrifices,
successes, and perseverance that constitute the Immigrant Dream;
&
Professor Jim Bruno (1940-2006) who made me believe that I could write about it.
iii

ACKNOWLEDGMENTS
Writing this dissertation has had the side benefit of highlighting the
considerable resources in my own social networks, and I am thankful to a number of
people who supported me through this project.
I thank my family for their unwavering support. This project was first
inspired by the sacrifices of my own parents, Cheryl and Ian, to come to the United
States from South Africa in 1994, so that my brother Jeff and I would have access to
all that this country has to offer. The journey, and the team it made of the four of us,
made me want to understand how the family dynamics would be different or similar
when language and other factors were part of the experience. I am grateful for Dad’s
patient editing and nightly emails, Mom’s lunchtime phone calls, and Jeff’s
drivetime chats, all of which were daily reminders that you were present, even at a
distance.
This dissertation bears the indelible mark of Dr. Sandra Ball-Rokeach, my
advisor, my mentor, and my role model. With tireless energy and dedication, she
shaped my graduate education, and the work I did in these past five years, into
something I am not only proud of, but could not have accomplished without her
guidance. I thank her sincerely for all that she has given me. Dr. Michael Cody, Dr.
Peggy McLaughlin, and Dr. Robert Rueda, have nourished this project in so many
ways, and I (and it) have benefited greatly from your guidance. Thank you for all
that you have taught me.
iv

Judy Manouchehri was my cheerleader, intrepid editor and celebrant of even
the smallest successes through the writing process. I am blessed to have your
friendship and support, in this project and in the larger one.
The past and present members of the Metamorphosis Project, which has been
my professional home for the past five years, have been my colleagues and friends.
My time at USC will always be synonymous with the Meta office, and with the
collaboration and support I received from its members.
Thank you to the colleagues who I am lucky to call my friends, and who
contributed to this dissertation with ongoing conversations, referrals to the ‘good
books,’ and reminding me that this project was interesting to people other than
myself: Lourdes Baezconde-Garbanati, Sarah Banet-Weiser, Manuel Castells, Tara
Chklowski, Johann D’Agostino, Marc Davidson, David Fitzgerald, Howie Giles,
Larry Gross, Carmen Gonzalez, Marjorie Orellana, Renee Reichl, Roger Waldinger,
and Jennifer Wilson. Additional thanks to David Fitzgerald for thoughtful editing
and challenges that improved this work.
Holly Wilkin was my statistical wizard: The strengths of Chapter 3 are
entirely attributable to your unending patience and expertise. The weaknesses and
errors are of course, my own.
I am grateful to Matthew Matsaganis, who has shared this journey with me,
and is my friend, my colleague, and finally, my co-author! Thank you for your
editing, support, and thoughtful interventions into this project.
v

Thank you to my dear friends who supported me, successfully feigned
interest in what I was doing when I was working through weekends and missing
special events, and very importantly, reminded me when it was time to take breaks:
James Ahn, Booby Armin, Zubin Chagpar, Pamela Cheng, Vanessa Hernandez,
Marcus Lam, Kami Manouchehri, Lucy Montgomery, Lior Ziv, and Leah & Gili
Tilinger, for meals that nourished me in many ways.
I am grateful to the Annenberg Foundation for their generosity in supporting
my graduate education with an Annenberg Fellowship, and to the Annenberg School
for Communication, whose seemingly unending support came in many forms. I am
especially grateful to Larry Gross, Randy Lake, Justin Acome, Anne Marie Campian,
Christina Lloreda, and Geoff Cowan for the myriad forms of support they have
granted me over the years.
The Urban Initiative at the University of Southern California supported the
research presented in this dissertation with a Dissertation Research Fellowship. I am
grateful to them for making this project possible.
Beni Gonzalez and Michelle Hawks, my superstar assistants, I owe a
gratitude I cannot find words for. I can only say that without your dedication,
companionship, and unending energy and enthusiasm for this project, it could never
have happened. Watching the two of you, child brokers yourselves, place yourselves
in this project, was one of the best parts of the whole experience for me.
vi

Finally, my thanks to the many who shared their stories and perspectives with
me in the schools, clinics, and social service offices of Greater Crenshaw. To the
parents and children who humbled me with their trust and opened their hearts and
homes to me, Beni and Michelle: les agredezco mucho la confianza que me dieron y
les doy mi respeto con todo mi corazon.
vii

TABLE OF CONTENTS

DEDICATION ii

ACKNOWLEDGMENTS iii

LIST OF TABLES xi

LIST OF FIGURES xiii

ABSTRACT xiv

INTRODUCTION 1
Who are These Children? 3
The Research Site 7
Overview of the Dissertation 11

CHAPTER 1: NETWORKS AND COMMUNITY: A THEORETICAL
FRAMEWORK AND HEURISTIC MODEL 14
Communication Infrastructure Theory 20
Resident Networks 21
Community Organizations and Local Media 27
Bridging and Bonding in the Storytelling Network 31
The Communication Action Context 32
Networks in Institutions 37
Domestic Infrastructure 41

CHAPTER 2: RESEARCH DESIGN 50
Research Questions 50
Telephone Survey Methods 56
Survey Samples: Spanish vs. English-Speaking Respondents 57
Survey Samples: High Brokering vs. No/Low Brokering Respondents 58
Survey Measures 61
Interviews 69
Data Analysis 77
Field Observation: Site Visits 79
Community Institution Interviews 79
Data Analysis 84
Employing Multiple Methods 84

viii

CHAPTER 3: SURVEY ANALYSES 91
RQ1: Composition of Respondents’ Informal Networks 91
RQ2: Respondents’ Connections to Community Organizations and
Local Media 96
RQ3: Integration of Respondents’ Storytelling Networks 105
RQ4: Respondents’ Connections to Community Institutions 105
Discussion 108

CHAPTER 4: INFORMAL NETWORKS AND DOMESTIC
INFRASTRUCTURE 120
Access to Media Artifacts 125
Brokering Activities and the Family’s Range of Storytelling Resources 130
Linguistic Brokering 131
Cultural Brokering 135
Technological Brokering 137
Quality of Family Social Arrangements 139
The Invisibility of Brokering 140
Parental Authority 141
Sibling Cooperation 142
Inhibiting Family Arrangements 143
Embedded Reward Structures for Brokering 147
From Family to Peer Networks 149
Child Brokering as a Resource in Peer Networks 150
Peer Networks as Resources for Child Brokering 154
Peer Networks that Bridge: Getting Ahead 156
Conclusion 158

CHAPTER 5: BROKERING AN EDUCATION: CHILDREN’S
EFFORTS TO CONNECT WITH SCHOOL NETWORK RESOURCES 159
Child Brokers’ Connections to Institutional Network Nodes 164
Embodied Capital: Enabling and Constraining Factors 165
Teacher-Student Relationships: Aesthetic and Authentic Caring 165
Teachers’ Capacity as Resource Brokers 171
Connections with Institutional Capital: Enabling and Constraining
Factors 174
After-School Programs 174
Objectified Capital: Enabling and Constraining Factors 175
Curriculum (Dis)Connnection 176
Educational Options as Objectified Capital 179
Individual vs. Collective Connections with Objectified Capital 181
ix

Parents on Campus: Child Brokering in School Contexts 183
Organizational Membership as Access to Institutional Capital 184
School Newsletters as Objectified Capital 185
Parents’ Access to Embodied Capital 187
Brokering Parent-Teacher Meetings 190
Conclusion 194

CHAPTER 6: BROKERING FOR FAMILY HEALTH AND WELLBEING:
CHILDREN’S EFFORTS TO CONNECT WITH HEALTH CARE AND
SOCIAL SERVICE INSTITUTIONS 196
Child Brokers’ Connections with Health Care and Social Service
Institutions 200
Connecting with Embodied Capital 202
Constraining Factor: Feelings of Embarrassment 207
Constraining Factor: Brokers’ Own Illnesses or Pain 208
Enabling Factor: Brokers’ Active Requests for Assistance 209
Enabling Factor: Parent-Broker Scaffolding 211
Enabling or Constraining Factor: Service Provider and Staff
Members’ Attitudes 212
Connecting with Objectified Capital 215
Connecting with Institutional Capital 218
Institutions as Resource Brokers 221
Funders’ Mandates to Broker Resources 221
Connections Initiated by Other Institutions 222
Resource Brokering by Embodied Capital 224
Conclusion 226

CHAPTER 7: THEORETICAL CONSIDERATIONS AND RESEARCH
IMPLICATIONS 227
Theoretical Considerations and Contributions 227
Summary of Findings 228
Methodological Contributions 232
Practical Implications: The Immigrant Bargain and its Relationship to
Parents and Children’s Aspirations 237
Policy Recommendations 243
Research Limitations 247
Suggestions for Future Research 249

REFERENCES 251

x

APPENDICES
Appendix A: Communication Action Context Features 267
Appendix B: Cultural vs. Social Capital 269
xi

LIST OF TABLES
Table 2.1: Demographics of full sample: Spanish and English-speaking
respondents 58

Table 2.2: Demographics of no/low brokering and high brokering
respondents 61

Table 2.3: Demographics of survey and WIC interview pools of broker’s
parents 73

Table 2.4: Summary of interview participants: Parents 75

Table 2.5: Summary of interview participants: Child brokers 76

Table 2.6: Community institution interview participants 83

Table 3.1: Topics frequently discussed with neighbors 94

Table 3.2: Location of respondents’ close friends and family, as
percentages 95

Table 3.3: Respondents who belong to one community organization 97

Table 3.4: Respondents who belong to more than one community
organization 97

Table 3.5: Two most important ways to stay on top of community news and
events: Spanish vs. English-speaking respondents 100

Table 3.6: Two most important ways to stay on top of community news and
events: High vs. no/low brokering respondents 101

Table 3.7: General media connection patterns of Spanish and
English-speaking respondents 103

Table 3.8: General media connection patterns of high and no/low brokering
respondents 104

Table 3.9: Feeling connected to community organizations: Spanish and
English-speaking respondents 106

xii

Table 3.10: Feeling connected to community organizations: Spanish and
English-speaking respondents 107

xiii

LIST OF FIGURES
Figure 1.1. The storytelling network 20

Figure 1.2. Resident networks disaggregated into family and peer
networks 23

Figure 1.3. Community organizations and local media as potential
bridging and/or bonding storytellers 30

Figure 1.4. The storytelling network set in the communication action
context 33

Figure 1.5. The communication infrastructure model from the child
broker’s perspective 35

Figure 1.6. Internal network of community institutions 37

Figure 1.7. Components of the domestic infrastructure 43

Figure 1.8. Domestic infrastructure’s links to peer networks through
child brokering 46

Figure 1.9. Child brokering connections happen at the expense of
their own connections to schools 47

Figure 3.1. Relative positions of English-speaking and Spanish-
speaking no/low and high brokering respondents on RQ1 measures 112

Figure 4.1. The domestic infrastructure 124

Figure 4.2. Links between the domestic infrastructure and peer
networks through child brokering 150

Figure 5.1. The internal network of community institution 161

Figure 6.1. The internal network of community institutions 199

xiv

ABSTRACT
Although children are generally absent from research on immigration and
adaptation, this research demonstrates that children of immigrants make daily
contributions to their families’ community connections. Immigrant Latino parents in
a Los Angeles community indicated close connections to friends and extended
family in their local area. However, the homogenous nature of these interpersonal
relationships effectively constrained Spanish monolingual parents’ knowledge of
available community resources. Children emerged as crucial linkages to community
information resources and institutions for their parents. Children brokered parents’
connections to other residents by translating conversations, and connecting with a
wide range of media for their parents as translators and interpreters of everything
from telephone calls to newsletters mailed to the home.
This research focuses on children’s brokering of community information
resources in family and peer networks, and in local media, and their efforts to
convert these resources into connections with important community institutions for
their parents: schools, health care, and social service institutions.
This research employed communication infrastructure theory (Kim & Ball-
Rokeach, 2006), integrated with domestic infrastructure (Livingstone, 2002) and
social capital theories (Bourdieu, 1986; Putnam, 2000) to inform a multi-level, multi-
method investigation of child brokering activities in larger community context.
Survey analyses, interviews with child brokers and their parents, interviews with
xv

service providers in community schools, health care, and social service institutions,
as well as field observations and site visits with families, collectively inform the
findings in this project.

1

INTRODUCTION
The impetus for this project grew from two primary sources: my own
immigration experience, and a conviction that started as a nagging observation in my
readings of the research on immigration in the United States. As I read article after
article where children’s roles in the immigration experience were ignored, or
mentioned only to detail their seemingly universal rejection of their parents and
family breakdown, I became increasingly convinced that there was an important
perspective on immigration that, with a few notable exceptions (Dorner, Orellana &
Li-Grining. 2006; Orellana, Dorner & Pulido, 2003; Park, 2002; Song, 1999; Thorne
et al., 2003; Tse, 1995, 1996; A. Valenzuela Jr., 1999) was being given short shrift.
Although immigration research in America has burgeoned in recent decades,
only in the past fifteen years have scholars seriously considered intra-family
differences in the migration experience. Feminist scholars have been at the forefront
(Hirsch, 2003; Hondagneu-Sotelo, 1994, 2001, 2003; Kibria, 1993; Pessar, 2003),
presenting a range of settlement contexts where husbands and wives have
experienced immigration differently. These same scholars have shown that different
migration contexts can result in spousal conflict, cooperation, or a combination of
both, in the acculturation experience. Much as female migrants were virtually
undetectable in the literature until recently, the immigration literature’s assumption
of a top-down relationship between parents and children (Portes & Rumbaut, 2000;
2

Zhou & Bankston, 1998), has rendered children’s contributions to the family’s
immigration and settlement, invisible.
The invisibility of children in research on immigrant adaptation is not,
however, the lived experience of immigrants themselves. Children of immigrants,
whether born in the Unites States or immigrants themselves, play essential roles in
the day-to-day survival of their families. Particularly as they enter middle childhood
and adolescence, children “broker
1
” or translate for their parents in many different
settings, mediate cultural differences, access information and media, contribute to the
household income, and/or perform domestic duties and caretaker work (Dorner,
Orellana & Li-Grining. 2006; Orellana, Dorner & Pulido, 2003; Park, 2002; Song,
1999; Tse, 1995, 1996; A. Valenzuela Jr., 1999). Past studies indicate that
adolescence
2
is often associated with a developmental maturity that allows increased
assumption of familial responsibility, ranging from caring for younger siblings, to
helping in a family business, to outside work that contributes to the family income
(Park, 2002, Song, 1999, A. Valenzuela Jr., 1999). Parents also report preferring to
entrust tasks of a sensitive nature, such as paying bills or accompanying a sick parent
to the doctor, to adolescents than to younger children, when they have a choice.

1
“Brokering” refers to the linguistic, cultural, and technological translating and interpreting that child
brokers perform in their efforts to connect their families to community institutions: namely, schools,
health care, and social services.
2
Child brokers included in interviews and ethnographic observation ranged from 11 to 19, meaning
that all participants were still in middle or high school and living in the family home. While 11-13
technically qualifies as ‘pre-teen’ years, developmental ranges differ, particularly in this population,
where many children mature more rapidly out of necessity, given their responsibilities.
3

Linguistic skills are likely more developed in older children, in terms of
actual bilingual capacity, cultural sophistication, understanding of meta-language
(Buriel & DeMent, 1998), and ability to reflect on their brokering behaviors (Dorner,
Orellana & Li-Grining. 2006; Orellana, Dorner & Pulido, 2003). The range of
bilingual capacity is considerable; some children exhibit these sophisticated
vocabularies and understandings, while others struggle to express themselves in
either language.
Adolescence is also popularly considered the period of development most
associated with child-parent conflict. My research project focused on Latino child
brokers in one Los Angeles community. My findings demonstrate extensive parent-
child cooperation in adolescence, challenging the general assumptions of parent-
child conflict so often emphasized in the immigration literature.
Who Are these Children?
Although my research focuses primarily on adolescents, I refer to brokers as
“children” to highlight their subjective positioning within the family home
3
. Thorne
et al. (2003) indicate that children of immigrant parents may remain “children” in the
eyes of their family after the legal age of adulthood (i.e. turning 18). This extended
childhood is in part a function of children’s ongoing responsibilities in the home, and
possibly as wage earners or workers in the family business (Park, 2001, 2002; Song,
1999; A. Valenzuela Jr., 1999). This extended childhood is juxtaposed with brokers’

3
Like Thorne (1993), I sometimes refer to child brokers as ‘kids’ since this is their primary self
reference.
4

activities that involve invoking adult voices, roles, and information at very early
ages. The tightrope that children walk requires them to take on adult responsibilities
and knowledge at very young ages, but to do so while maintaining culturally
appropriate deference to parents and elders. In many ways, these children become
adults prematurely, and are expected to remain children beyond America’s legal
boundary of adulthood. The tensions between child brokers’ seemingly
contradictory patterns of compressed and extended childhoods are the ‘double bind’
of immigrant childhood.
The compression and extension of childhood in these families highlight the
imported, culturally defined expectations that immigrant parents have of their
children. Referring to brokers as children emphasizes ‘childhood’ as culturally
defined. Ariès (1962) and Zelizer’s works (1985) are among the most celebrated
analyses of Western childhood as a social construction: a time of passivity, “moral
pricelessness,” and of “economic uselessness” (Zelizer, p.6). Abel Valenzuela
challenges these formulations of American childhood, charging that “in the family
context, be their contributions economic, symbolic, or both, immigrant children are
anything but ‘useless” (1999). The findings from the current project show that child
brokers are not only “anything but ‘useless,” but that their contributions have real
consequences for their families.
But who are these children? I interviewed children ranging in age from 11 to
19. Whether the main child broker in the family is 11, 14, or 17, the family
5

configurations and needs determine their levels of familial responsibility more than
their own chronological age. I spent a great deal of time with an 11 year old whose
sensitivity and devotion to helping his family with the difficulties of raising his
epileptic brother had forced him to grow up early. In him, and in so many others, I
saw that the various forms of aid these children provide, or try to provide, on a daily
basis for their families, become a source of self-identification and pride. Their
identity as helpers is integrated into the family as a reward structure, and the positive
feedback they receive resulted in many of these children’s defining themselves
primarily by what they give others. The intuitiveness of many child brokers made
them so attuned to the needs of their parents that they often acted on and obviated
their parents’ needs before the parents knew they needed help. Connections between
child brokers and their parents were close and almost inextricably intertwined; the
attendant pressures of these close relationships were evident in the wide number of
ways that these children sacrificed their own needs—such as completing their
homework, for example—to fulfill the needs of others.
Being home to help their parents, however, was in no way the cure-all for the
realities of living in neighborhoods burdened with concentrated disadvantage
(Sampson, Morenoff & Gannon-Rowley, 2002). Lopez (2003) and Zentella (1997)
found that Dominican and Puerto Rican girls, respectively, exhibit higher
educational attainment, bilingual proficiency, and cleavage to parents’ cultural
values than their brothers, because they are more tied to the home and to family
6

responsibility. There is no doubt that I also saw family responsibility shielding
children from the attractions of street life and enhanced parental control over
children’s behavior. However, this shield was far from impermeable. I interviewed
teens that help their parents, but were also routinely suspended from school for
fighting. I interviewed a pregnant teen whose heavy familial responsibilities seemed
to have fueled her desire to have something ‘just for herself,’ as she said. Another
respondent felt compelled to isolate himself from his neighborhood pals and devote
his attention to his mother’s needs, to prove that he wouldn’t break her heart by
falling into gangs and drug dealing as his older brother had.
The link between helping at home and educational attainment was also not
evidenced in this group, although others have reported one (Dorner, Orellana & Li-
Grining, 2006). Most of the children I interviewed performed very poorly in school,
and the range of linguistic capacity was stunning. Of course, I did encounter a few
honor roll child brokers, and these children exhibited relatively strong bilingual
capacity. But, for the most part, I was struck by the non-standard and limited
English that most of the children used, and in a number of cases, that they were not
able to express themselves adequately in either Spanish or English—essentially,
handicapped in two languages. These low rates of school achievement and linguistic
capacities raised important questions about the quality of the translating that children
are performing for their parents, even with the best of intentions.
7

And, of course, there were times that kids admitted to altering translations in
ways that served their own interests. A few of the kids who were not doing well in
school admitted to minimizing or softening teachers’ complaints when translating in
school meetings. They were well aware of the constraints faced in stretching the
truth; as Teresa,
4
a 15 year old broker told me, “you can’t change what [the teachers
are] saying completely, because your mom is still seeing the teacher’s face.” These
children also interpret and change what their parents say to minimize negative
outcomes parents might face if their language makes them appear intractable. In
tense situations or when brokering arguments, children often soften their parents’
words or explain the reasons for their dissatisfaction in great detail, to put their
parents in the best possible light. This is also the reason why many parents take their
children to translate in school meetings, doctor’s offices, and social services, even
when translators are present. As Hilda, one of the mothers, said: “My daughter
knows my heart, and knows how to speak it.” For selfish or unselfish ends, child
brokers are not disinterested filters; they are intimately invested in the outcomes of
the interactions in which they participate.
The Research Site
This research project focused on residents of one community in Los Angeles,
which the Metamorphosis Project has defined as “Greater Crenshaw.” Traditionally
a predominantly African American community, the area has experienced a dramatic

4
All names mentioned in the text are pseudonyms to ensure confidentiality.
8

demographic shift over the last two decades as large numbers of Central American
immigrants have made the area their home. The study area comprises nine zip
codes, located roughly between the Civic Center and University Park, the area
around the University of Southern California. According to the 2000 Census,
Latinos comprised 37% of the total population in the study area. Of these Latino
residents, 38% report household incomes below the 200% poverty level ($20,000),
compared with 27% of Latinos in Los Angeles County and 28% of Latinos
nationwide who fall into this income bracket
5
(US Census Bureau). The Latinos in
this area are relatively young; 37% are under the age of 18, well over the County and
national averages (28% and 26%, respectively).
With regard to language, 37% of study area households that report speaking
Spanish are linguistically isolated,
6
according to the 2000 Census. The
Metamorphosis telephone survey of Greater Crenshaw allowed respondents to
choose the language in which they wished to be interviewed. Fifty four percent of
Latino respondents (n = 170) elected to answer the questions in Spanish, and 46% in
English (n = 143), with most first-generation respondents opting for Spanish.
Census questions on English speaking proficiency also reveal a divide between first
and second generation Latinos. Of study area residents who reported speaking
Spanish in the home, 85% report that children in the home (i.e. under the age of 18)

5
Note: For the overall population, 23% of Los Angeles County and 22% of Americans nationwide
fall below the 200% poverty line
6
According to the US Census Bureau, “A household is classified as ‘linguistically isolated’ if no
household members age 14 years or over speak only English, and no household members age 14 years
or over who speak a language other than English speaks English ‘Very well."
9

speak English “very well” or “well,” as opposed to 15% reported as speaking
English “not well” or “not at all.” For Spanish speakers over 18, only 54% report
that they speak English “very well” or “well.” These estimations of spoken ability
are likely exaggerated. Alba, Logan, Lutz & Stults (2002) and Lopez (1996) provide
evidence that immigrants’ self-ratings of language ability on a Census are usually
overstated, so the percentage of Spanish speakers with high verbal proficiency in
English is likely lower in practice. Lopez also notes that spoken proficiency is a
poor marker of language proficiency, since competency in reading comprehension
and writing are more difficult to acquire (p.145).
Given that this is not a long-settled residential Latino community, fewer
institutions in the area effectively cater to Spanish speakers, relative to more
established communities such as, for example, East Los Angeles. Previous
Metamorphosis research findings indicate that the constraints and opportunities of a
particular geographic location can account for demographically similar communities’
residents exhibiting very different behaviors, or geo-ethnicity (Kim & Jung, 2003;
Kim, Jung & Ball-Rokeach, 2006). How accessible Greater Crenshaw schools,
health care, social services, and commercial outlets were to residents varied
noticeably within the study area itself—Latino residents living in Census tracts that
are still predominantly African American depended more heavily on their children’s
brokering capacities than residents even a few blocks away. Overall, reliance on
children’s capacities as brokers was evident in our survey results. Almost half (44%)
10

of the Latino respondents who chose to answer the Metamorphosis telephone survey
in Spanish indicated that their children routinely translate English language
television, radio, or newspapers, mail, and receive or make phone calls requiring
English (N=84). Interviews with parents, children, and English-speaking community
service providers made it clear that reliance on children extends beyond the home;
children participate actively in their family’s interactions with community
institutions as well.
For the sake of project manageability, it was necessary to select what the
immigration literature suggests are the most important community institutions for
new immigrants. Schools are the most obvious site choice. Both quantitative and
qualitative immigration research have found that enrolling children in school is
usually the very first action taken by new immigrant parents (Brittain, 2003; Reese,
2001). Other Latino residents inform newly arrived families that enrolling children
in schools immediately is a legal imperative (Menjìvar, 2000; Valdés, 1996).
Research also indicates that even the most time-strapped parents will make efforts to
keep up with school events, even if they exhibit no formal connections to community
institutions (Fielding, 2004; Rueda, Monzó & Arzubiaga, 2003).
Following from research findings that new immigrant parents connect to
community institutions primarily for their children’s needs (and, as this project
argues, through their children’s help), health care and social welfare services have
commonly been documented as initial community connections for low-income new
11

immigrants (Baquedano-López , 1998; Coutin, 2000; Jones-Correa, 1998; Menjìvar,
2000). Even immigrants who do not make regular office visits (many do not, leading
to the kinds of health disparities summarized in Wilkin, 2005) are more vigilant
about maintaining their children’s health. These efforts lead to interactions with
health care facilities as well as social services such as Woman, Infants and Children
(WIC), a free federal program that provides pre- and post-natal care and nutritional
assistance to low income mothers and children, regardless of residency status.
Overview of the Dissertation
This research project was primarily concerned with the ways in which
interpersonal connections between immigrant residents—both in their family, and in
their settlement community—can be ‘translated’ into meaningful connections with
the community institutions with which they need to maintain ongoing contact:
schools, health care facilities, and social services. Specifically, this project focuses
on the children’s efforts to make or maintain these connections for their families, and
explores reasons for the relative success or failure of these efforts.
Chapter 1 details the theoretical model that guides this study, drawing
primarily from communication infrastructure theory and prior research (Ball-
Rokeach, Kim & Matei, 2001; Kim & Ball-Rokeach, 2006). Domestic infrastructure
theory (Lievrouw & Livingstone, 2005; Livingstone, 2002), and the social capital
literature (Bourdieu, 1986; Coleman, 1988 & 1990; Putnam, 2000) are employed to
12

describe micro-level processes related to child brokering that have not previously
been foci of Metamorphosis Project research.
Chapter 2 details the research design of the project, and includes an
explanation of how the multiple methods employed in this project contributed to
analysis of findings. I interviewed parents, children, and service providers in Greater
Crenshaw, conducted extensive site visits, accompanied families to their
appointments with these three institution types, and conducted quantitative analyses
using the Metamorphosis survey data. Chapter 2 therefore also gives attention to the
points of convergence and contradiction within data collected through different
methods.
Chapter 3 is devoted to survey analyses, comparing Latino residents of the
Greater Crenshaw community on key indicators, according to their language
preferences and reported frequency of child brokering. Although these data were
collected from adults, and the focus of this project is children, comparisons between
sub-groups of Latino residents gives a sense of where child brokers’ families are
placed within the larger social milieu.
Chapter 4 focuses on the family set in its domestic infrastructure, with
particular attention to the brokering activities, media connections, and attendant
social arrangements that affect family access to resources in their efforts to connect
with community institutions. The second part of this chapter investigates the
influence of child brokering on connections to peer networks, community
13

organizations, and locally available media, as well as the resources in peer networks
that can accessed by children to improve or supplement their brokering activities.
In chapters 5 and 6, I shift from focusing on the resources in family and peer
networks, local media and community organizations, to evaluate successful and
unsuccessful attempts at converting those resources into social capital that has
currency in the community institutions that families attempt to connect with most
often. Chapter 5 focuses on connections with schools as community institutions, to
the individuals embedded in these institutions (i.e. teachers and administrators), and
the roles that educators and educational institutions can play as ‘resource brokers’ to
connect families with institutions and resources beyond the schools. Chapter 6
follows the same format, but focuses on families’ connections with community
institutions devoted to health care and social services. Again, connections with the
institution itself, with associated individuals (doctors and other health care staff, and
social service providers and advocates), and the potential roles of these institutions as
resource brokers, will be considered. These chapters also consider how child
brokering in health care and social service institutions affects child brokering and
success in schools.
Finally, Chapter 7 focuses on the consequences of brokering for the children
who perform these tasks, and for their family’s community connections.
Contributions to theory are discussed along with policy suggestions and directions
for future research.
14

CHAPTER 1
NETWORKS & COMMUNITY:
A THEORETICAL FRAMEWORK AND HEURISTIC MODEL
It is well established in the literature that social networks are both cause
(Massey, Durand, & Malone, 2002) and consequence (Hondagneu-Sotelo, 1994;
Suárez-Orozco & Suárez-Orozco, 2001) of Central American migration. Social
networks among Latinos have routinely been celebrated (summarized in Kibria,
1993, p.18 and Menjívar, 2000, p.33; Stanton-Salazar & Spina, 2003 p. 241) in
research that insinuates that dense networks alleviate or eliminate the structural
inequalities that characterize daily life in many new immigrant communities. Other
researchers, notably Menjívar (2000) in her study of Salvadorans in San Francisco,
make a strong case for social networks being ‘fragmented,’ and a source of
disagreements, stress, and isolation amongst immigrants whose personal resources
are stretched too thin to share much with each other. Menjívar’s contribution
challenged future researchers to differentiate between types of interpersonal
networks to understand the utilities and costs associated with particular connections.
Menjívar also emphasized the local neighborhood context as essential to
understanding how social networks affect their members.
Dense interactions and connections between family members, relatives, and
friends in the neighborhood can and do serve a number of deeply important functions
for new immigrants as they adjust to their new community. Having people to trust,
15

watch your kids, come to your celebrations, and listen to your heartaches are all
essential to an immigrant’s mental and emotional health (McKenzie, Whitley, &
Weich, 2002), as well as day-to-day survival (Domínguez & Watkins, 2003).
However, the value of these resident networks is necessarily limited if they include
only horizontal relationships (Granovetter, 1973; Monkman, Ronald. & Théramène,
2005). For example, Waldinger & Der-Martirosian (2001) and Menjívar (2000)
found that being embedded in a close-knit social network may constrain occupational
opportunities for exceptional immigrants. Close personal connections can help as
well as handicap an individual’s chances of success in other social realms as well.
The most advantageous balance for the immigrants would be a strong set of
horizontal interpersonal relationships that serve their survival functions—in addition
to vertical connections, or what Granovetter (1973) referred to as the ‘strength of
weak ties.’ How does an immigrant family navigate the unfamiliar territory of key
community institutions—schools, health care, and social service resources—or even
find out that such resources exist? Making and maintaining (vertical) connections
with these institutions and the individuals who work in them are essential to
supporting children’s educations and maintaining the health and wellbeing of the
family.
Children emerge as essential brokers for their families, in both making and
maintaining family connections to community institutions. Children’s daily
interactions with schools mean that they often have more contact and familiarity with
16

the English language and with American cultural mores
7
than their parents do
(Rueda, Monzó & Arzubiaga, 2003). The work that children do, brokering language,
culture, and a range of media and communication technologies, is often a primary, or
the only way that families can make or maintain connections with community
institutions.
A small number of prior studies indicate that children’s brokering in a variety
of settings is a key contributor to their family’s daily problem solving activities and
community connections. Particularly as adolescents, children “broker” or translate
for their parents in many different settings, mediate cultural differences, access
information and media, contribute to the household income, and/or perform domestic
duties and caretaker work (Dorner, Orellana & Li-Grining. 2006; Orellana, Dorner &
Pulido, 2003; Park, 2002; Song, 1999; Tse, 1995, 1996; A. Valenzuela Jr., 1999). I
extend Tse’s (1995; 1996) definition of “brokering” beyond translating to include a
larger range of communicative tasks. In this case, “brokering” refers to tasks that
children of immigrants perform in efforts to convert resources in their informal
networks into formal connections with community institutions. Brokering tasks can
be:
• “Linguistic brokering” including all verbal, written, and mediated tasks related to
translation/interpretation of language;

7
I do not mean to insinuate that children’s language or cultural capacities are perfect—in fact, I will
argue the opposite. For the meantime, I am arguing that children have relatively greater linguistic and
cultural capacity than their parents, although in some families, the distinctions between parents and
children proved almost nonexistent.
17

• “Cultural brokering,” referring to tasks that require mediating at least two sets of
cultural norms (for example, Mexican and American); and/or
• “Technological brokering,” referring to ways in which these children broker their
parents’ connections to and understanding of a range of media and
communication technologies.
A particular brokering task may implicate one, two, or all three of these brokering
types.
The existing research on children’s brokering activities and observable
outcomes is scarce and scattered across a number of disciplines. Because these
research projects lack a set of common definitions and terms (Orellana, Dorner &
Pulido, 2003), few accumulated lessons can be drawn from previous work. In
addition, these studies have tended to be atheoretical and focused directly on
educational outcomes of child brokers (Buriel & DeMent, 1998; Dorner, Orellana &
Li-Grining. 2006). The absence of a multi-level theoretical orientation that can
handle multiple dimensions and implications of child brokering has also contributed
to a literature that (a) has little crossover between studies (Orellana, Dorner &
Pulido, 2003, p.423), and (b) does not situate child brokering efforts in a larger
context. Additionally, for a phenomenon clearly rooted in communication
processes—interpersonal, mediated, and family communication, media
connectedness patterns, and organizational communication—communication studies
18

perspectives have not previously featured in systematic analyses of child brokering
activities.
Child brokering has potentially transformative properties for community
communication. Child brokers can be the links that instantiate their parents into the
local community by connecting them with local media resources, community
organizations, other residents, and key community institutions. This project focuses
on child brokers’ conversions, attempted or successful, of resources in their family
and peer networks into meaningful connections with local institutions. However,
this dissertation does not intend to cast children’s contributions as a panacea for the
very real, and often debilitating, layers of difficulty many families experience in their
day-to-day attempts to ‘get by’, let alone ‘get ahead.’ Rather, this research is offered
as a nuanced description of the efforts of parents and children to connect with and
maintain valuable connections with community schools, health care providers, and
social services.
Theoretical contributions
In this chapter, I develop a heuristic model which places child brokering
activities in a larger community context. This model draws primarily on
communication infrastructure theory (Ball-Rokeach, Kim & Matei, 2001; Kim &
Ball-Rokeach, 2006a), utilizing core concepts from social capital theory (Bourdieu,
1986; Coleman, 1988 & 1990; Putnam, 2000) to detail the resources being
exchanged at the interpersonal level, and the relative value of those resources for
19

connecting with community institutions. I define social capital as Ream (2005) does;
“Social capital is the aggregate of the actual or potential resources embedded in
social networks that may be converted, via social exchange, into other manifestations
of capital” (pp. 11-13). Those resources are exchanged through community
storytelling, which communication infrastructure theory defines as any type of
communicative action that addresses residents, their local communities, and their
lives in those communities (Ball-Rokeach, Kim & Matei, 2001; Kim & Ball-
Rokeach, 2006a).
Social capital is also useful for disaggregating the network of resources
within institutions to which child brokers attempt to connect their families
(Bourdieu, 1986). Livingstone’s (2002) domestic infrastructure theory is also
incorporated into this heuristic model to facilitate a focus on the home-based
brokering activities and social arrangements of the family around local media.
Children’s daily brokering efforts at home are the context within which efforts to
connect with community institutions occur. Child brokering in the home is also key
to understanding how communication with other residents is incorporated into family
strategies to connect with local institutions.
The heuristic model is introduced here in an accumulative fashion, beginning
with the basic structure and becoming increasingly complex.

20
Communication Infrastructure Theory
Communication infrastructure theory (Ball-Rokeach, Kim & Matei, 2001;
Kim & Ball-Rokeach 2006a & 2006b) conceives of (1) resident networks, (2)
community organizations, and (3) local media, as an interlinked set of nodes that
constitute the storytelling network of a local community:

LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS

Figure 1.1. The storytelling network
When the storytelling network nodes interact meaningfully with each other to
‘storytell’ the local area, it is considered integrated. An integrated storytelling
network means that community storytelling is occurring within and between network
nodes. Residents talk with their neighbors, family, and friends in the area,
potentially sharing news and resources through their stories. The community
organizations can be locations for resident storytelling, and can also be storytellers
themselves, by communicating community information to the residents that they
serve, and communicating with other community organizations to improve their
mutual community outreach. If local media are acting as storytellers, committed to
covering community events and sharing information resources, community

21

organizations can benefit from having these media disseminate their message to the
community. Local media are concurrently served by having the opportunity to
provide the residents with resources that they want and need, increasing their
relevance to their audience. In an integrated storytelling network, residents connect
with these media, and share what they have heard, read or seen with their family,
peers, and other residents in their interpersonal network.
Previous Metamorphosis Project research indicates that the integration of
these three storytelling network nodes is a strong litmus test of community
communicative capacity. The connection between an integrated storytelling network
and higher levels of civic engagement (Ball-Rokeach, Kim & Matei, 2001; Kim,
Jung & Ball-Rokeach, 2006) has also been linked to positive outcomes for children
through higher levels of family interaction (Wilkin, Katz & Ball-Rokeach, in
preparation), more frequent visits to libraries, and participation in outdoor activities
in newcomer Latino communities in Los Angeles. The active roles children play in
this storytelling network have not previously been studied, but this research will
demonstrate that children are often key links between their families, local media and
resident networks, and can also be the impetus for becoming involved with
community organizations.
Resident Networks
Communication infrastructure theory suggests that resident networks carry
the burden of community storytelling (Ball-Rokeach, Kim & Matei, 2001),
22

particularly in communities where the storytelling network is not well integrated.
Residents have agency to ‘storytell’ their community and can activate resources in
their interpersonal networks to agitate successfully for community change, even
when community organizations and local media are not fully integrated into the
storytelling network (Katz & Matsaganis, 2004
8
). Resident networks can achieve a
great deal if they are able to access resources that can be successfully converted into
desired community connections. The issue, of course, is whether the horizontal
kinds of relationships evident in Greater Crenshaw have the necessary resources to
make these kinds of conversions.
Previous applications of communication infrastructure theory have not
investigated how storytelling, and the resources that are or are not shared in these
communication acts, might vary in different kinds of interpersonal networks. This
research follows Ream’s (2005) disaggregation of resident networks into family and
peer networks. As Figure 1.2 shows, the family network is connected to two primary
kinds of peer networks: bonding peer networks, which connect residents to others
much like themselves, and to bridging peer networks, which can facilitate residents’
access to a more diversified set of community information resources. Bridging peer
networks are rare in Greater Crenshaw, and are therefore indicated by a hyphenated
arrow in Figure 1.2.

8
This refers to the city of South Gate in South-East Los Angeles where residents successfully
challenged a Los Angeles Unified School District announcement that they would not build five new
schools in the city, as they had promised to do. The parents had a diversified set of resources within
their resident networks, including some second generation council members who helped them to force
a reversal of LA Unified’s decision.
23

LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Family
Network
Bridging Peer Networks
Bonding Peer Networks
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Family
Network
Bridging Peer Networks
Bonding Peer Networks
Bridging Peer Networks
Bonding Peer Networks

Figure 1.2. Resident networks disaggregated into family and peer networks.

Family networks refer to the members of the household itself, whether they
are formally related or are members of a household who act like family (Weston,
1991), such as informal ‘adoptions’ of a spouse’s children. Peer networks include
friends, neighbors, and extended family that live in the local community. I include
relatives who do not live in the household in peer networks because interview
respondents indicated that their relationships with extended family were on par with
their relationships with close friends, but substantively different from the family
relationships within the home. The phenomenon of ‘family as friends’ is well
documented in the social capital literature (see Domínguez & Watkins, 2003 for a
summary).

24

Family Networks
The family unit is the primary storytelling unit of the community (Willkin,
Katz & Ball-Rokeach, in preparation). Community storytelling within the family
network means that parents and children have access to each others’ community
information resources, which differ because members have the opportunity to talk
with different people in the course of their daily routines. Family interaction around
a dinner table, for example, can prompt a discussion about community news or
resources by someone asking, ‘How was your day?’ Family members can also
activate their storytelling resources for each other as they become salient to solving a
particular problem. Family storytelling about the community can accumulate a
synergistic range of information resources that can be pooled among family members
(Kibria, 1993) and used to communal advantage. The larger the range of these
resources, the more likely that child brokers and their families will be able to convert
interpersonal resources into meaningful connections with schools, health care, and
social services.
Peer networks
Children’s peer networks are primarily composed of school mates, who can
be sources of academic support or of academic disconnect (Lopez, 2003; Olsen,
1997; A. Valenzuela, 1999; Zentella, 1997), but are seldom resources for community
information, or for brokering support. For some child brokers, cousins, extended
family members, and family friends serve as support for brokering activities, either
25

by modeling successful (and sometimes unsuccessful) brokering that kids can learn
from, or by supplementing children’s brokering when parents’ needs go beyond the
linguistic, cultural, or maturity levels of the child brokers themselves. In this sense,
these older sources of support can supplement children’s efforts to link their families
to community institutions and resources, and can encourage the refinement of
children’s own brokering skills for more successful independent brokering in the
future. Stanton-Salazer & Spina (2003) and Tierney & Vengas (2006) found, as I
did, that very few adolescents have the benefit of such networks.
When parents know and interact with the parents of their children’s
schoolmates, social control over children’s collective wellbeing is more likely
(Bould, 2003; Coleman, 1988). Parents are more likely to trust that another parent
will notice when a neighborhood kid (maybe their own) is getting into trouble, and
will be more likely to do something about it (Putnam, 2000, p.302). Connections to
other parents can encourage community storytelling about resources for children, and
can also encourage connections to community institutions: “Parents’ participation in
social networks that include the parents of other similarly aged school children may
provide feedback on effective child-rearing strategies and make available crucial
information about school policies, teachers, and many of the students’ peers”
(Horvat, Weininger & Lareau, 2003).
Parents’ peer networks serve important ‘survival’ functions. Domínguez &
Watkins’ (2003) excellent study of the social networks of low income single Latina
26

mothers details the ways in which friends and family provide essential support—
sometimes a place to stay, or a source of child care that allows women to go to
work—as well as emotional support. These relationships are based on what Putnam
(2000) calls specific reciprocity, meaning that people help each other in a tit-for-tat
fashion (Axelrod, 1985)—that is, you help me today and you can count on me when
you need help. For most of the parents in Greater Crenshaw, the friends, family, and
neighbors who constitute their ‘survival’ network are other Central American
immigrants
9
, and interactions are based on specific reciprocity. Resources available
in these peer networks is necessarily constrained by having homogenous
memberships of individuals who are limited financially and temporally, in their
abilities to connect meaningfully with community institutions and resources, which
would make them resources for each other (Domínguez & Watkins, 2003;
Hondagneu-Sotelo, 2001; Menjívar, 2000; Nelson, 2000).
Putnam (2000) calls the resources in these kinds of ‘survival networks’
bonding social capital, which refers to the networked ties of a homogenous group
that reinforce group solidarity. A tightly bonded network that consists exclusively of
horizontal ties can compromise the convertibility of those network resources into
social capital, because these residents will have a relatively narrow set of resources
to share with each other through storytelling. Bridging social capital, on the other
hand, links individuals and families to a larger, heterogeneous network where

9
Interviewees did not differentiate between country of origin when it came to their peer networks.
Most reported friendships with immigrants from other countries, and many of the parents I
interviewed indicated that their spouse was from a different Central American country.
27

members may have a wider range of information resources that can be shared with
other residents. Waiting for children outside the school gates can be natural
opportunities for parents of different races, ethnicities, or classes to communicate
with each other and possibly share stories about community resources for their
children and families. Constrained by language barriers and racial/ethnic tensions,
these communicative opportunities often go unrealized. Presence of bridging and
bonding social capital in resident networks is not necessarily an either/or proposition.
Individuals or families might be connected to peer networks with varying levels of
bridging and bonding social capital. In this research, bridging peer relationships
were the exception to the general rule. Most parents and children reported peer
networks that were primarily bonded.
Community Organizations & Local Media
Community organizations have the potential to connect members with other
community resources through storytelling, by giving members a place to talk and
exchange information among themselves, and/or because the organization
consciously connects its members with other resources in the local area.
10
The kinds
of stories being told in these organizations depend on a few key factors. First, the
members of that organization can potentially reify bonded peer networks, or can
provide opportunities to develop bridging social capital. Some community
organizations are practically extensions of parents’ peer networks, and as such, tend

10
Community organizations can also storytell the community with other community organizations,
and one local media outlet may have similar connections to another outlet. These inter-organizational
storytelling links are, however, beyond the scope of this research.
28

to exhibit bonded memberships. For example, the recreational soccer organizations
in which a number of fathers played were little more than extensions of informal
peer networks. There was no communication system, no head of the organization—
just a basic understanding to show up to the field at a particular time on a Saturday or
Sunday if you could play. Although these recreational organizations are primarily
resources for bonding social capital, even a casual conversation between players
during half time could be an opportunity for storytelling the community, should a
player mention information about a social service or local program previously
unknown to all the players.
Other organizations, like a Parent-Teacher Association (PTA), can provide
opportunities to develop both bonding and bridging relationships. PTA meetings
would give parents a chance to chat and tell community stories with each other,
possibly pooling and exchanging information resources helpful to other parents.
These would also be informal opportunities to interact meaningfully with teachers,
which may improve the chances that child brokers’ efforts to connect themselves and
their parents to community institutions will be successful.
While few organizations fall cleanly into either ‘bonding’ or ‘bridging’
categories, many organizations tend to be primarily storytellers of either bonding or
bridging stories, as indicated in Figure 1.3. Previous Metamorphosis research
divides community organizations into four main types (Wilson, 2001):
29

cultural/ethnic/religious, sports/recreation, neighborhood/homeowner and
political/educational.
Cultural and ethnic organizations are self-explanatory in their bonding
function, since they are explicitly created along ethnic lines. While religious
organizations can potentially serve more than one racial/ethnic group, Chàvez’s
(2006) research in Greater Crenshaw indicates that Spanish-language services and
organizations tend to be ethnically self-contained. Sports/recreation organizations
might be bridging organizations in some communities, but in this area, the sport and
the ethnicity of the people who play it are synonymous. In my observations of South
Los Angeles, the basketball courts are almost uniformly African American social
property and the soccer fields exclusively Latino.
Neighborhood/homeowner and political/educational organizations are more
likely to draw people based on a particular characteristic (owning a home, being a
Democrat or a Republican, or belonging to a Parent-Teacher Association) than along
ethnic lines. It is telling, therefore, that Latinos in Greater Crenshaw who do report
belonging to one or more community organizations, overwhelmingly belong to
cultural/ethnic/religious, and sports/recreation organizations. Nevertheless,
community organization membership of any kind can be an important indicator of
community involvement, and provide opportunities for individuals to exchange
stories and information about their community that can potentially be converted into
social capital by other members of the organization. Community organizations can
30
also be resources by virtue of their connections to other community organizations, or
their direct affiliations with community institutions (for example, Parent-Teacher
Associations & schools).
Figure 1.3 disaggregates community organizations and local media to draw
attention to their potential roles as storytellers that foster bridging and/or bonding
relationships among community residents.

LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers

Figure 1.3. Community organizations and local media as potential bridging and/or
bonding storytellers.

Local media can also be storytellers that contribute to reifying the bonding
features of residents’ peer and family networks. Ideally, local media would tell
stories about community resources and news that would increase child brokers and
parents’ chances of successfully connecting with community institutions. By doing
so, local media stories would serve a bridging function for the residents that connect
with them. These media could also be complementing, and perhaps compensating,
for resources missing from resident’s peer and family networks. The content of local
stories can also prompt discussion between family members and neighbors, so that

31

information resources are discussed and exchanged with other residents, reinforcing
interpersonal connections between them.
The programming that Latino residents actually connect with constrains local
media’s capacity to be community storytellers. Assuming that these media have both
bridging and bonding resources embedded in their content, residents who connect
with more local media content will have access to more media storytelling about the
community. Adult residents who connect with limited types of media or a limited
range of content constrain local media storytelling when they watch, for example
telenovelas
11
but not the local news. Many of the parents interviewed for this project
indicated limited connections with local media, and were therefore not benefiting
from local storytelling those media may be doing. Interviews indicate that children
are the primary connectors with local media for their families, and therefore are
crucial to their families’ connections with local storytelling in these media, ranging
from television and radio, to school newsletters and other mail.
Bridging and Bonding in the Storytelling Network
Child brokers and families that have access to bridging resources in their
storytelling network would be more likely to have connections to a heterogeneous
resident network, community organizations with diverse memberships, and local
media connections that tap a wider range of local resources. Child brokers in these
kinds of storytelling networks would be able to access a wide range of resources that

11
Telenovelas can be and often are valuable resources for health-related information and attendant
efficacy beliefs (Beck et al., 2006). However, most of these programs do not offer links to local
resources to help with the health concerns raised in the shows (Wilkin & González, 2006).
32

could be shared through storytelling within family and peer networks. This
storytelling network integration would enable child brokers’ efforts to parlay
resources into social capital that is recognized in community institutions. Families
who have more bonded storytelling networks are more likely to have connections
with homogeneous resident networks, ethnically/racially-specific community
organizations, and local media with limited community storytelling that may confine
their coverage to issues affecting a particular racial/ethnic group only. If resident
networks are primarily sources of bonding social capital, community organizations
and local media have the potential to compensate for those limitations by storytelling
community resources to which child brokers and their families would otherwise have
little access.
The Communication Action Context
The storytelling network is embedded in a larger set of features that can act to
constrain or enable the efforts of immigrant families, set in their storytelling
network, to form meaningful connections with community institutions. These
features (see Appendix 1 for a full list) include neighborhood safety, racial/ethnic
diversity, and neighborhood institutions like schools, health care, and social services.
The communication action context is conceived as dynamic, meaning that one
feature can constrain or enable other feature(s), and that the enabling capacity of one
feature may mitigate the constraining effects of other features. This dynamic
33
capacity also draws attention to residents’ agency as they navigate the community
infrastructure.

LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers
Family Network
Schools
Health
Care
Providers
Social
Services
Safety
Ethnic/Racial
Diversity
Communication Action
Context
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers
Family Network
Schools
Health
Care
Providers
Social
Services
Safety
Ethnic/Racial
Diversity
Communication Action
Context

Figure 1.4. The storytelling network set in the communication action context.

It was necessary to limit focus to particular communication action context
features for this project: schools, health care and social services, and peripherally,
safety and racial/ethnic diversity (indicated in italics in Figure 1.4), since
respondents repeatedly referenced these two features as salient to their efforts at
institutional connections. For example, the racial/ethnic diversity of a local area
might mean that Latino immigrants avoid community institutions they perceive as
serving or favoring other ethnic groups, if relations are strained. The number of, and

34

attitude towards, Spanish speakers in a particular school can constrain or enable
connections to institutions by influencing parents’ feelings of comfort in spending
time on campus, going to school meetings, and meeting with teachers. Similarly, if
health care facilities are not well maintained and the staff are abrupt, patients may be
less inclined to return for follow-up visits or to comply with the treatments. On the
other hand, features of the communication action context may enable families’
connections with community institutions. If a health care facility manages a
successful community outreach campaign, families might be encouraged to pursue
that institutional connection. If teachers make an effort to make parents feel
welcome in the classroom, families may be encouraged to attend voluntary school
meetings or to ask that teacher for assistance that may ultimately impact children’s
present and future educational achievements.
The model presented in Figure 1.4 presents the communication infrastructure
of Greater Crenshaw from the parent’s perspective, which will be investigated
further in survey analyses in Chapter 3. Parents’ own connections provide context
for understanding what child brokering adds to the picture.
35
Figure 1.5 presents the child brokers’ communication infrastructure model.

LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers
Family Network
Schools
Health
Care
Providers
Social
Services
Safety
Ethnic/Racial
Diversity
Communication Action
Context
LOCAL
MEDIA
RESIDENT
NETWORKS
COMMUNITY
ORGS
Primarily Bridging Storytellers
Primarily Bonding Storytellers
Family Network
Schools
Health
Care
Providers
Social
Services
Safety
Ethnic/Racial
Diversity
Communication Action
Context

Figure 1.5. The communication infrastructure model from the child broker’s
perspective.

Figure 1.5 demonstrates that child brokers are actively contributing to their
family networks within the home, and also are connections to their parents’ peer
networks. Children’s brokering can be a way for parents to return or incur favors
from their neighbors, and peer networks can also be resources for child brokering.
Children actively enable their families’ connections to a range of local media, and in
many homes, are the only way that parents can meaningfully connect with many
information resources.

36

Connections to community organizations are indicated with a broken line
because these are connections that parents usually make on their own, with little
assistance, if any, from their children. At best, children are an incentive to become
involved with a community organization (for example, a Parent-Teacher Association
or similar school-related organization), but children seldom, if ever, broker these
connections for their parents. The connection between local media and community
organizations is similarly indicated with a broken line, since child brokers obviously
do not perform these functions.
The large arrows between the family network and community institutions
indicate that child brokers’ relative successes in navigating community stories and
resources within the storytelling network can constrain or enable their abilities to
connect with institutions for themselves (in the case of schools) and for their families
(in the case of health care and social services). As in Figure 1.4, safety concerns and
the salience of racial/ethnic differences are not direct objects of study here, but were
mentioned frequently enough that their constraining and enabling effects on
children’s efforts to broker institutional connections were clear.
Community institutions are not monolithic entities, but rather, dynamic
networks themselves. The nature of these networks, and particularly, the
characteristics of the individuals who work in and act as gatekeepers to the resources
within those institutions, can affect the success or failure of child brokers’ efforts to
connect with the institutions that they represent.
37
Networks in Institutions
The networks embedded in community institutions have three distinct forms
of capital that can be advantageous to residents who have and maintain meaningful
relationships with them. The first is institutional capital, which refers to the
authority or resources contained within the community itself. For example, a
hospital has institutionalized capital by virtue of being overseen by the state’s
requirements and confirmed as an authority on human health and wellbeing. Second,
the people who represent and work within that institution are embodied capital, for
example, doctors, teachers, administrators, and social service workers. Finally,
objectified capital refers to the physical objects associated with that institution, such
as the media produced there. The community information in school newsletters
would be an example of objectified capital (Bourdieu, 1986, p.243; see Appendix 2).
These three kinds of capital can be conceived as a network of their own:

INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
Figure 1.6. Internal network of community institutions
Here, we see that connections to one part of an institution can be a gateway to
connections with other sources of capital within that institution. As in the
38

storytelling network, the people working in these institutions bear the primary
burden for connecting with the residents they serve. Well-informed individual
providers can be sources of embodied capital for child brokers and their families
even if they can offer little access to other forms of capital within the institution.
However, when the internal network of an institution is integrated:
• Service providers share stories with each other about the neighborhood and
current service recipients, and pool their experience so that newer employees
learn from more established ones, and experienced employees can learn new
techniques and strategies from more recently trained staff. When staff are
storytelling in these ways, they are more likely to be resources as individuals (i.e.
embodied capital) for child brokers and their families because they know more
about the area, and more about their own organization.
• Staff members share institutional knowledge with each other and individual
employees are better able to access institutional capital on behalf of service
recipients.
• Employees have reliable access to objectified capital, including program
deadlines, and other new opportunities for their own development and for their
recipients. When objectified capital is a reliable conduit of institutional
resources for the often time-strapped people working there, they will be more
aware of available opportunities, and more likely to optimize institutional
resources for the benefit of recipients.
39

Integration of this internal institutional network also makes each institution
more likely to be a “resource broker” (Small, 2006), meaning that an institution can
connect residents with other institutions when residents have needs that are beyond
the scope of their own institution. When the internal network of an institution is
integrated, it is easier for employees in one institution to feel supported in locating
resources in other community institutions for their recipients. Integrated institutions,
and the individuals who work in them, are also more likely to see resource brokering
as an essential part of their mission, making these sources of institutional and
embodied capital all the more valuable for child brokers and their families who are
connected to them. Institutions with integrated networks are not simply fixtures in
the community architecture. They become active and meaningful parts of their
communities, and real resources for residents, thus enabling greater integration of the
storytelling network.
Sources of embodied capital carry the primary burden in their internal
institutional networks because an interpersonal connection is most often the
institutional entry point for immigrants. New immigrants are generally distrustful of
institutions, particularly if they are undocumented. Connections with individuals,
however, can be the entrée for a new immigrant who comes to trust that institution
by transposing the trust they feel for a particular employee (Katz, 2005). Feelings of
trust are associated with higher educational achievement in children (Stanton-
Salazer, 2001; A. Valenzuela, 1999), increased parental involvement in the schools
40

(Monkman, Ronald, & Théramène, 2005), greater patient compliance with treatment
routines (Street, 1991; Watson & Gallois, 2004), and repeated regular appointments
(Chong, 2002; Erzinger, 1999; Kim, Kaplowitz & Johnston, 2004). The ‘promotora’
model deployed in a public health effort in Pico Union in Los Angeles by the USC
School of Dentistry
12
follows the same logic—creating interpersonal connections
between immigrants and representatives of clinics, and a loyalty that in time could be
transformed into a connection with the institution itself. If the connection remains at
the interpersonal level, the family is vulnerable to losing that connection if that
individual leaves the institution. Considering the high rates of staff turnover in many
Greater Crenshaw institutions, the likelihood of losing such a connection is
considerable.
Community institutions vary in terms of the resources they possess. Pierre
Bourdieu (1986)’s dimensions of access, range, and quality (p.249; Portes, 1998, p.4)
are useful for evaluating these variations. Access draws attention to the admission an
institution accords their recipients to their resources. Institutions can offer access not
only to the capital embedded in their own institution, but have the capacity to be
‘resource brokers’ for other community institutions as well. Once the level of access
to a particular institution has been established, it is possible to evaluate the range of
resources available in that institution that can be helpful to child brokers and their
family. The quality of those resources is equally important. A family may have

12
Retrieved at: http://www.usc.edu/hsc/dental/update/august05/community_04.htm on February 22,
2007.
41

access to a range of resources in a local institution, but will not connect with them
because they believe those resources to be of low quality. Quality of resources also
includes a consideration of how sustainable that connection is. To ensure continuing
access to an institution’s resources, the family has to move from the specific
relationship with the individual they first connected with to a more generalized
connection with the institution. That way, if the particular teacher, doctor, or social
service provider should leave their job, the family does not lose their connection with
the institution entirely.
Connecting with community institutions is a process that begins in the home.
To properly understand how children broker in doctor’s offices, social services, and
in the schools, the ways in which those connections were initially made must be
understood. How family members interact on a day-to-day basis affects which child
shoulders primary brokering responsibility, and sense-making activities observed
between parents and children during appointments reflect understandings of each
other that are born in the space of the family home. Livingstone’s (2002) domestic
infrastructure facilitates theoretical connections between the home as an
infrastructure, and the larger community infrastructure.
Domestic Infrastructure
The home is the primary site of child brokering. Home-based activities are
usually the first step in connecting with community institutions; child brokers make
appointments, read newsletters, go online to find a local service, and pay bills. All of
42

these activities are necessary precursors to making connections with institutions
and/or to maintaining them.
Livingstone (2002) characterizes the home as a domestic infrastructure,
proposing that “the diffusion and appropriation of media into the practices of
everyday life plays such a key role in defining the home, in spatial terms, and daily
life, in temporal ones, that domestic media have become part of the infrastructure of
family life” (p.67). The domestic infrastructure construct is useful because it draws
attention to how the media artifacts present in the home—ranging from newsletters
that arrive in the mail, to the telephone, Internet, newspapers, and television—
produce an ecology of connection choices for satisfying a particular goal (Ball-
Rokeach, 1985), such as connecting with a health provider.
Combining Livingstone’s (2002) domestic infrastructure concept with later
development of the concept of ‘infrastructure’ generally (Lievrouw & Livingstone,
2005), I conceive of the domestic infrastructure as three interrelated elements, set in
an infrastructural framework:
43

ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure
ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure
Figure 1.7. Components of the domestic infrastructure

Media artifacts are the new and old communication technologies present in
the home, the activities are the ways in which family members interact with those
media to achieve particular goals, and the social arrangements are the patterns of
activities and media connection that become embedded habits in the family home
(Lievrouw & Livingstone, 2005). To understand the impact of child brokering in the
home, domestic infrastructure’s media artifacts, activities, and social arrangements
have been integrated with Bourdieu’s three dimensions of social capital: access,
range, and quality.
The media artifacts available in the home reflect the range of media that
brokers can access to find resources for their families. Access is constrained by
family finances, in that these families do not have the means to purchase an optimal
range of artifacts from which children can broker information resources about the
community that help their families. The media artifacts accessible in the home are

44

also constrained by what is available in the community’s local media—if local media
are not telling local stories, child brokering efforts will be constrained by this limited
information. Parents will be able to negotiate some of these communication
resources on their own, but children’s brokering activities can compensate for
parents’ cultural, linguistic or technological limitations and increase the range of
information resources to which their families have access. A wider range of
available resources increases the chances that families will have resources available
to convert into social capital that can link them with community institutions. Finally,
the social arrangements of family members that become embedded and ingrained
within the domestic infrastructure affect the quality of resources to which children
are able to connect their parents. Considering the home as an infrastructure that can
enable or constrain the efforts of brokers is helpful in understanding how social
arrangements can affect children’s efforts. If family relationships are strained by
illness, addiction, or divorce, the quality of children’s brokering will be diminished
by the time that these family instabilities take away from their brokering efforts. On
the other hand, family cooperation, sharing of skills and resources, and storytelling
the community, can enhance children’s brokering activities.
Although the relationships between media artifacts, brokering activities, and
social arrangements are dynamically related, they do become “routine, established,
institutionalized, and fixed to various extents, and so become taken for granted in
everyday life” (Lievrouw & Livingstone, 2005, p.4). Conceiving of the family home
45

as an infrastructure provides a micro-level complement to communication
infrastructure theory. Domestic infrastructure accounts for human agency within a
framework that can constrain and enable that human agency, as communication
infrastructure theory does for the community.
Since family members are agents within the domestic infrastructure,
brokering activities also affect the relationships between family members
themselves. In most cases, child brokers want to help their parents, and actively look
for opportunities to do so. Many of the children interviewed seemed to have an
intuitive sense of when their parents would need them (which I witnessed on
numerous occasions when they completed tasks without their parents’ having to ask
them to help), and almost all children reported positive feelings and a sense of pride
in their capacity to help their families. In fact, family social arrangements have
reward structures embedded in them that make children identify very strongly with
their contributions to their families. These children’s sacrifices of time and effort for
the collective good are often at the expense of their own individual achievement and
educational attainment. This is even more evident when one considers that children
do not only broker in their own homes and for their own families—they perform the
same kinds of services for their parents’ friends, neighbors, and for their extended
family. Older cousins, siblings no longer living at home, and family friends in these
peer networks can also be resources for child brokers, as sources of support for or
models of effective brokering. These connections are indicated by the dual arrow
46
connecting activities to peer networks in Figure 1.8. Since modeling by individuals
in peer networks can result in brokers adapting their brokering strategies, social
arrangements of family members, in relation to children’s brokering, can similarly be
altered by peer network connections. This arrow is also dual-sided in Figure 1.8
because the social arrangement of the family can affect child brokers’ ability to
access resources in the peer networks. For example, illness within the family could
constrain children’s abilities to connect with peer network resources, or could be an
impetus for making these connections, depending on the context.

ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure
Parents and Brokers’
PEER NETWORKS
CHILD
BROKERING
ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure
Parents and Brokers’
PEER NETWORKS
CHILD
BROKERING
Figure 1.8. Domestic infrastructure’s links to peer networks through child brokering

Brokering can have limiting effects on children’s human capital development
when family obligations overshadow their personal needs and commitments,
particularly to their educations. Teachers report absences, incomplete homework,
and fatigue as a result of students’ family responsibilities. Children also report that
emergencies or stressful brokering events, usually in health care settings, can
negatively affect their relationships with teachers who may not know or care about

47
the reasons for their lack of attendance or their incomplete homework. Brokering
activities therefore have the potential to decrease children’s human capital and
simultaneously constrain their efforts to convert limited human capital into
meaningful relationships with their teachers. These missed connections mean a
disconnect from teachers as embodied capital and as the gateway to sources of
institutional and objectified capital that can yield real advantages for children’s
individual development (Monkman, Ronald & Théramène, 2005; Stanton-Salazer,
2001; A. Valenzuela, 1999). Children’s identification with what they are to other
people—rather than who they are as individuals—means that brokering in
institutions where others will be the beneficiaries of their efforts, such as health care
and social services, are prioritized by the children themselves, even at the expense of
their own development, as indicated in Figure 1.9. In this sense, brokering
commitments, including those in health care and social services, can detract from
child brokers’ own educational attainment. This weakened link to the schools is
indicated by a hyphenated arrow in Figure 1.9.

CHILD
BROKERING
CONNECTING
TO
SCHOOLS
CONNECTING
FAMILIES TO
HEALTH CARE AND
SOCIAL SERVICES
CHILD
BROKERING
CONNECTING
TO
SCHOOLS
CONNECTING
FAMILIES TO
HEALTH CARE AND
SOCIAL SERVICES

Figure 1.9. Child brokering connections happen at the expense of their own
connections to schools.

48

Child brokering in the domestic infrastructure, and connections brokered to
peer networks and local media indicated in Figure 1.5, are the focus of chapter 4.
Chapter 4 is key to understanding the evidence presented in chapters 5 and 6,
because brokering activities in the home affect and are affected by connections to
community institutions. Chapter 5 focuses on child brokers’ connections to the
schools, for themselves and for their parents, paying particular attention to factors
that constrain and enable child brokers’ connections to school networks of embodied,
institutional, and objectified capital. Chapter 6 focuses on child brokers’ connections
to the institutional networks in health care and social services. Children broker
connections to health care and social services primarily for the benefit of others,
whereas the children themselves are the primary beneficiaries of connections they
broker to the schools. Chapter 7 returns to the heuristic model, and discusses the
potentially disadvantageous consequences of child brokering in the home and
community for child brokers themselves, in terms of limiting their own meaningful
connections with schools and their own development of human capital.
Chapters 2 and 3 provide the foundation for these later chapters. Chapter 2
details the research design and methods employed in this study, as well as the
research questions that guided this investigation. Chapter 3 is devoted to analyses
from the Metamorphosis Project survey of Greater Crenshaw. Survey respondents
who indicated frequent brokering are compared to those who indicated no/low
brokering. As an added layer of comparison, Latino respondents who opted to
49

answer the survey in English are compared to those who answered in Spanish as an
undifferentiated group (i.e. regardless of their reported frequency of brokering).
These analyses provide clues to where child brokers’ families, like those interviewed
for this research, are placed within the community with regard to key demographic
information and community indicators that have been developed and tested in past
Metamorphosis Project research.
50

CHAPTER 2
RESEARCH DESIGN
This chapter presents the research design and research questions that guided
this study.
Research Questions
The first four research questions guide the quantitative analyses in chapter 3,
drawing from data collected in the Metamorphosis Project’s telephone survey of
Greater Crenshaw. These data reflect responses from Latino residents over the age
of 18 (N=300). Although these questions do not reflect the direct experiences of
child brokers, they do provide a sense of the community context in which brokering
activities occur, as well as the relative social standing of child brokers’ families
within their community.
“Frequency of brokering” refers to respondents’ self-reports of the frequency
with which they require brokering assistance with English-language television, radio,
and newspapers, bills, and phone calls. Only respondents who opted to answer the
survey in Spanish were asked these questions (n=170). Latino respondents who
opted to answer in English (n=130) are included in chapter 3 as an additional
reference group. Although English-speaking Latinos are not the population of
interest in this study, their responses provide an additional layer of richness to these
analyses because even in cases where no significant differences between groups were
observed, there are some trends toward significance that are worth noting. Each of
51

the research questions below tested for differences between English and Spanish-
speaking respondents, and between ‘no/low’ and ‘high brokering’ Spanish-speaking
respondents. The operational definitions of these terms are detailed later in this
chapter.
The first four research questions explore the nature of the storytelling
network in Greater Crenshaw, to see if there are differences in its composition
between groups of Latino respondents.
The first set of research questions explores availability of resources in
residents’ informal networks of family and peers. The questions that comprise RQ1
illustrate, within the constraints of the dataset, the character of respondents’ resident
networks. Levels of family interaction indicate the quality of communication within
the family network. Reported levels of community belonging and confidence in
fellow residents’ ability to come together and solve a common problem (i.e.
collective efficacy), along with reported frequency of talking with one’s neighbors,
give an indication of the quality of relationships with neighbors within peer
networks. Proximity of family and friends detail these respondents’ peer networks
further. The last two questions included under RQ1 serve less obvious purposes.
Comparing Latino respondents’ evaluations of previous encounters with African
Americans is a rough estimate of the likelihood that peer network connections
include residents who are not Latino, since the Greater Crenshaw area is almost
entirely African American and Latino. If peer networks include African Americans,
52

there is a possibility of more heterogeneous resources of information on which
children might draw to broker community connections for their parents. The final
question under RQ1 measures the convertibility of the resources in these family and
peer networks into desired information: in this case, respondents’ ratings of their
difficulty in accessing advice on how to raise children. An indication of close peer
and family connections but difficulty accessing advice, would support the theoretical
assertion that bridging peer networks are good for ‘getting by,’ but bonding social
networks are necessary for ‘getting ahead’ and beyond basic survival needs.
RQ1: How does the composition of respondents’ informal networks affect
the social capital resources embedded in them?
RQ1a: Do respondents’ reported levels of family interaction vary by
frequency of brokering?
RQ1b: Do reported levels of community belonging and collective
efficacy vary by frequency of brokering?
RQ1c: Does frequency of discussion with neighbors vary by
frequency of brokering? What topics of conversation are reported
most often?
RQ1d: Does frequency of brokering vary by proximity of close
friends and extended family ?
RQ1e: Does reported comfort with African American residents vary
by frequency of brokering?
53

RQ1f: Does reported difficulty of accessing advice on child rearing
vary by frequency of brokering?
The second set of questions examines the storytelling roles of community
organizations and local media as resources for residents. As detailed in the previous
chapter, community organizations can serve as primarily bonding or as bridging
storytelling sites. I emphasize ‘primarily’ because even the most homogenous
community organization might be a source of bridging community storytelling
resources. Survey respondents were asked to identify the names of particular
community organizations of which they are members, which permitted comparison
of membership patterns between groups of Latino respondents.
Bridging and bonding storytelling in local media cannot be divided according
to the nature of that medium, such as whether that outlet is in English or Spanish.
However, comparisons of respondents’ media connection patterns for finding out
about their community, as well as their more general connection patterns, can
illustrate the range of media that respondents are able to access. That is, respondents
who can connect with media in both English and Spanish potentially have access to a
wider range of storytelling resources than respondents who are limited to Spanish-
language media only.

54

RQ2: What is the nature of respondents’ connections to the storytelling
network?
RQ2a: Are respondents’ community organization memberships
primarily sources of bridging or bonding storytelling?
RQ2b: Do respondents’ connections with media for learning about
their community vary by frequency of brokering?
RQ2c: Do respondents’ general media connection patterns vary by
frequency of brokering?
The third research question explores respondents’ integration into the
storytelling network, meaning the scope and intensity of their connections to other
residents, to community organizations, and to local media. The integration into the
storytelling network (ICSN) measure has been found to be significantly related to
community belonging, political participation, and collective efficacy (Kim & Ball-
Rokeach, 2006b) and family interaction (Wilkin, Katz & Ball-Rokeach, in
preparation). RQ3 investigates potential differences in storytelling network
integration based on reported frequency of brokering.
RQ3: Do levels of storytelling network integration vary by frequency of
brokering?
The fourth set of research questions shifts focus from the storytelling network
to examine residents’ evaluations of their access to community institutions. These
survey questions asked respondents to rate their feelings of connections to local
55

schools, churches, businesses, places to exercise, and police. Additional questions
asked respondents to rate their difficulty in accessing health care for themselves and
for their children. Taken together, these questions illustrate respondents’ general
connections to community institutions, and their self-reported difficulties in
accessing health care. RQ4 explores whether these connections and difficulties vary
according to frequency of brokering.
RQ4: Do respondents’ connections to community institutions vary by
frequency of brokering?
RQ4a: Do respondents’ evaluations of their connections to
community institutions vary by frequency of brokering?
RQ4b: Do respondents’ evaluation of access to health care for
themselves and for their children vary by frequency of brokering?
Research questions 5, 6, and 7 guide the qualitative analyses based on parent, child,
and service provider interviews as well as field observations in community
institutions. The fifth research question guides the analyses in chapter 4, by
exploring the mediated and interpersonal resources in the domestic infrastructure that
can affect children’s attempts to connect their families with schools, health care, and
social service institutions.

56

RQ5: How do domestic infrastructure components: access to media artifacts,
range of brokering activities, and quality of family social arrangements,
constrain and/or enable child brokers’ efforts to connect their families with
the community?
Chapters 5 and 6 are guided by two research questions. RQ6 focuses on child
brokers’ interactions with the different nodes of community institutions’ internal
networks. RQ7 draws comparisons between children’s efforts to broker connections
to institutions where they are likely to be the primary beneficiary of their brokering
successes (i.e. in the schools) versus contexts where their families are likely to be the
primary beneficiaries (i.e. in health care and social services).
RQ6: How do community institutions’ internal networks of embodied,
institutional, and objectified capital affect child brokers’ efforts to make and
maintain high quality connections with them?
RQ7: How does the nature of child brokering differ depending on the
primary beneficiary of their brokering efforts?

Telephone Survey Methods
Survey data were collected as part of the Metamorphosis Project’s multi-year,
multi-method investigation of new and old immigrant communities in Los Angeles.
The telephone data used for this dissertation come from a survey of Latino
13

13
303 African American residents of Greater Crenshaw were surveyed in the same time period, but
are not included in these analyses.
57

residents (N=300) in the Greater Crenshaw study area. The survey was conducted in
the second half of 2005.
Random digit dialing (first adult contacted) was used by a well-established
commercial survey research organization to contact residents 18 years or older for
the survey. The survey was administered in either Spanish or English, according to
the preferences of the respondent. The survey lasted an average of 53 minutes in
English and 57 minutes in Spanish. It was introduced as a survey of residents’
thoughts about their community.
Survey Samples
Spanish vs. English-Speaking Respondents
The first stage of data analysis separated the total number of Hispanic
respondents (N=300) into those who had elected to answer the survey in Spanish
(n=170) and those who had answered in English (n=130). The first set of analyses
compared Spanish and English-speaking respondents on all relevant measures.
There were clear differences within the Latino population based on primary
language, many that are likely tied to immigrant generation. The majority of
English-speaking respondents was second generation (65%) and the majority of
Spanish-speaking respondents was first generation (69%). The English-speaking
population was younger, with a median age of 28, compared with 36.5 for Spanish
speakers. The median household income of English-speaking respondents was
$27,500, and Spanish-speaking respondents reported a median income of $17,500.
58
Most English-speaking respondents were either married/living with a partner (45%)
or were never married/single (45%). The Spanish-speaking respondents were most
likely to be married/living with a partner (69%). Educational attainment also
revealed stark differences: only 5% of English-speaking respondents had less than an
8
th
grade education, compared with 47% of Spanish speakers. Both had a median
household size of 4, and roughly half of both samples was female.
Table 2.1 provides the survey sample characteristics.
Table 2.1: Demographics of full sample: English and Spanish-speaking respondents

English-speaking Spanish-speaking
Respondents Respondents
Sample Size N=130 N=170
Age (median) 28 36.5
Immigration Generation
First Generation 19.5% 68.9%
Second Generation 65.0% 24.6%
Female 50% 44.7%
HH income $27,500 $17,500
<=$35,000 57.4% 94.5%
>=$60,000 10.0% 1.2%
HH size (median) 4 4
Education
<=8
th
Grade 5.4% 46.5%
<=High School 48.1% 40.6%
>=Some College 46.5% 12.9%
Marital Status
Never married 45.4% 20%
Married/Living w/partner 45.4% 68.9%

High Brokering vs. No/Low Brokering Respondents
Respondents who opted to answer the survey in Spanish (n=170) were asked
about the frequency of brokering activities in their homes. They were asked to think
about times that they might have needed help with speaking, writing, reading, or

59

understanding English. Five questions were asked: “How often (very often, often,
sometimes, rarely, or never) does someone help you…understand English-language
television or radio?”, “…understand newspapers written in English?”, “…understand
mail that comes to your house (bills, newsletters, etc.)?”, “…make or return phone
calls that require speaking English?, and “…understand information that your
child(ren) bring home from school, including homework and report cards?
14

The last question was discarded due to the discovery of variation between the
community schools’ policy on sending information home to parents; some schools
send all materials in English, other schools provide varying amounts of translated
information and newsletters. Therefore, a ‘never’ answer to this last question could
be due to two very different realities: the parent may not need help because the
document is in Spanish, or because the parent is able to understand it in English—a
crucial distinction, considering the focus of this study.
The four brokering questions were recoded, summed and averaged to create a
brokering scale. Tabachnik & Fidell’s (2001) procedure was used to correct for a
moderate negative skew, by taking the square root of a constant (largest score plus 1)
minus the current score. This corrects for the skew, but the resulting index has to
once again be recoded so that higher values indicate a higher need for brokering.

14
A follow-up question asked for the most common source of brokering help. The answers were not
read aloud, and interviewers coded spontaneous responses into predetermined categories: child,
spouse, friend, neighbor, other family member, co-worker/employer, and other.

60

Cronbach’s alpha was calculated to determine the reliability of this new measure
( α =.84).
The sample was split at the scale midpoint to ensure roughly equivalent
groups for analysis. The ‘no/low brokering’ group included brokering scale scores
from 0.0 to 2.53 (n=86), and the ‘high brokering’ group included scale scores from
2.54 to 5.0 (n=84). A dummy variable was created with no/low brokering as 0 and
high brokering coded as 1.
Although the demographic differences between the no/low brokering and
high brokering groups were not as stark as those observed between Spanish- and
English-speaking respondents, there were still some noteworthy differences. High
brokering respondents were more likely (74%) than no/low brokering respondents
(64%) to be first generation immigrants. High brokering respondents were also more
likely to have less than an 8
th
grade education (52%) than no/low brokering
respondents (41%). There were small differences in the median ages of no/low
brokering respondents (37.5 years) and high brokering respondents (36 years).
Annual household income was equivalent between groups, with 95% of high
brokering and 94% of no/low brokering respondents indicating an income below
$35,000. The majority of both groups: 67% of no/low brokering respondents, and
70% of high brokering respondents reported being married or living with a partner.
Approximately half of each group was female, although the no/low brokering group
was skewed female (55%) compared with the high brokering group (44%).
61
Table 2.2 provides the characteristics of the no/low brokering and high
brokering groups of Spanish-speaking respondents.
Table 2.2. Demographics of no/low brokering and high brokering respondents

No/Low High
Brokering Brokering
Sample Size N=86 N=84
Age (median) 37.5 36.0
Immigration Generation
First Generation 64.0% 74.1%
Second Generation 30.2% 18.5%
Female 54.7% 44.0%
HH income
<=$35,000 93.8% 95.1%
>=$60,000 1.2% 1.2%
HH size (median) 4 4
Education
<=8
th
Grade 40.7% 52.4%
<=High School 45.3% 35.7%
>=Some College 14.0% 11.9%
Marital Status
Never married/Single 23.3% 16.7%
Married/Living w/partner 67.4% 70.3%

Survey Measures
Informal Networks
Family Networks (RQ1a):
Family Interaction Index. The family interaction index (FII) has been developed and
tested previously in Metamorphosis project research (Wilkin, Katz & Ball-Rokeach,
in preparation). Six questions measured the frequency of family communication in
the home, as well as the frequency of talk about particular topics and family
activities. The FII consisted of six questions, all measured on a 5-point Likert scale
where response options were ‘very often,’ ‘often,’ ‘sometimes,’ ‘rarely,’ and ‘never.’

62

The questions ask respondents: “How often does your family…sit down and eat
together?”, “…sit down and talk together?”, “…discuss issues that are happening in
the neighborhood?”…“…discuss work-related issues?”, “…participate in
neighborhood and community activities as a family?”, and finally, “…participate in
activities outdoors together like sports, hiking, going to parks, etc.?” Cronbach’s
alpha for this measure was calculated ( α = .75) (Wilkin, Katz & Ball-Rokeach, in
preparation).
Peer Networks (RQ1b to RQ1e):
Neighbors: Belonging Index. Tested in previous Metamorphosis Project
research (Ball-Rokeach, Kim & Matei, 2001; Kim & Ball-Rokeach, 2006b), the
belonging index explored the degree to which respondents report connections with
their neighbors. Respondents were asked to indicate their responses to the following
questions on a 5-point Likert scale, where the options were ‘strongly agree,’ ‘agree,’
‘neither agree nor disagree,’ ‘disagree,’ or ‘strongly disagree’: “You are interested
in knowing what your neighbors are like,” “You enjoy meeting and talking with your
neighbors,” “It’s easy to become friends with your neighbors,” and “Your neighbors
always borrow things from you or your family.” The second part of the belonging
index asked respondents to indicate a specific number of neighbors that they know
well enough to ask them to: “…keep watch on your house or apartment,” “ask for a
ride,” “talk with them about a personal problem,” and “ask for their assistance in
making a repair.”
63

Neighbors: Collective Efficacy: The collective efficacy index measured
residents’ confidence that their neighbors could come together to improve the
community in which they all live. Residents were asked: “How many (all, most,
some, few, none) of your neighbors do you feel could be counted on to ‘do
something’ if…a stop sign or a speed bump was needed to prevent people from
driving too fast through your neighborhood?”, “…there were dangerous potholes on
the streets where you live?”, “…the sports field that neighborhood kids want to play
on has become unsafe due to poor maintenance or gangs, for example?”, “…you
asked them to help you organize a holiday block party?”, “...a child in your
neighborhood is showing clear evidence of being in trouble, or getting into big
trouble?”, and “…the trees along the streets in your neighborhood were uprooting the
sidewalks, making them unsafe?”
Communication with Neighbors. A set of questions asked respondents to
indicate how frequently they talk with their neighbors about the community, and to
specify the topics that they discuss most often. The initial question asked: “How
often do you have discussions with other people about things happening in your
neighborhood? Using a 10-point scale where “1” means “never” and “10” means all
the time, where would you place yourself?” The follow-up question asked residents
to specify the topics they discuss with their neighbors: “What kinds of things do you
talk with your neighbors about?” Multiple answers were accepted, and the list of
categories was not read to respondents. Interviewers coded spontaneous responses
64

into predetermined categories, which I divided thematically into: informal networks
(“Your/their children,” “your/their families,” “mutual friends or other neighbors,”
“your/their health,” and “personal issues”); general community conditions (“The
weather,” “traffic/road conditions,” “cost of living,” and “location/quality of goods
& services”); community institutions (“Schools,” and “church/church events”); and
news (“Current events and news,” “politics,” “sports,” and “neighborhood events).”
“Neighborhood problems,” “crime/gangs/drugs,” and “employment” received
enough responses as single categories to be included independently.
Location of Close Friends and Family. Respondents were asked about the
proximity of their close friends and family. Respondents were asked, “Where do
most of your close friends and family live?” The following fixed responses were
provided: “In your neighborhood,” “In the Los Angeles area,” “In California, but
outside of the L.A. area,” “In the United States, but not in California”, or “In another
country.”
Neighbors: Intergroup Interactions: The preceding questions (location of
friends/family, talking with neighbors, the belonging index, and collective efficacy)
possibly collected information on relationships Latino residents have with other
Latinos only. Respondents were asked to evaluate their past experiences with
African Americans. This question served as an indirect indicator of residents’
likelihood of voluntarily interacting with African Americans in the future, building
access to bridging social capital in the process. Respondents were asked to rate their
65

personal experiences with African Americans on a scale from a ten-point scale, with
“1” indicating “extremely bad experiences” and “10” indicating “extremely good
experiences.” If a respondent indicated having no contact with African Americans at
all, their answers were coded as “0.”
Access to Resources in Informal Networks (RQ1f):
Respondents’ evaluation of difficulty in finding advice on child rearing was
chosen as an indicator of access to social capital resources in informal networks.
Respondents’ reported difficulty accessing needed advice would indicate that the
resources in residents’ informal networks do not convert into needed social capital:
“Thinking about this child, how easy or difficult is it for you to get good advice on
how to handle problems that come up in raising the child?” Respondents were read
four options: very difficult, somewhat difficult, somewhat easy, and very easy.
Community Organizations (RQ2a):
Community organizations can be sites for resident storytelling, and/or key
community storytellers themselves. Respondents were asked, “Do you or does
anyone in your household participate in any clubs or organizations – that is,
organizations that have to do with your culture, schools, hobbies, sports, politics,
religion, or your neighborhood?” Those who responded positively to this question
were asked the following questions: “Do you or does anyone in your household
participate in any…sport or recreational organizations or clubs?”, “…cultural, ethnic,
or religious or church organizations or groups?”, “…neighborhood group or
66

homeowners association?”, “…political or educational organizations?”, and “…other
organizations or groups?” If respondents indicated that someone in the household
did belong to one of these organization types, the respondent was asked to name the
most important organization of that type. These organization names made it possible
for me to ascertain if respondents’ organizational membership were primarily
sources of bridging or bonding community stories.
Local Media (RQ2b & 2c):
Community Information. Respondents were asked to indicate the primary
ways that they gather information about their community: “Thinking about all of the
different ways of communicating and getting information - using television, radio,
newspapers, books, magazines, movies, the Internet, talking with other people or any
other way, what are the two most important ways for you to stay on top of what’s
happening in your community?” The list of possible responses was not read, and
interviewers coded the spontaneous responses into the following categories: “Talking
with other people/telephone;” “Internet;” “Movies;” “Books and magazines;”
“Television/cable TV/satellite TV;” “Radio;” “Newspapers;” “Organizations;” and
“Leaflets & folders (other media).”
If respondents reported television as one of their primary sources of
information, a follow-up question asked if most of this viewing was “major English
commercial channels”, “public television channels”, or “other television or cable
channels that target your area or that are produced for your ethnic group.”
67

Respondents were given examples of each kind of television station if they needed
help distinguishing between these categories. Equivalent follow-up questions were
asked if respondents indicated radio and newspaper as primary resources for
community information.
Time with Ethnic Media. Another question was asked to determine how
much time respondents spent with ethnic media generally, rather than for a particular
goal. Respondents were asked, “Approximately how many hours did you spend last
week…reading newspapers produced for your area or for your ethnic group?”,
“…listening to radio stations that target your area or are produced for your ethnic
group?”, and “…watching television and cable channels that target your area or are
produced for your ethnic group?”
Time with English-language Media. A similar set of questions was asked
regarding respondents’ time with English-language media, but asked for an
estimation of the day before, not the past week: “Approximately how many hours did
you spend yesterday…watching major English language television channels?”,
“…listening to major English language radio stations?, “…reading major English
language newspapers like the LA Times?”, and “…reading English language books
or magazines?”
Storytelling Network Integration (RQ3):
The storytelling network is more than the sum of its three nodes; the relative
level of integration is an indication of the network capacity of the storytelling
68
network. Kim (2003) developed the integrated connection to a storytelling network
indicator (ICSN) to measure the extent to which individuals’ connections to key
neighborhood storytellers (neighbors, local media, or community organizations)
stimulate connections to the other storytellers. The measure includes interaction
terms between the scope of connections to local media, scope of connections to
community organizations, and the intensity of interpersonal neighborhood
storytelling—
ICSN = LC OC OC INS INS LC × + × + × (LC = local media connectedness,
INS = intensity of interpersonal neighborhood storytelling, and OC = scope of
connection to community organizations) (Kim & Ball-Rokeach, 2006b).
Formal Networks
Connections to Community Institutions. The survey did not ask respondents
about their feelings of connection to social services directly, but did ask questions
about health care, schools and some other key community institutions. Together,
these five questions provide clues to respondents’ overall sense of connectedness key
community institutions. Respondents were asked to rank how connected they feel to
“public schools in your neighborhood”, “a church or temple,” “local businesses,”
“local gym or place to exercise/work out,” and “the police who work in your area.”
A five-point Likert scale was provided, with the instruction that “1” meant “not at all
connected” and “5” meant “very connected.”

69

Health Access Measures. Two questions on the survey asked respondents to
rate the difficulty they have experienced in finding health care for themselves, and
for their children. These two questions framed access to health care institutions in
terms of ease or difficulty making connections. The first question asked, “Overall,
how easy or difficult is it for you to get medical care for the child we’ve been
discussing when they need it? Would you say it is very difficult, somewhat difficult,
somewhat easy or very easy?” Later in the survey, respondents were asked to make
the same evaluations of their own efforts to make connections with health care
institutions: “Overall, how easy or difficult is it for you to get medical care when you
need it? Would you say it is very difficult, somewhat difficult, somewhat easy or
very easy?”
Interviews
Survey respondents who indicated (a) that their child(ren) are their most
frequent source of brokering help (as opposed to a friend, relative, or other adult),
and (b) had at least one child between ages 11 and 19, were considered eligible for
participation in in-depth interviews. For interviews, the brokering group was limited
to those who had an average of 3.0 (‘sometimes’) to 5.0 (‘very often’) on the
brokering scale (n = 34). The difference between the quantitative and qualitative
thresholds for inclusion is due to the different purposes of the two methodologies.
The survey data analyses presented the magnitude of group-level differences (i.e.
‘high’ brokering vs. ‘no/low’ brokering) on community-level indicators. Recruiting
70

interview respondents who need brokering help most frequently (those with a 5.0 on
the brokering scale were contacted first, moving down the scale to 3.0) was
important to ensure the richest possible dataset from the interviews, since people
who depend on their children every day would have more experiences to draw from
than someone whose children helps him or her only occasionally.
Past Metamorphosis recruitments for focus groups were my guide for likely
recruitment rate from the high brokering respondent pool, since the Project has never
contacted individuals for interviews. Past focus group recruitment indicated a 50%
inclusion rate, at best. To ensure an adequate number of participants, I also recruited
participants from a Women Infants and Children (WIC) office in the Greater
Crenshaw study area. This particular site was chosen for two primary reasons. First,
WIC has been identified in the immigration literature as an important social service
institution for Latino immigrants (Tse, 1995; A. Valenzuela Jr., 1999). The second
reason was even more pragmatic; I have a strong working relationship with WIC’s
local leadership, which afforded me easy access to a WIC Center site within the
confines of Greater Crenshaw.
During October and November, 2006, two bilingual undergraduate assistants
conducted 10-minute screening surveys at the WIC center with 27 potential
participants identified by WIC employees. Another seven individuals were
interviewed in March, 2007. The screening survey was designed to assess if that
person would be an appropriate respondent. Since WIC caters to pregnant women
71

and cares for children’s nutritional needs until their fifth birthday, we interviewed a
number of people who would have been good candidates, but did not have a child in
our age range (11 to 19 years old). The screening survey instrument collected
demographic information (household income, occupation, marital status, household
size, US and neighborhood tenure, and education) to ensure baseline similarities
between WIC recruited participants and Metamorphosis telephone survey
respondents.
Table 2.3 presents demographics of the telephone and WIC screening survey
participants from which interviewees were selected. Since WIC services target
pregnant women and young mothers, the gender skew was expected. On all other
demographic indicators, the two groups were roughly equivalent. The mean age for
telephone survey respondents was 36 years, compared with 35 years for WIC survey
respondents. Both telephone and WIC survey respondents were overwhelmingly
first generation (74% and 85%, respectively). Over half of each sample were of
Mexican origin (62% and 59%, respectively), and about a quarter of each sample
(21% of survey respondents and 26% of WIC respondents) indicated Guatemala as
their country of origin. The majority of both telephone and WIC survey respondents
reported an annual household income below $35,000 (94% and 87%, respectively),
and were married or living with a partner (71% and 85%, respectively). Lower rates
of employment reported in the WIC survey sample (62%) than in the telephone
survey sample (79%), was likely attributable to the gender skew in the WIC survey
72

participants, and to respondents having at least one child between the ages of 0 and 5
years old
15
. Telephone survey respondents were more likely to have at least some
high school education (52%) than WIC survey respondents (26%), but almost half
(42%) of telephone respondents still reported less than an 8
th
grade education, which
was the educational attainment of 70% of WIC survey respondents.

15
WIC serves pregnant women and mothers with children between the ages of 0 and 5 years old.
73
Table 2.3. Demographics of survey and WIC interview pools of broker’s parents

Source Telephone WIC Screening p-value
Survey Survey
Sample Size N=34 N=34
Age (mean) 36.30 34.67 .72
Immigration Generation
First Generation 73.53 85.19
Second Generation 20.59 14.81 .55
Country of Origin
Mexico 61.76 59.26
Guatemala 20.59 25.93
El Salvador 11.76 7.40 .26

Languages Spoken at Home
Spanish Only 29.41 48.15
Spanish & English 70.59 51.86 .13
Female 55.80 92.60 .001*
HH income
<=$35,000 93.94 86.96
>=$60,000 0.0 4.3 .90
Employment Status
Employed/Self Employed 78.8 61.5
Homemaker 15.15 31.62 .39
HH size (mean) 4.91 4.81 .80
Education
<=8
th
Grade 42.42 70.38
<=High School 51.52 25.93
>=Some College 6.06 3.70 .21
Marital Status
Never married/Single 17.65 0.0
Married/Living w/partner 70.59 85.19 .16
Years in the United States (mean) 20.56 17.54 .25
Years in Los Angeles (mean) 13.88 12.44 .40
Years in Neighborhood (mean) 8.35 6.81 .31

The WIC screening survey also contained the brokering scale questions from
the telephone survey so that interviewees could be selected on the same criteria,
regardless of whether the connection had been made via the telephone or WIC
screening survey. At the time of the in-person interview at WIC, the respondent
indicated whether they would be interested in participating in an in-depth interview.

74

If they met the criteria (i.e. a child/children between 11 and 19 as their main source
of brokering support, and a brokering scale average of 3.0 or higher), the respondent
was contacted by telephone by an undergraduate assistant to set a time for an
interview in their home.
Between January and April, 2007, in-depth interviews were conducted with
20 parents and their child who brokers for them most often (n=21). In one case, a
mother indicated that her two eldest children broker with equal frequency, so both
children were interviewed.
Tables 2.4 and 2.5 summarize the characteristics of the parents and children
interviewed for this project. Pseudonyms are used for all participants. Table 2.4
reveals that the gender skew evident in Table 2.3 was also reflected in the eligible
parents who ultimately became interview respondents. The skew toward women was
partly the result of sampling at WIC. However, there was a systematic gender
distinction operating independent of the sampling process. Men who were contacted
for interviews were more likely to insist that their children do not broker for them,
despite their responses in the telephone survey. There are two possible reasons for
this insistence: the first option is that men need less assistance because many
brokering tasks are domestic—tending to children’s educations and the family’s
health is traditionally feminized in immigrant Latino families. “Even working
mothers continue to be the primary caretakers of their children…and as such, come
into contact with a whole set of institutions through their children” (Jones-Correa,
75
1998, p.327). The second possibility is that men may need just as much help as
their wives, but refused to admit it to two young women who had come to their home
to ask them questions that exposed this vulnerability (see Menjívar, 2000, for similar
gendered experiences).
Table 2.4. Summary of interview participants: Parents
Name Country of Origin Education Age Years in U.S Survey

Ileana Guatemala <=8
th
Grade 38 17 WIC
Marta Mexico Some HS 41 22 WIC
Paula El Salvador <=8
th
Grade 41 20 WIC
Hilda Mexico Some College 38 20 WIC
Rosa Mexico
16
<=8
th
Grade 36 4 months WIC
Mauricio Mexico

<=8
th
Grade 18 WIC
Ana Mexico 2
nd
Grade 39 15 WIC
Carolina Mexico <=8
th
Grade 35 15 Telephone
Carlos Mexico Some HS 35 13 Telephone
Virginia Mexico Some HS 39 18 Telephone
Tulio Mexico <=8
th
Grade 43 20 Telephone
Anabel Guatemala <=8
th
Grade 54 22 Telephone
Alejandro Guatemala Some HS 51 27 Telephone
Estela Guatemala 3
rd
Grade 42 18 WIC
Carmen Mexico Some HS 40 23 Telephone
Guadalupe Mexico Some HS 40 9 WIC
Gabriela Mexico College Grad 30 9 months WIC
Hector Mexico Some HS 45 25 WIC
Armando Mexico 6
th
Grade 37 7 WIC
Aracely Mexico 8
th
Grade 33 18 WIC
Marita Mexico Some HS 27 1 WIC

16
Rosa & Mauricio are married and were interviewed together; having just arrived, she was nervous
to be interviewed alone.
76
Table 2.5. Summary of interview participants: Child brokers

Name Age Parent’s Name Generation
Graciela 13 Ileana Second
Alicia 12 Marta Second
Maria 19 Paula Second
Julio 17 Paula Second
Sonia 14 Hilda Second
Rolando 17 Rosa Immigrant (>2 years in US)
Mauricio
Luis 11 Ana Second
Hernando 12 Carolina Second
Liliana 16 Carlos Second
Regina 14 Virginia Second
Aurora 16 Tulio Second
Evelyn 15 Anabel Second
Milagro 13 Alejandro Second
Juana 13 Estela Second
Teresa 15 Carmen Second
Vida 18 Guadalupe Immigrant (9 years in US)
Yanira 14 Gabriela Immigrant (6 months in US)
Victor 16 Hector Second
Luisa 15 Armando Second
Gus 14 Aracely Second
Ruth 11 Marita Second

Interviews lasted approximately one hour, and were conducted in
respondents’ homes. Parental consent to interview their primary child broker(s) was
secured at the time of the parents’ interview, and parents agreed to their child being
interviewed in private. On this point, this researcher was relatively unsuccessful,
since the cultural differences in terms of ‘privacy’ became immediately apparent,
coupled with the reality that limited space in most families’ homes usually precluded
a totally private conversation. However, either because the child was being
interviewed in English, or for another unknown reason, most children seemed
unaffected by the presence of their parents, and gave rather candid responses.
Interviews were often conducted with younger siblings running around the living

77

room, televisions in the background, and family members walking in and out, which
provided a number of impromptu opportunities to observe the children brokering in
the middle of an interview that I was conducting. I was also able to observe how
other family members (i.e. those not being interviewed at the time) interacted in their
own space.
I conducted parents’ interviews with one of my bilingual undergraduate
assistants who took the lead in the actual interview because of her greater fluency
and cultural convergence with the participants,
17
and because this afforded me the
opportunity to record concurrently occurring interactions as well as participate in the
interview by adding my own questions to the standard interview protocol. I returned,
usually on a different day, to interview the child(ren) in English on my own. All
interviews were audio recorded and then transcribed. My undergraduate assistants
and I also compiled comprehensive field notes cooperatively immediately after each
interview.
Data Analysis
Data analysis of interviews was conducted both deductively and inductively.
Deductive criteria based on communication infrastructure, domestic infrastructure,
and social capital theories meant emphases on: child brokering tasks; resources in
informal networks; nature of connections with community organizations and local

17
I attended a language immersion program in Antigua, Guatemala in October and November, 2006
to improve my own Spanish ability, so that I did not have to rely too heavily on the brokering
capacities of my undergraduate assistants in conducting this research. I passed the US State
Department’s Foreign Service Exam before returning to the United States.
78

media; artifacts, activities and social arrangements in the domestic infrastructure; and
experiences in past efforts to convert social capital into connections with community
institutions. I coded specific instances and examples of interpersonal, organizational,
and mediated community resources identified by children and parents (as well as
those identified as missing from their networks), and examples of (in)effective
conversions of social capital into institutional connections, and was able to map these
thematically across interviews. The interview format allowed encouragement of
particular recollections by parents and children of interactions that they evaluated as
(un)successful, and their feelings about these interactions and brokering activities
generally.
Inductive analysis was based on emergent themes which, as Ream (2005)
indicated, are “anticipated but not specifically foreseen in the original
conceptualization of [the] study” (p.63). The particular constraining and enabling
factors in the home that affected child brokering efforts constituted an emergent set
of themes. Constraining factors included but were not limited to: children’s limited
linguistic capacity in both languages, domestic instability (including one case of
abuse), safety concerns, and limited parental attention due to large numbers of
siblings (particularly in homes with newborns). An unforeseen enabler of child
brokers was the strategy of teamwork to broker effectively, meaning that brokers
collaborated with children of parents’ friends or older siblings or cousins who could
79

be consulted on ‘grown up stuff,’ as Aurora, one of the brokers I interviewed, termed
tasks that were beyond her understanding.
Field Observation: Site Visits
From the interview participants, three parent-child broker pairs agreed to
participate in further observation. We returned to talk with these families, and
explained that we wanted to shadow them to meetings and appointments at schools,
health care, and social services, and observe the interactions. We explained that we
would not be there to assist with translating, but rather, to observe the interactions as
naturally as possible. One of my undergraduate assistants contacted each family
twice a week to find out if they had scheduled appointments, and to arrange where to
meet them for the observation. Each family was shadowed on two or three occasions
over the course of a two month period from March to May, 2007, for a total of 8
observations. We shadowed families to doctor’s appointments, drop-in times at the
clinic, teacher meetings, a school award ceremony, appointments at WIC, social
security, and to a social worker who was helping one of the families to collect
disability benefits.
Community Institution Interviews
Interviewees who indicated a connection, past or present, to health care,
education, and social service institutions were asked to provide the name of that
institution. There was overlap between the institutions mentioned by the
respondents. They provided informal evaluations of these institutions in terms of
80

their accommodations to Spanish speakers. Many of these organizations did have
some Spanish-speaking staff (some had no language services at all), but there was a
wide range in how effective residents thought these efforts at linguistic
accommodation were. For example, respondents indicated that many clinics had
Spanish-speaking receptionists who were called in to translate for doctors, but these
women were often stretched thin (since they are not officially translators) and
parents indicated that they were distinctly aware of the short-temperedness of the
staff on busy days. More often than not, parents opted to take their children with
them because there were no translating services at the institution, or because they
rated their children’s help as of higher quality and/or less stress-inducing.
I contacted two of each institution type, based on the feedback collected from
participants. If more than one respondent had mentioned a particular institution, I
made it a priority to interview service providers at that institution. I interviewed
between two and five individuals who were not native speakers of Spanish at each of
these six institutions, to understand more fully the ‘other side’ of the relationships
that immigrant families form with these institutions. I interviewed a total of 22
individuals across two middle schools, private medical practices, a local health care
clinic, a WIC Center, and a state-funded health insurance insitution. These
interviews lasted 20 to 40 minutes, and were conducted at the institution where the
respondent worked. Questions varied slightly depending on the particular role of
that individual in the institution (i.e. principal vs. teacher). The interviews focused
81

on personal evaluations of how language barriers affected their efforts to perform
their professional duties, and the stances they took on these challenges. Respondents
were asked to identify other people in their organization who might be
accommodating to immigrant families, as well as those who were more hostile to
these newcomers. I wanted to understand the full range of reactions, as these
institutional gatekeepers influence the relative success or failure of children and
family’s efforts to convert social capital into institutional connections. I also asked
these providers to detail any connections they had to other institutions in the
community, and the degree to which they viewed themselves (as embodied social
capital) or their institutions as resource brokers for their service recipients.
Table 2.6 details the institutions I visited and the people I interviewed at each
location. The pseudonym of the institution is listed, along with the ‘names’ and job
descriptions of the people I interviewed. I indicated the number of years that each
individual had worked there to emphasize the relatively rapid turnover at many of
these institutions, which is a risk to families’ efforts to convert their social capital
into sustainable connections with community institutions. The ‘level of language
accommodation’ column refers to (immigrant parent) interviewees’ subjective
evaluations of these institutions’ accommodations to Spanish-speakers.
I interviewed more people at the schools than at other institutions because
teachers’ daily free period made them regularly accessible in ways that health care
and social service providers were not. I also used time before and after each
82

interview for field observation, spending time in the schools, for example, during
classes and during breaks to view the natural interactions among students, and
between students and teachers.
83
Table 2.6. Community institution interview participants.
Institution Name Profession Years at Level of
Institution Language
Accommodation

Bellum Middle Low
School Mrs. Rhodes English as Second Language 3
teacher (6-8
th
grade)
Ms. Hernandez ESL Coordinator; 8
th
Grade 4
history teacher
Mr. South 7
th
grade social studies 6
Ms. Nelson Assistant Principal 4
Ms. Gibson 7
th
Grade Dean 6
Ms. Wagner 7
th
and 8
th
grade math 7
Tyler Middle Medium
School
Mrs. Thomas Administrator 2
Mr. Donald Teacher 3
Mr. Rathan Teacher 4
Ms. Lynwith Teacher 5

Family Medicine Practice High
Dr. Zimba Pediatrician 2

Children’s Care, Inc.
Medium
Dr. Truthe Pediatrician 3
Dr. Victor General Practitioner 4

Union Clinic Medium
Dr. Meeren Pediatrician 5
Ms. Banther Nurse Practitioner 2
Dr. Levande General Practitioner 3
Dr. Banzer Radiologist 4

WIC
High
Ms. Fenstein Nutritionist 1
Mrs. Dorsey Administrator 9
Ms. Brinkley Site Manager 4

Commission for Health Insurance High
Ms. Jaramillo Case Worker 3
Ms. Choza Project Manager 1

84

Data Analysis
Following the data analysis strategy employed in the parent and child broker
interviews, I employed a deductive-inductive process in analyzing these interviews.
Deductively, I focused on the service providers as sources of embodied capital for
the institutions they represent, as well as their capacities as gatekeepers to the
institutional and objectified capital in their institutions, based on their favorable or
unfavorable characterizations of interactions with child brokers and their immigrant
parents. I also evaluated their self-reported potential as resource brokers, both as
individuals and as representatives of their institutions.
Inductively, a number of anticipated but not completely foreseen themes
arose from these interviews. Many service providers characterized Latinos (either
favorably or unfavorably) in contrast to African Americans, regardless of their own
ethnicity or race. The ways in which providers expressed the differences they
perceived gave telling insights into (a) environmental impact on the probability that
Latino residents would have connections with African American neighbors as a core
part of their informal networks, and (b) the treatment of both groups by various
institutions and the individuals who work in them.
Employing Multiple Methods
The literature on multiple methods rests primarily on the assertion that all
methods have their weaknesses and limitations. Multiple methods are an effective
way to complement the weaknesses of one method with the strengths of another,
85

(a) by using ‘triangulation’ to increase confidence in a particular finding, or (b) to
provide partial or full explanations for unexpected results. Denzin (1978) described
triangulation as ‘the combination of methodologies in the study of the same
phenomenon” (p.291), drawing the term from navigation, where multiple reference
points are used to locate an object’s exact position. Denzin describes four types of
triangulation—here, I have employed data trangulation, by drawing from a variety
of data sources for this study, and methodological triangulation, by using multiple
methods to study a research problem (in Johnson et al., 2007, p. 114).
I employed multiple methods in this study not only to increase confidence in
my findings, but also to ground child brokering and social capital formation in a
larger, articulated context. Survey results gave a sense of the prevalence of
particular phenomena, and the ‘larger picture’ of where high brokering families were
socially placed, relative to other families in the community. It would have been an
incomplete picture, however, to study brokering and social capital formation—both
being processes that come to exist through communicative interaction—without
having a sense of how these interactions and negotiations take place in everyday life.
The qualitative data collected in this study, through parent, child, and service
provider interviews, as well as field observations, were designed to complement the
strengths of quantitative analysis and inform those findings, just as the quantitative
findings informed the selection of interview participants and the questions asked in
these interviews.
86

A research design where one method feeds into the next (what Morse [1991]
called sequential triangulation) also means that the strengths of one method can be
used to mitigate common issues with another. For example, by drawing interview
participants from the pool of telephone survey respondents (and expanding that pool
with a demographically equivalent group), the methods complement each other more
naturally than if interview participants had been selected separately, or through
snowball sampling, for example. Sieber (1973, in Johnson et al., 2007) and Nickel et
al. (1995) also integrated methods by using survey data to identify a desired sample
for interviews. Jick (1979) argues that quantitative methods can systematize
qualitative research (as in my project), and that qualitative findings can contribute to
the interpreting of statistical analyses by helping to clarify puzzling findings.
No matter how integrated the qualitative and quantitative elements of a
research project may be, there are still serious limits to generalizability. The
research on mixed-method design is often too self-congratulatory, focusing on the
strengths of integrating various methodologies without paying enough attention to
the persistent limitations of these research designs. Mixed-methods designs are very
unlikely to be replicable, which is one of the standards by which (particularly
quantitative) research marks progress (Bryman, 2007). Replication is possible, and
more likely when multi-method research is guided by a comprehensive and well-
articulated theoretical foundation, so that conducting an equivalent study in a
different immigrant community, for example, would have an equivalent grammar
87

and set of central organizing principles by which to compare and contrast the two
cases. However, replication is less likely with multi-method projects than with
single-method projects. This means that most multi-method studies are deep
investigations in a particular phenomenon in a particular site, but are highly limited
in terms of their exportability.
Other factors can be equally limiting. The gender imbalance among
interview participants is a clear limitation of this study’s generalizability. However,
this study is intended as a textured, ‘thick’ (Geertz, 1973) investigation of brokering
and social capital formation in a single community, not as a set of findings that can
be cleanly exported to another site—although some lessons may be learned from it.
Employing multiple methods was also a tool for understanding the nature of the
gender skew. Even if the pool of interview possibilities had been half men, it is still
likely that more women would ultimately have agreed to be interviewed. Field
observations of parents in community institutions—and the differences between male
and female interviewees’ willingness to be observed in these places at all—spoke
volumes about the gendered constraints on males requesting aid. The general
discomfort of males during interviews corroborated these field observations; many
men completed the interviews without making eye contact with us, and with their
shoulders squared away while speaking words that made them vulnerable. Older
male children exhibited equivalent body language when answering questions that
88

required retelling of vulnerable incidents, such as times when they had felt frightened
or had been unable to broker effectively.
The above discussions of multi-method projects’ limitations also reflects their
strength; researchers are more likely to identify their blind spots, since multiple
methods readily reveal inconsistencies that need to be addressed. Denzin (1978)
discusses three possible types of findings from mixed methods projects—
convergence, inconsistency, and contradiction (p.14). The major contradiction in
this study was revealed early because I had employed multiple methods. Initially, I
had framed this study with the hypothesis that child brokers were their parents’ links
to the community that compensated for the ‘fact’ that their parents were likely to be
relatively isolated from interpersonal networks, as Census data on linguistic isolation
had suggested. Survey analyses quickly revealed inconsistencies in this hypothesis,
indicating that high brokering families had close family and friends in the area, and
were still most likely to report difficulties connecting with community resources,
particularly institutions. Interviews quickly shifted ‘inconsistency’ into
‘contradiction.’ Interviewing children whose linguistic capacities were very limited,
for example, made it clear that investigating this phenomenon required a more
nuanced theoretical approach. I realized that the better approach was to disaggregate
interpersonal networks into family and peers, identify resources present and missing
from those networks, and to focus on the factors that constrained and enabled child
89

broker and parent efforts, successful or no, to convert resources into social capital
that links them to community institutions.
Convergence of findings is closely related to triangulation. Results are
strengthened when two or more methods indicate congruent results, since the
variation observed can be more confidently attributed to the phenomenon under
study, rather than to the nature of the method itself (Jick, 1979, p.602).
Triangulation is useful (to carry the navigation metaphor further) not only to confirm
the ‘location’ of the phenomenon itself, but to identify and investigate relevant
surrounding points that impact the phenomenon under study. In this case, the
surrounding ‘points’ were the field observations and interviews with service
providers in schools, health care, and social services. Field observations of
interactions and interviews with service providers expanded the focus from parents
and children and gave perspective on the other side of interactions that can determine
families’ access to institutional capital.
To understand why parents and children experienced particular difficulties in
making connections with community institutions, it was essential to understand how
individuals in positions of power, relative to institutional capital, perceived them.
Characterizations of the family from the ‘other side’ can show, for example, that an
unsuccessful attempt to convert social capital may have had little to do with the
family and everything to do with the institution’s gatekeeper. Employing multiple
methods to document interactions from multiple perspectives can also illuminate
90

more subtle reasons for unsuccessful attempts to link with community institutions.
For example, as was commonly the case, all sides of an interaction may have had a
convergent goal, such as that children do well in school. Multiple methods make it
possible to document the specific miscommunications that can unravel the best of
convergent intentions—and more importantly, may be able to offer suggestions of
how to ameliorate these situations.
The next chapter reports the telephone survey analyses guided by the first
four research questions. These analyses are followed by a discussion of the results,
including points of convergence and contradiction between the results of survey
analyses and findings from qualitative methods that are presented in chapters 4, 5,
and 6.

91

CHAPTER 3:
SURVEY ANALYSES
While child brokers are the primary focus of this research project, exploring
intra-group differences among Latino adults in Greater Crenshaw provides a sense of
the larger community context in which child brokering occurs. These analyses
provide a sense of the relative social positioning of high (N=84) and no/low
brokering residents (N=86) of Greater Crenshaw, and provide additional context by
including comparisons between English-speaking respondents (N=130) and Spanish-
speaking respondents (N=170).
The first set of research questions explores the composition of residents’
family and peer networks. The goal of these analyses is to determine the range of
interpersonal connections that respondents have in the local area, and with RQ1f, to
assess indirectly the social capital resources that may be embedded in those
relationships.
RQ1: How does the composition of respondents’ informal networks affect the
social capital resources embedded in them?
RQ1a: Do respondents’ reported levels of family interaction vary by
frequency of brokering?
The family interaction index (Wilkin, Katz, & Ball-Rokeach, in development
did not indicate significant differences between Spanish-speaking (M=2.89) and
English-speaking respondents (M= 2.81), (p=.323). High brokering respondents
92

indicate significantly higher family interaction (M=3.03) than no/low brokering
respondents (M=2.76), t(1,168)= -2.41, p=.02
RQ1b: Do reported levels of community belonging and collective
efficacy vary by frequency of brokering?
Spanish (M=22.98) and English-speaking residents (M= 23.56) did not
exhibit significant differences in belonging levels, t(1,298) = .28, p= .60. High and
no/low brokering respondents did not indicate significant differences in belonging
levels (p=.51), although high brokering respondents did report a higher level of
belonging (M=23.44), than no/low brokers (M=22.54).
Spanish-speaking residents (M= 3.37) indicate significantly higher levels of
collective efficacy than English-speaking residents (M= 3.08), t(1,298) = -2.69, p <
.01. Levene’s test of equal variances showed a significant difference in variances for
this item, F(1, 298)=12.96, p<.001.
High and no/low brokering respondents are not significantly different with
regard to collective efficacy (p=.12), although high brokers did report a slightly
higher level of collective efficacy (M=3.48), than no/low brokers (M=3.27).
RQ1c: Does frequency of discussion with neighbors vary by
frequency of brokering? What topics of conversation are reported
most often?
A significant difference was found between Spanish (M=5.70) and English-
speaking respondents’ (M=4.79) in terms of how much they engage their neighbors
93

in discussion, t(1,297) = -2.5, p=.02. High and no/low brokering respondents did not
exhibit significant differences in the frequency with which they speak to their
neighbors (p=.19), although high brokering respondents indicated talking more
frequently with their neighbors (M=6.04) than no/low brokers (M=5.37).
Table 3.1 presents the topics respondents report discussing most often with
their neighbors. Respondents were asked to provide multiple spontaneous responses,
which were coded into 25 categories, resulting in low sample sizes for each category.
Categories with fewer than five responses in total are not included here. The
remaining 18 are grouped thematically in Table 3.1 to illustrate general trends in
residents’ responses. “Neighborhood problems,” “crime/gangs/drugs,” and
“employment” were the three categories that received the largest proportion of
responses. Therefore, they are included in Table 3.1 as independent categories.
Several other answers have been grouped into the following four broader theme
categories:
• Informal networks includes the following topics of discussion with neighbors:
“Your/their children,” “your/their families,” “mutual friends or other neighbors,”
“your/their health,” and “personal issues;”
• General community conditions: “The weather,” “traffic/road conditions,” “cost
of living,” and “location/quality of goods & services;”
• Community institutions: “Schools,” and “church/church events;” and
94
• News: “Current events and news,” “politics,” “sports,” and “neighborhood
events.”
Table 3.1.Topics frequently discussed with neighbors
High brokering No/low brokering English-
respondents respondents speaking
respondents
N=84 N=86 N=130
Informal networks 31 24 33
General community conditions 13 13 17
Community institutions 14 11 9
Neighborhood problems 18 24 32
Crime/gangs/drugs 28 27 43
News 7 11 18
Employment 8 10 4

The three categories most frequently reported across all groups were informal
networks, neighborhood problems, and crime/gangs/drugs. The frequency of
discussion between neighbors and the topics of those discussions have important
implications for resident community involvement. Discussions about negative topics
can either discourage resident engagement in the community, or may encourage
residents to collectively attempt to improve the situation.
RQ1d: Does frequency of brokering vary by proximity of close
friends and extended family ?
No significant difference was obtained between Spanish and English
speaking respondents’ location of close friends and family. Seventy two percent of
Spanish-speaking and 70% of English-speaking respondents indicated that their close
friends and family live in the neighborhood or in Los Angeles County. High and
no/low brokering respondents also did not indicate a significant difference in the

95
locations of their close friends and family. Eighty percent of high brokering and
63% of no/low brokering respondents have close family and friends in the
neighborhood or within Los Angeles County as indicated in Table 3.2.
Table 3.2. Location of respondents’ close friends and family, as percentages
High brokering No/low brokering English-
respondents respondents speaking
respondents
N=84 N=86 N=130
In the neighborhood 12 10 7
In Los Angeles 68 53 63
In California, outside Los Angeles 12 23 21
In the US, outside California 4 7 7
Other countries 4 7 2

RQ1e: Does reported comfort with African American residents vary
by frequency of brokering?
. On average, English speakers (M=6.95) rated their past experiences with
African Americans more positively than their Spanish-speaking counterparts (M=
6.35). The difference between the groups was tested via an independent samples t-
test and was found to approach statistical significance, t(1,291) = 1.74, p=.08. High
brokering respondents rated their past experiences with African Americans more
positively (M=6.71) than no/low brokers (M=5.99). The difference between
brokering groups was not significant on this measure, (p=.13).

96

RQ1f: Does reported difficulty of accessing advice on child rearing
vary by frequency of brokering?
Spanish-speaking residents reported significantly more difficulty accessing
advice on child rearing (M= 3.44) than their English speaking neighbors, (M= 2.75),
t(1,96) = -4.02, p<.001. No significant difference was indicated between high and
no/low brokering respondents (p=.48), although high brokering respondents are
slightly more likely to report difficulty finding advice (M=3.52) than no/low
brokering respondents (M=3.37).
The second set of questions examines the storytelling roles of community
organizations and local media connection patterns as resources for respondents.
RQ2: What is the nature of respondents’ connections to the storytelling
network?
RQ2a: Are respondents’ community organization memberships
primarily sources of bridging or bonding storytelling?
Spanish speakers indicate significantly lower membership in community
organizations (M=1.12) than English-speaking respondents (M=1.66), t(1,105) =
3.11, p=.002. No significant difference was indicated between no/low and high
brokering respondents’ community organization memberships (p=.44). However, the
analysis did show that no/low brokering respondents have slightly higher
membership rates (M= 1.20) than high brokering respondents do (M=1.06).
97
Tables 3.3 and 3.4 illustrate respondents’ membership patterns. All Latino
respondents are most likely to report belonging to sports/recreational and/or
cultural/ethnic/religious organizations, whether they belong to one or multiple
community organizations. Most high and no/low brokering respondents belong to
one community organization, while more than half of English-speaking respondents
belong to more than one community organization.
Table 3.3. Respondents who belong to one community organization
High brokering No/low brokering English-
respondents respondents speaking
respondents
N=84 N=86 N=130
Sports/recreation 9 3 7
Cultural/ethnic/religious 11 9 12
Neighborhood/homeowner 1 1 0
Political/educational 1 0 1
Other 0 0 1
Total 22 13 21

Table 3.4. Respondents who belong to more than one community organization
High brokering No/low brokering English-
respondents respondents speaking
respondents
N=84 N=86 N=130
Sports/recreation 6 7 16
Cultural/ethnic/religious 5 5 21
Neighborhood/homeowner 0 2 9
Political/educational 1 2 9
Other 0 1 7
Total 6 8 24

If survey respondents indicated that they (or a family member) were members
of one or more community organizations, they were asked to specify the name of the

98

organization they felt was most important to them. The names of the organizations
provide clues as to whether these organizations, as community actors, are performing
a primarily bridging or bonding storytelling role. The following patterns were
gleaned from their responses:
• Sports/recreation organizations provided by Spanish-speaking were
overwhelmingly soccer clubs with Spanish names, and/or names that indicated
Central America in the title of the organization. A few simply said, “grupo de
amigos,” indicating that their primary community organization ‘membership’ is a
loosely organized group of friends who meet to play a sport, presumably soccer.
• English-speaking Latinos only mentioned soccer organizations a few times, and
usually referenced local American Youth Soccer Organization (AYSO) leagues.
These respondents provided a neutral list of organization names, including gym
chains, volleyball, swimming, softball, and basketball in public parks and
recreation areas, and school football teams.
• More than half of the Spanish-speaking respondents who reported a
cultural/ethnic/religious organization as most important to them named an
evangelical church. The other half indicated Catholic places of worship and
organizations associated with their country of origin.
• The majority of English-speaking respondents named a Catholic church as their
primary cultural/ethnic/religious organization.
99

• There was only one instance of overlap between the Catholic church names
mentioned by English and Spanish-speaking respondents.
• Spanish-speaking respondents who named a primary neighborhood/homeowners
organization indicated a police-related organization and two different
neighborhood groups with Spanish names.
• All but one English-speaking respondent who named a
neighborhood/homeowners’ organization indicated membership in a
neighborhood watch program.
• Only two Spanish-speaking respondents indicated a primary political/educational
organization, and one of these was a parent association with a Spanish name.
• English-speaking respondents indicated political/educational organization
memberships in various student associations, women’s groups, the Democratic
caucus, and school site committees.
RQ2b: Do respondents’ connections to media for learning about their
community vary by frequency of brokering?
Residents were asked to indicate the two most important ways that they find
out about community news and events. Chi-square tests were conducted on these
measures. Table 3.5 illustrates a number of significant differences between the
connections of Spanish-speaking and English-speaking respondents to find out about
the community:
100
• Spanish-speaking respondents are most likely to depend on television,
particularly ethnically targeted (i.e. Spanish-language) programming. Ethnically
targeted radio is the second most common resource mentioned by respondents,
followed by interpersonal communication, and ethnically targeted newspapers.
• English-speaking residents are also most likely to depend on television for
community information, but most watch English-language programming. The
Internet is the second most commonly reported connection for English speakers,
followed by interpersonal connections and English-language newspapers.
Table 3.5. Two most important ways to stay on top of community news and events:
Spanish and English-speaking respondents

Spanish-speaking English-speaking X
2
P
respondents respondents value
N=170 N=130
Talking with others 51 43
Internet*** 15 44 29.20 <.01
Television/cable/satellite ** 142 92 6.99 <.01
English language*** 28 61 32.74 <.001
Public station 16 21
Ethnically targeted*** 111 20 74.60 <.001
Radio*** 58 18 16.00 <.01
English language 6 5
Public station 6 8
Ethnically targeted*** 49 6 28.83 <.001
Newspapers 39 36
English language*** 8 30 22.48 <.001
Community 9 5
Ethnically targeted*** 27 4 13.04 <.001
Movies 0 1
Books/Magazines 1 4
Organizations 0 1
Leaflets/folders 1 2

101
Table 3.6 illustrates high and no/low brokering respondents’ most important
ways for staying on top of community news and events. No significant differences
were indicated between these two groups’ connection patterns. Regardless of
reported frequency of brokering, both groups are most likely to connect with
ethnically targeted television, followed by radio (usually ethnically targeted stations),
interpersonal connections, and (usually ethnically targeted) newspapers.
Table 3.6. Two most important ways to stay on top of community news and events:
High and no/low brokering respondents

High brokering No/low brokering
respondents respondents
N=84 N=84
Talking with others 24 27
Internet 8 7
Television/cable/satellite 68 74
English language 11 17
Public station 8 8
Ethnically targeted 57 54
Radio 32 26
English language 4 2
Public station 6 0
Ethnically targeted 26 23
Newspapers 19 20
English language 5 3
Community 6 3
Ethnically targeted 12 15
Movies 0 0
Books/Magazines 1 1
Organizations 0 0
Leaflets/folders 1 0

102

RQ2c: Do respondents’ general media connection patterns vary by
frequency of brokering?
This question moves beyond media connections for a particular purpose—
finding out about community news and events—to consider the general media
connection patterns for Latinos in greater Crenshaw. Respondents were asked to
report how much time they had spent connecting with English-language media
yesterday, and with media produced for their ethnic group in the past week.
The differences between Spanish and English-speaking respondents are stark.
Table 3.7 presents the results of independent samples t-tests on these measures.
Significant differences were found in every comparison, indicating that Spanish
speakers connect with all ethnic media forms significantly more than English-
speaking respondents do. The opposite is true for English-speaking respondents who
connect with all forms of English-language media significantly more than Spanish-
speaking respondents.
103
Table 3.7. General media connection patterns of Spanish and English-speaking
respondents

Spanish-speaking English-speaking df t-value p
respondents respondents value
N=170 N=130
Mean time spent with
ethnic media 10.20 6.01 296 -7.29 <.001
in the last week
Television 4.69 2.92 297 -6.76 <.001
Radio 3.70 2.19 298 -5.37 <.001
Newspapers 1.82 .89 297 -4.13 <.001
Mean time spent with
English-language media 2.17 5.93 298 11.74 <.001
yesterday
Television 1.09 3.18 298 10.50 <.001
Radio .89 1.88 298 5.08 <.001
Newspapers .18 .87 298 7.38 <.001
Books/media .40 1.20 298 5.82 <.001

Generally, high brokering respondents (M=11.04) report spending
significantly more time with ethnic media than no/low brokering respondents
(M=9.39), t(1, 166)= -2.23, p=.03. Within that general pattern, high brokering
respondents spend significantly more time with ethnically targeted radio (M=4.07)
than no/low brokering respondents (M=3.34), t(1,168)=.199, p=.05. Levene’s test of
equal variances showed a significant different in variances for this item, F(1,
168)=4.28, p=.04. As Table 3.8 shows, there are no other significant differences in
general media connection patterns between high and no/low brokering respondents.

104
Table 3.8. General media connection patterns of high and no/low brokering
respondents

High brokering No/low brokering df t-value p
respondents respondents value
N=84 N=86
Mean time spent with
Ethnic media 11.04 9.39 166 -2.23 .03*
in the last week
Television 4.92 4.47 167 -1.49 .14
Radio 4.07 3.34 168^ 1.99 .05*
Newspapers 2.05 1.59 167 -1.40 .16
Mean time spent with
English-language media 2.04 2.30 168 .71 .48
yesterday
Television .92 1.27 168 1.71 .09
Radio .87 .92 168 .22 .83
Newspapers .25 .12 168 -1.77 .08
Books/media .29 .51 168 1.77 0.8

^ Equal variances not assumed.

The first two research questions explored the nature of respondents’
connections to resident networks (both peer and family) and to community
organizations as bonding and/or bridging storytellers. I also examined the media
connections people rely on to stay on top of what is happening in their community,
as well as general daily and weekly connections to English and Spanish-language
media. The third research question focuses on the integration of the three nodes of
the storytelling network for each group of Latino respondents (high and no/low
brokering, as well as English-speaking) to explore potential differences in
storytelling network integration based on reported frequency of brokering:

105

RQ3: Do levels of storytelling network integration vary by frequency of
brokering?
The Integrated Connection to a Storytelling Network (ICSN) is the weighted
total of the interaction terms between local media connectedness, scope of
connections to community organizations, and intensity of interpersonal
neighborhood storytelling. The ICSN captures residents’ connections to resident
networks, community organizations and local media, and infers the integration of the
storytelling network as a whole from integration of individual residents into it. The
degree of integration for English-speaking respondents is lower (M=37.63,
SD=22.92) than for Spanish-speaking residents (M=50.59, SD=27.29). High
brokering respondents are part of a more integrated storytelling network (M=54.92,
SD=26.96) than no/low brokering respondents (M= 46.36, SD=27.10).
The fourth set of research questions examine the effects of reported brokering
frequency on residents’ evaluations of their access to community institutions.
RQ4: Do respondents’ connections to community institutions vary by
frequency of brokering?
RQ4a: Do respondents’ evaluations of their connections to
community institutions vary by frequency of brokering?
Spanish-speaking respondents reported feeling more connected to local
schools, local church/temple, local businesses, and the police that work in the area,
than their English-speaking counterparts. Spanish speaker respondents’ (M =2.86)
106
indicated feeling higher levels of connectedness to local schools than English-
speaking respondents (M =2.51).This difference approached significance, t(1, 297) =
-1.87, p=.06. Spanish-speaking respondents (M =3.12) also indicated a higher sense
of connectedness to local businesses than English speaking respondents (M =2.78).
This difference approached significance, t(1,298) = -1.86, p=.06. Spanish-speaking
respondents (M=2.63) indicated a significantly greater sense of connection to the
local police than English-speaking respondents, (M=2.13), t(1, 294) = 18.41, p=.007.
Levene’s test of equal variances showed a significant difference in variances on this
measure, F(1,294)=20.67, p<.001. Table 3.9 shows the results of these analyses.
Table 3.9. Feeling connected to community organizations: Spanish and English-
speaking respondents.

Spanish-speaking English-speaking df t-value p
respondents respondents value
N=170 N=130
Public schools 2.86 2.51 297 -1.87 .06
Church/temple 3.18 2.85 297 -1.65 .10
Local businesses 3.12 2.78 298 -1.86 .06
Local gym/place to exercise 1.81 2.09 296^ 1.65 .10
The local police 2.63 2.13 294^ -2.71 .007*
^ Equal variances not assumed.

High brokering respondents indicated higher levels of connectedness to local
institutions than no/low brokering respondents across all five measures. High
brokers indicated a significantly stronger connection to local churches or temples
(M=3.54) than no/low brokers (M=2.83), t(1, 168)=-2.82, p=.005. High brokers are
also significantly more connected to local police (M=2.95) than non-brokering

107
respondents (M=2.33), t(1,165)=-2.38, p=.02. Table 3.10 shows the results of these
analyses.
Table 3.10. Feeling connected to community organizations: Spanish and English-
speaking respondents.

High brokering No/low brokering df t-value p
respondents respondents value
N=84 N=86
Public schools 3.04 2.70 168 -1.35 .18
Church/temple 3.54 2.83 168 -2.82 .005*
Local businesses 3.19 3.06 168 -.547 .59
Local gym/place to exercise 1.92 1.71 167 -.996 .32
The local police 2.95 2.33 166 -2.38 .02*

RQ4b: Do respondents’ evaluation of access to health care for
themselves and for their children vary by frequency of brokering?
Spanish-speaking respondents indicated significantly greater difficulty in
accessing health care resources for themselves (M= 3.89) than English speaking
respondents did, (M=3.09), t(1,294) = -7.08, p<.001. Levene’s test of equality of
variances showed a significant difference in variances for this measure, F(1,
294)=.483, p=.03. Accessing health care for children is significantly more difficult
for Spanish-speaking respondents (M= 2.91) than for English-speaking respondents
(M=2.35), t(1,96) = -3.98, p<.001. High and no/low brokering respondents are not
significantly different with regard to how challenging they find it to get the health
care they need for themselves (p=.611). However, no/low brokers find it slightly
more difficult (M=3.93) then high brokering respondents (M=3.86) to get access to
the health resources they need. Both groups rated accessing health care for their

108

children as easier than locating these services for themselves, although the
differences between no/low and high brokering respondents was not significant on
this measure (p=.36). No/low brokering respondents did indicate slightly more
difficulty (M=3.00) than high brokering respondents on this measure (M=2.81).
Discussion
The four sets of research questions presented in this chapter document the
differences between respondents’ storytelling network connections, overall
integration of their storytelling networks, and their connections with community
institutions in the communication action context. A number of significant
differences between groups, on language preference (Spanish or English-speaking)
or on frequency of brokering within Spanish-speaking respondents (no/low and high
brokering) were reported in this chapter. There were also many instances when the
differences between group means indicated a general trend, even though statistical
tests did not yield significant differences. This was due in part to the low sample
sizes of the no/low (N=86) and high brokering groups (N=84) used in these analyses.
In this discussion section, I examine some of the general trends observed between
these groups, and connect the results presented in this chapter to findings collected
through other methods.
RQ1a indicated that high brokering respondents report significantly more
family interaction than their no/low brokering neighbors. Given that brokering is
dependent on family interaction and communal problem solving, this finding
109

converges with the results of interviews and observations of families with child
brokers. The significant difference between brokering groups on this measure also
supports the idea that families where children broker frequently are substantively
different from families where children do not, particularly since these two groups of
respondents are essentially equivalent on all other socio-demographic indicators.
RQ1b measured respondents’ belonging levels and collective efficacy.
Belonging levels measure respondents’ level of interdependence with their
neighbors, as well as the relative importance that residents place on having close
relationships with their neighbors. Collective efficacy measures the convertibility of
connections with neighbors into confidence that residents can come together to solve
common problems. In a sense, collective efficacy is an indirect measure of the
resources respondents believe they have in their networks. High brokering
respondents indicated the highest mean score on this measure, relative to English-
speaking and no/low brokering respondents.
RQ1c results indicated that Spanish-speaking respondents are significantly
more likely to talk frequently with their neighbors. When the main topics of talk
were examined, neighborhood problems and crime/gangs/drugs were the two largest
categories of responses. These findings are all the more remarkable because these
were individual categories and the others were composites. This converges with
qualitative findings that safety is a highly salient communication action context
feature in this community.
110

The relationship between the frequency of neighborhood discussion and the
topic of conversation is less clear. If most of respondents’ discussions with their
neighbors focus on neighborhood problems and crime, there are two potentially
negative consequences to consider. First, time spent heightening each other’s fear
about the neighborhood is time not being spent sharing information resources that
could help respondents and their families to better connect with their community.
Second, these conversations could have a chilling effect on residents’ desire to spend
time in the community. They may opt, for example, to seek medical care in other
areas and bus their children to schools outside their community to limit time spent in
the local area. On the other hand, the relatively high reports of collective efficacy,
particularly by Spanish-speaking respondents, might mean that the frequent
discussions of neighborhood detractors actually prompt neighbors to come together
and improve the situation. While I did not witness or hear about a large-scale effort
in this vein in my interviews and observations, more modest improvements,
cooperatively performed by residents, could have been happening below my radar
screen. Where neighbors are communicating regularly and storytelling the
community, there is the possibility of coordinated action.
RQ1d indicated that most respondents have close friends and/or family
within Los Angeles County, but few indicated that these individuals are living in the
local neighborhood. Interviews revealed the same pattern. Family members and
friends generally live in other parts of Los Angeles, and spend time with each other
111

on weekends. Very few interviewees from either the telephone survey or the WIC
screening survey indicated close friends and family living in their neighborhood.
While RQ1e did not directly test the presence of African Americans in Latino
residents’ peer networks, past experiences with their closest neighboring group is a
clue to whether these relationships are likely to form in the future. Individuals who
have had very negative past experiences—or no experience at all—are not likely to
seek even casual conversation with African Americans. It is interesting, therefore,
that high brokering respondents report more favorable experiences with African
Americans than no/low brokering respondents, considering that the former are most
likely to have language barriers to contend with in these interactions. Perhaps
children’s brokering efforts to connect their parents to African American neighbors,
as reported in interviews, contributes to these favorable reports of prior contact by
high brokering respondents.
Figure 3.1 presents the relative standings of English-speaking and Spanish-
speaking no/low and high brokering respondents on the measures in RQ1a-e. On all
five measures, a higher mean would have potentially positive outcomes for
community storytelling in interpersonal networks, and potentially for the range of
resources embedded in those networks.
112

English
No/low
HIGH
No/low
HIGH
English
English
No/low
HIGH
English
No/low
HIGH
No/low
English
HIGH
No/low
HIGH
English
Family interaction
Belonging level
Collective efficacy
Frequency of
discussion with
neighbors
Proximity of close
family and friends
Ratings of African
Americans
- -
+ +
English
No/low
HIGH
No/low
HIGH
English
English
No/low
HIGH
English
No/low
HIGH
No/low
English
HIGH
No/low
HIGH
English
Family interaction
Belonging level
Collective efficacy
Frequency of
discussion with
neighbors
Proximity of close
family and friends
Ratings of African
Americans
- -
+ +
Figure 3.1. Relative positions of English-speaking and Spanish-speaking no/low and
high brokering respondents on RQ1 measures.

Figure 3.1 shows that high brokering respondents report the highest levels of
family interaction, collective efficacy, discussion with neighbors, and proximity of
close friends and family in the Los Angeles area. High brokering respondents’ mean
belonging scores and ratings of past experiences with African Americans are higher
than no/low brokering respondents, but lower than English-speaking respondents.
No/low brokering respondents seem to be at a disadvantage compared to both high
brokering and English-speaking respondents. An interesting question therefore
emerges about no/low brokering respondents: do some people in this group actually
need brokering assistance, but either refuse to ask for help, or have no one to ask?
High brokering respondents seem fairly well placed within informal networks,
relative to the two other groups of Latino respondents examined here. The results of
RQ1f, however, reveals a caveat: when reporting difficulty accessing advice on child
rearing, high brokering respondents report the most difficulty, followed by no/low
brokering and English-speaking respondents.

113

Consideration of RQ1f in relation to the results of RQ1a-e in Figure 3.1
indicates convergence between qualitative and quantitative analyses in this study.
Although high brokering respondents report strong links to their family and peer
networks, the limited resources in those networks constrain residents’ efforts to
convert those resources into desired connections with community resources—in this
case, accessing advice on raising children.
The second set of research questions examined respondents’ connections to
community organizations and local media. Spanish-speaking respondents indicated
significantly lower rates of connection to community organizations, and the
memberships that they do report are almost entirely of the bonding variety. Spanish-
speaking respondents belong to informal, ethnically-bound sports/recreation (mainly
soccer) organizations, ethnic/cultural organizations associated with their country of
origin, and churches that are inwardly focused. Chàvez’s (2006) research on the
evangelical and Catholic churches in Greater Crenshaw revealed that evangelical
churches (which accounted for more than half of the cultural/ethnic/religious
organizations reported by Spanish speakers) are homogeneously immigrant and
focus on the religious mission of the church rather than the community in which the
church is located. Chavez found that Catholic religious leaders are more likely to be
community storytellers than evangelical leaders.
While community storytelling may be issued from the pulpit, churches that
conduct services in multiple languages tend to create sub-congregations that have
114

little contact with each other, even though they share the same building (Chafetz &
Ebaugh, 2000). That there was almost no overlap between the names of the churches
specified by English and Spanish-speaking respondents supports the argument that
Spanish-speaking respondents probably attend segregated religious services.
Spanish-speaking respondents indicate very low connections to the more likely
bonding storytellers: homeowner/neighborhood and political/educational
organizations, particularly respondents who indicate only one community
membership. Respondents who belong to more than one organization may benefit
doubly from their increased access to organizations, and from being connected to
different types of organizations. These respondents are more likely to be connected
to sources of both bonding and bridging storytelling than the vast majority of
respondents.
English-speaking respondents indicate more neutral and varied community
organization membership patterns conducive to increasing the range of community
information resources these residents have available. English speakers report
connections to a wide range of sports/recreation organizations, and a number indicate
membership in local gym chains. Most specified a connection to a Catholic church,
meaning a greater likelihood of community storytelling by the organization itself,
and among congregation members, than in evangelical churches (Chàvez, 2006).
These residents also report connections to a range of political/educational
organizations, and to a lesser degree, to homeowner/neighborhood organizations.
115

English-speaking respondents are likely to have access to a wider range of both
bridging and bonding storytelling resources relative to Spanish-speaking
respondents.
Local media cannot be presumed to function as bridging or bonding
storytellers based on the data collected in this survey. Ethnically targeted media are
not predestined to tell only bonding stories, and English-language media are not
necessarily bereft of bonding storytelling resources. The results to RQ2b and RQ2c
presented here do give some clues as to how media connectedness patterns could
affect connections to the storytelling network and access to needed resources. For
second generation Latinos or immigrants with some English proficiency, ethnic
media will be part of a wider range of possible media connections they can make to
satisfy a particular goal. Respondents with low or no English language proficiency
are more likely to connect only with ethnic media forms. This means when local
news is missing from ethnic media, these respondents are less likely to get
information that can help them adapt to their new community. These limitations are
further exacerbated by residents’ narrow choices in programming that were
evidenced in interviews and field observations. Most parents I interviewed said that
their day-to-day media connections are primarily entertainment, with radio for
music, and with television for telenovelas and rented movies, not for community
information and news. These reciprocal limitations to local media storytelling reflect
Wilkin (2005) and McDaniel’s (2006) findings that Latinos who connect primarily
116

with Spanish-language media have more barriers to health care access and to the
political process, respectively.
A significant number of English-speaking respondents indicated that the
Internet is a key resource for community news, compared with very few Spanish-
speaking respondents. This converges with interview findings that most parents
avoid the computer, but do have their children find information for them online when
they need it. Chapter 4 devotes attention to the access, range and quality of mediated
connections that children broker for their parents.
Only one person in the total sample referenced a community organization as
one of the primary ways they find out what is going on in the community (RQ2b),
which raises questions about the integration of storytelling network nodes.
To answer RQ3, storytelling network integration was calculated for each
group of Latino respondents. High brokering respondents indicated connections to a
more integrated storytelling network than no/low brokering or English-speaking
respondents. This measure indicates the integration of individual residents into
resident networks, a range of community organizations, and connections to local
media. Why high brokering respondents indicate a more integrated storytelling
network than the other groups of Latino respondents is an interesting question with
many partial answers. First, these levels of integration have to be understood as
relative to each other; how integrated these residents’ storytelling networks are on an
objective scale is questionable. Second, qualitative analyses would conclude that
117

high brokering respondents’ storytelling network integration is both modest and a
group effort. While parents tend to connect with community organizations on their
own, child brokers are the family’s key connectors to local media, particularly for
community information and news. Connections to resident networks are the result of
a combined effort by child brokers and their parents. Child brokers have the
potential to compensate for their parents’ linguistic, cultural, and technological
difficulties in connecting with other residents and local media to contribute to their
parents’ integration into the storytelling network.
The fourth set of research questions explored respondents’ connections to a
range of community institutions. These questions provide a foundation for
understanding the communication action context by asking respondents to rate their
enabling (feelings of connection with local institutions) and constraining (difficulties
accessing health care) factors in making and maintaining connections with
community institutions.
RQ4a results complement the RQ3 finding that high brokering respondents
have the most integrated connections to their storytelling network. High brokering
respondents indicate highest mean feelings of connection to four of five community
institutions: schools, local churches, local businesses, and local police, relative to
no/low brokering and English-speaking respondents. High brokering respondents’
relatively strong feelings of connection to local institutions may enable their
118

connections to other institutions (including health care and social services) if these
institutions are resource brokers for residents.
English-speaking respondents indicate the highest mean connection to a local
gym or place to work out, which complements findings in RQ2a that English
speakers are more likely to be members of local gym chains. As RQ2a findings
indicate, English-speaking respondents are more likely to belong to exercise-related
institutions, like gym chains, than Spanish speaking respondents who fulfill these
needs with informal arrangements or participation in local soccer organizations.
RQ4b results indicate that Spanish-speaking respondents indicate
significantly greater difficulty accessing health care for themselves and for their
children than English-speaking respondents, and that no/low brokering respondents
indicate more difficulty on both measures than high brokering respondents. Again,
the identity of these no/low brokering respondents raises questions. Are these
respondents more isolated than those who can admit that they need help or have
people to help them when needed? Respondents reporting less difficulty accessing
health care for their children provide some evidence for the success of free and
reduced-cost programs to provide health insurance and health care for children in
Los Angeles County that have been introduced or expanded in recent years. These
results converge with qualitative data. Parent interviews indicate that their children
benefit from services these insurance programs provide and see doctors regularly.
Parents do not see doctors often (or for many years at a time) for a number of
119

reasons. Parents’ general lack of health insurance is a serious barrier to care (Wilkin,
2005), as is the cost of care. Unless parents have an acute or serious chronic health
concern, most prioritize spending money on their children’s care at the expense of
their own.
As parents make decisions and sacrifices to help their children, so too do
their children. Children’s brokering activities in the home, peer networks, and in
community institutions often involve prioritizing the needs of family members over
their own. In the next chapter, we turn our attention to the domestic infrastructure of
the family home, to examine the day-to-day brokering activities of children that are
key to connecting with, and maintaining links to, local media, peer networks, and
ultimately, to community institutions.

120

CHAPTER 4:
INFORMAL NETWORKS AND DOMESTIC INFRASTRUCTURE
Understanding child brokering must begin with a thorough exploration of the
family network and home-based brokering activities that are the precursor to
brokering activities in community spaces, including schools, social services, and
health care institutions. Children’s efforts to connect with community resources that
can be converted into successful connections with local institutions encourage a
renewed focus on the storytelling network.
Connecting the family to the storytelling network is a cooperative effort by
parents and children. Parents connect with community organizations with little help
from their children. Child brokers are the family’s primary connectors with local
media. Parents and children have connections to different peer networks, but
children’s brokering activities can enhance parents’ connections to their peers when
brokering serves as a way of returning or invoking favors and feelings of goodwill.
Children can also broker conversations between parents and African American
residents that would otherwise be precluded by language differences.
Children’s brokering efforts in the home, connecting with peer networks, and
with local media, comprise the set of storytelling network resources they can attempt
to convert into connections with community schools, health care, and social service
institutions. This chapter focuses on children’s brokering activities in peer and
family networks. Brokering activities in the home often involved connecting with
121

different forms of local media, ranging from mail, school newsletters, and telephone
calls, to television and the Internet. These local media forms all have potential
community storytelling capacity, and children’s connections with these media
facilitate further instantiation of the brokers’ families into the storytelling network.
The data presented in this chapter are the result of interviews with 20 parents
in Greater Crenshaw, and 21
18
children that parents identified as their primary child
brokers. The distinctiveness of each family network was evident in the interviews,
field observations, and site visits conducted with three of these families. Each family
displayed a different constellation of factors that constrained and/or enabled
children’s efforts to connect with community storytelling resources, and to convert
those resources into connections with community institutions. Many of these
enabling and constraining factors were common across families with whom we had
contact.
Livingstone’s (2002) concept of domestic infrastructure, combined with
Lievrouw & Livingstone’s (2005) definition of ‘infrastructure’ as an ensemble of
artifacts, activities, and social arrangements, helps explain both the patterns and
uniqueness of child brokering in the family network. Applied to child brokering in
the domestic infrastructure, artifacts refer to the range of media forms in the home
that children can access in their efforts to connect their families to community
storytelling resources; activities refer to the brokering and consequent family sense-

18
In one family, the mother indicated that her two oldest children broker for her and her husband
equally, so both children were interviewed.
122

making that result from connecting with those media; and social arrangements refer
to routinized relationships between family members that affect child brokering
activities.
The relationships between artifacts, activities and social arrangement in the
domestic infrastructure should be considered, according to Lievrouw & Livingstone,
as an “ensemble” rather than linear (2005, p.4). Although the interrelations are
dynamic, the term “infrastructure” conveys that the relations are not infinitely
flexible. Rather, the relationships between media artifacts, activities, and social
arrangements become “routine, established, institutionalized, and fixed to various
extents, and so become taken for granted in everyday life” (p.4). This chapter
provides insight into the fixed routines of parents and children in their family
networks, with particular focus on how routinized activities and social arrangements
can constrain and enable children’s brokered connections to community storytelling
resources.
Since a focus on child brokering necessitates a discussion of community
resources—whether they are shared through interpersonal networks, community
organizations, and/or local media storytelling—differentiating between the kinds of
resources child brokers access and their relative value is essential. Domestic
infrastructure’s components: media artifacts, activities, and social arrangements are
complemented by Bourdieu’s (1986) dimensions of social capital: access, range, and
quality, respectively. The media artifacts to which child brokers have access
123

depends on a number of structural considerations, and access is the necessary first
step to understanding the relative successes or failures of child brokers’ efforts to
connect with community storytelling resources. Children’s brokering activities
increase the range of content that the family network can access through connections
with media artifacts, because child brokering can compensate for parents’ linguistic,
cultural, and technological limitations to accessing these media on their own.
Finally, social arrangements of family members around brokering activities and
media artifacts determine the quality of connections that family members have to
different community information resources. Parental retention of authority and
cooperation between siblings can enable the primary child brokers’ capacity to
connect with local media and other community storytelling resources. On the other
hand, family instability caused by conflict, illness, addiction, and abuse can constrain
the quality of children’s efforts to connect with community resources.
Figure 4.1 provides a heuristic model of the domestic infrastructure. In this
illustration, the interrelationships between access to media artifacts, range of
(brokering) activities, and quality of (family) social arrangements are indicated with
dual sided arrows. The establishment of routines and structures related to brokering
activities—i.e. the infrastructure of the family home—is indicated by the hyphenated
enclosure.
124

ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure
ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure

Figure 4.1. The domestic infrastructure.

Domestic infrastructure is useful for exploring immigrant family networks
because it provides a complementary, micro-level framework to communication
infrastructure theory. Communication infrastructure theory accounts for human
agency, within a framework that can constrain and enable that human agency.
Domestic infrastructure achieves the same balance between agency and structure for
investigating the family home.
19
Livingstone (2002) considers both parents and
children as agents that negotiate connections with media artifacts in a domestic
context that can constrain or enable connections to resources for community
information. The recursive relationship between the domestic infrastructure and the
agents within it implicate the uniqueness of each family situation. RQ5 explores the

19
In fact, Livingstone (2002, p. 74)) explicitly relates her theoretical development to communication
infrastructure (Ball-Rokeach, Kim & Matei, 2001) in her initial formulation, although she did not
fully develop these links.

125

family network differences revealed in interviews with parents and child brokers, as
well as field observations of these families interacting at home:
RQ5: How do domestic infrastructure components: access to media artifacts,
range of brokering activities, and quality of family social arrangements,
constrain and/or enable child brokers’ efforts to connect their families with
the community?
Access to Media Artifacts
The term ‘media artifacts’ refers to two distinct types of media to which child
brokers regularly connect their parents. “Media” draws attention to the traditionally
defined local media connections that children broker for their families, including
television, radio, Internet, newspapers, and community newsletters. “Artifacts”
expands the definition of local media to include artifacts such as bills, pamphlets,
and telephone calls that children regularly brokered for their families. Bourdieu’s
emphasis on access as the gateway to evaluating resources—in this case, those
community resources that children connect with and attempt to convert into
connections with community institutions— draws attention to structural constraints
in the domestic and community infrastructures that affected families’ general access
to local media.
Although parents reported making independent media connections, these
were almost always entertainment-focused choices. Newspapers were purchased for
sports scores, television programming was generally limited to game shows and
126

telenovelas, and radios were usually turned on for Spanish or Christian music. Some
parents did listen to popular morning radio programs on Spanish-language stations
that have proved their capacity for community storytelling. Three of these disk
jockeys were instrumental in mobilizing the Latino community to participate in the
2006 immigration rallies that took the English-language media and police by surprise
(Félix, González & Ramírez, 2007). In general, however, parents’ primary media
connections were not oriented toward gathering community information.
Connections that child brokers reported making to local media artifacts were
generally their families’ primary contact with these community storytelling
resources. There were a few key structural factors that made each home a distinctive
context in which child brokers were forging these media connections. The first
factor was the range of media the family could afford to have in the home. The
second was the location of these media within the home, and the final factor was the
community-level availability of local media, which necessarily circumscribes what
children can connect with from the family home.
Livingstone’s (2002) emphasis on the range of media artifacts available in the
home also draws attention to the media artifacts that are not available in those homes
(Ball-Rokeach, Cheong, Wilkin & Matsaganis, 2004). The working-class
immigrants who participated in this research could not afford to own any media
artifact they might desire. Therefore, the artifacts they reported owning reflected the
family’s media priorities. Every home we visited had at least one television, radio,
127

and telephone. Most homes had a computer (although fewer had Internet access,
most citing cost as a constraint), and only a few parents reported regularly
purchasing newspapers or magazines. The range of media artifacts that these parents
made available in their homes necessarily constrains the range of local media
artifacts their children could access to connect with community storytelling
resources.
Television was the dominant artifact in these homes—both literally and
figuratively. In a number of homes, the size of the television most closely resembled
a movie screen, often taking most free space in small living rooms. Every home we
visited had cable as well. Many parents cited cost as their main reason for not
having access to the Internet. Since cable is more costly than Internet access, this
choice reflected a clear priority on entertainment. Internet access might have served
child brokers’ efforts to locate community resources more readily than television,
because the Internet can accommodate customized access to information (December,
1996) in ways that are impossible with television. Parents’ choices to spend their
limited discretionary income on cable rather than Internet access potentially
constrained their children’s efforts to connect with community storytelling resources.
Most of the homes (17 of 20) we visited had at least one computer, although
about half were not connected to the Internet. The computers were generally the
exclusive domain of child brokers and their siblings. Juana commented, “the only
time my mother touches the computer is to clean it.” Most parents viewed the
128

computer, but not Internet access, as a necessary investment in their children’s
educations. In families that had Internet access, parents proved capable of limiting
their children’s Internet access, despite not knowing how to use the computer
themselves. Two parents had discontinued their Internet service because their
children were viewing material they considered inappropriate. Other parents
indicated that they punish their children with restrictions on using the computer if
they can see that they are online to talk with their friends rather than doing their
homework.
Access also depends on the location of the media artifact in the home.
Livingstone provides the example of a computer placed in a child’s bedroom playing
a different role in the domestic infrastructure than a computer placed in a communal
area where all family members have equal access to it (p. 138-9). Parents were
asked to specify the location of media artifacts that we could not see in the room
where the interview was taking place. In general, families were most likely to have
two televisions; one was usually in the living room, and another in the parents’
bedroom. The location of the television also often determined the primary language
of programs; the parents’ bedroom was universally a Spanish-only zone, whereas the
living room tended to be more devoted to the English-language content preferred by
child brokers and their siblings. In homes with one television, parents tended to
dictate content when they were viewing; children reported watching Spanish-
129

language television alongside their parents, and English-language content among
themselves.
The location of computers was evenly split between the living room and a
child’s bedroom. Most parents said that they preferred to have the computer in a
public place because they could more easily monitor their children’s online
behavior. In families where the computer was in a child’s (usually shared) bedroom,
children were more likely to claim the computer as ‘theirs.’ The actual setup of the
computer also gave clues as to its use. Some families had set up a mini-office for the
computer, and children’s responses indicated their understanding that the computer
was intended primarily for work rather than play. In other homes, the computer was
tucked into a corner of the living room or on the corner of a dining room table,
making it less conducive to serious attention to homework or accessing online
community information. In one home, the PC was set up on the floor in front of the
television. With five children in this one-bedroom apartment, the television was
always on. Clearly, this computer was unlikely to be a tool for concentrated
attention to schoolwork or searching for community resources.
The range of artifacts available in a particular domestic infrastructure
depends on more than the family pocketbook. Different communities have different
local media ecologies (Ball-Rokeach, Cheong, Wilkin & Matsaganis, 2004),
meaning that each community has access (or lacks access) to a different constellation
of local media options. The range of locally available media can affect the content
130

of a particular domestic infrastructure as much as families’ prioritizing ownership of
certain media artifacts over others. In Greater Crenshaw, families generally had
access to a range of English and Spanish television programming and radio. Few
families regularly purchased print media. School newsletters and other mailings
were often not linguistically accessible for parents, particularly in areas of Greater
Crenshaw that were predominantly African American.
Children’s brokering activities, such as translating mail, making phone calls,
translating news coverage and going online for their parents, has the potential to
enhance the family networks’ range of connections to media artifacts and to the
community storytelling resources they may contain. With regard to the Internet,
children’s activities can also connect parents to resources they would otherwise
never access at all—in this case, children are the sole conduits of parents’ access to
online community resources.
Brokering Activities & the Family’s Range of Storytelling Resources
The previous section indicated that child brokers’ access to community
information resources was constrained by media artifacts available in the home and
local media available in the community. Interviews indicated that children’s
brokering activities facilitated their families’ connections to available media
artifacts, and had potential to increase the range of local media storytelling to which
their families have access. Child brokers connected parents with a range of
131

community resources to which they would otherwise not have access because of
linguistic, cultural, or technological constraints.
Linguistic Brokering
Children’s most common brokering activities centered on translating and
interpreting verbal, written and mediated communications for their families. These
linguistic brokering activities facilitated communication with community
organizations and institutions, understanding the mail, and parents’ ability to connect
with their employers. Children’s linguistic brokering therefore facilitated parents’
connections to community resources, and to their economic resources as well.
A great deal of child brokering revolved around the telephone. Many parents
reported that they do not ever answer the phone themselves. When their children
were not home, most parents screened phone calls on the answering machine to
avoid the difficulties of conversing in English. Child brokers routinely made phone
calls to their parent’s employers, negotiating terms, salary, availability, and even
conflicts. Children were the primary links to community institutions and
organizations by making and returning phone calls to schools, doctor’s offices, and
to resolve discrepancies on utility bills. In some families, these brokering activities
followed a back-and-forth format; the parent stood close by and children used them
as a touchstone at each step of the phone call. In other cases, the parent and child
discussed the problem beforehand, and the child made the phone call independently.
Calls received from community organizations and institutions were usually in
132

English. Only one family specifically reported an institutional connection (a
doctor’s office) that had the capacity to consistently contact patients in Spanish.
Many families mentioned that WIC always calls to confirm appointments in Spanish.
Cellular phones were not the norm in this community, and if anyone in the
family had one, it was usually the father. For the most part, cellular phones served
the same purpose as in most families. Cell phones made it possible to check in with
children and make sure they were safely home from school, for example. For a
couple of families, however, cellular phones also served as a mobile translating
service. Parents reported calling their children from stores and appointments when
they could not find someone to translate for them. In these families, children’s
linguistic brokering had utility even when the children were not immediately present.
A number of parents indicated that bills were less challenging for them than
other mailed artifacts. Numbers, they told me, are a universal language, and they
have learned the format of the bills over time so that they can anticipate the format
and meaning of the different sections. Many parents still used their kids as a
touchstone to make sure that they have filled out the sections correctly. Some
children were formally in charge of paying the family bills; for example, Maria had
full access to her parents’ bank account and signing power on their checks. Some
child brokers read the bills aloud while their parents wrote checks. In Evelyn’s case,
her grandmother read the bills, but Evelyn wrote the checks because her grandmother
133

was unsure where all the details should be written
20
. Letters and other mail were
generally difficult for parents, and for their child brokers, to understand. My
interview with Julio afforded me the opportunity to observe a naturally occurring
brokering activity. Julio had just received a mailed letter from the landlord and
called his father’s cell phone to translate it to him. I watched him struggle with
many of the words, and his repeated attempts to clarify his meaning. He mentioned
to me in his interview that he had a difficult time reading in English, even though he
was born and raised in the United States.
Three primary constraints on children’s linguistic brokering emerged from
these interviews. Sometimes materials were beyond a broker’s developmental stage;
what Aurora referred to as “grown-up stuff.” A number of child brokers mentioned
that completing immigration, naturalization, and health insurance documents was
particularly stressful, both because the content was technically advanced and because
the consequences of making a mistake were so great.
In many cases, however, the constraint on children’s brokering capacities was
even more basic—they lacked the linguistic capacity to successfully complete the
task at hand. All but three of the child brokers I interviewed failed to understand at
least one question I asked, even though questions were worded simply. In some
interviews, we code-switched between Spanish and English, and it quickly became

20
Still other child brokers literally pay the bills with their jobs; Rolando worked 35 hours a week after
school as a busboy to pay the family utility bills. Others did not earn money directly, but went to
work with their parents to help them finish the work more quickly. Julio went to his father’s weekend
gardening jobs, and Regina and Evelyn went with their mothers to clean houses when they had school
vacations.
134

obvious that these children were fluent in neither language, even though they were
their parents’ primary source of English support. “Getting stuck” was an emergent
theme from the child broker interviews; four children used that term to describe
times when their language capacity was inadequate for the task. Maria commented,
“When I talk Spanish I get confused…I get stuck.” Others described similar
moments of anxiety. Juana said, “Sometimes it’s difficult [to translate], because you
get mixed up with words and stuff like that and someone says, ‘no, that’s not right.’
So then I try to tell [my mom] in other words until she understands.” Child brokers
reported considerable anxiety in situations where their parents were depending on
them and constraints, either developmental or linguistic, made their brokering efforts
unsuccessful.
Finally, child brokers were acutely aware of how their social positioning as
children constrained their abilities to act as brokers for their families. Invoking their
parents’ voices, particularly in an argument, was particularly difficult for children.
Graciela recalled having to resolve a dispute with one of her mother’s employers. “I
was arguing between adults and I was like, ‘I’m a kid here!’ [laughs] and it was kind
of awkward for me. But then, like I knew what to say, because my mom said he
can’t do anything [to hurt me] by (sic) the phone.” Regina recalled a similar phone
call with more negativity. Her mother had canceled cleaning an employer’s house
twice because she had been sick. In anger, the woman raised her voice at Regina,
and Regina responded with “don’t talk to me like that,” at which point the woman
135

collected herself. But, Regina said, crossing her arms defensively, the woman didn’t
apologize because “Americans don’t ever say they’re sorry.” Most children were
more likely to report humility in these situations, even if they had done nothing
wrong. When I asked Evelyn if her grandmother’s employers ever got upset with her
when she canceled cleaning appointments on her grandmother’s behalf, she said,
‘yeah, that has happened…but you know, I don’t want to talk back, you know, cuz
she can lose her job for that.” She recalled times when she had struggled to be
polite, even when she knew that the employer was out of line. Differential responses
to conflicts with mothers’ employers reflected children’s negotiations of cultural
appropriateness to facilitate a resolution favorable to their parents.
Cultural Brokering
Cultural brokering was distinguished from linguistic brokering in stories
children told about times when parents did not understand the significance of an
event, even when coverage was in Spanish. Children therefore brokered between
sets of cultural norms to explain the cultural significance of events to their parents, at
least to the best of their abilities. These cultural brokering activities were not one-
way experiences. Rather, parents and children reported cooperatively negotiating the
meanings of particular events to reach shared understandings. Parents reported that
they scaffold
21
their children’s understanding of a situation (Vygotsky, 1978) by
introducing them to more sophisticated Spanish vocabulary and necessary contextual

21
Scaffolding refers to activities where interactants support and supplement one another’s
understanding, much as a scaffolding literally supports a building under construction (Orellana,
Dorner & Pulido, 2006; Vygotsky, 1978)
136

information. Child brokers scaffolded their parents’ understanding of important
materials or events by drawing on their relatively more sophisticated understanding
of American language use and cultural norms.
For example, Felipe, Alicia’s father, was one of the few parents who
regularly watched the local news. Occasionally, Felipe would switch from the
Spanish-language station to the mainstream English broadcast on Channel 7, and
Alicia would explain what he could not understand. When I asked her if she
translated only the words themselves, or helped him understand the (cultural)
significance of the story as well, she replied, “he explains [the significance] to me.”
This example demonstrates that although Alicia had the capacity to scaffold her
father’s understanding, his limited English did not preclude his ability to scaffold her
12-year old understanding of the world around them. Their interaction is indicative
of the general patterns described by child brokers and parents in their interviews.
Children indicated that they brokered television content differently depending on the
language of the program. If their parents were watching television in English, their
children would broker both language and the cultural significance of the program.
Even when programming was in Spanish, many parents and children indicated that
cultural brokering is still necessary, since most parents, unlike Felipe in the previous
exemplar, often did not understand why a local news item was significant. Cultural
and/or linguistic brokering actively connected parents to local media storytelling,
regardless of the language in which that media production was communicated.
137

Scaffolding in cultural brokering also provided natural teaching opportunities
for parents to introduce their children to more advanced Spanish vocabulary. Most
child brokers welcomed these natural teaching opportunities, which they felt were
essential to their cultural educations (Durham, 2005). A number of children also
indicated positive feelings about these scaffolding sessions because they knew their
parents “felt really good about teaching us something,” as Graciela said. However,
not all parents’ scaffolding efforts were welcomed. Luis confirmed what his mother
told us in her interview; her attempts to improve his Spanish vocabulary by having
him translate his schoolwork to her were usually met with refusal or irritation. Luis
said that his mother’s attempts to teach him are “a waste of his time” that he thinks
he could be spending on homework or helping his mom take care of with his
brothers. Luis’ attitude was essentially a refusal to make optimal use of his mother’s
time and language proficiency to further his bilingual development. Families where
parents and children took advantage of these natural learning opportunities could
potentially enhance children’s brokering efforts. Improved mastery of Spanish
would have enabled children to help their parents more effectively, as a direct result
of their parents having invested in their Spanish language development.
Technological Brokering
In any home, children are likely to be the family authorities on new
communication technologies, and may broker connections to these media for their
parents (Banet-Weiser, 2004; Buckingham, 2000; Livingstone, 2002). In this case,
138

parents’ limited (or nonexistent) new media literacy was compounded with
linguistic, cultural, and other literacy constraints as well. Earlier discussion of
children brokering their parents’ connections to communication technologies like the
telephone, television, mail, and bills are testament to the fact that these children
brokered more technologies than most children do.
Child brokers’ technological savvyness depended on whether they had an
Internet connection at home (Cranmer, 2006). Children who only accessed the
Internet occasionally in public access points like libraries and schools knew little
about resources available online. I emphasize the Internet in this section because it is
the only medium parents did not access at all unless their children did it for them. In
this sense, child brokering facilitated both access to and the range of the family’s
Internet connection.
For child brokers who did have Internet access at home, there was a great
deal of variation in how much of their online time was committed to connecting their
families with local resources. Most of these child brokers reported that they often
went online when their parents asked them to look up directions. Fewer reported
going online to find stores that sold an item the family needed, or to check the hours
a local business was open. One mother reported that she and her daughter had
planned her quincea ňera
22
almost entirely online the year before. Another parent
had children look for vacant apartments to see if the family could afford to move.

22
A quincea ňera is a traditional ‘coming out’ party for a Mexican girl’s 15
th
birthday.
139

Evelyn regularly connected with Prensa Libre, one of Guatemala’s major
newspapers, so that her grandmother could read the local news from ‘home.’
Generally, interviews indicated that children’s online brokering activities connected
their families to a number of community resources, but the range of resources was
generally narrow. Even in the most connected families, children were unaware that
they were accessing limited community storytelling resources in their web searches.
These linguistic, cultural, and technological brokering activities clearly had
potential to strengthen family relationships, but this was contingent on the social
arrangement of the family around brokering practices that could constrain or enable
children’s efforts to connect with community resources.
Quality of Family Social Arrangements
The social arrangement of family members draws attention to the divisions of
responsibility that exist and persist within a family. The child or children that bear
primary responsibility for brokering in the home and community affect the
arrangement of the family. The social arrangement of family members can also
constrain or enable brokering activities in the home and children’s abilities to
connect with community storytelling resources. These interviews indicated that
retained parental authority and cooperative brokering between siblings are social
arrangements that enable primary child brokers’ efforts to connect their families with
their communities. On the other hand, family instability caused by events ranging
from illness to divorce can constrain child brokers’ efforts to connect with the
140

community, because energy spent compensating for these distracters saps time from
efforts to connect with storytelling resources. The primary issue in family social
arrangements is the normalization and consequent invisibility of child brokering.
The invisibility of what children do, even to themselves, can negatively affect child
brokers’ own development. This invisibility also gives rise to reward structures,
embedded in the functioning of domestic infrastructure, that result in children’s
prioritizing helping others at the expense of their own development.
The Invisibility of Brokering
Lievrouw & Livingstone (2005) indicate that the social arrangements in a
domestic infrastructure become “routine, established, institutionalized, and fixed to
various extents, and so become take for granted in everyday life” (p.4). Previous
research (Dorner, Orellana & Li-Grining. 2006; Orellana, Dorner & Pulido, 2003)
indicates that child brokers and their parents usually take their children’s brokering
activities for granted as a near-invisible aspect of everyday family functioning.
When we asked Marta if her daughter translated for other people in the
neighborhood, she looked surprised by the question and responded “Why? They
have their own children.” Her reaction reflected those of other parents when we
asked how they feel about their children brokering for them. Children’s brokering
was not visible to family members unless it was failing to provide needed aid (Star &
Bowker, 2002). The invisibility of children’s brokering contributions made it
difficult for child brokers to articulate a need for help with difficult tasks, and/or to
141

locate appropriate sources of support when needed. The invisibility of brokering
also resulted in little recognition—by parents or the children themselves—of how
attendant pressures affected child brokers’ development.
Buriel & DeMent’s (1998) interviews of second-generation college students
demonstrated the routinized nature of family social arrangements around brokering
activities. Their analysis indicated that children’s brokering become so ingrained in
family social arrangements that these practices persist even when they should be
obviated by parents’ acquisition of cultural and linguistic proficiencies. Their
interviews revealed that, “for some children, the role of cultural broker is a lifelong
responsibility that continues even after parents have lived in the United States for
many years and have learned English….[child brokers are] still required to make
doctor appointments, translate documents and negotiate business transactions for
parents” (p.190).
Some families’ social arrangements effectively tempered children’s brokering
responsibilities, while others exacerbated them. Two patterns of family social
arrangements emerged as enablers of child broker’s efforts to connect with the
community: retention of parental authority and cooperative brokering by siblings.
Parental Authority
Stories retold by children and parents in interviews and data from field
observations revealed that children frequently returned to their parents as a
touchstone and for assurance that they had brokered accurately, whether at home or
142

in the community. These instances highlighted that brokers are, in fact, young
children, still dependent on and desirous of parental authority. The continued and
clear authority of parents, even when children are invoking parents’ voices in adult
situations, belies the assumptions in the immigration literature that as children
become “experts” on American culture, they thoroughly undermine traditional
parent-child authority structures and parents lose control (Buriel & DeMent, 1998;
Menjívar, 2000; Ong, 2003; Zhou, 1997). Certainly, inversion of authority is a
possible scenario, but the ‘power’ of children and desire to undermine parental
authority was not evidenced in this research. Even when children reported brokering
at times they would have preferred not to, clear parental authority was evidently a
social arrangement desired by parents and children alike. Parental authority
facilitated parent-child scaffolding, increased children’s confidence in the accuracy
of their brokering, and facilitated parents and children working cooperatively to
solve familial problems.
Sibling Cooperation
Similar to the cooperative relationships observed between parents and
children in some families, some families indicated that siblings work cooperatively
on tasks that require brokering. Milagro and her older sister had a clear division of
labor that offset Milagro’s developmental and linguistic constraints by taking
advantage of her time after school to help at home:

143

The real important things [like health insurance forms] we let my sister do
because she’s more experienced…all the doctor’s appointments, prescriptions
and stuff, I do it…She’s always helped us economically, and I do the
translating and stuff at the house…[my parents and sister are] working so I
try to make it easier for everybody…It’s easier for me [to be at home] than
for my sister, ’cuz she’s always liked to work, and I’ve always liked being
around [home].

The synergistic division of labor allowed both sisters to contribute to the family’s
resources in accordance with their strengths. Pyke’s (2005) study of Korean and
Chinese college students demonstrated similar findings: that cooperative efforts
among siblings with different linguistic strengths and human capital can harness their
strengths for a common outcome, such as supporting the success of a family
business. Sibling cooperation on brokering tasks could potentially facilitate a wider
range of connections to community storytelling resources that can connect the family
to community institutions, and/or stronger efforts to make these connections based
on synergy between siblings.
A number of families indicated family arrangements that detracted from child
brokers’ efforts to connect with community storytelling resources for their families.
Negative social arrangements turned families inward, so that child brokers’ time was
split between holding things together and attempting to connect with the community.
Inhibiting Family Arrangements
Many children revealed (or hinted) at negative family social arrangements.
Two fathers had battled with alcoholism; Julio & Maria’s father used to drink
heavily, and Julio intimated that he was abusive when he had been drinking. Julio
144

said his father had reached a point where he would drink anything with alcohol in it,
including mouthwash. Alejandro broke down in tears during his interview
describing his alcohol addiction and the harm done to his family and his health over
the years. Both families described the strain on family relations and finances
resulting from these addictions.
Illness had a similar inhibiting effect on family social arrangements.
Alejandro has been out of work for over a year dealing with complications from his
diabetes. His younger daughter Milagro’s after school activities were circumscribed
by explicit family expectations that she would devote that time to caring for her
father: “Most of my time, I just give it to my dad.” Alejandro’s illness had also been
a source of strain for Milagro’s older sister, who was obliged to drop out of high
school and work to support the family.
23
Luis faced similar pressures; his 7-year old
brother’s epilepsy and consequent brain damage had profoundly changed family
functioning. Luis took on an enormous amount of responsibility, providing not only
brokering but emotional support for his mother. He frequently missed school or
sacrificed his own homework to take care of his family, constraining his own
chances of making and maintaining successful connections with his teachers and
peers.
Divorce had constraining effects on child brokering in some families and
enabling effects in others. An emergent finding in these interviews was the number

23
Happily, this was a temporary situation. She completed her GED, and now attends community
college while working to help support her family.
145

of parents who had informally adopted a spouse’s children and recreated an
integrated family. These adopted pairs had genuine emic experiences of the other as
family, referring to each other openly as ‘my daughter’ and ‘my dad.’ Child brokers
in these families only temporarily suffered the potential detracting effects of divorce.
In other families, unstable family arrangements following a divorce inhibited child
brokers’ efforts to connect with community storytelling resources. When single
parents were inevitably working long hours, the oldest child became a surrogate
parent to younger siblings, effectively limiting them to the confines of the home. In
one case, divorce had also been coupled with (or perhaps caused by) sexual abuse of
the oldest child. The trauma that Regina had suffered at the hands of her father was
inadvertently revealed in our interview. In this case, the financial fallout that
accompanied all the divorces in these families was coupled with emotional and
psychological constraints that affected Regina’s daily functioning, and consequently,
her ability connect her family to community resources.
Gang crimes had touched the lives of one family, and had a similar
psychological effect on Hernando’s brokering capacity. Carolina’s oldest son had
been in and out of jail and detention for years because of his gang and drug-related
crimes. She had not been able to work for extended periods because of his behavior.
Hernando, her younger son, was so determined not to break his mother’s heart that
he was afraid to leave the house, lest he be drawn into the same life as his brother.
146

His emotional turmoil made it impossible for him to help his mother navigate the
courts
24
and constrained his capacity to broker in community spaces generally.
Conflicts did not have to be chronic to potentially affect child brokers’
attempts to help their parents connect with the community. A number of child
brokers reported times when they have been frustrated or fought with their parents
while trying to help them. When I asked Luis if he ever felt frustrated or did not
want to broker, he answered yes before the words were fully out of my mouth.
Sometimes, he said, his dad made him read the same letter or bill over and over to
make sure Luis had translated it correctly. Luis was clearly irritated by this practice,
either because he found the repetition boring, or because he interpreted his father’s
insistence on repetition as a lack of confidence in Luis’ abilities.
When Milagro accompanied Alejandro on his errands in the neighborhood,
she told me that they often bickered with each other: “Yeah, we fight, ’cuz he gets
nervous. And like, I understand him, but sometimes we get mad at each other, and
I’m like, ‘well I’m not going to know every answer.’ And then he gets mad ’cuz he
needs to know. But it’s just in the moment,” meaning that the conflict was fleeting,
as in most cases where children reported occasional frustrations with brokering.
Children’s ability to ‘read’ their parents’ thoughts and needs was an emergent theme
in the stories children volunteered in their interviews. The children who brokered

24
When I asked if he had ever gone to court when his brother is in trouble to help his mother, he said
yes. But he said that he couldn’t walk into the courtroom when he got there. He told me, “I don’t like
going to those places, and when I go I stay outside with my dad”. His shoulders hunched, he started
shifting and talking in a low voice. When I asked him if it hurts to talk about it, he just nodded his
head and could not make eye contact with me.
147

most often were also those who had an almost adult understanding of their parents,
excused their embarrassment and frustrations, and were able to put their ‘work’
ahead of their emotions at times when their parents might have snapped at them
because they were nervous for an appointment.
Embedded Reward Structures for Brokering
Although previous studies of child brokering suggested that gender might be
the key factor in determining the primary child broker (Buriel & DeMent, 1998;
Orellana, Dorner & Pulido, 2003; A. Valenzuela Jr., 1999), these interviews
indicated that age and personality
25
were more influential factors. Most often, the
oldest child (or the oldest child still living in the home) was the primary broker,
regardless of gender. Pyke (2005) suggested that eldest children’s pride in their
brokering responsibilities may reinforce a reward structure that would dissipate if
they emphasized personal goals instead of prioritizing familial needs (p.502).
Aurora was one of many child brokers who commented, “Maybe I’m too nice. I like
to help a lot, but I don’t know; that’s just me.” For many children, male or female,
brokering for their mothers was embedded in a general willingness to contribute to
house chores. My interview with Alicia revealed a striking example of a child
broker self-identifying by her usefulness. My interview with her mother evidenced a
family reward structure that emotionally rewarded Alicia for helping—so much so

25
In the few cases where a younger child was the primary broker, parents’ explanation for the
aberration was that their oldest child was “lazy.”
148

that she even helped clean up the house when she’s ‘bored’ or ‘upset’ to make
herself her feel better:
A: Yesterday, I was like really bored and I felt, like, really sad because I
didn’t know what to do. So I asked my mom if she needed help with
anything cuz (sic) I was bored…[she cleaned the fridge, cleaned the other
fridge, cleaned up the house, dishes, vacuumed, polished the TV
table]…and after that I wasn’t bored anymore.
V: You felt better?
A: Yeah, I felt a lot better.

Child brokers were often so attuned to their parents’ emotional states that they often
preempted their parents’ requests for brokering assistance. For example, Juana’s
father spent a lot of time online
26
because he was working toward a degree in
computer maintenance, but often needed her help with the computer. Her prescience
in anticipating his need was clear in a story she told:
With him and the computer…[I think], okay, I know you’ll need my help, so
I’ll go be right there for him. He doesn’t usually have to call me. Because I
know that there’s [sic] going to be parts that don’t make sense. He’ll say,
like, ‘I was going to call you!’ And I’ll say, ‘well, you know I’m here.’

Many parents reported that they usually did not have to ask their children for
brokering help; their kids just seemed to “know” when they are needed.
That child brokers identified primarily by their usefulness to others is a cause
for concern, not celebration. Children made very real trade-offs to help their families
with brokering tasks, as educational requirements like homework and studying for
exams took second place to fulfilling family needs. Reese (2001) documented the

26
Juana’s father was one of two parents who was an exception to the no-Internet policy pervasive
among parents I interviewed.

149

ambivalence that many Mexican immigrant parents felt about raising their children
in the United States. On one hand, they hoped for their children’s academic and
professional success. On the other, they worried that the individualistic values of the
American education system would draw their children away from the family. The
parents in this study expressed the same ambivalence, and the selflessness their
children displayed, even at the expense of their educational attainment, reflected
children’s internalization of this ambivalence. Child brokers would routinely forgo
activities related to their schooling, because, as Hernando phrased it, “I need to be in
the house.” Their mastery of English suffered as a result of their limited connections
to schools. Ironically, this developmental constraint, a direct result of brokers’
family commitments, ultimately limited child broker’s capacities to help their
families in the ways that they tried to.
From Family to Peer Networks
Children’s brokering responsibilities often extended to their parents’ peers as
well. Children brokered for their parent’s friends and neighbors in many of the same
ways they did for their own parents, and brokering sometimes served as a resource
that parents used to pay back or incur favors from other residents, as detailed below.
Members of peer networks were also resources for some child brokers by
supplementing children’s knowledge and acting as models of successful brokering
from which children can learn new skills and strategies. Child brokers who had
150
these kinds of support in peer networks had a real advantage over those who lacked
access to support they sorely needed.
Figure 4.2 illustrates these relationships between the domestic infrastructure
and peer networks. The dual-sided arrow between the activities node and peer
networks indicates that children routinely brokered in peer networks, and that
individuals in peer networks had the potential to be resources that improved
children’s brokering skills. The dual sided arrows between the social arrangements
node and peer networks indicates that brokering responsibilities to individuals in
peer networks affected and was affected by brokering responsibilities in the home.

ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic

Figure 4.2. Links between the domestic infrastructure and peer networks through
child brokering.

Child Brokering as a Resource in Peer Networks
Many of the children I interviewed regularly brokered for extended family
and neighbors, ranging from sporadic help to consistent aid that involved large
amounts of time and effort. Children who had consistent brokering responsibilities
for neighbors and parents’ friends usually indicated that their parents were

Infrastructure
Parents and Brokers’
PEER NETWORKS
CHILD
BROKERING
ACTVITIES
(Range)
MEDIA
ARTIFACTS
(Access)
SOCIAL
ARRANGEMENTS
(Quality)
Domestic
Infrastructure

Parents and Brokers’
PEER NETWORKS
CHILD
BROKERING
151

approached to solicit their help, rather than the children being asked directly. This
was consistent with Menjívar’s (2000) observation that parents would “lend out”
their children to neighbors and friends as a way to repay or obligate favors from
other community members (p.214). In this community as well, children’s brokering
efforts could facilitate connections to the storytelling network and community
institutions for parent’s peers who did not have children, or whose children were still
too young to help them.
A number of child brokers helped cousins’ or neighbors’ children with their
homework on a daily basis. Hernando helped his 6-year old cousin with her
homework and translated notes and newsletters that she brought home from school
for his aunt. Teresa helped the son of a neighbor with his homework everyday, and
accompanied the mother to school meetings to translate. Aurora helped so many of
the neighborhood kids with their homework that one of the parents bought her a
whiteboard for her ‘lessons’ as a token of his gratitude.
Many children routinely facilitated access to media artifacts for neighbors,
aunts, uncles, parents’ friends, and church members, by translating their mail and
helping them to write checks to pay bills. If children found a mistake in the bill, they
were also responsible for calling the billing department to resolve the problem.
Aurora reported that she routinely accompanies neighbors to pay bills at the local
utility company offices or when they have appointments where they do not anticipate
finding Spanish-speaking staff.
152

Child brokers’ desire to, as Graciela said, ‘just help out,’ extended the
considerable commitments that many brokers already had to their own family into
their parents’ interpersonal networks. These commitments diverted brokers’
attention from their own schoolwork. Time Aurora spent in front of her whiteboard
was, by her own admission, time she could have been poring over her science
textbook in an effort to improve her lowest grade.
Child brokers also facilitated occasional conversations between their parents
and their African American neighbors. Juana reported that she often translated
pleasantries between her mother and African American parents outside of her middle
school. Liliana served as a bridge between her uncle, a landlord, and his African
American tenants. She routinely helped him collect rent and discuss maintenance
issues with his residents. Luis translated conversations between his mother and an
elderly African American neighbor who comes downstairs each afternoon to ‘check
up’ on the family.
27
While child brokering had the potential to facilitate real
connections between their parents and African American residents, which would
widen their parents’ peer networks to potentially include sources of bridging social
capital, this potential was seldom realized. Aside from Luis’ elderly neighbor, only
one other parent indicated interactions with African American neighbors that went
beyond the occasional smile and nod. These interviews indicated that children’s

27
Children provide technological brokering for neighbors as well. While I was interviewing Luis, the
same neighbor came downstairs to inform him that her cable was out, again. He immediately jumped
up and went to fix it, apologizing for his absence when he returned five minutes later.

153

brokering was primarily a resource for increasing bonding social capital in peer
networks by inspiring specific reciprocity between the parents of the broker and the
adult benefiting from their support.
When Carmen and her husband first married, they rented a one-room
apartment behind her husband’s sister’s house. Her sister-in-law’s children were a
great help to Carmen during those years. Because they were older, her nieces helped
Carmen care for her children and with all her brokering needs. Her nieces filled out
forms for WIC, Medi-Cal, and the schools, accompanied Carmen to the parent-
teacher meetings when Luz and Teresa started school, and helped the girls with their
homework. At the time of my interviews, Luz and Teresa were serving those same
functions for their neighbors; they helped Carmen’s friends with their mail and with
forms for insurance and Medi-Cal. Teresa also tutored Carmen’s friend’s elementary
school-age son and went with her to all school and teacher meetings.
Carmen’s story indicates that bonding social capital in peer networks has
clear survival value in peer and family networks. Her story also indicates that child
brokering can potentially invoke and return favors for parents (i.e. specific
reciprocity), and can encourage generalized reciprocity. Putnam (2000)
characterizes generalized reciprocity as “I’ll do this for you without expecting
anything specific back from you, in the confident expectation that someone else will
do something for me down the road…and I will do it for you even if you don’t
[return the favor]” (p. 21). Putnam argued that generalized reciprocity creates the
154

bridging social capital that is the “bread and butter of community development”
(p.22). Carmen’s nieces helped her without the expectation of personally recouping
their investment, and her daughters paid back her ‘debt’ with their brokering services
to others in the community. A number of child brokers also indicated that they
broker for strangers. For example, Gabriela told us that her son had made her stop
the car the week before so that he could jump out and translate for someone he could
see was struggling. These relatively invisible acts of community when people most
need it have potential, still largely untapped, for building social capital between
community residents.
Peer Networks as Resources for Child Brokering
Children’s friends were not generally brokering resources for each other.
When I asked children if their friends helped their parents in the same ways that they
do, some said they had no idea. Others said yes, but all indicated that they don’t talk
about their contributions to their families with their friends. The invisibility of
brokering in daily family functioning worked to deny most children access to
potential resources in their peer networks. If brokering activities were more visible,
children might have been able to discuss and pool their brokering experiences to
learn from each other’s mistakes, and enhance future brokering efforts. Milagro was
one of two child brokers who indicated having access to her friends’ brokering
experiences:

155

We talk about [brokering]; it just comes up in the subject [that we’re talking
about]. Like my friend Lola, she helps a lot in her house and stuff like that
and we just talk about what we do…we exchange ideas and stuff like, “Oh,
probably next time you’ll want to ask more about this or that” or something
like that. Even though it doesn’t always seem like it, we’re always learning
something [from] each other.

Because most child brokers I interviewed did not have support from their
schoolmates, extended family and family friends tended to fill the void. Juana said
that her mom was very close to a neighbor with a 9-year old daughter. The mothers
often pooled their resources by making appointments to go to the clinic at the same
time, so that Juana could translate for both of them, with the 9-year old helping
where she could. Juana told me that this 9-year-old neighbor as her main source of
brokering support. Ronaldo reported depending on an 11-year-old cousin for help
with translating and with his own homework. Juana and Rolando’s sources of
support were far from optimal, since younger peers are even more likely to face
developmental and linguistic constraints in their brokering than the brokers asking
them for help.
Evelyn and Aurora reported more favorable situations. Evelyn had an aunt
who moved to the United States as a teenager and completed her high school
education here. She was able to help with brokering tasks that were beyond Evelyn’s
capacities and also served as a resource for Evelyn to learn from her aunt’s
successful brokering efforts. Aurora had a 27-year old cousin who was her primary
source of brokering support:
156

Yeah, he helps me with…I call it ‘grown up stuff’ that I just don’t get, he
comes and helps them…for me, it’s car insurance and banking [that’s too
hard]. When my cousin comes, I’m usually there so I can see what he
explains and how…that way, if in another case the same thing comes, then at
least I can help my dad better than I did that time.

Evelyn and Aurora’s extended family members provided the support the girls needed
to help their families effectively even when the task went beyond their own
capacities. They also had the opportunity to improve their own skills by modeling
their behavior on these sources of support, to be able to handle similar situations on
their own in the future.
Most of the children I interviewed, however, had no clear source of support
for their brokering efforts. These kids depended on scaffolding with their parents or
looking words up in a dictionary. These limited resources made their efforts to
broker their parents’ connections to the storytelling network and to community
institutions that much more difficult.
Peer Networks that Bridge: Getting Ahead
Bonding social capital can be the impetus for successful connections with
community organizations. For Alejandro, the other Latino members of his church
social group were instrumental in his recovery from alcoholism. It was a church
couple’s retreat that convinced him to stop drinking and provided the support he
needed to be successful. Church members gave him nonjudgmental support and
connected him with a local Alcoholics Anonymous chapter where he drew support
from other Latino immigrants recovering from the same problem.
157

Community storytelling with other Latino residents also connected many
newly arrived immigrants to Women Infants and Children (WIC). Most mothers
indicated that they had connected with WIC as a direct consequence of community
storytelling with other immigrant women. Parents who feared making themselves
visible because of their immigration status gathered the necessary confidence to
connect with WIC through the recommendations of neighbors in the same position as
themselves. In this case, bonding social capital resources, shared through neighborly
storytelling, successfully converted into links with a community institution. One
connection begot another; a few mothers indicated WIC had been an effective
resource broker, connecting them with local health care institutions and child health
insurance programs.
Carmen, a mother who has been in Los Angeles since high school, was the
only parent who indicated that she has close friendships with her African American
neighbors. She told us that her neighbors had children the same age as her own, and
that their mutual interest in the schools led to their exchanging information about the
school and teachers. Having been in the US from an earlier age than most of the
parents, she understood a good deal of English, although she had difficulty speaking.
She said that she and her neighbors could usually manage to understand each other,
and Teresa or Luz brokered for her if their discussion exceeded her comprehension.
She was also close to other Latino families on her block, where she had lived for the
last decade. Storytelling over the backyard fence with a diverse set of neighbors is
158

an ideal way for residents to share bridging and bonding social capital resources that
can be used to mutual benefit, and can better connect parents and child brokers
themselves to the local schools.
Conclusion
This chapter documented the primary ways that child brokers attempt to
connect their families with community resources embedded in peer networks and the
local media. The home is conceived as a domestic infrastructure, where children’s
access to media artifacts, range of activities, and quality of family social
arrangements form an infrastructure that is recursively related to children’s brokering
efforts. Child brokering activities in peer networks extend children’s burdens of
brokering responsibility but potentially provide support for children’s efforts to
connect their families to community storytelling resources that can successfully be
converted into connections with local schools, health care, and social services. In the
following chapter, children’s efforts to connect with the local schools are considered
in context of the connection they attempt to make to health care and social service
institutions, which are the focus of chapter 6. These chapters illustrate the
problematic consequences of children identifying primarily with what they are to
other people, even to the detriment of their own development and access to
opportunities.
159

CHAPTER 5:
BROKERING AN EDUCATION:
CHILDREN’S EFFORTS TO CONNECT WITH
SCHOOL NETWORK RESOURCES
The previous chapters provide a sense of the storytelling network resources
that child brokers can access and try to convert into quality connections with
community institutions. This chapter focuses on child brokers’ efforts to connect
meaningfully with schools. Children’s connections to the schools—for themselves
and those they broker for their parents—were fundamentally different to connections
to other institutions, because successful or failed attempts to broker in the schools
directly affected outcomes for the children themselves. Chapters 5 and 6 explore the
continuities and distinctions between child brokering efforts in different community
institutions. Brokering responsibilities often constrained children’s abilities to
connect with schools’ institutional resources that were instrumental to their own
development.
Communication infrastructure theory conceives of institutions, including
local schools, as part of the communication action context of the community. On
one hand, access to institutions is constrained and/or enabled by storytelling network
resources that children can activate in these connection efforts. Chapter 4 explored
the resources that children accessed in their domestic infrastructure, peer networks,
and local media in their efforts to connect with local institutions. On the other hand,
160

the degree to which institutional connections are successfully made and maintained
can constrain and/or enable children’s brokering efforts in their family and peer
networks. Child brokers’ connections to schools can impact brokering activities in
more ways than connections to other institutions, because connections to schools
have direct consequences for brokers’ bilingual capacity development.
Community institutions, including schools, are not undifferentiated entities.
Rather, they are dynamic networks that can constrain and/or enable child brokers’
efforts to connect meaningfully with the resources embedded in them. Bourdieu’s
(1986) conception of capital as three interrelated types is useful for this purpose.
Embodied capital refers to the resources that teachers, administrators and other
school personnel have to share with and invest in students and parents. Objectified
capital refers to resources embedded in mediated artifacts, including curriculum
content, school newsletters, and application forms for charter schools, magnet
programs, and other opportunities for families to optimize children’s educational
choices. Institutional capital refers to resources embedded in the institution itself
that are distinct from embodied and objectified capital. In the case of schools,
institutional capital refers to resources including after-school programs and parent
organizations.
161

INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
Figure 5.1. The internal network of community institutions

My data indicated that the internal networks of Greater Crenshaw schools
were minimally integrated. An integrated school institutional network would be
advantageous for child brokers because connections to one node in the network
would have easily linked them and their parents to the other sources of capital
embedded in the institution. Interviews with parents, children, and service providers,
and field observations of the schools indicated that the minimal integration of
schools’ institutional networks constrained each nodes’ capacity to connect parents
and children with resources. Teachers were generally unaware of their students’
home lives and responsibilities that might have affected their attendance or
homework completion; sources of objectified capital were not effectively distributed;
and opportunities to connect directly with programs the school provided were
constrained by brokers’ responsibilities and safety concerns that drew them off
campus immediately after school and occasionally during the school day. The
minimal integration in schools’ institutional networks made it very difficult for child

162

brokers to connect themselves or their parents with school resources. RQ6 explored
child brokers’ interactions with these internal network nodes for themselves and for
their parents:
RQ6: How do community institutions’ internal networks of embodied,
institutional, and objectified capital affect child brokers’ efforts to make and
maintain high quality connections with them?
This chapter draws on interviews with parents, children, teachers and school
administrators, as well as field observations in two community schools. Convergent
findings from these sources revealed that child brokers were in a catch-22 situation
with regard to their schooling. Most brokers did not have the time to participate in
school programs (such as after school homework help) because of their family
responsibilities. Nonparticipation in these programs and absences related to
brokering constrained children’s access to these sources of institutional capital, as
well as meaningful connections with teachers as embodied capital. Child brokers’
family commitments served to limit their access to institutional and embodied capital
that would have encouraged their human capital development—which, ironically,
would have resulted in brokers being better able to help their families. Educational
institutions place value on individualism and personal achievement (Gibson, 1988);
child brokers’ tendencies to place the others’ needs ahead of their own generally
constrained their abilities to access forms of capital associated with schools. RQ7
163

addresses the differences between children’s brokering activities in the schools, and
in health care and social service institutions:
RQ7: How does the nature of child brokering differ depending on the
primary beneficiary of their brokering efforts?
My data revealed that child brokers’ disconnections from school network
resources had two primary causes. These failures to connect were most often the
direct result of brokering responsibilities, including absences from school, having to
go home directly after school (and therefore not being able to participate in
institutional resources like after school homework help), and unfinished homework.
The second reason was children’s difficulties in proactively resolving their
educational difficulties. Most child brokers navigated educational systems
independently, or with little active input from parents. For many brokers, the
invisibility of their brokering activities, to their teachers as well as their parents, was
exacerbated by their silence about difficulties they were having in school. Weisner
et al. (2001) reported that low- achieving Latino adolescents talked less and less with
their parents about school as they became increasingly disconnected from their
educations. High-achieving adolescents, on the other hand, kept their parents
updated and engaged with their school progress. Stanton-Salazar (2001) argued that
most Latino adolescents do not actively seek academic support from their teachers
and other sources of embodied capital when they need it due to fear of rejection and
similar factors. Child brokers in this study added support to these findings. The high
164

achievers knew where to find help and how to ask for it. The low achievers had
sunk, as Stanton-Salazar said, into a “quiet despair,” struggling through with few, if
any, direct sources of support for their educations.
The first part of this chapter focuses on child brokers’ primary
(dis)connections from each of the three nodes in the school’s institutional network.
The minimal integration of schools’ institutional networks were revealed in
interviews as a considerable constraint not only for children’s efforts to connect, but
for teachers’ capacity to serve as embodied capital and as gatekeepers to other
resources embedded in the institutional network. The second part of this chapter
reports parents’ connections to schools’ institutional, objectified, and embodied
capital. These connections were most commonly made through their children’s
brokering assistance. The independent connections that some parents made to
institutional capital by attending parent-teacher organization meetings had potential
to enable children’s connections to teachers as embodied capital.
Child Brokers’ Connections to Institution Network Nodes
Interviews with child brokers, teachers, and school administrators indicated
that access to teachers’ embodied capital was often constrained by children’s silence,
and teachers who did not ask questions to break that disconnect. Teachers were
generally the gatekeepers to the institutional network and its embedded resources.
The discussion of students’ (dis)connects from embodied capital are intertwined with
teachers’ general lack of awareness of children’s lives outside of the classrooms.
165

Student absences from class and from after-school programs may be excused but the
reasons were not fully understood by teachers and administrators. Most child
brokers did not have access to institutional capital because their responsibilities took
them off campus as soon as the school day was complete. Finally, the connections
between objectified and embodied capital took two important forms. First, children’s
limited English proficiency constrained their ability to connect with some parts of
the curriculum, but had little effect on others, particularly math. The subject matter
was important, but so was the way that the class was taught. Child brokers were
sensitive to teachers’ tactics that indicated their disinterest in the material or in the
students themselves. Second, child brokers who had access to embodied capital
were also the most likely to report successful connections with objectified capital,
such as applications for charter and magnet programs. Although the school networks
were not optimally integrated, connections between the nodes were still evidenced in
some cases.
Embodied Capital: Enabling and Constraining Factors
Teacher-Student Relationships: Aesthetic and Authentic Caring
Angela Valenzuela’s (1999) study of a Latino-majority high school in Texas
indicated that teacher-student disconnects were often related to a misalignment of
expectations. Teachers expected students to emphasize aesthetic caring about
schooling, meaning “ideas or practices that purportedly lead to achievement.” Latino
students, on the other hand, expected authentic caring from their teachers, meaning
166

an emphasis on “relations of reciprocity between teachers and students” (p.61). The
potential for misalignment of expectations to lead to student disconnect was
evidenced in these child brokers as well. Brokering responsibilities often meant that
children did not finish their homework, or came to class tired or distracted. Since
most teachers did not know how much of child brokers’ ‘free time’ was committed
to their families and peer networks, some teachers took these in-class behaviors as
evidence of students’ lack of aesthetic caring. These perceptions could affect
teachers’ decisions as gatekeepers to the institutional network. On the other hand,
some brokers indicated that teachers’ behaviors demonstrated a lack of authentic
caring that led to their own disconnect from class materials.
A review of the data made it obvious that some teachers’ limited
demonstrations of authentic caring emanated directly from environmental constraints
related the safety and racial/ethnic diversity features of communication action
context. Gang initiations and African-American/Latino fighting were symptomatic
of the widespread disciplinary problems at Bellum, where the (unofficial) school
motto was: “Our first job is to keep the kids safe; the second is to educate them.” I
heard some version of this motto in every teacher and administrator interview. Ms.
Hernandez told me that a student had come into class dripping blood from a knife
gash he incurred when he tried to break up a fight between two other students. She
said, “It’s not just coming to school and doing your work. You have to go outside.
Hang out with your friends. Deal with boys and girls. Worry about not getting
167

pushed around or jumped, you know.” Teachers were forced to devote substantial
energy to enforcing rules and meting out punishments, and therefore had less time
available to focus on the students who were not making trouble. Mr. South said that
when the parents of his “good kids” come to parent-teacher night, “What I really
want to tell them is the truth. That [your kid] is great, does all her work, never
makes problems. And because of that I’ve ignored her the entire year because I’m
too busy dealing with hooligans.” Teachers having to emphasize discipline and
order constrained their abilities to demonstrate authentic caring to the rest of their
students.
Safety concerns motivated many teachers to minimize interaction with
students beyond curriculum content to avoid being too close to the violence. Child
brokers frequently reported that teachers emphasized aesthetic caring through
“teaching to the test” and other practices that tended to disconnect all but the most
motivated students from the curriculum, and certainly from the teachers themselves.
For most children that I interviewed, high and low-achievers alike, their teachers
were simply presences in their lives—not usually overtly negative (although there
were those too), but not overtly positive either.
Most teachers were surprised or unaware of the family responsibilities,
including brokering, that many of their students assumed. Only one of the teachers I
interviewed seemed to make the connection that family responsibility might have
explained some students’ absences or incomplete homework. Luis told me that there
168

were “bad days” with his epileptic younger brother
28
that might involve late nights at
the emergency room, cleaning up a wrecked living room, or comforting his mother
and putting his youngest brother to bed. These were times when homework was his
last priority. He told me that after a “bad day” the week before, he went to his
teachers before school with his incomplete homework to ask for help. Two teachers
responded, “You were supposed to do this at home.” His third teacher asked instead,
“What happened that you were not able to finish your homework?” The teacher who
displayed authentic caring at the appropriate time received an understandable answer
and insight into Luis’ difficult family situation. His other teachers, emphasizing
aesthetic caring, were left to erroneously ascribe his incomplete homework to
laziness or disinterest, and these kinds of strained student-teacher relationships were
unlikely to yield quality real connections with embodied capital.
29
Lareau & Horvat
(1999) made the same observation, pointing out that individuals act as gatekeepers to
resources: “[Bourdieu] has not always been sufficiently aware of variation in the
ways in which institutional actors legitimate or rebuff efforts by individuals to
activate their resources” (p.38). Ms. Hernandez’s characterization of her own
response to certain student divulgences illustrated the rebuffs to which Lareau &

28
In the time we spent with this family, “bad days” were frequent occurrences even though Ana kept
telling us that he is much improved, compared to how he used to be.
29
In the time we spent with Luis and his family, it became clear that the invisibility of his brokering
contributions to his family functioning, coupled with his silence at school about the pressures he faced
at home, weighed heavily on this 11-year old’s educational attainment. That he managed to make the
honor roll during that period had everything to do with his natural academic ability, and little to do
with seeking help when he needed it.
169

Horvat alluded. She explained to me how she knew when something was going on
at home:
The only thing I see is…[that] their faces change sometimes [when
something is wrong], and sometimes they’ll tell me...[But] I don’t want to
hear if it’s like domestic violence I don’t want to get involved. I’ll just say
[motioning as if this is the point at which she walks away from the student],
“Okay. Hope everything gets better.”

Ms Hernandez’s response demonstrated that even when children divulged
information in active attempts to access embodied resources, a response that lacked
authentic caring would effectively close off access to much-needed resources. A
rebuff in an interaction like this could also discourage students from future attempts
to connect with their teachers and their embodied capital.
Generally, child brokers constrained their own access to teachers by not
vocalizing their needs and concerns. Only a couple of child brokers indicated that
they actively seek out help when they need it. Stanton-Salazar (2001) refers to these
Latino adolescents as the “silent majority”:
While the most obviously disengagement students ‘disrupt classes [or] skip
them…most disengaged students behave well in school and generally attend
class. Yet researchers suggest that beneath the surface, this silent majority
experiences a lack of psychological investment in learning, a lack of
intellectual mastery of the curriculum, and a nagging sense of disconnection,
if not alienation, from the core social fabric of the institution (p.13).

Generally, students’ silent disconnection from school network resources mirrored the
invisibility of their brokering contributions. Teachers had little idea of the potential
constraints that brokering responsibilities had on students’ classroom performance
and homework completion, and students did little to inform them. A few teachers
170

said that they had learned to ask students directly when they felt something was
going on. Even when teachers did ask, brokers could have many reasons for evading
a direct answer. A desire to protect their families’ privacy because of immigration
status, or because a full explanation would involve divulging sensitive family
information kept many child brokers quiet. Child brokers’ capacity to connect
meaningfully with their teachers’ embodied capital was therefore constrained when
teachers emphasized aesthetic caring, but could also be limited even when teachers
demonstrated authentic caring. In these cases, concerned teachers found themselves
in a catch-22 situation as well.
My field observations in the local schools revealed that authentic caring
varied with race/ethnicity. I observed that African American teachers and school
staff (the majority at Bellum) tended to display authentic caring with their African
American students, more so than with Latino students. Whether in the hallway, in
disciplinary meetings where parents were present, or in an award ceremony, African
American teachers talked with African American students in an almost familial way.
In a parent conference, Ms. Wagner chastised the African American student in
question, saying “Girl, can’t you see you breakin’ your Daddy’s heart?” My field
notes revealed repeated authentic caring displays between African American teachers
and students in the classrooms and hallways. Even when students were being
disciplined, there was a mutual understanding that this was authentic caring, that the
discipline was being meted out in their best interest. I do not think that these
171

teachers communicated differently with Latino students intentionally, but the lack of
common cultural (and sometimes linguistic) ground made African American teachers
more professional and less familial in their interactions with Latino students.
Convergent with this finding, child brokers indicated that they generally
received more aesthetic than authentic caring from their teachers, particularly
because Latino teachers and staff did not affect an equivalently familial tone with
Latino students. One teacher I observed told me that he was called “La Migra”
30
by
his Latino students, and time in his classroom gave me little doubt why. My
interview with him revealed his deep disaffection with the teaching profession, and
this was communicated to his students in his emphasis on “just teaching them what
the District says I need to.” He was an extreme case, certainly, but indicative of an
emergent pattern of teachers displaying aesthetic caring at the expense of authentic
caring. This imbalance ran the risk of disconnecting students from course content,
and discouraging students’ efforts to broker quality connections between themselves
and their teachers.
Teachers’ Capacity as Resource Brokers
While some teachers did manage to sustain high quality connections with
individual students, their ability to broker connections to resources beyond the
school itself was limited. High turnover of teachers in Greater Crenshaw constrained
children and parents’ abilities to connect with them. As in many low-income areas,

30
“La Migra” is a pejorative slang term used in Latino communities to refer to the US Border Patrol.
172

most teachers were relatively inexperienced and not familiar with the community or
its resources because they did not live in that area. The schools where I conducted
interviews had high teacher turnover; Mr. South indicated that in his six years at
Bellum, at least four teachers had left each year, usually within their first two years.
New teachers indicated that they viewed their positions at these schools as
temporary. As such, they had little incentive to invest the time necessary to become
effective community storytellers and resource brokers for students and their families.
Most teachers admitted that they did not know enough about the area to give parents
or children advice on where to go for help or services if they asked. Most teachers,
however, were fairly well versed on homework help and other campus-bound
opportunities.
Two of the teachers I interviewed, however, had not only grown up in the
community, but had attended the school themselves as students. Both indicated that
they took pride in their ability to be resource brokers for their students, and both had
many additional roles within the school that facilitated their access to a wider range
of objectified capital. The additional range of these two teachers was evident in Ms.
Wagner’s summary of her after-class commitments to the school:
The reason I get the resources is because I’m the students’ success team
coordinator, and I’m also Title 7 designee…I work with the bridge
coordinator, and I get the resources from him, and that’s the way I know, and
you know, from doing it so many years I know of the resources.

In general, however, most teachers reported feeling uninformed about resources that
could help their students. The constraints individual teachers faced in accessing
173

informational resources that can help their students and their families reflected
similar constraints at the institutional level. Ms. Gibson, the 7
th
grade dean of
students, volunteered a multiple level explanation:
I don’t think [the] administration knows the complete number of resources
that are available, and because they don’t know, we don’t get information.
You know, [administration] should be the conduit, the link, you know…but if
we don’t know what’s out there— Okay, the classroom teachers are one
level. If we don’t know what’s out there and the next level doesn’t know
what’s out there, then the teachers can’t send a kid so that we get the
assistance, and I just think the community resources aren’t letting schools
know how helpful they could be in helping kids.

Ms. Gibson’s comments highlighted the difficulties that teachers and school
administrators experienced because the school’s institutional network was minimally
integrated. Time-strapped teachers and administrators struggled to compensate for
these network constraints, and child brokers’ efforts to connect themselves and their
families to individual network nodes were often unsuccessful because of the minimal
integration of school network resources. My interviews and observations indicated,
as Stanton-Salazar & Spina (2003) found in their study of Latino immigrant students,
that quality connections with the schools and with teachers are “structured around
fortuity and individual access, not group access” (p.250). Although a number of
child brokers indicated access to school programs and resources, the sustainability of
their fortuitous, individual connections to an individual node in a minimally
integrated network are questionable at best.

174

Connections with Institutional Capital: Enabling and Constraining Factors
After-School Programs
Child brokers were generally absent from after-school opportunities to access
institutional capital, because they were needed at home. Juana told me, “I can’t stay
after school because I have to help my little sister with her homework, and like, I
need to help my mom if people are talking to her and she doesn’t understand.”
Chapter 4 detailed the myriad responsibilities that child brokers had to their families
and parents’ peer networks that generally took up their afternoons. Some brokers
added that that they were not allowed to stay after school for homework help or to
use the libraries, even on days when they did not have family obligations, because
their parents feared for their safety.
Only five of the 21 children I interviewed participated in after-school
homework programs, either regularly or occasionally. Hernando had started going to
an after school homework program twice a week the week that I interviewed him.
He told me that having access to a science, history and English teacher for that time
meant that he would “do a lot more homework and you can read and stuff”. These
programs also provided low-pressure opportunities for students to connect with
teachers, enabled by students’ demonstrating their commitment to their educations
by attending these programs.
Connections to the institutional capital in one after-school program could
connect students to other resources in the school network. One child broker’s
175

integration into her school’s institutional network was exceptional, and also enabled
by the full support of her mother. Graciela’s story illustrated that successful
connections to capital in one school network node had the potential to beget
connections to capital in the other nodes as well. Graciela was involved in an
accelerated after-school program that prepared middle school children for advanced
high school courses, and another program that paired college students with middle
school children for activities and homework help. She also worked as a student
translator in the Administration office, giving her access to a diversified social
network with numerous sources of embodied and institutional capital. It was no
coincidence that she had successfully navigated the necessary bureaucracy to gain
entry into a highly competitive magnet high school. Her mother told us that she and
Graciela have formulated a schedule that allows Graciela the time necessary to
participate in these programs and access these resources, and help her mother when
she gets home. Her mother told us that her daughter’s educational achievements
were more important to her than her own needs. They were the only family we
interviewed who had made child brokering visible in order to manage the possible
subtractive effects that brokering could have had on Graciela’s education.
Objectified Capital: Enabling and Constraining Factors
The school curriculum was child brokers’ daily opportunity to connect with
objectified capital embedded in their lessons. Language emerged as a key
determinant of access to curricular content, constraining access to certain subjects
176

and enabling access to others. Successful connections with other forms of objectified
capital potentially meant access to a wider range of educational options, such as
magnet and charter schools. Without a source of embodied capital to guide the
bureaucratic process, child brokers were often vulnerable to missing deadlines to
access these resources, and parents’ laissez faire approach left children to make
independent decisions when they may have needed adult guidance.
Curriculum (Dis)connection
The content of the curriculum—textbooks, exams, notebooks, and the like—
were potential sources of objectified capital for students. “Students—in either weak
or good schools—have not only the ability to learn intended academic knowledge,
such as algebra and Shakespeare, but also the potential to learn the symbolic lessons
that form social capital” (Tierney & Venegas, 2006, p. 1689). The “funds of
knowledge” embedded in family homes and practices (Moll et al., 1994; Rueda,
Monzó& Arzubiaga, 2003) and brought into the classroom by students affect their
ability to learn these symbolic lessons. For child brokers who are successful, by
conventional definitions of success—completed homework, class attendance, and
demonstrated mastery of the material—successful connections with objectified
capital can be then entrée to enhanced relationships with teachers, and can
potentially open doors to other opportunities, such as charter schools and magnet
programs.
177

For many of the brokers I interviewed, their limited language capacity was
the primary constraint on their ability to connect with objectified capital, evidenced
by the relatively consistent patterns in terms of the subjects that child brokers found
most challenging and most accessible. History was usually brokers’ worst subject.
Twenty of twenty-one brokers volunteered that math was their strongest subject, and
reports on science were mixed. High achieving child brokers were the exception,
rather than the norm. Most brokers told me that they had a “D” in at least one class,
and many were failing at least one subject.
History was the weakest subject for almost every broker I interviewed.
Ronaldo summarized the reason best: “History, it’s just sentences,” meaning that the
material is low context. Having been in the American school system for less than a
year, he was familiar with neither the content of American history, nor the language.
His teacher had them read and copy the textbook rather than actively engaging the
material, which further compounded his difficulty. “I didn’t learn nothing [in my
history class]…when the teacher was writing in the blackboard…yes, I was copying
all the sentences there but I didn’t learn.” In general, teachers’ aesthetic caring,
evident from the emphasis on rote learning, compounded by the exclusive emphasis
on language proficiency, made history curricula very difficult for brokers to access.
Ronaldo’s silent non-comprehension was emblematic of many brokers’ response to
inaccessible material. Their low grades and silence may have had consequences for
teachers’ impressions of their academic ability and attitude. A disconnect from
178

objectified capital could therefore result in a disconnect from embodied capital as
well.
Math, on the other hand, was highly accessible to child brokers for a number
of reasons that directly connected resources in the family network to expectations in
the classroom. For many child brokers who lacked full proficiency in English, math
provided an opportunity to display knowledge with minimal language constraints.
Regardless of parents’ level of formal education, most brokers indicated that their
parents are real resources for their math homework. Aside from numbers being
“another language,” as many of the parents explained, Mexican-origin parents had an
advantage when it came to math. Zu ňiga & Hamann (2006) summarized evidence
that Mexico’s elementary math programs are more advanced than programs in the
United States. This meant that a Mexican-origin parent with a 6
th
grade education
was not only able to help their children in math because there is no language barrier,
but that the utility of their 6
th
grade math education may extend into their children’s
8
th
or 9
th
grade math content in the United States (Zu ňiga & Hamann). A
considerable number of parents and children indicated that parents teach children
math concepts from their own school textbooks, which they brought with them when
they emigrated from Central America. When I expressed surprise at hearing this
again, Graciela shrugged and said, “My dad really loves math. He came here only
with what he could carry—but he brought his textbook.” Parents’ demonstrated
commitment to their children’s development in math enhanced children’s access to
179

the objectified capital in this curriculum, and also played into the reward structures
in family social arrangements that were discussed in chapter 4. Learning math was a
mutually positive experience in that parents felt relevant to their children’s
educations, and children felt that they were not only doing their homework, but
encouraging their parents’ confidence to participate actively in their educations.
Since mastery of math is generally the prerequisite to understanding science
content, the mixed reviews that child brokers gave their ability to access objectified
capital in science curriculum was intriguing. High-achieving brokers usually said
that science was one of their best subjects; as Graciela said, “math and science just
go together.” Most child brokers were struggling in science, even though math came
easily to them. The reason for this disconnect likely lay less in the subject itself than
in the economic constraints surrounding instruction. California’s chronic shortage of
science teachers and underfunded labs effectively removed the demonstrative
elements of science learning. As a result, child brokers with limited English literacy
could not translate their math skills into achievement in science. In many child
brokers’ schools, science too had been reduced to “just sentences.”
Educational Options as Objectified Capital
Interviews with child brokers and teachers revealed that access to information
about other schooling options were highly valued, but relatively inaccessible,
objective capital resources. Transferring kids to other schools required familiarity
and skill with navigating many forms of objectified social capital. First, notices from
180

the school itself had to be understood, followed by filling out necessary forms, and
taking tests (for magnet programs). Successfully completing the necessary
paperwork usually required a trip to the Los Angeles Unified School District main
office downtown, and familiarity with attendant bureaucracy. In the few instances
where families had successfully managed these conversions—and in cases where
they hadn’t—access to embodied capital was the key factor. For example, Liliana
used to go to a poorly rated local middle school, but found out that a new charter
school was opening in the area. Her close connection to her teacher became a
valuable source of embodied capital. Not only did this teacher inform her of the
opportunity, but played an active role in Liliana and her father’s successful
navigation of the necessary application forms and tests.
Even for proactive, highly motivated students like Aurora, navigating
objectified capital without the active guidance and support of parents, teachers, or
administrators was extremely difficult. She had hoped to apply to a local magnet
high school that an older friend attended. Her friend told her about the program, but
did not have the resources to inform her about the application deadlines, which she
missed. Aurora had an unusually strong network of older peers, but these resources
did not successfully convert into a connection with this magnet school because her
friends lacked crucial information about deadlines, and her teachers were unaware of
the program when she asked.

181

Individual vs. Collective Connections with Objectified Capital
Talking with Juana exemplified the difficult and sometimes
counterproductive choices children made when left to their own devices. She was
earning straight A’s, and had successfully completed all the forms and tests
necessary to be admitted to the magnet program within her middle school. The
problem with going into the magnet program, she told me, was that she would be
tracked to a different (and superior quality) high school than her friends. As we
talked about this issue, it became clear that this very talented girl was going to opt
out of an opportunity to accelerate her own learning so that she could stay with her
friends. Juana’s story also highlighted the problematic practice of allowing ‘good
kids’ to make their own school decisions, common among parents we interviewed.
Making independent choices with little or no parental guidance meant that Juana
would be able to selectively present her choice to her parents in order to minimize
their resistance to her plan to stay with her friends. If her parents were more
involved in the process and more aware of the opportunities she was turning down,
they would certainly make her reconsider her choice. Juana’s story indicated that
children’s prematurely independent decision-making could constrain their human
capital development. Likewise, parents’ limited presence or absence from decision-
making could constrain their ability to convert their own resources—in this case,
parental authority—into enhancing Juana’s educational trajectory.
182

Juana’s parents’ disconnection from the information contained in her school’s
objectified capital clearly had potentially deleterious effects on their daughter’s
education. Sometimes the consequences of parents’ failure to connect with objective
capital were not quite so dramatic, but still worked to reinforce parents’ feelings of
disconnect from the school’s institutional network. I attended Luis’ award night with
his parents, where he received a certificate and award in recognition of making the
school’s honor roll. His father left work early, and the whole family was in
attendance. There were extensive speeches in English, which the considerable
number of Latino families in the audience were not able to follow. The program,
invitation, and all other materials were only available in English. Even when Luis
was called up to the stage and honored, while his parents clapped enthusiastically, I
could see that Ana was removed from the proceedings. She leaned over to me and
said, “I am ashamed that even when I am proud of my son’s hard work, I cannot
understand anything. I sit here and the words go over me, and I smile. But I am
ashamed.” Luis’ parents’ presence and his father’s effort to rearrange his work
schedule all communicated their pride in his achievement. But his mother’s
comment to me also showed that their inability to fully connect with the school’s
objectified, institutional, and embodied sources of capital made their full
participation in Luis’ education very challenging.

183

Parents on Campus: Child Brokering in School Contexts
Parents indicated two primary reasons for their general lack of connection
with their children’s schools. The first, not surprisingly, was linguistic difficulties in
communicating with teachers, understanding school meetings, and connecting with
school newsletters. Children’s brokering activities to connect their parents to all
three nodes of the institutional network were often parents’ only real conduit for
school-related information. Children’s efforts were often constrained by the second
reason parents had for not connecting meaningfully with the schools; most of the
parents interviewed for this study had little formal education. Women in particular
indicated that they had been sent to work at an early age—sometimes even at eight or
nine years old, to help support their families. Similar to the parents in Valdés’
(1996) research, these parents volunteered feelings of vulnerability and helplessness
related to their children’s education. Many were acutely aware that their children
had already surpassed their own levels of education. Parents commented that they
had been more actively involved in their children’s early years of education, but no
longer felt able to help. Limited human capital stemming from limited education
was compounded for these parents by a general lack of confidence and assertiveness
when it came to their children’s education. These parents indicated that they were
often confused about school meetings and newsletters sent home, but few made
inquiries (either personally or through their children) to clarify, and therefore did not
take advantage of potential resources in these objectified forms of capital.
184

Inevitably, the more educated parent was the one deputized to attend school meetings
and functions, with the other parent indicating that they had nothing to say or
contribute in a school meeting.
Parents with more formal education were likely to take more active roles in
their children’s schoolwork and in making connections with the schools—a fact
observed by school teachers and administrators as well. For example, Gabriela, a
college-educated mother from Oaxaca who had been in the United States for nine
months, went to her daughter’s school weekly and visited her son’s teachers almost
daily, because she was concerned about his progress in an English-only classroom.
Gabriela told us that she felt her education partially compensated for her lack of
English proficiency, because she understood the importance teachers placed on
active parental involvement.
Organizational Membership as Access to Institutional Capital
Teachers confirmed that the parents who attended school-wide meetings or
monthly meetings for parents of bilingual students were different from the majority.
Almost all participants were women, and Ms. Hernandez observed that they were
usually more educated than most Latino parents. One of the group leaders was a
teacher in Mexico, “so I think she knows how important it is to be involved, and she
is not scared to come into a school.” Ms. Hernandez & Mrs. Rhodes observed that
regular attendees were also the most actively involved with their children, and theirs
tended to be the highest achieving students. Whether this impression was based on
185

children’s actual achievements, or teachers’ favorable impression of these students
because of their parents’ active school involvement, the consequences were similar.
These children were likely to experience improved access to embodied and
institutional sources of capital as a result of their parents’ involvement. None of the
parents I interviewed indicated that they regularly participated in school
organizations, consistent with the very low involvement with education-related
community organizations indicated in the survey data reported in chapter 3.
School Newsletters as Objectified Capital
School newsletters and other objectified capital that comes home in
children’s backpacks can potentially be the impetus for parents to step onto campus.
In some Greater Crenshaw neighborhoods, school newsletters were only printed in
English. In other areas, where the majority of students were Latino, schools sent
home newsletters in English and Spanish—at least, most of the time. Aurora made
an insightful observation about the newsletters from her school: “Information about
what’s going on in school [i.e. school events] is mostly in English. And then stuff
like, we’re getting out early this day, or there’s not going to be school this day, is in
Spanish [and English]. And report cards are in English too.” Aurora’s observation
highlighted a subtle but essential distinction. Spanish-speaking parents in her
neighborhood were receiving basic information about the schools, but were subtly
excluded from school events and opportunities for parental engagement. There was
an interesting discrepancy in reports by administrators and by parents in this regard.
186

Ms. Nelson insisted that all newsletters sent to parents are bilingual. Parents and
children indicated otherwise, and some parents from Bellum Middle School showed
me examples of letters that had only come in English. As Aurora had indicated, the
English-only newsletters generally contained more technical information, and details
about opportunities for which students could apply.
I do not mean to imply that schools intentionally constrained Spanish-
speaking families’ access to valuable resources. Few schools had fully bilingual
staff who are hired for the purpose of translating information and meetings. These
more sophisticated newsletters likely did not get translated because school staff
lacked full bilingual fluency, or simply did not have the time to translate these
media. Ms. Hernandez, the bilingual coordinator at Bellum, revealed multiple points
at which family access to objectified capital might be constrained:
MH: It’s really hard to break all that information down, and even if -– like
they send memos out, and the memos will flow through the school.
There is no telling if the kid will get that information to the parent; and
when they do get it, the parent will often call up and say, “Well, I don’t
understand it. What does this mean?” If the District can’t send the
information out directly, I don’t think.
VK: They’re the more connected ones, the parents calling to say they don’t
understand?
MH: Yeah. Most parents who read it probably think, “I don’t understand”
and don’t call…[If] they do call, they say, “Well, I got this letter, and I
don’t understand it,” and it might just be, “What am I supposed to do
with it?”

Ms. Hernandez’ observations indicated that parents who were already more
connected with the schools were most likely to call and clarify the meaning of the
187

newsletter, effectively turning that artifact into objectified capital when they were
able to access the resources it described. Parents who were less connected did not
make those clarifying phone calls, and were likely to become even more
disconnected from school network resources. Parents who already had access to
capital in school networks were favorably placed to accrue more capital—often
through the direct assistance of their children, who were making the clarifying phone
call for them.
Parents’ Access to Embodied Capital
Some schools’ institutional networks had scant capacity to accommodate
non-English-speaking parents. At Bellum Middle School, none of the staff teaching
bilingual classes spoke Spanish. Mrs. Rhodes completed her degree in Utah, with
minimal bilingual training, and admitted she had a difficult first year when she was
placed at Bellum. The only bilingual staff were in the front office and were by all
accounts overworked since translating was not part of their official job requirements.
Despite the school’s meager linguistic resources, Mrs. Rhodes and other teachers
managed to communicate with parents by pooling their own Spanish-language
resources, having children phone home to arrange meetings, and having students
translate their own or other students’ parent-teacher meetings. Mrs. Rhodes’
interaction with one student’s mother demonstrated that high quality connections
were possible even with linguistic difficulties, assuming the teacher was willing to
188

allow access to embodied capital, and the parent or student was willing to ask for
that assistance:
One mom…actually [used to] read her daughter’s textbook, and she is trying
to learn. And then at one of the parent meetings she asked me if she could
borrow the CD that goes with the book, like if there was an extra teacher’s
copy…At the last meeting she is like: ‘You know, I finished with her book.
You know, do you have the next level?’ And so I gave her another textbook,
and I like that that she feels comfortable enough [with me] to explain to me
that she wants to learn English, and she is trying; and, you know, can I help
her with that. I’m like, “That’s wonderful. I think that’s great.” And she’s
such a good example for her daughter. Her daughter is the hardest working
kid in the class. I mean the fact that she finds English so valuable, it really
rubs off on the girl. I appreciate that.

This incident demonstrated an appropriate blend of aesthetic and authentic caring as
a prerequisite to this mother’s successful attempt to access Mrs. Rhodes’ embodied
capital. Mrs. Rhodes reinforced school-mandated standards of academic
performance by assisting this mother, and together they demonstrated the importance
of these standards to the student. Her comment, “I like that she feels comfortable
enough [with me]” indicated authentic caring that likely encouraged the mother’s
request in the first place.
Younger, more assimilated, and/or more educated parents were more likely to
indicate this kind of alignment with teachers’ expectations of active parent
involvement. These parents usually reported that they connect regularly with their
children’s teachers. Carlos walked his 5-year old son into his classroom and greeted
his teacher every morning—and much to Liliana’s adolescent dismay, he routinely
did the same with her.
189

Most parents, however, did not regularly see or communicate with their
children’s teachers. There were many reasons that parents might not attend
meetings, and generally, teachers seemed accepting and aware of the time constraints
that work and other responsibilities placed on parents’ capacity to be on campus on a
regular basis (Ream, 2005; Reese, 2001). For most of the parents I interviewed, the
primary reason for not attending meetings was more straightforward. In keeping
with their own education experiences in Central America, most parents did not see
any reason to meet with teachers unless their child was in trouble. For many parents,
the added embarrassment and difficulty of having to talk through their children
meant that they did not go to meetings unless they were mandatory. For other
parents, commitments to babies and younger siblings effectively ended their
attendance of meetings. Julio told me that his mom used to come to parent meetings,
but was too busy with the baby to attend recently. When he was hauled into the
school office for fighting, which seemed like a regular occurrence (“I’m a
troublemaker,” he shrugs), his father was usually the one who responded to the
school’s calls.
Schools varied widely in terms of bilingual staff, translators on site, and other
accommodations available to facilitate parent-teacher meetings. Some parents
indicated that they meet privately with their children’s teachers (if the teacher is
bilingual) or with a school-sponsored translator. Sometimes parents took their
190

children with them “just in case” they needed them. In most cases, however, child
brokers were the conduits that facilitated meetings between parents and teachers.
Brokering Parent-Teacher Meetings
Of course, children were not disinterested brokers on the subject of their own
educations. Much is made of children’s potential to mislead or lie to their parents in
these meetings (Menjívar, 2000), but both child brokers and teachers indicated that
children translated faithfully, although they occasionally required a little
prompting.
31
Sànchez & Orellana’s (2006) conversation analyses of child brokering
in parent-teacher meetings indicated that children were more likely to downplay their
successes than their failures. American teachers tended to describe students’
accomplishments in glowing terms that were culturally inappropriate for children to
translate verbatim to their parents. Graciela indicated that translating criticisms
(such as talking too much in class) was easier than translating praise. If a teacher
was too effusive, she said, “sometimes [my mom] won’t believe me, or she’ll get
mad—you never know the reaction.” As Sànchez & Orellana suggested, a cultural
divide was evident when brokering praise to parents, and the child brokers I spoke
with sometimes downplayed their teachers’ words to align with parents’ expectations
of appropriate self-presentation. Juana, another high achiever, even couched her
explanation to me in self-minimizing language, so that telling me how well she was
doing at school would be appropriate. She said, “sometimes I get nervous, ’cuz I

31
Some teachers did indicate a preference for having a student’s sibling or another unrelated student
as the broker in a school meeting, just to ensure that the messages are not being filtered. Only one
teacher could give an example of an incident where a student had lied about their poor performance.
191

don’t know if [the teacher is going to say] something good, or bad, or if I’m gonna
get in trouble. Sometimes my teacher can tell I’m nervous and she’ll say, ‘don’t be
nervous, you’re doing fine, you’re like the best one in the class.”
Generally, child brokers indicated that they had more experience with
relaying teacher criticism than compliments. Teachers felt that most students didn’t
lie to their parents, but might avoid certain details that would get them into more
trouble. Ms. Hernandez said when she had students call their parents to say that they
had detention or that their parents had to come for a meeting,
They don’t lie, no…but they’re reticent to say something. So the parent
might prod, and so you might hear them add something; but they don’t -- I
haven’t heard them twist anything. “I have detention.” And [the parent
asks], “Why do you have detention?” And then you hear a silence. “Because
I didn’t do my work in class.”

Alicia, like many of the low-achieving brokers, indicated feeling regretful and
“ashamed” when she had to tell her mom that she hadn’t completed assignments.
When I asked her if she had ever wanted to make less of what she’s done wrong so
that her mother wouldn’t be angry, she looked at me wide-eyed and shook her head
vigorously. She said that even if she wanted to minimize her teacher’s criticism, her
mom would know from looking at the teacher’s face that she was lying. Consistent
with Sànchez & Orellana’s (2006) findings, brokers usually relayed teachers’
criticisms honestly to their parents, aware of their limited capacity to filter their
teachers’ meanings.
192

Most teachers reported that their interactions with Latino parents were
usually positive, despite the language barrier. Latino parents, like their children,
were more likely to internalize blame for their child’s poor performance than to
blame the teacher. Mrs. Rhodes said:
We sit around the table, because I don’t want them to feel like [I’m superior],
because I think Latino culture already does that. They already put teachers
up, which I appreciate, because the parents look at their kid and say, ‘What
are you doing?’ You know, they don’t look at the teacher and say: ‘Why are
you screwing up?”

None of the teachers felt that child brokering prevented effective communication
with parents. Mrs. Rhodes continued:
I have never really had an interaction where I felt like communication didn’t
happen, and the kids are usually very appropriate in that situation. They
know that they’ve got two people looking at them, and they have to do what’s
appropriate. I think the parents tend to know enough of what I am saying,
and I know enough of what they are saying.

Teachers and children generally indicated positive or neutral feelings about
children’s brokering in school meetings. Teachers appreciated brokers’ help, and felt
that they were honest conduits of meaning between themselves and their students’
parents. None of the teachers indicated that language barriers prevented them from
communicating with parents and the contributions of child brokers was generally the
factor that enabled their ability to fulfill this important part of their jobs. In this
sense, child brokers’ efforts to connect their parents with their teachers were
favorably received and had the potential to connect that child and their parents with
teachers’ embodied capital.
193

When child brokers were asked to describe incidents when they had felt
strong emotions (there were separate questions for embarrassment, fear, nervousness,
and pride), brokering in schools almost never evoked these strong feelings. There
were two primary reasons why school-related brokering would be less memorable.
First, schools were a more familiar environment for child brokers than other
institutions. They knew both adults for whom they are translating, and the
vocabulary was fairly familiar. The setting was a known entity, as were the social
scripts that ordered the meetings. Child brokers also had a good idea whether they
were likely to be receiving praise or castigation for their schoolwork. Children likely
experienced less ambiguity when brokering in school settings than they did when
brokering in other institutions. An emergent theme from interviews with teachers
and children was the awareness that brokering in schools was different to translating
in other settings. Mrs. Rhodes told me:
If [the kid is]really having a problem, I ask the parent to come in and then I’ll
have the kid sit there, and the kid will often have to translate, which I know is
difficult for the kid, but a familiar language for most of them. School
language. It’s not like taking them to the bank with, you know, [or] go[ing]
to a doctor’s office [to] translate in that situation.

Mrs. Rhodes was unique amongst the teachers I interviewed in her conscious
accommodation of child brokers. She told me that she had attended a conference at
UCLA about child brokers:

194

It is a pressure [on kids], and like I kind of try to be more sensitive. You
know, I…try to just discuss what they understand more comfortably, and I
feel sad sometimes. I know that one of my students was gone because they
had to take their parent to the doctor and translate, and I think -– and I would
think that that would be more awkward for the kid than the school situation.

The second reason that school brokering was less memorable and not tied to
strong emotions was that child brokers defined themselves primarily by what they
were to other people. When they translated at school, the primary potential
benefactors of their brokering success were themselves. Given how much these
children identified with their helping roles, and the family reward structures
associated with their assistance, it is not surprising that translating for other family
member was more likely to involve stronger emotions, including fear of failure. The
fact that no child recalled a school-related event when I asked for a time when he or
she had been proud of brokering was consistent with brokers’ generally collectivist
orientation. If these children had more individualist tendencies, they would be more
likely to feel proud of translating in settings where they were a primary beneficiary
of their efforts.
Conclusion
This chapter detailed child brokers’ own efforts to connect with schools’
sources of embodied, institutional, and objectified capital, the primary reasons that
enabled connections or caused disconnect from these individual network nodes, and
from the network as a whole. Children’s brokering activities connected their parents
to the schools, particularly to teachers as embodied capital, in parent-teacher
195

meetings. Parents’ active involvement in the schools also had potential to enable
children’s connections to the schools, by helping with (math) homework,
participating in school organizations, and engaging actively with school newsletters,
applications for other schooling options, and other forms of objectified capital.
Children’s brokering responsibilities at home, in peer networks, and in other
community institutions constrained their abilities to connect meaningfully with after-
school programs and other resources of institutional capital. Teachers were also
unlikely to recognize when family responsibility was the root of incomplete
homework, absences and visible fatigue, and children exacerbated these constraints
to their school connections by silently disconnecting from school, rather than
actively seeking and requesting help when they needed it. Child brokers’ general
practice of not actively seeking help and support meant the quality of their
connections to their schools’ institutional network was questionable. Their inability
to verbalize their needs for guidance was exacerbated by parents’ general practice of
allowing children to make independent educational choices. These decisions
betrayed the fact that child brokers were, in fact, still children, and did not always
have the necessary resources and experience to make optimal decisions on their own.
In the next chapter, we turn our focus to children’s brokering activities in health care
and social service institutions, where their families were usually the primary
beneficiary if children managed to make and/or maintain successful institutional
connections.
196

CHAPTER 6
BROKERING FOR FAMILY HEALTH AND WELLBEING:
CHILDREN’S EFFORTS TO CONNECT WITH HEALTH CARE
AND SOCIAL SERVICE INSTITUTIONS
Research on Latino populations generally indicates low levels of access to
health care and social service institutions. Massey et al.’s (2002) longitudinal study
of Mexican immigrants in the United States consistently found that Latinas in
particular underutilize government provided services that they qualify to access.
Mohanty et al.’s (2005) nationwide study found that immigrants received less than
half of the health care services that native-born Americans do. Latino immigrants’
annual health expenditures averaged $962 per person, which was half the average for
U.S.-born Latinos ($1,870) and less than one-third of U.S.-born whites ($3,117).
Analyses presented in chapter 3 reflected these national trends, indicating that
Spanish-speaking respondents, particularly high brokering respondents, reported
more difficulty in accessing health care resources than English-speaking
respondents. Given these disparities, child brokers’ efforts to connect their families
to health and social service institutions were potentially a step toward mitigating
these kinds of disparities. On the other hand, instances when child brokering was
less than successful were potentially part of the untold story of how these disparities
came to be in the first place.
197

In 2003, California lawmakers considered AB292, which would have
forbidden child brokering in health care settings and would have censured
institutions in violation of this law with loss of their state funding. The bill passed in
the California Assembly, but not in the Senate.
32
Many privately owned health care
institutions, however, took this proposition as an indication of a coming trend, and
officially banned child brokering in their facilities. The ‘illegality’ of child
brokering notwithstanding, every child I spoke with indicated that they regularly
broker for their families in doctors’ offices and clinics, and every health care
provider confirmed that children frequently came prepared to broker for their
parents, even when the office had bilingual staff.
Brokering in health care and social service institutions differed from
brokering in the schools in several important respects. First, child brokers did not
interact with health and social service institutions independently of their parents.
Every time child brokers connected with one of these institutions, whether in person,
over the phone, or translating written materials and forms, was at the direct request
of their parents. Even when children were the recipients of services or medical care
themselves, their connections with these institutions were oriented toward helping
their parents understand every step of the process, at least to the best of their
abilities. Second, because interactions with health care and social services were not
everyday events (in contrast to going to school), child brokers were less aware of the

32
Information retrievable on State of California website: http://www.legislature.ca.gov/cgi-bin/port-
postquery?bill_number=ab_292&sess=0304&house=B&author=yee
198

procedures and social scripts that attended interactions in these institutions. This was
the case even in families where chronic illnesses meant that brokers regularly
interacted with doctors. Third, the vocabulary that child brokers were expected to
translate, particularly in health care settings, was more likely to be out of the
brokers’ reach, and the consequences of a mistake were necessarily more dire.
Unfamiliar vocabulary, coupled with family members’ fear or distress, made these
interactions among the most stressful that child brokers had experienced. RQ7
explores these differences between brokering in familiar situations where child
brokers are the beneficiaries of their own efforts (i.e. schools) and less familiar
environments, where successful connections serve the family network as a collective
entity (i.e. health care and social services):
RQ7: How does the nature of child brokering differ depending on the
primary beneficiary of their brokering efforts?
The previous chapter’s framework also guides this chapter’s analysis of child
brokers’ successful and unsuccessful attempts to connect with health care and social
services. The internal networks of these institutions are again considered as
interrelated nodes of embodied, objectified and institutional capital (Bourdieu,
1986), and reproduced in Figure 6.1:
199

INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
INSTITUTIONAL
CAPITAL
EMBODIED
CAPITAL
OBJECTIFIED
CAPITAL
Figure 6.1. The internal network of community institutions

Embodied capital encompassed the resources that personnel in health care
and social services personally possessed and could share with child brokers and their
parents. Objectified capital included flyers on clinic bulletin boards and information
tables, health insurance paperwork, and the ubiquitous forms that determined if
families qualified to access an institution’s network of resources. Institutional
capital included resources embedded in the institution itself, including those
mandated by the institutions’ funding directives. Common examples of institutional
capital were informational sessions, emergency room services, and state-provided
benefits for legal residents. These analyses are guided by the same research question
as the previous chapter:
RQ6: How do community institutions’ internal networks of embodied,
institutional, and objectified capital affect child brokers’ efforts to make and
maintain high quality connections with them?
This chapter explores health care and social service institutions in tandem
because, in reality, they were difficult to separate. Health care and social services

200

were frequently resource brokers (Small, 2006) for each other. For example,
Alejandro’s doctor encouraged him to contact a local social service office to collect
disability benefits while he was unable to work, and WIC centers were mandated to
connect their recipients with health care services. The integration of one institution’s
internal network had direct consequences for its ability to act as a resource broker for
its participants. The last part of this chapter explores the connections between health
care and social service institutions as a consequence of the integration of each
institution’s internal network, and examines the constraining and enabling effects of
these institutional links on child brokers’ and their families’ access to community
resources.
Child Brokers’ Connections with Health Care and Social Service
Institutions’ Networks
Health care and social service institutions had the potential to act as
storytelling sites for residents in Greater Crenshaw. McRoberts (2003) found that
institutions’ deliberate functions are only one part of their potential contribution to
the lives of their recipients. Waiting rooms and institution-sponsored events and
presentations are natural opportunities for residents to connect with each other and
share resources through community storytelling. Observations at a WIC site
revealed two mothers discussing the merits of one local school versus another.
Although many of the observed interactions between parents in the waiting room
were racially/ethnically bound, there were examples of African American and Latino
201

parents exchanging pleasantries while they watched their children in the play area. If
the Latino parents have the capacity to communicate in English, either on their own
or through their broker children, these waiting rooms could be opportunities to
develop bridging social capital as well.
As in the schools, individual sources of embodied capital usually bore the
burden of being immigrants’ primary entry point into their institution’s internal
network. As Coutin (2001) and Mahler (1995) indicated, immigrant parents were
generally distrustful of institutions, particularly if they were undocumented. A
trusting relationship with a particular doctor or nurse could potentially encourage
ongoing connections with that institution’s internal network, and possibly with other
institutions if resource brokering was effective.
Child brokers accompanied their parents to health care and social service
institutions in Greater Crenshaw whether there were accommodations for Spanish-
speakers or not. Most children indicated having to perform linguistic brokering
frequently in facilities that did not have Spanish-speaking staff members, or where
the Spanish speakers were stretched thin. Sometimes children accompanied their
parents in order to fill out forms and help their parents understand instructions in the
waiting room, and a professional interpreter or bilingual staff member translated
during the appointment. Still other times, children went along as emotional support,
particularly to emergency rooms or stressful appointments at the request of their
parents, just in case they were needed. Numerous staff members indicated that
202

children sometimes brokered even when their services were not required, by
providing additional explanations for their parents. I observed this attentiveness and
readiness to broker in my field observations at a WIC site where accommodation to
Spanish-speakers meant that dependence on child brokering was low. Regardless of
actual need for child brokering in these institutions, interviews and field observations
demonstrated that embedded family reward structures and children’s strong
identification with their brokering roles made their attendance necessary even when
their brokering capacities were not.
Connecting with Embodied Capital
Parents and children who indicated regular contact with a service provider
were most likely to have access to that person’s embodied capital, and through that
connection, to resources in the other nodes of that institution’s internal network as
well. Domínguez & Watkins (2003) reported that consistent connections to service
providers were considerable sources of embodied capital for low-income single
mothers who had undifferentiated family and peer networks (p. 121). Wellman &
Gulia (1999) argued that relationships fortified by frequent contact are most likely to
be sources of support, whether those relationships are familial or professional.
Programs like WIC where families were obliged to make monthly contact were best
at creating these consistent connections. Likewise, families that had connected with
the same institutions for a long period of time indicated higher levels of comfort and
203

satisfaction with their doctors and other service providers than those who did not
have these long-standing connections.
Street (1991) found that long-term relationships between doctors and patients
become more personable over time. Service providers’ familiarity with a particular
family’s needs made it easier to offer relevant resources when they became available.
For example, when WIC staff members were aware that a family member had a
physical disability, they offered to make phone calls from the office to help connect
that family with special services in the community. Aside from volunteered
embodied capital, child brokers and parents sometimes felt more comfortable
directly requesting help or resources when a relationship was already established. In
many cases, however, brokers were unlikely to make direct requests for help,
resources or clarification, regardless of the length of time they had known the service
provider.
Parent interviews revealed a distinction between social service and health
care institutions in terms of what forms of embodied capital worked best. In social
services like WIC, similarity between recipient and provider had proved extremely
effective in getting people through the door, and to encouraging their retention in the
program. Mrs. Dorsey, a regional WIC administrator, told me that WIC centers hire
from their own participant pools:

204

Many of our employees are either former or current [WIC] participants…we
try to do very community-based hiring….We just post the job announcement
on the door of the clinic. And then the staff often—you know, staff get to
know the clients over time or they have friends, and I would say, you know
90 percent of WIC both in terms of participants getting through the door and
the hiring, is word of mouth…We are 90 percent peer-professional staff, and
these folks come, you know, from the street to the clinic. They tend to live
within a 3 to 5-mile radius of their work site.

WIC occupied a special place in the lives of many parents we interviewed. For
many, WIC was their first connection with community institutions, and generally,
they were encouraged to connect by people in their peer networks. For
undocumented immigrants in particular, this first connection was crucial to
connecting with other social services and health care institutions. WIC’s practice of
hiring within the community facilitated easy communication with ‘peer professional’
staff, who were low-income immigrant mothers like themselves. These interactions
did not require brokering and also inspired trust. Since trust is usually a serious
constraint to immigrants connecting with local institutions (Mahler, 1995; Small,
2006, p.287), similarity between provider and recipient was an effective strategy for
encouraging participation. Although brokering was not necessary at WIC centers
where the staff were generally Spanish-speaking immigrants, WIC nutritionist Mrs.
Fenstein noted that children often come with their parents and volunteer their
translating services, sometimes supplementing or rephrasing her basic Spanish for
their parents, even if it was clear that the parent understood her.
Generally, parents’ appreciation for similarity between provider and recipient
in a social service like WIC stood in contrast to their expectations of health care
205

providers (Katz, 2005). Most parents emphasized an appreciation for the
professionalism of their children’s doctors, and for those who saw health care
providers on a regular basis, of their own providers as well. Many compared the
medical care in this country to their country of origin. Gabriela came from Mexico
with her children 9 months ago to join her husband. Her son had been placed on
medication in Mexico, and the doctors here determined that the medication was
doing more harm then good. Gabriela said she was “amazed” by the number of tests
they ran, and the precision with which they were conducted. She said, “Back in
Mexico they were so careless with exams and lab work; the doctors are so much
better trained here.” Parents’ appreciation for medical professionalism seemed
enhanced by the fact that most doctors spoke no Spanish. Child brokers, therefore,
were instrumental in connecting their parents and health care professionals.
Children were not only brokering language in these interactions, but
managing cultural understandings as well—with varying levels of success. Aurora
said that sometimes her parents told her to translate something, and she thought,
“That’s not how you say it, but I don’t really tell them nothing. I just let them talk
and then I say it in my own words.” For example, Aurora said that her family had
had a problem with their health insurance recently, and their coverage was not being
recognized by the doctor’s office. Her parents were very upset and directed their
frustration directly at the office staff and doctors. Aurora altered their speech in
accordance with her understanding that the problem lay with the insurance company,
206

not the doctor’s office. Her approach resulted in the office manager taking
responsibility for clearing the problem personally. Aurora’s active involvement in
the transmission of the culturally appropriate response led to successfully accessing
the office manager’s embodied capital to achieve her parents’ desired goal.
Children’s efforts to broker for their parents in actual medical consultations
were usually less successful. Interviews with parents, children, and service
providers, as well as observations of actual encounters indicated that parents tended
to be reticent in their communication with doctors. Particularly for parents who
seldom see doctors, the unfamiliar environment and social scripts made them
unlikely to volunteer information, particularly sensitive details. Where a
professional interpreter may have had the meta-analytic capacity to tell doctors to
ask certain questions directly (Hseih, 2004), child brokers were generally too
embedded in their own cultural structures—and/or too aware of their social
positioning as children in those interactions—to be these kinds of strong advocates
for their parents. Parents’ embarrassment to admit sensitive information in front of
their children clearly contributed to their not volunteering details, and children were
clearly aware of their place, and constrained by their own maturity levels from
prodding their parents to reveal necessary details. Embarrassment was one of a few
key constraints to child brokering in health and social service institutions that
emerged from interviews and field observations. The primary constraints and
207

enablers of child brokers’ efforts to access embodied capital in these institutions are
detailed in the sections that follow.
Constraining Factor: Feelings of Embarrassment
Feelings of embarrassment were a constraint for some parents and child
brokers in accessing embodied capital. A few girls indicated that brokering for their
mothers at gynecological appointments were usually an embarrassing experience for
them both. Juana told me she goes to the gynecologist with her mom, and
occasionally with her neighbor. She rationalized her embarrassment about these
visits, telling me, “It’s okay if I know [about gynecological visits] ’cuz one day I’m
going to have to know it anyway, right?...and they’s not [embarrassed for me to hear
it] ’cuz they want me to learn.” By framing the embarrassing events as a learning
experience, Juana was able to control her own feelings to broker where she was
needed. Tulio was the only parent who told us that he wished he had a son to go
with him to the doctor, because he was uncomfortable to reveal personal information
in front of his daughter. Alejandro’s strategy was to take his 18-year old daughter
rather than Milagro, his main broker, when he anticipated having to share sensitive
information. He had a better understanding with the older daughter, he told us, and
felt that Milagro was too young to be privy to certain information about her father,
even though she was essentially his primary caregiver. Menjívar’s (2000)
informants revealed similar discomfort with having to communicate details of their
physical complaints through their children.
208

Constraining Factor: Brokers’ Own Illnesses or Pain
From child brokers’ perspectives, having to perform brokering tasks for their
parents in situations where they were patients themselves, were challenging
experiences. Children told me that they had felt obligated to calm their parents’ fears
and clarify their own treatments when they had to be taken to a doctor or emergency
room with an acute condition. Aurora said,
There was this one time that I was [so sick] I couldn’t talk; I couldn’t think.
And [my mom] was like, “No, help me talk to the doctor” and I was like,
“Just tell them I don’t feel good,” and my mom was getting more and more
nervous. But thankfully there was this other woman there whose daughter
was sick too and she could speak English so she translated for my mom. And
I was like, “Thank god, finally” because I was feeling so bad.

Graciela recounted an incident that occurred when she was 8 years old. She had
been bleeding (I did not press for more details) and her mom, Ileana, immediately
took her to the emergency room. Because Graciela was in pain, and hardly able to
speak, Ileana was left largely on her own to contend with her anxiety about her
daughter’s condition. Finally, after a couple of unsuccessful attempts to
communicate with the doctor directly, Ileana found a janitor who spoke some
English to help her, but the ordeal was memorable, and did not resolve until the early
hours of the morning. These stories demonstrated the precarious positions that
parents found themselves in when their children were incapacitated by their own
illness and discomfort. These examples reflect a larger number of children’s stories,
all of which indicated that parents’ connections to health care professionals as
sources of embodied capital were severely constrained when their child brokers were
209

the patients. Since these children often performed emotional support functions as
well, a number of brokers indicated that they had made efforts through their pain to
reassure their parents that it was not so bad, and that they would be fine.
Enabling Factor: Brokers’ Active Requests for Assistance
In more routine interactions when brokers were not suffering from acute
conditions, fevers, or pain, they enabled their own brokering efforts if they actively
sought assistance when they needed it. Similar to findings about child brokers’
behaviors in schools, those who indicated that they actively requested assistance
tended to be the high-achieving brokers. Graciela told me that when she was unsure
of a service provider’s meaning or how to fill out a form correctly, she explained her
confusion to the relevant employee and enlisted his or her help in resolving the issue
to her satisfaction. She giggled and said, “I don’t let them leave until I know I got it
right.” Aurora’s efforts to enlist active assistance were sometimes constrained by her
own inability to understand the sophisticated vocabulary associated with medical
care. She said,
Sometimes I can’t…I’m just like, “I don’t understand what you’re
saying”…I’ll go like, “Can you explain better or say it in other words,
describe it to me so I can better translate it to my parents.” And they try to,
but sometimes I just don’t get it…In a way I feel sad ’cuz I can’t help my
parents; I try to understand the doctors but I can’t and I feel like I have to
learn more and more so I can help my parents more. (her emphasis)

Aurora’s experience indicated that service providers obliged her direct requests for
assistance, but that her limited medical vocabulary constrained her capacity to
convert this assistance into her parents’ enhanced comprehension of the situation.
210

Her words also reinforced that children were not disinterested parties in these
interactions. She felt sad when she was not able to fulfill her parents’ need for help.
Her reaction was to make a redoubled commitment to her helping role by improving
her ability to connect effectively with embodied capital for her parents.
Despite the frequency with which Luis brokered in health care institutions, he
was unlikely to request clarification or direct assistance, and as a result, reported
feeling a great deal of pressure and anxiety when he brokered for his parents and his
brother’s doctors. Much of his unwillingness to request assistance likely was related
to his only being eleven years old, and also his having become accustomed to
struggling through brokering in emergencies so often with no assistance at all. He
told me that the times he had been most afraid to broker had all been in hospitals,
brokering between his parents and his brother’s doctors: “They use all these
terms…it’s just so hard for me. [They] like, load this stuff on me so I can translate
it.” His characterization of having brokering tasks “loaded onto him” demonstrated
the constraints that brokering without active assistance placed on those children who
did not ask for help.
Other child brokers compromised, not requesting direct assistance but
attempting to enhance their own abilities. Liliana and Ronaldo both told me that
they took a dictionary everywhere they went in order to supplement their ability to
broker conversations with service providers for their parents. Children connected
with health care and social service institutions in conjunction with their parents, who
211

could also enable children’s ability to broker for them, if they could work
cooperatively to enhance each other’s understanding.
Enabling Factor: Parent-Broker Scaffolding
Parents and child brokers’ ability to work as teams and ‘read’ each other was
an important enabler for connecting with embodied capital. Chapter 4 discussed
scaffolding between parents and children as a process of cooperatively reaching
shared understandings, and the embedded patterns of interaction developed in the
domestic infrastructure made it possible for parents and children to read each other in
institution-based interactions.
We observed Luis brokering for his mother, Ana, in a surgeon’s office.
33

The physician’s assistant was not accommodating to Luis, as she spoke quickly and
in complicated medical terminology. Luis struggled to find a pause in the
conversation to attempt a translation. He summarized the information, leaving out
many details. When he struggled with a particular term, such as the word
‘neurologist,’ Ana nodded her head to indicate that she understood that part, and he
was free to move on. My assistant’s field notes reveal the general pattern between
Luis and Ana:
Luis does this all the time. This was evident by how well he and his mom
worked together when it comes to translating. He knew when to start and stop
based on her facial expressions. If she nodded, then it meant that she
understood what was being asked. She responded and he translated her

33
The visit was a check up to ensure that his younger brother, who had epilepsy and had suffered
brain damage from his convulsions, was healthy enough to be placed under general anesthesia.

212

answer. When she didn’t understand something she asked him what was
meant, and he repeated or tried to rephrase.

Although this cooperation between Luis and Ana facilitated some connection with
the physician’s assistant, her non-accommodative manner made it very difficult for
Luis to translate. The limits of this interaction were evident when Ana tried to show
the physician’s assistant what medication her son was taking. We watched her
shaking the vials (which experience had taught her to bring with her to
appointments) and saying ‘night’ and ‘morning’ in English, but it still took numerous
demonstrations before a tentative understanding was reached.
Juana described a more accommodating relationship with her mother’s
doctor. She described to me how the three of them scaffold each other’s
understandings to develop shared meanings:
I’ll try to figure out how to say it [in Spanish] first and I’ll say it to [the
doctor] how I think it goes and sometimes she’ll be like, “It doesn’t go that
way.” And I’ll say, “Okay” and we’ll work on it until we know we
understand, and then I’ll tell my mom. I’ll make sure I understand what my
mom says back and translate it to English for [the doctor].

Juana’s ability to connect with the doctor’s embodied capital—in this case, the
doctor’s medical knowledge—is enabled by the doctor’s understanding and
accommodative attitude. Juana was the only child broker I interviewed who
indicated that she had a touchstone in each language to ensure that her translations
had fidelity.
Enabling or Constraining Factor: Service Provider and Staff Members’ Attitudes
213

The last two examples suggested that service provider and staff members’
attitudes toward child brokers could either enable or constrain the outcomes of
children’s brokering efforts. A number of brokers indicated that staff in health care
and social service facilities gave them positive reinforcement for their efforts by
acknowledging their contributions.
Juana said that she, her mother, and the clinic staff laughed together on
occasions when there was disagreement over the right word to use: “They say [to my
mom] ‘you’ve got a great daughter right here.’” Child brokers who had had health
care and social service providers acknowledge their contributions displayed more
confidence in connecting with those individuals than brokers whose contributions
were not verbally reinforced. The reward structures embedded in family social
arrangements that encouraged children to identify themselves by their helping roles
for their families were again reinforced, now by outside sources.
Staff attitudes could also constrain brokers and their parents’ abilities to
effectively connect with sources of embodied capital. Alejandro’s bad experiences
with impatient staff in the past meant that he never went to that particular office
without Milagro. He scheduled appointments for days that she did not have school
or late in the afternoon so that she could translate for him instead of the staff.
Milagro told me that none of the doctors there spoke Spanish, and the office policy
was to bring in the overextended receptionists to translate:
M: Yes, but sometimes [the receptionists] are not having a good day, and it’s
understandable, but…they’re not that polite. And I’ve noticed that a lot.
214

V: What do they do?
M: They’ll try to rush through things, and just try to get through as fast as
they can…and that’s the thing about my dad, he feels bad when he has to
go through that, you know. He prefers for me to go over there.

Milagro’s framing of the receptionists’ hurried attitudes reveals two observations
common to brokers’ understanding of past brokering experiences. Her demonstrated
understanding of her father’s negative experience reflected child brokers’ general
attentiveness to their parents’ needs and feelings. Her desire to be the solution to this
discomfort was also the general response of child brokers in such situations. The
second pattern was a little more unsettling. When brokers or their families were
treated poorly by an institution’s staff, child brokers tended to justify staff members’
behaviors, rather than fighting back. Orellana, Dorner & Pulido (2003) observed that
child brokers are aware of their social positioning as children operating in adult
situations, and that “these factors may affect how they and their families are viewed
and treated, as well as how entitled they felt to ask questions, make demands, or
speak on behalf of their families” (p.522).
Luis’ behavior in the surgeon’s waiting room provided a clear example of
this phenomenon. His mother told him to go to the counter and ask if their name had
been called while they were in the pharmacy, and to demand that their name not be
moved to the bottom of the pile as a result. Luis walked to the counter and asked
only if their names had been called, but did not request that the family not lose their
priority to be seen. Ana could obviously see his reticence, and marched him back up
to the counter to make the request as she had stated it. Luis was clearly
215

uncomfortable making a demand of adults. He made the request in a small voice
with downcast eyes.
Disaggregating institutions into different components was useful for
identifying institutional stratification that affected brokers and their parents’ access
to embodied capital. Generally, doctors did not speak Spanish, so other staff
members were deputized to act as translators. Nurses, if they were present, often
spoke Spanish but were spread thin. Pulling receptionists into consultations was
common practice, but parents were generally uncomfortable with this practice and
preferred their children’s brokering. Parents’ access to their primary care provider
was therefore usually constrained by language barriers, but accommodations that
some doctors made were instrumental to the success of children’s efforts to broker
the language gap.
Interviews revealed another stratification in connecting with institutional
networks. Although front office staff were usually Spanish-speakers, they were not a
resource for filling out forms and understanding instructions from the doctor.
Therefore, even in contexts where adequate language accommodations were
available for actual encounters with health care providers, child brokers usually came
with their parents to fill out forms and navigate doctor’s instructions, including
filling prescriptions.
Connecting with Objectified Capital
216

Child brokers’ connections with objectified capital were less successful than
their connections with sources of embodied capital, because the vocabulary and
technical knowledge required to complete these documents was often beyond their
grasp. Many child brokers reported that they were primarily or solely responsible for
filling out forms in health care and social service institutions. A number of parents
indicated that their limited literacy constrained their ability to fill out forms even
when institutions made them available in Spanish. These forms were sources of
objectified capital, in that they were used to determine access to benefits provided by
that institution, and potentially by other local institutions through resource brokering
and mandated referrals. Knowing that their mistakes could cost their families access
to considerable assistance made completing these forms anxiety-inducing
experiences for brokers. Liliana told me:
You know how sometimes you have to fill out papers? Oh my god, [my
mom] gets mad when I make mistakes with her forms. “You know what it
says,” she tells me, and she makes me check it [in the dictionary] three times
to make sure I got it right.

Maria’s uncle had been injured at the construction site where he worked, and she had
been his main source of brokering assistance for filling out forms at various doctors’
offices and social service centers where they determined if he was eligible for certain
disability benefits. She said that a lot of these forms were really difficult to
understand, and she was often nervous that she would make a mistake and cost him
his benefits because of her limited abilities in English.
217

Navigating health insurance benefits, and particularly prescriptions, were
very stressful experiences for many child brokers. Evelyn said:

To me when [my grandmother] needs me to order pills, that’s the worst ’cuz I
think, what if I say the wrong name and they give her something
else?…Yeah, even though I’ve done it for a while I still feel like, you know,
it’s the first time.

Alicia had a similarly fear-filled memory. She recalled being most afraid to broker
when her youngest sister was born and her father sent her to the pharmacy to pick up
a prescription on her own. She was worried that she would “mess up the words” and
get the wrong medication. Child brokers generally had difficulty with family
members’ prescriptions because these brokering activities required specialized
medical vocabulary, a working knowledge of health insurance systems and
compensation plans that were beyond their grasp. They also had to perform these
duties without their parents present to confirm their understanding. In addition, the
potentially devastating consequences of making a mistake—either by accepting the
wrong prescription or incurring considerable costs to their families by taking
something that was not covered by insurance—weighed very heavily on these child
brokers. Even those who usually asked for help and assistance from English
speakers were unsure of how to phrase their questions related to health insurance and
prescriptions, since these were such complex subjects.
218

Dr. Zimba, a pediatrician in Greater Crenshaw, said most parents she saw had
very little understanding of what their children’s state-provided health insurance
actually covered:

No, they don’t understand how the system works, and then I think a lot of
times…they feel they can’t ask for change. I don’t know if they know they
have a lot of choices…They call me and say, “Well, why is my one kid’s
[primary care doctor] changed,” or they’ll say, “Why am I not with you this
month?” I called the [insurance provider] and they said, “Oh they have forms
they have to fill out.” If they don’t fill it out in a timely fashion, they drop
them or change them, and this is news to me too.

Dr. Zimba expressed deep dissatisfaction about the constraints that health insurance
technicalities and loopholes placed on her ability to connect consistently and
meaningfully with her patients and their parents. Her comments indicated that not
only do her patients’ parents have trouble navigating their children’s insurance, but
she was unsure of the system herself, and therefore could not advise them or
champion their efforts to understand their benefits. Dr. Zimba also commented on
parents’ reticence to ask questions or advocate aggressively for change or choices.
In these contexts, child brokers were well out of their realms of both linguistic and
cultural understanding, and usually unable to actively assist their parents.
Connecting with Institutional Capital
Most of child brokers’ connections with institutional capital were a direct
result of successfully connecting with embodied capital. Generally, service
providers were gatekeepers that controlled access to the state and federally mandated
219

resources that constituted institutional capital in health and social service institutions.
Institutional capital at WIC included food coupons, pre-natal care and nutritional
counseling, which parents usually connected with on their own. In other institutions,
particularly hospitals, social security offices, and government offices, child brokers
were their parents’ primary means of connecting with institutional resources.
Alejandro was the one parent who indicated that his daughter routinely
brokered his connections to institutional capital in a social service institution. He
had trouble with his work visa a while ago, he said, and Milagro remembered having
to broker for him on multiple occasions at the then-Immigration and Naturalization
Services (INS) office. Alejandro had been very nervous, and had snapped at her for
not understanding every word. She told him, “Well, I don’t understand everything,
but I’m doing my best.” Two other brokers indicated that brokering immigration and
visa-related materials and visits was very difficult for them as well, partly because
their parents were markedly less sure of themselves in these particular institutional
contexts. When parents were more hesitant, child brokers were not able to scaffold
meaning with them according to their established routines, which constrained
children’s abilities to broker at the times when their assistance was most needed.
Accessing institutional capital in emergency rooms produced high levels of
anxiety for many child brokers; in some cases, even telling the stories induced a
noticeable change in demeanor and tone. The clear costs of making a mistake and
the anxiety of their family members were often compounded by a lack of available
220

bilingual staff members for brokers to request assistance from. Julio’s only
admission of vulnerability was to say that he was “nervous” when his uncle was hurt
on a construction job three years before, and had been rushed to the emergency room
with Julio and his father. He told me with a voice that shook with the memory, that,
“Those doctors, they speak another language, man. I don’t understand the words
they tell me and I couldn’t make them give my uncle a bed and what he needed.” It
was, by Julio’s recollection, a long time until a professional translator was found to
supplement Julio’s brokering capacity, which was among the most limited of the
brokers I interviewed.
Alejandro suffered from multiple complications with his diabetes, and his
daughter Milagro served as his primary caretaker, even injecting his insulin multiple
times a day. Her proudest brokering experience had been when Alejandro had been
hospitalized for a recent eye surgery, and she had been “in charge”:
My mom was there but she didn’t know nothing; she really needed someone
to figure out [everything]…I had all the information on my dad, and like
everything…paperwork, …little notes, appointments, his history, and um,
how he’d been hospitalized before. So I organized everything, and I’m the
only one there who knows everything….my mom really was [appreciative].
She even said, “If you hadn’t been there I don’t know what I woulda done.”

Milagro’s successful connection with the hospital’s internal network of resources by
brokering, keeping detailed notes, and employing her extensive knowledge of her
father’s conditions, won not only her mother’s gratitude and recognition, father’s
doctors’ as well. For Milagro, who was so invested in her role as her father’s
caretaker, this opportunity to showcase her accumulated knowledge certainly
221

reinforced the reward structures embedded in her family’s social arrangements that
have formed around her father’s illness and her brokering activities.

Institutions as Resource Brokers
Interviews with parents, child brokers, and service providers indicated that
health and social service institutions were generally and consistently connected with
other community institutions’ resources.
34
These inter-institutional connections
were great enablers for child brokers because a successful connection to one
institution or source of embodied capital could be parlayed into multiple institutional
connections with little effort on the brokers’ part.
According to Small (2006), institutions become resource brokers for three
reasons: by their funders’ mandate; because another program initiated a connection
of mutual benefit to both institutions’ recipients; or by the volition of individual
employees who felt that connecting recipients with additional resources was the right
thing to do.
Funders’ Mandates to Broker Resources
WIC was the only institution I connected with that had a formal mandate to
connect their recipients to other resources. WIC had to offer their recipients contact
information for a minimum of three local health care providers. Connections to
other resources were generally the prerogative of the staff. Although connecting

34
These institutions also demonstrated more integrated internal networks than the schools where I
conducted fieldwork.
222

families with free or reduced-cost health insurance for their children was not
technically part of the WIC mandate, these referrals had become so embedded into
the internal network of the institution that service providers treated child health
insurance as a mandate as well. Child brokers’ health insurance coverage meant that
they had frequent contact with doctors at regular checkups. For many parents, their
children’s doctors were their only consistent access to health care information and
resources, because parents generally indicated infrequent medical care for
themselves. When I asked Juana where her father goes to the doctor, she gestured at
the bottle of cough syrup on the kitchen table and said, “That’s his doctor right
there.” Some health care providers recognized that family contact with health care
institutions was generally limited to children’s care. Dr. Zimba told me that she kept
information on free and reduced-cost adult health care in her office, and made a point
of encouraging parents to take advantage of these services for their own care.
Connections Initiated by Other Institutions
Community organizations and institutions targeted these health care and
social service sites as places to connect with these low-income populations. During
our time in the WIC site, we observed organizations handing out Spanish and
English books to parents to encourage them to read with their children. Staff
members from child health insurance and subsidized pre-school programs were also
regular features in this WIC site. Mrs. Dorsey, a regional administrator for WIC,
said that WIC’s goal was to be a “gateway to community resources” for their
223

recipients. WIC fulfilled this objective in the lives of parents we interviewed as a
consequence of outreach to organizations such as Universal Preschool
35
, and
encouragement of appropriate organizations that approached them as well. Ms.
Fenstein, a nutritionist at the WIC site, said:
We also give out referrals for housing, food stamps, Medi-Cal. Anytime you
have a program who wants to put out flyers who [sic] target our population as
long as it’s a legit program, we’re happy to, you know, set it out on our
information take. We have referrals for doctors, for free immunization
programs. Any referral lead we can get we try to get and give to our
participants.

Referral books also constituted objectified capital for recipients by facilitating
service providers’ abilities to be resource brokers to other institutions. WIC
provided their employees with referral books with resources ranging from local
housing assistance to job training. Health insurance companies provided doctors
with referral lists to connect their patients with local specialists when they needed to.
Dr. Zimba had taken the initiative to visit specialists in the local area that she needed
to have strong referrals for, including a pediatric orthopedic surgeon and a pediatric
dentist. Her referral lists therefore became a vehicle for increasing her embodied
capital.
Dr. Zimba’s office also revealed numerous flyers and newsletters from
health-related organizations, taped to the walls or at the front desk that provided
parents with information for their children’s immunizations and health care in both
languages. The walls were also plastered with bilingual reminders that they were

35
Additional information is available at the Universal Preschool website: http://www.laup.net/
224

legally entitled to translators. Another local health institution, Union Clinic, had
fewer of these resources immediately available for their recipients. Union Clinic did
not have bulletin boards or information tables that would have facilitated child
brokers or their parents’ ability to browse through available resources and programs
without actively requesting assistance from staff members.
Resource Brokering by Embodied Capital
Resource brokering by individual service providers enhanced their capacities
to act as embodied capital for their recipients. Individual efforts to connect with
other institutions can be the most resilient form of resource brokering, because
changes in funding are less likely to affect individual providers’ practices when they
believe they are providing an important service to their recipients (Small, 2006). On
the other hand, because embodied capital bore the primary burden of connecting
brokers and their families to their institutions, these connections were vulnerable; if
that provider left the institution, their embodied capital would leave with them.
Mrs. Dorsey indicated that much of WIC’s resource brokering capacity
depended on the motivations of their peer-professional staff, who acted as strong
links between WIC’s institutional resources, and the community storytelling network
in which they participated as residents:
There is some training [about community resources] because part of the WIC
mandate is referral to health care...Everything else, which is really everything
else, from child abuse prevention stuff to food banks is not formally trained
on when people are hired…[At least half of] of our staff who have been with
WIC for, you know, ten, fifteen years…So those are the sort of folks who
often get pretty unsatisfied after a year or two with the same old work
225

everyday and will branch out and really take on finding out what other
resources are there in the community, and making sure families can find
them...It benefits them [too]. You know, they learn things about their own
community for themselves and their families and then they share that.

WIC’s successful integration into their local communities and role as a “gateway to
resources” was facilitated by the commitments and long tenure of their staff
members. The practice of hiring from within the community drew natural
opportunities for community storytelling into their sites, and meant that knowledge
of their resources would be disseminated organically through talk between
neighbors. That WIC never had to do more than minimal formal advertising was
testament to the success of their approach. When peer-professional staff made
connections with other institutions to serve their own families’ needs, they had the
opportunity to share these connections with brokers and their families who had
similar needs. WIC’s hiring practices made their institution an active part of the
local community, and a real enabler for the efforts of child brokers to connect their
families to local institutions.
Effective resource brokering connected families directly from one institution
to another. For child brokers, resource brokering by institutions effectively released
them of the responsibilities of connecting their families to each institution separately.
Resource brokering could therefore substitute for storytelling resources missing from
peer and family networks that would have constrained brokers’ efforts to connect
their families to each institution separately. Ms. Fenstein commented:
226

Being brand new to the country, I think a lot of people aren’t aware of what is
actually offered to help, and even speaking with friends and family they don’t
necessarily learn about everything until they come to a program that can give
them a few extra numbers [to make connections with specific institutions].

Resources gathered through successful connections with one or more institutions
could also be shared in peer and family networks through community storytelling,
enabling others’ efforts to make these connections as well.
Conclusion
This chapter explored the connections that child brokers make to health care
and social service institutions for their parents, and identified the major constraining
and enabling factors that child brokers face in making these connections. My data
indicated that most child brokers’ connections to these institutions, when successful,
are to particular sources of embodied capital, who act as gatekeepers to resources
embedded in their institution’s internal network, and potentially, to resource brokers
who can connect families to other local institutions as well.
The final chapter considers the theoretical implications of the data collected
and presented in this research, as well as the implications of this research for parents
and child brokers themselves. The chapter concludes with a discussion of this
study’s limitations, and suggested directions for future research.

227

CHAPTER 7
THEORETICAL CONSIDERATIONS AND RESEARCH IMPLICATIONS
This final chapter discusses this dissertation’s theoretical and methodological
contributions, practical implications of this research for child brokers and their
parents, as well as policy considerations. The chapter concludes with a discussion of
the project’s limitations, and suggestions for future research that emanated from this
project.
Theoretical Considerations and Contributions
This research was guided by communication infrastructure theory (Ball-
Rokeach, Kim & Matei, 2001; Kim & Ball-Rokeach, 2006a) which facilitated a
multi-level inquiry of child brokering set in larger community context. Child brokers
were an important component of their families’ integration into the storytelling
network. Children facilitated connections within the residents’ networks node of the
storytelling network, and to local media. Parents generally managed connections
with community organizations independent of their children. This project
investigated child brokers’ efforts to integrate their families into the storytelling
network by connecting with interpersonal and mediated resources that could
facilitate connections to important community institutions: schools, health care, and
social services. The current project contributed to the development of
communication infrastructure theory in three main ways:
228

(1) Utilizing social capital theories (Bourdieu, 1986; Putnam, 2000) to
disaggregate resident networks into family and peer networks, and
differentiating between types of resources that might be embedded in or
missing from those networks;
(2) Incorporating Bourdieu’s (1986) evaluation of resources into domestic
infrastructure theory (Lievrouw & Livingstone, 2005; Livingstone, 2002) to
explain how child brokering helps to incorporate their families into the
storytelling network; and
(3) Utilizing Bourdieu’s (1986) three types of capital to disaggregate community
institutions’ own internal networks allowed evaluation of the types of
resources embedded in those institutions. Conceiving of Bourdieu’s types of
capital as an internal institutional network facilitated a nuanced
understanding of child brokers’ efforts to convert resources embedded in
their family and peer networks into connections with community institutions.
Summary of Findings
Since child brokering is a micro-level activity that occurs in interpersonal
interactions, it was necessary to disaggregate the resident network node of the
storytelling network and differentiate between different interpersonal networks that
affect and are affected by the brokering activities of children. To this end, the family
network was separated from peer networks, which included neighbors, friends, and
extended family who lived in the local community. This research indicated that
229

child brokers were a key connection between their families and their parents’ peers,
performing many of the same tasks for parents’ peers as for their parents. In this
sense, child brokering encouraged specific reciprocity between their parents and
individuals in their parents’ peer networks.
Disaggregating peer networks also allowed a deeper investigation into the
types of resources that were present and missing in these relationships. Data revealed
that families generally had close relationships and stores of bonding social capital
that were useful for “getting by,” but constraining in terms of “getting ahead.”
Children were also key enablers of their parents’ efforts to connect with their African
American neighbors. These connections could potentially give brokers and their
families access to a wider range of storytelling resources by developing bridging
social capital. Communication between African American and Latino neighbors
were generally limited to pleasantries at the school gate, but even these fleeting
interactions could function as storytelling opportunities under the right conditions.
The storytelling resources embedded in, or missing from, peer networks worked to
enable or constrain child brokers’ efforts to connect their families to community
institutions.
Since most brokering occurs during daily activities within the home,
integrating domestic infrastructure theory (Lievrouw & Livingstone, 2005;
Livingstone, 2002) with communication infrastructure theory facilitated a deep
investigation of child brokers’ connections with a wide range of media artifacts and
230

local media for their parents. Parents’ access to community information resources
and local media were facilitated by their children’s brokering activities that increased
both the range and quality of their parents’ connections. Considering the home as
infrastructure contributed to understanding how family social arrangements
developed around child brokering activities. These embedded relationships made it
clear why these children identified so deeply with their helping roles, and also, why
child brokering practices were largely invisible to families, unless children were
failing to provide needed assistance (Star & Bowker, 2002).
Taken together, the integration of social capital and domestic infrastructure
theories contributed to the development of communication infrastructure by
extending the theoretical framework to support deeper investigations at the micro-
level of analysis. This analysis was the first extensive development of the micro-
level of communication infrastructure theory, and the demonstrated ability of this
combined framework to support multiple-method research, at multiple levels of
analysis, indicates that this was a fruitful extension of the existing theoretical
framework.
At the meso-macro level (Kim & Ball-Rokeach, 2006a), community
institutions have been theoretically understood as part of the communication action
context, capable of constraining or enabling residents’ communication in their
storytelling network. Past applications of communication infrastructure theory have
generally considered institutions as holistic entities. Bourdieu’s three types of
231

capital (embodied, institutional, and objectified) extended the current conception of
institutions in the communication infrastructure by demonstrating how institutions
can constrain or enable residents’ abilities to connect with their resources.
Bourdieu’s distinctions between types of capital also demonstrated that residents’
connections to institutions result from communicative processes. Service providers
were generally the gatekeepers to their own embodied capital and to resources
embedded in the rest of the institution’s network. Service providers could decide
who to invest resources in based on their interactions (Stanton-Salazar, 2001) with
child brokers and their parents. Service providers also communicate with other
employees in their institution to find out about resources for institutional and
objectified capital that can serve residents connected to those institutions.
Considering community institutions as networks where embodied capital
carries the major burden of connecting with child brokers and their families also
demonstrated that the integration of these internal institutional networks has
important outcomes. First, a minimally integrated network results in service
providers feeling less connected to and knowledgeable about the resources their
institution provides, which constrains their ability to accommodate brokers’ efforts to
connect with these resources. The local schools illustrated the constraints that a
minimally integrated internal network places on individuals working in that
institution, and the people who wish to connect with them. Second, when
institutions’ internal networks were integrated, these institutions were more likely to
232

be effective resource brokers, and could connect their recipients to other institutional
networks in the area. The previous chapter reported on the potential of health and
social service institutions to facilitate each other’s efforts to connect with potential
recipients. For child brokers, community institutions that acted as effective research
brokers eased the burden for them to broker each institutional connection separately.
Methodological Contributions
This exploration of child brokers’ attempts to connect their parents with
schools, health care, and social service institutions was best served with a multiple
method approach. The survey analyses presented in chapter 3 gave a sense of the
social placement of child brokers’ families in the larger community. High brokering
respondents were compared with no/low brokering and English-speaking Latino
respondents in terms of the composition of their family and peer networks,
connections with community organizations, local media, and community institutions,
and their integration into the storytelling network. These analyses were valuable
because they completed the community picture in a way that would have been
impossible with qualitative methods alone. For example, since most parents
connected with community organizations independently of their children’s help,
these connections would not have been considered in the qualitative inquiries in this
study.
The survey data also facilitated the identification of a subject pool for parent
and child broker interviews (Morse, 1991). This participant pool was expanded with
233

equivalent subjects identified at a Greater Crenshaw WIC site. As Jick (1979)
argued, quantitative methods can systematize qualitative research, and that
qualitative findings can contribute to the interpretation of statistical analyses by
helping to clarify puzzling findings. The data analyses presented in this dissertation
demonstrate the strengths of triangulation, since convergence of data collected
through different methods increase confidence that results reflect what is actually
happening, not that the method itself is responsible for observed outcomes.
Interviews with service providers and observations of their institutions, as
well as observations of child brokering in schools, health care, and social service
institutions completed the picture by providing the ‘other side’ of brokering
interactions. These interviews and observations also indicated that integrating
Bourdieu’s (1986) types of capital explained a great deal about the internal
conditions of institutions to which child brokers were trying to connect their families.
These interviews and observations identified key enablers and constraints that could
render even child brokers’ best attempts to connect to that institution, unsuccessful.
Despite some limitations discussed later in this chapter, this research
demonstrated that communication infrastructure theory, necessarily expanded by
social capital and domestic infrastructure theories, was capable of supporting and
explaining data collected through multiple ways of seeing this particular
phenomenon.

234

Contributions to the Literature
Research on child brokers (Buriel et al., 1998; Buriel & DeMent, 1988; Tse,
1995 & 1996), who are sometimes called translators (Song, 1999), immigrant child
mediators (Chu, 1999), interpreters (Valenzuela, 1999), or ‘para’-phrasers (Dorner,
Orellana, & Li-Grining, 2006; Orellana, 2002) is spread over a variety of disciplines
and lacks a set of common terms and theoretical orientations that would make it
easier for these studies to be cumulative (Orellana, Dorner & Pulido, 2003, p. 423).
Some researchers report that brokering causes stress for children (Parke & Buriel,
1995); Weisskirch & Alva’s (2002) study of 36 5
th
graders reported that girls are less
likely to experience stress and more likely to exhibit bilingual proficiency than boys
are. Song (1999) and Park (2002) indicate that child brokers in Asian families report
stress and a conflict between their individual desires and their family responsibilities.
Some of the research, however, emphasizes the positive effects of these
activities for children, particularly for Mexican-origin Latinos, versus Asian-origin
youth. The reasons for these differences, found in large (Chao, 2006) and small
(Tse, 1995) quantitative studies, are not clear. Tse (1995), Eksner & Orellana (2005)
and Orellana, Dorner and Pulido (2003) discuss the attachment that Mexican-origin
child brokers have to their identities as helpers, emphasizing children’s pride in and
desire to help their families. While my own findings also indicate this strong
identification with their helping roles, convergence of multiple methods also show
that this is not necessarily something to be celebrated.
235

Yet some researchers do indeed celebrate child brokering as a uniformly
positive experience for children. Valdés (2002) contended that brokering constitutes
a kind of “giftedness” because of metalinguistic competancies, and Malakoff &
Hakuta (1991) make similar, but not quite as strong, claims.
Two studies (Buriel et al., 1998; Dorner, Orellana & Li-Grining, 2006) use
regression analyses to demonstrate that child brokering is positively related to
educational attainment. Buriel et al. argued that the metalinguistic capacity of child
brokers is directly related to school achievement. Dorner, Orellana & Li-Grining
report that child brokering and standardized test scores are related. My own research
does not entirely contradict these findings, but certainly challenges the presumption
that (a) children are functionally bilingual, and (b) brokering is good for children’s
educational development.
Why the discrepancy? One possibility is a difference in methodology. Both
of these publications were quantitative in nature. While mine was a multiple method
study, there were no quantitative analyses conducted on child brokers themselves,
and I cannot claim generalizability from my interview sample. Beyond
methodology, the differences in our results may simply be due to differeneces in
operationalization and research focus. Dorner, Orellana & Li-Grining measured
school performance by standardized test scores. My own study focused on
children’s and parents’ sense of connection to schools’ internal networks, including
to teachers and classroom material.
236

Another possibility is the difference in the locations of these studies. Dorner,
Orellana & Li-Grining (2006) collected data at one elementary school in Chicago,
where most of the children were Mexican-origin and families had moved to this
more settled immigrant community after living elsewhere (p.12). By contrast, the
families in my study were Central American, but over a third were from Guatemala
and a number were from El Salvador and Nicaragua. The civil wars in these
countries often interrupted and prematurely ended parents’ education, which affects
not only children’s connections to school, but parents’ willingness to connect with
schools, as discussed in chapter 5. The Latino residents of Greater Crenshaw were
part of a more recent settlement, which also could impact their own and their
children’s connections to the community, and integration into the schools. It is also
likely that Latinos in California, given the political climate and attitude toward
Latino immigrants, experience a different local reality than Latinos in a Chicago
suburb.
Additionally, the ages of the children raises questions. Dorner, Orellana &
Li-Grining (2006) focused on 5
th
and 6
th
grade students; could it be that early gains
dissipate in later years of education? The question of developmental stages, as well
as the difficult school climate observed in middle and high schools in Greater
Crenshaw, and similar schools (Stanton-Salazar, 2001; A. Valenzuela, 1999) could
make a follow-up study with those students interesting, to see if the observed gains
diminish in later years.
237

Finally, my project was guided by a theoretical framework that facilitated an
investigation of child brokering in a larger community context, and at multiple levels
of analysis. The research presented here placed children’s connections to schools,
for themselves and for their parents, in context of brokering in other community
settings that impact, and as these results show, limit children’s abilities to connect
with resources in the schools. Therefore, Buriel et al.’s (1998) and Dorner, Orellana
& Li-Grining’s (2006) findings are not contradicted by my own. Rather, my
research raises the question of what brokers’ reported gains can be measured against.
Buriel et al. and Dorner, Orellana & Li-Grining found that child brokers experience
an increase in scores relative to their non-brokering classmates. This may well be
the case in my sample as well. The question, however, is: Are the gains child
brokers are making as large as they can be? My research suggests that there is still a
great deal to be done in this respect.
Practical Implications: The Immigrant Bargain and its Relationship to
Parents and Children’s Aspirations
The theoretical framework and multiple methods employed here provide a
great deal of insight into the lives and relationships of the child brokers and parents
who participated in this project. Smith (2002) describes an “immigrant bargain” that
forms between parents and children, “whereby the parents’ sacrifice in leaving home
is redeemed by the success of their children in the United States” (p. 151). Most of
the parents we interviewed said that they felt the United States was a better place to
238

raise children than their home countries, and most cited the opportunities available
here as the main reason. Gabriela said, “Your children can be anything here,
anything they want, and this is why it is worth our sacrifice for them to be here.”
And yet, parents’ hopes for their children were somewhat conflicted. Reese’s (2001)
research with Mexican-origin immigrants revealed that parents did want their
children to ‘succeed’ in the United States, but not at the expense of their connections
to the family as a collective unit. Many parents had felt difficulties with the schools’
emphasis on individual achievement and attainment, particularly because these
achievements had to be made with little parental input. Most parents in my study felt
disconnected from their children’s educations; even the most educated among them
were not able to participate fully in their children’s homework, studying for exams,
and other school-related events designed to encourage their children’s performance.
Because they were unable to interact meaningfully with the objectified
capital in their children’s school curriculum (except for math), many parents
emphasized that they were sources of “moral capital” for their children (Reese,
2001). Parents indicated that they were their children’s teachers of morality and
appropriate behaviors, and that they took these roles very seriously. A number of
parents criticized their neighbors who had raised kids who “fell off the path,” as
Paula said, and saw these wayward children as a sign that their parents lacked moral
fortitude. Virginia acknowledged the difficulties of being a single mother, but said
that they had more here than they would have had in Mexico. She, like many of the
239

parents we spoke with, felt that if she had open and loving relationships with their
children, she could keep them on the “right path.”
Reese (2001) found that parents considered instilling children with a sense of
being a “good person” with respect for their elders and for maintaining close family
ties (p.465) required “cultural maintenance and values that some parents see as being
at odds with American values” (p.460). Many of the parents we spoke with feared
that losing control over their children would lead to family disintegration. Family
social arrangements that gave rise to reward structures for children’s active
contributions to their families were consistent with parents’ desire for close family
relationships and interdependence. Brokering activities also gave parents
opportunities to scaffold their children’s understanding of Spanish vocabulary and
sophisticated social situations. These were also natural opportunities to provide their
children with moral guidance. For the most part, child brokers recognized and
appreciated their parents’ roles as their moral teachers. Graciela said, “My dad will
give me the money to go here and do this and that, and tells me to make the right
choice. My mom makes the right choice” (her emphasis).
When asked what hopes they had for their children’s futures, most parents
framed their responses in terms of their children’s moral characters—that they hope
that their children will be good people, and good to their families. Only a few made
explicit references to academic achievement and college degrees. Most said that
they hoped their children would find jobs where they did not have to work as hard as
240

their parents did. Children’s answers mirrored their parents’ hopes. Most said that
they hope to get a good job that allows them to, as Julio said, “do something in the
life you know, be something…be a help to my parents, if I could.” Even a number
of children who originally indicated a desire for a profession were clearly tempted by
the idea of being able to help their families immediately after school rather than the
delayed gratification of a college degree. Rolando said that he would like to go to
college, but knew nothing of the process, and admitted that not working full time to
help his parents seemed “selfish, ’cuz they really need me, you know.”
There was a distinction between the high and low-achieving child brokers
when it came to their future plans. Only Luis, the youngest of the high achievers at
eleven years old, did not have a clear future plan. Most of these brokers indicated a
desire to go to UCLA or USC so that they could balance their personal aspirations
with their commitments to their families. All of these children, when I asked if they
had any questions for me at the end of the interview, were full of questions about
what college was like. Most were concerned that university would be too difficult
for them, and only one was aware that it was possible to receive merit scholarships.
Generally, it was clear that even the most motivated of these child brokers would
face an uphill battle to get the necessary support and information they needed to
successfully apply for university. During one of our visits, Liliana told me how
nervous she was about taking the SAT the next month. I asked her if she had a book
or any study materials. She looked at me wide-eyed and said, “They have books
241

where you can practice and stuff?” The limited resources that these children had in
their family and peer networks, and their constrained capacity to connect with
schools’ institutional resources, put them at a marked disadvantage relative to the
average college-aspiring high school student.
In general, child brokers were acutely aware of the immigrant bargain. They
had a clear understanding of the hardships their parents experienced in America,
particularly those brokers whose parents had given up professions and skilled trades
to immigrate and were now working as housekeepers, gardeners, and manual
laborers. Julio and Maria’s father could not afford the classes he would need to
obtain his certification to work as a mechanic in the United States. He worked at a
car dealership washing cars just to be around them, because he missed his profession
as a mechanic so much. Children’s brokering activities developed into their way of
reciprocating their parents’ hard work and sacrifices. Graciela said, “If she needs
help we do it—she’s our mom. She’s always been there for us and she just needs a
little bit in return.”
When this “little bit in return” meant sacrifices of their own, particularly in
relation to their schoolwork, these were often made without parents even being
aware of their child’s choice to place their family’s needs above their own. The
child brokers who were excelling in school were generally doing so in spite of their
brokering activities, not because of them. These were the children who were able to
make up for deficits caused by brokering commitments—such as a missed
242

homework assignment or a low test score—by working harder on the next one.
Child brokers who lacked these natural academic abilities were not able to recover so
easily, as evidenced by the number of them who were failing, or in danger of failing,
one or more classes.
That children’s brokering responsibilities constrained their abilities to
connect meaningfully with the schools was both ironic and counterproductive,
because children’s development of their language and academic skills would have
enhanced the quality of their brokering efforts. And yet, the data presented here
show that children taking up their end of the immigrant bargain with their brokering
responsibilities to their families and their parents’ peer networks had a subtractive
effect on their school connections. The collectivist-individualist tension that children
have to navigate is the direct result of two different definitions of success: schools
emphasized individual achievement, while their parents’ definitions of success
included individual attainment only to the extent that these achievements did not
jeopardize children’s connections to the family network.
Children’s willingness to act in accordance with their parents’ definitions of
success also reflected families’ general lack of awareness of education-related
resources that were available to children. Brokers’ generally limited connections to
schools’ internal institutional networks, and the lack of role models in the
community to guide their choices, meant that many of these brokers equated
scholastic achievement and college attendance with loneliness. Even the high-
243

achieving brokers intimated that the roads to higher education would be ones they
were forced to walk alone. It is not surprising that many of them, even those with
considerable academic promise, were tempted by the warmth and familiarity of the
family setting and settling for adequate employment after high school. For all intents
and purposes, educational resources and opportunities that brokers were not made
aware of, might as well not have existed. The challenge, in terms of policy change,
at the interpersonal, community, and larger levels, will be to optimize child brokers’
academic attainment without sacrificing their commitments to their families.
Policy Recommendations
This research indicates that child brokers’ efforts are an important part of
how their families connect with—or fail to connect with—community institutions
that are essential to family wellbeing. Findings in the previous chapter suggest that
well-integrated internal institutional networks can parlay families’ connections with
one institution into multiple institutional connections. The importance of that first
institutional connection is consequently heightened by its potential to become a
general set of community connections. Since child brokers’ efforts emerged as an
important factor in how this first connection is made, brokering could ultimately play
a role in diminishing the considerable health and service utilization disparities
currently evident in Latino immigrant communities.
At the same time, this research also made it clear that children’s brokering
efforts often occurred at the expense of their schoolwork. This project is a
244

contribution toward understanding how the internal dynamics of immigrant families
affect their connections to institutions, as well as outcomes for child brokers
themselves. These findings could form the basis for community workshops with
local service providers that present ‘best practices’ for communicating through child
brokers. These workshops could be particularly valuable for teachers, who would be
made aware of the considerable responsibilities that children silently shoulder at
home, and teachers could be taught how to ask questions sensitively to ascertain the
real reasons that children might be feeling disconnected or overwhelmed by
schoolwork. Enhancing teachers’ understanding and skills could be a considerable
contribution toward enhancing children’s sense of connections to their teachers, their
schools’ internal network of resources, and consequently, to their educations. For
example, teachers might amend their teaching style in subjects that are most difficult
for children with limited English capacity to understand, such as history and science.
Science in particular can be taught in active, demonstrative ways that can facilitate
student understanding without language acting as a barrier. Teachers could also be
shown ways to develop an environment that encourages children to ask questions
and actively participate in classroom activities.
Improved connections to educational institutions would also be a first step
toward improving child brokers’ bilingual capacities. The stunted development in
two languages that I found in most child brokers is the most basic, and most
pervasive, constraint to the efficacy of their brokering efforts. California’s
245

inconsistent policies on bilingual education (Crawford, 1999) have had negative
effects on untold thousands of English Language Learner’s (ELL)
36
abilities to
successfully transition to English while still maintaining and developing their skills
in their first language. Since these bureaucratic difficulties are unlikely to resolve
soon, smaller-scale intervention programs in local schools that facilitate meaningful
connections between children, parents, and teachers, may be more effective for
mitigating child brokers’ short term difficulties with schoolwork, and for
encouraging their long-term language development.
Any ‘solution’ to the ‘problem’ of brokering activities’ effect on children’s
schoolwork has to address the collectivist-individualist tension these children
experience as they make and prioritize daily choices. Emphasizing children’s
individual scholastic achievement at the expense of their commitments to their
families’ needs will not only fail to connect, but will reinforce these brokers’
impression that individual achievement requires disconnecting from their families.
Creative ways to ease children’s responsibilities to enhance their educational
achievement, while taking their identification with their helping roles and family
reward structures around brokering, seriously, would be most effective. Skills
training for parents and children could be helpful in this regard. If children’s skill
sets were increased so that connections with community information resources and
institutions could be more successfully managed, with less negative feeling, in less

36
Children who enter schools speaking a language other than English are referred to as “English
Language Learners” by the Department of Education.
246

time, this could be a good compromise. Teaching children how to articulate their
needs for assistance in these institutional settings, proactively, would go a long way
toward this goal. Similarly, teaching service providers how to accommodate their
speech to facilitate easier brokering—by talking more slowly and in clear language,
for example—would be the essential counterpart to teaching children how to
connect.
Parents could also be encouraged to become more active in their own
connections with needed resources. Buriel & DeMent (1998) indicated that parents
often continue to depend on their children’s brokering skills after they have
developed sufficient English-language capacity to navigate these connections on
their own. The continued enactment of established patterns of interaction likely
results from parents’ limited confidence in their own abilities, more than an actual
need for their children’s assistance. Parents and children could both be served by
increasing parents’ confidence and feelings of self-efficacy in connecting with the
local community.
Legislation that forbids child brokering, such as AB292, fails to recognize the
situation on the ground. Although professional linguistic accommodations in all
institutions that serve immigrant populations may be desirable, placing these
responsibilities on institutions that are already time and finance-strapped make it
unlikely that these changes could come to fruition. Findings in this research suggest
that parents often prefer their children’s brokering to that of professional translators.
247

Legislators would be well served to find out why this should be the case. Legislation
like AB292 fails to take into account patients’ need for patience and for a sense of
connection with the people who are responsible for their care. Creating a more
accommodating environment for child brokers, supplemented by real linguistic
accommodations—i.e. trained staff for this specific purpose, not receptionists
stretched by serving double duty—would be a better, and more realistic, solution to
the current situation than banning the contributions of child brokers without a visible
replacement for their services.
Research Limitations
There were several limitations to this research project. First, the parents we
interviewed were mostly female, which means that these data may have failed to
identify all the ways that brokering for mothers might differ from brokering for
fathers. Men were less likely to admit that they needed their children to broker for
them, and often couched their answers (if they even agreed to be interviewed) in
terms of their wives’ needs for brokering assistance. The research focus may also
have contributed to this gender skew. Jones-Correa (1998) indicated that even when
both parents work, immigrant mothers bear the primary burden for connecting with
community institutions, since their husbands perceive social services and school
meeting attendance to be women’s work. My data supported Jones-Correa’s finding
that mothers were over-represented in accessing social service and health care
institutions, but I did not find evidence for the same phenomenon in the schools. The
248

parent who felt more comfortable communicating with English speakers generally
handled school meetings, meaning that fathers often assumed this responsibility. The
limited number of fathers in this study might mean that we did not capture the full
picture of child brokering in school contexts.
Second, interviewing service providers, particularly in health care
institutions, was very challenging. Many were too busy to be interviewed at all.
That it was so difficult to make contact with the most time-strapped providers might
mean that I made contact with more integrated and efficient health care institutions,
and did not achieve a full understanding of health and social service institutions that
might be minimally integrated. Child brokers and their families, therefore, might be
experiencing even greater constraints in connecting meaningfully with service
providers in health care institutions than I have reported here.
Finally, the survey analyses reported in chapter 3 had certain limitations for
the purposes of this project. The survey measures I included in these analyses to
compare groups, particularly measures regarding the informal network research
questions (RQ1) were not all optimal for this purpose. For example, rating previous
experiences with African Americans was not the equivalent of a question where
respondents would have indicated if they have close friends who are African
American. It would be useful to repeat these analyses with more sensitive measures.
Likewise, the brokering scale that was created to differentiate between no/low
brokering and high brokering survey respondents was based on frequency of
249

brokering English-language television, radio, newspapers, mail, and phone calls.
These measures only reflect a limited range of brokering activities, which means that
these questions may not have failed to include families where other kinds of
brokering were common
37
.
Suggestions for Future Research
A number of possible research avenues emanated from the current study.
Two of these potential future projects are outlined here.
In interviews with parents and children—and occasionally with service
providers as well—references were made to child brokering in legal institutions.
Participants referred to children brokering in meetings with court-appointed
individuals or attorneys, or filling out legal documents for their parents. This would
be a fruitful and as-yet largely unstudied dimension of child brokering activities,
since legal concerns are central to the lives of immigrant populations for a range of
reasons. Connections to legal institutions are also a necessary precursor to applying
for citizenship, so understanding how child brokers might play a role in connecting
their parents not only to the local community, but to the larger national entity would
be an important extension of the current project.
The second direction for future research would be to replicate this study in
another community, with another group of immigrants. This research design and
theoretical framework is a contribution toward unifying studies of child brokering

37
The survey also included questions about brokering in schools, health care, and social service
institutions, but the ns for these questions (because of filtering) were too low to conduct meaningful
analyses.
250

that are currently scattered across multiple disciplines, by providing a blueprint for
future studies that could easily be compared with each other. These future efforts
would contribute toward understanding how much of child brokering activities are
specific to the conditions in a particular local environment. These studies would also
identify continuities and contrasts between child brokering activities in different
immigrant groups. Understanding how much of the child brokering experience is
locality- and group-specific, and how much of the experience is part of a more
universal immigrant story, would be a real contribution to the immigration literature,
which has largely ignored the roles of children up to this point.
This project demonstrates that child brokers, whether American-born or
immigrants themselves, are not “morally priceless and economically useless”
(Zelizer, 1985). Child brokers made real and sustained contributions to their
family’s daily functioning and community connections. Ronaldo said: “Oh, I help
them. I help with phone calls, I help my dad at work, or at the store or at the
movies—I talk for them.” If children’s speaking for their parents is helping to
integrate their families into their new communities as this research suggests, their
seat at the ‘big table’ may be long overdue.

251

REFERENCES
Abrams, J. R., Eveland, W. P., & Giles, H. (2003). The effects of television on group
vitality: Can television empower non-dominant groups? In P. K Kalbfleish
(Ed.),Communication Yearbook 27, 193-219. Thousand Oaks, CA: Sage.

Alba, R., Logan, J., Lutz, A., & Stults, B. (2002). Only English by the third
generation? Loss and preservation of the mother tongue among
grandchildren of contemporary immigrants. Demography (39)3, 467-484.

Amabisca, A. (1999). Alone together: Social integration of immigrant students at the
middle school level. California Association for Bilingual Education.
Available online: http://www.bilingualeducation.org/1999_1_10.html

Ariès, P. (1962). Centuries of childhood. (trans. R. Baldick). New York: Vintage
Books.

Astone, N. & McLanahan, S. (1991). Family structure, parental practices and high
school completion. American Sociological Review 51, 403-12

Astone, N., Nathanson, C., Schoen, R. & Kim, Y. (1999). Family demography,
social theory, and investment in social capital. Population and Development
Review 25(1), 1-31.

Axelrod, R. (1985). The evolution of cooperation. New York: Basic Books.

Ball-Rokeach, S. J., Cheong, P. H., Wilkin, H. A., & Matsaganis, M. D. (2004,
May). A map to the multiethnic communication landscape of Los Angeles
immigrant communities, old and new. Paper presented at the 54
th

International Communication Association Conference, New Orleans, LA:
May 28, 2004.

Ball-Rokeach, S. J., Kim, Y. C., & Matei, S. (2001). Storytelling neighborhood:
Paths to belonging in diverse urban environments. Communication Research,
28(4), 392-428.

Ball-Rokeach, S., Gibbs, J., Guiterrez-Hoyt, E., Jung, J.Y., Matei, S., Wilson, M.,
Yuan, Y., & Zhang, L. (2000). Community storytelling, Storytelling
community: Paths to belonging in diverse Los Angeles residential areas
(White paper 1). Los Angeles: Annenberg School for Communication
Technology & Community Program, Metamophosis Project.

252

Banet-Weiser, S. (2004). Surfin’ the net: Children, parental obsolescence, and
citizenship. In M. Sturken, D. Thomas, & S. Ball-Rokeach (Eds.),
Technological visions: The hopes and fears that shape new technologies.
Philadelphia: Temple University Press (270-292).

Barry, C. (2002). Multiple realities in a study of medical consultations. Qualitative
Health Research 12(8), 1093-111.

Bassani, C. (2007). Five dimensions of social capital theory as they pertain to youth
studies. Journal of Youth Studies 10(1), 17-34.

Bazeley, P. (2003). Teaching mixed methods. Qualitative Research Journal 3, 117-
126.

Beck, V., Wilkin, H., Valente, T., Murphy, S., Cody, M. & Huang, G. (June, 2006).
The Effects of a Telenovela Storyline on Breast Cancer Knowledge and
Behaviors Among Hispanic/Latino Audiences. Presentated at the
International Communication Association Annual Conference (Dresden,
Germany).

Boissevain, J. (1974). Friends fo friends: Networks, manipulators, and coalitions.
New York: St. Martin’s Press.

Bould, S. (2003). Caring neighborhoods: Bringing up the kids together. Journal of
family issues, 24 (4), 427-447.

Bourdieu, P. (1977). Outline of a theory of practice (R. Nice trans.). Cambridge:
Cambridge University Press.

Bourdieu, P. (1986). The forms of capital. In J.G. Robinson (Ed.), Handbook of
theory and research for the sociology of education (pp. 241-258). Westport,
CT: Greenwood Press.

Brittain, C. (2003, May). Crossing borders in the school yard: The formation of
transnational social spaces among Chinese and Mexican immigrant students
(Working Paper No. 76). San Diego: University of California, San Diego,
Center for Comparative Immigration Studies.

Bryman, A. (2007). Barriers to integrating quantitative and qualitative research.
Journal of Mixed Methods Research 1, 8-22.

253

Buckingham, D. (2000). After the death of childhood: Growing up in the age of
electronic media. Cambridge, UK: Polity Press.

Burawoy, M. (2003). Revisits: An outline of a theory of reflexive ethnography.
American Sociological Review, 68, 645-679.

Buriel, R. (1993). Childrearing orientations in Mexican American families: The
influence of generation and sociocultural factors. The Journal of Marriage
and the Family, 55, 987-1000.

Buriel, R. & De Ment, T. (1998). Immigration and sociocultural change in Mexican,
Chinese, and Vietnamese American families. In A. Booth, A. Crouter, & N.
Landale (Eds.). Immigration and the family (pp. 165-200). Mahwah, NJ:
Lawrence Erlbaum.

Buriel, R., Perez, W., DeMent, T., Chavez, D. & Moran, V. (1998). The relationship
of language brokering to academic performance, biculturalism, and self-
efficacy among Latino adolescents. Hispanic Journal of Behavioral Sciences
20(3), 283-297.

Burt, R. (2001). Structural holes vs. network closure as social capital. In N. Lin, K.
Cook, & R. Burt (Eds), Social capital: Theory & research (pp. 31-56). New
York: Aldine de Gruyter.

Chao, R. (2006). The prevalence and consequences of adolescents' language
brokering for their immigrant parents In M. Bornstein & L. R. Cote (Eds.),
Acculturation and parent-child relationships: Measurement and development
(pp. 271-296). Mahwah, New Jersey: Lawrence Erlbaum Associates,
Publishers.

Chaskin, R., Brown, P., Venkadesh, S. & Vidal, A. (2001). Building community
capacity. New York: Aldine de Gruyter.

Chávez, C. & Ball-Rokeach, S. (under review). Exploring the role of religious
affiliation on civic engagement amongst new immigrant Latinos in Los
Angeles: A storytelling perspective.

Cheong, P. (2004). Media relations, threat and health problem solving behaviors:
Extension and application of communication infrastructure theory and
research. Dissertation, Annenberg School for Communication, University of
Southern California, Los Angeles, CA.

254

Chong, N. (2002). The Latino patient: A cultural guide for health care providers.
Yarmouth, ME: Intercultural Press.

Chu, C. (1999). Immigrant children mediators (ICM): Bridging the literacy gap in
immigrant communities. The New Review of Children’s Literature and
Librarianship, 85-94.

Coutin, S. (2000). Legalizing moves: Salvadoran immigrants’ struggle for U.S.
residency. Ann Arbor: University of Michigan Press.

Cranmer, S. (2006). Children and young people’s uses of the Internet for homework.
Learning, Media, and Technology 31(3), 301.315.

Coleman, J.S. (1990). Foundations of social theory. Cambridge: Harvard
University Press.

Coleman, J.S. (1988). Social capital in the creation of human capital. American
Journal of Sociology 94, S95-S120.

DeBell, M. & Chapman, C. (2006). Computer and Internet use by students in 2003
(NCES 2006-065). U.S. Department of Education. Washington, D.C.:
National Center for Education Statistics.

December, J. (1996). Units of analysis for Internet communication. Journal of
Communication 46 (1), 14–38.

Denzin, N.K. (1978). The research act (2
nd
ed.). New York: McGraw-Hill.

Dika, S. & Singh, K. (2002). Applications of social capital in education literature: A
critical synthesis. Review of Educational Research 72(1), 31-60.

Documét, P. & Sharma, R. (2004). Latinos’ health care access: Financial and
cultural barriers. Journal of Immigrant Health 6(1), 5-13.

Domínguez, S. & Watkins, C. (2003). Creating networks for survival and mobility:
Social capital among African-american and Latin-American low income
mothers. Social Problems 50(1), 111-135.
Dorner, L. Orellana, M. & Li-Grining, C. (2006). “I helped my mom and it could
help me: Translating bilingual skills into improved standardized test scores."
The American Journal of Education.
255

Durham, M. (2004). Constructing the “New ethnicities”: Media, sexuality, and
diaspora identity in the lives of South Asian immigrant girls. Critical Studies
in Media Communication 21(2), 140-161.

Ebaugh, H. & Chafetz, J. (2000). Structural adaptation in immigrant congregations.
Sociology of Religion (61)2, 135-153.

Eksner, H.J. & Orellana, M. (2005). Liminality as linguistic process: Mediation and
contestation by immigrant youth in Germany and the U.S." In J. Knörr (Ed.),
Childhood and migration: From experience to agency. New Jersey: Bielefeld
& Somerset, Transcript & Transaction Publishers."

Emerson, R. Fretz, R., & Shaw, L. (1995). Writing ethnographic fieldnotes.
Chicago: University of Chicago Press.

Ermisch, J. & Francesconi, M. (2001) Family matters: Impacts of family
background on educational attainments. Economica 68(270), 137–156.

Engeström, Y. (1999). Activity theory and individual and social transformation. In
Y. Engeström, R. Miettinen & R. Punamäki (Eds.), Perspectives on activity
theory (pp.19-38). Cambridge: Cambridge University Press.

Erzinger, S. (1999). Communication between Spanish-speaking patient and their
doctors in medical encounters. In G. Ma & G. Henderson (Eds.), Rethinking
ethnicity and health care: A sociocultural perspective. Springfield, IL:
Charles C. Thomas Publisher.

Félix, A., González, C. & Ramírez, R. (2007). Today we march, tomorrow
we…naturalize? The role of political protests and media in the path to
citizenship. Presented at the UCLA Immigration Seminar.

Fischer, C. (1982). Personal community. In To dwell among friends: Personal
networks in town and city. Chicago: University of Chicago Press.

Fishman, J. (1999). Sociolinguistics. In J. Fishman (Ed.), Handbook of language
and ethnic identity. Oxford: Oxford University Press (152-163).

Fitzgerald, D. (2006). Toward a theoretical ethnography of migration. Qualitative
Sociology,29(1), 1-24.

Flores, G. et al. (2003). Errors in medical interpretation and their potential clinical
consequences in pediatric encounters. Pediatrics 111(1), 6-14.
256

Furstenberg, F.F., Cook, T. & Eccles, J. (1999). Manging to make it: Urban famlies
and adolescent success. Chicago: University of Chicago Press.

Geertz, C. (1973). The Interpretation of cultures: Selected essays. New York: Basic
Books.

Gibson, M. (1988). Accommodation without assimilation: Sikh immigrants in an
American high school. Ithaca, NY: Cornell University Press.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology
78(1), 360-380.

Hagan, J., MacMillan, R. & Wheaton, B. (1996). New kid in town: Social capital
and the life course effects of family migration on children. American
Sociological Review 61, 368-385.

Heath, S.B. (1983). Ways with words: Language, life and work in communities and
classrooms. Cambridge: Cambridge University Press.

Hirsch, J. (2003). A courtship after marriage: Sexuality and love in Mexican
transnational families. Berkeley, CA: University of California Press.

Hochschild, A. (1983). The managed heart: Commercialization of human feeling.
Berkeley: University of California Press.

Hondagneu-Sotelo, P. (2003). Gender and immigration: A retrospective and
introduction. In P. Hondagneu-Sotelo (Ed.), Gender and U.S. immigration
(pp. 3-19). Berkeley: University of California Press.

Hondagneu-Sotelo, P. (2001). Doméstica: Immigrant workers cleaning and caring
in the shadows of affluence. Berkeley: University of California Press.

Hondagneu-Sotelo, P. (1994). Gendered transitions: Mexican experiences of
immigration. Berkeley: University of California Press.

Hseih, E. (2004). Conflicts of interpreter’s role performance in health care settings.
Presented at the National Communication Association Annual Meeting
(Chicago, IL).

257

Huss-Keeler, R. (1997). Teacher perception of ethnic and linguistic minority
parental involvement and its relationships to children’s language and literacy
learning: A case study. Teaching and Teacher Education 13(2), 171-182.

Jick , T.D. (1979). Mixing qualitative and quantitative methods: Triangulation in
action. Administrative Science Quarterly 24, 601-611.

Johnson, R.B. & Onwuegbuzie, A.J. (2004). Mixed methods research: A research
paradigm whose time has come. Educational Researcher 33(7), 14-26.

Johnson, R.B. & Onwuegbuzie, A.J. & Turner, L.A. (2007). Toward a definition of
mixed methods research. Journal of Mixed Methods Research 1, 112-133.

Jones-Correa, M. (1998). Different paths: Gender, immigration, and political
participation. International Migration Review 32(2), pp. 326-349.

Kao, P. (2004). Social capital and its relevance to minority and immigrant
populations. Sociology of Education 77(2), 172-175.

Katz, V. (2005). Communicating care: A communication infrastructure approach to
health and child care choices of new immigrants in Los Angeles.
Unpublished M.A. Thesis.

Kibria, N. (1993). Family tightrope: The changing lives of Vietnamese Americans.
Princeton, NJ: Princeton University Press.

Kim, S., Kaplowitz, S. & Johnston, M. (2004). The effects of physician empathy on
patient satisfaction and compliance. Evaluation and the Health Professions
27(3), 237-251.

Kim, Y. C. & Ball-Rokeach, S. J. (2006a). Civic engagement from a communication
infrastructure Perspective. Communication Theory, 16, 1-25.

Kim, Y.C. & Ball-Rokeach, S.J. (2006b). Community storytelling network,
neighborhood context, and civic engagement: A multilevel approach. Human
Communication Research 32 (4), 411-439.

Kim, Y. C., Jung J. & Ball-Rokeach, S. J. (2006). Geo-ethnicity and civic
engagement: From a communication infrastructure perspective. Political
Communication, 23 (3).

258

Kim, Y. C., & Jung, J.-Y. (2003). "Geo-ethnicity" and neighborhood engagement: A
communication infrastructure perspective. Top Four Student Paper. Political
Communication Division. Paper presented to the International
Communication Association Conference. San Diego, CA: May 23-27, 2003.

Kowaleski-Jones, L., Dunifon, R. & Ream, R. (2006). Community contributions to
scholastic success. Journal of Community Psychology 34(3), 343-362.

Lareau, A. & Horvat, E.M. (1999). Moments of social inclusion and exclusion:
Race, class and cultural capital in family-school relationships. Sociology of
Education 72, 37-53.

Leduc, N. & Proulx, M. (2004). Patterns of health services utilization by recent
immigrants. Journal of Immigrant Health 6(1), 15-27.

Lieberson, S. (1981). In Language diversity and language contact. Stanford, CA:
Stanford University Press (1-19; 158-172).

Lievrouw, L. & Livingstone, S. (Eds.) (2005). Introduction to the student edition.
The handbook of new media: Social shaping and consequences of ICTs.
London: Sage Publications (pp. 1-24).

Lin, N. (2001). Social capital: A theory of social structure and action. Cambridge:
Cambridge University Press.

Lipsit, M. (2003). Newcomers left behind: Immigrant parents lack equal access to
New York City’s schools. New School University: Center for New York
City Affairs.

Livingstone, S. (2002). Young people and new media. London: Sage Publications

Lopez, N. (2003). Hopeful girls, troubled boys: Race and gender disparity in urban
education. New York: Routledge.

Lopez, D. (1996). Language: Diversity and assimilation. In R. Waldinger & M.
Bozorgmehr (Eds.). Ethnic Los Angeles (pp. 139-159). New York: Russell
Sage.

Mahler, S. (1995). American dreaming: Immigrant life on the margins. Princeton,
NJ: Princeton University Press.

259

Malakoff, M. & Hakuta, K. (1991). Translation skill and metalinguistic awareness in
bilinguals. In E. Bialystock (Ed.), Language processing in bilingual
children. Cambridge: Cambridge University Press.

Massey, D., Durand, J. & Malone, N. (2002). Beyond smoke and mirrors: Mexican
immigration in an era of economic integration. New York: Russell Sage.

McDaniel, M. (2006). Media use, linguistic preference, and social capital in the
Hispanic community. Unpublished M.A. thesis. Manship School of Mass
Communication, Louisiana State University.

McKenzie, K., Whitley, R. & Weich, S. (2002). Social capital & mental health.
British Journal of Psychiatry, 280-283.

McRoberts, O. (2003). Streets of glory: Church and community in a Black urban
neighborhood. Chicago: University of Chicago Press.

Menjívar, C. (2000). Fragmented ties. Berkeley: University of California Press.

Merrifield, J., Bingman, M.B., Hemphill, D., Bennett-deMarrais, K.P. (1997). Life
at the margins: Literacy, language, and technology in everyday life. New
York: Teachers College Press.

Mohanty, S., Woolhandler, S., Himmelstein, D., Pati, S., Carrasquillo, O. & Bor, H.
(2005). Health care expenditures of immigrants in the United States: A
nationally representative analysis. American Journal of Public Health 95(8),
1-8.

Moll, L. C. (1994). Mediating knowledge between homes and classrooms. In D.
Keller-Cohen (Ed.), Literacy: Interdisciplinary conversations (pp. 385-410).
Cresskill, NJ: Hampton Press.

Monkman, K., Ronald, M. & Théramène, F.D. (2005). Social and cultural capital in
an urban Latino school community. Urban Education 40(1), 4-33.

Monzó, L. (2003). Latino children and families neogitating identities and cultural
productions: A socioculutral perspective on cultural contact. Unpublished
dissertation, Rossier School of Education, University of Southern California.

Morse, J. (1991). Approaches to qualitative-quantitative methodological
triangulation. Nursing Research 40, 120-123.

260

Nelson, M. (2000). Single mothers and social support: The commitment to, and
retreat from, reciprocity. Qualitative Sociology 23 (291-317).

Nickel, B., Berger, M., Schmidt, P. & Plies, K. (1995). Qualitative sampling in a
multi method survey. Quality and Quantity 29 (223-240).

Noguera, P. (2004). Social capital and the education of immigrant students:
Categories and generalizations. Sociology of Education 77(2), 180-.

Olsen, L. (1997). Made in America: Immigrant students in our public schools. New
York: The New Press.

Ong, A. (2003). Buddha is hiding. Berkeley: University of California Press.

Dorner, L., Orellana, M. & Li-Grining, C. (2006). “I helped my mom” and it helped
me: Translating the skills of language brokers into improved standardized test
scores. American Journal of Education

Orellana, M., Dorner, L. & Pulido, L. (2003). Accessing assets: Immigrant youth’s
work as family translators or “para-phrasers”. Social Problems (50)4: 505-
524.

Orellana, M. (2001). The work kids do: Mexican and Central American immigrant
children’s contributions to households and schools in California. Harvard
Educational Review, 71(3): 366-389.

O’Reilly, K. (2005). Ethnographic methods. New York: Routledge.

Park, L. (2002). Asian immigrant entrepreneurial children. In L. Võ & R. Bonus
(Eds.), Contemporary Asian American communities: Intersections and
divergences. Philadelphia: Temple University Press.

Park, L. (2001). Between adulthood and childhood: The boundary work of
immigrant entrepreneurial children. Berkeley Journal of Sociology 45, 114-
135.

Parke, R.D. & Buriel, R. (1995). Socialization in the family: Ethnic and ecological
perspectives. In N. Eisenberg (Ed.), Handbook of social psychology: Social,
emotional, and personality development (3). New York: John Wiley & Sons.

261

Pessar, P. (2003). Engendering migration studies: The case of new immigrants in the
United States. In P. Hondagneu-Sotelo (Ed.), Gender and U.S. immigration
(pp. 20-42). Berkeley: University of California Press.

Portes, A. (1998). Social capital: Its origin and applications in modern sociology.
Annual Review of Sociology 24, 1-24.

Portes, A. & Rumbaut, R. (2000). Legacies: The story of the immigrant second
generation. Berkeley: University of California Press.

Portes, A. & Rumbaut, R. (1996). Immigrant America: A portrait. Berkeley:
University of California Press.

Putnam, R. (2000). Bowling alone: The collapse and revival of American
community. New York: Simon & Schuster.

Pyke, K. (2005). “Generational deserters” and “black sheep”: Acculturative
differences among siblings in Asian immigrant families. Journal of Family
Issues 26(4), 491-517.

Ream, R. (2005). Uprooting children: Mobility, social capital, and Mexican
American underachievement. New York: LFB Scholarly Publishing, LLC.

Reese, L. (2001). Morality and identity in Mexican immigrant parents’ visions of
the future. Journal of Ethnic and Migration Studies (27)3: 455-472.

Rueda, R., Monzó, L.D., & Arzubiaga, A. (2003). Academic instrumental
knowledge: Deconstucting cultural capital theory for strategic intervention
approaches. Current Issues in Education [On-line] 6(14). Available:
http://cie.asu.edu/volume6/number14/index.html

Rueda, R., & Moll, L. C. (1994). A sociocultural approach to motivation. In H. F.
O'Neill & M. Drillings (Eds.), Motivation: Theory and research (pp. 117-
136). Hillsdale, NJ: Lawrence Erlbaum Associates.

Ruffin, J. (2002, April 12). Statement on the National Institute of Health’s role in
decreasing health disparities. Retrieved from
http://www.hhs.gov/asl/testify/t020412b.html on June 2, 2007.

Sampson, R., Morenoff, J. & Earls, F. (1999). Beyond social capital: Spatial
dynamics of collective efficacy for children. American Sociological Review
64, 633-660.
262

Sampson, R., Morenoff, J., & Gannon-Rowley, T. (2002). Assessing “neighborhood
effects”: Social processes and new directions in research. Annual Review of
Sociology 28, pp. 443-478.

Sànchez, I. & Orellana, M. (2006). The construction of moral and social identity in
immigrant children’s narratives-in-translation. Linguistics and Education.

Sanders, J. & Nee, V. (1996). Immigrant self-employment: The family as social
capital and the value of human capital. American Sociological Review 61,
231-249.

Schudson, M. (1999). Good citizens and bad history: Today’s political ideals in
historical perspective. Paper preseted at “The Transformation of Civic Life”
Conference, Nashville and Murfreesboro, TN, November 12-13.

Schudson, M. (1998). The good citizen. New York: Free Press. Shah, D., McLeod,
J. & Yoon, S. (2001). Communication, context, and community: An
exploration of print, broadcast, and internet influences. Communication
Research (28)4, 464-506.

Servon, L. & Pinkett, R. (2004). Narrowing the digital divide: The potential and
limits of the US Community Technology Movement. In M. Castells (Ed.),
The network society: A cross-cultural perspective (319-340). Northampton,
MA: Edward Elgar Publishing.

Small, M.L. (2006). Neighborhood institutions as resource brokers: Childcare
centers, interorganizational ties, and resource access among the poor. Social
Problems 53(2), 274-292.

Song, M. (1999). Helping out: Children’s labor in ethnic businesses. Philadelphia:
Temple University Press.

Stack, C. & Burton, L. (1993). Kinscripts. Journal of Comparative Family Studies
24, 157-170.

Stanton-Salazar, R.D. (2001). Manufacturing hope and despair: The school and kin
support networks of U.S.-Mexican Youth. New York: Teachers College
Press.

263

Stanton-Salazar, R.D. & Dornbusch, S.M. (1995). Social capital and the
reproduction of inequality: Information networks among Mexican-origin high
school students. Sociology of Education 68, 116-135.

Stanton-Salazar, R.D. & Spina, S.U. (2003). Informal mentors and role models in
the lives of urban Mexican-origin adolescents. Anthropology & Education
Quarterly 34(3), 231-254.

Star, S. & Bowker, G. (2002). How to infrastructure. In L. Lievrouw & S.
Livingstone (Eds.), The handbook of new media: Social shaping and
consequences of ICTs. London: Sage Publications (pp. 151-162).

Star, S. & Ruhleder, K. (1996). Steps toward an ecology of infrastructure: Design
and access for large information spaces. Information Systems Research (7):
111-133.

Street, R. (1991). Accommodation in medical consultations. In H. Giles, J.
Coupland, & N. Coupland (Eds.), Contexts of accommodation: Developments
in applied sociolinguistics. Cambridge: Cambridge University Press.

Suárez-Orozco, C., Suárez-Orozco, M. (2001). Children of immigration. Boston:
Harvard University Press.

Tashakkori, A. & Teddlie, C.B. (1998). Mixed methodology: Combining qualitative
and quantitative approaches. Thousand Oaks, CA: Sage Publications.

Thorne, B. (1993). Gender play: Boys and girls in school. Piscataway, NJ: Rutgers
University Press.

Thorne, B. et al. (2003). Raising children, and growing up, across national borders.
In P. Hondagneu-Sotelo (Ed.), Gender and U.S. immigration (pp. 241-262).
Berkeley: University of California Press.

Tierney, W.G. & Venegas, K.M. (2006). Fictive kin & social capital: The role of
peer groups in applying and paying for college. American Behavioral
Scientist 49(12), 1687-1702.

Tse, L. (1995). Language brokering among Latino adolescents: Prevalence, attitudes,
and school performance. Hispanic Journal of Behavioral Sciences 17: 180-
193.

264

Tse, L. (1996). Who decides?: The effect of language brokering on home-school
communication. The Journal of Educational Issues of Language Minority
Students 16: 225-233.

Turley, R.L. (2002). Is relative deprivation beneficial? The effects of richer and
poorer neighbors on children’s outcomes. Journal of Community Psychology
30(6), 671-686.

US Census Bureau (2000). Retrieved from www.census.gov.

Valdés, G. (2002). Expanding definitions of giftedness: The case of young
interpreters from immigrant countries. Mahwah, N.J.: Lawrence Erlbaum &
Associates.

Valdés, G. (1996). Con respeto: Bridging the distances between culturally diverse
families and schools. New York: Teachers College Press.

Valenzuela, Abel, Jr. (1999). Gender roles and settlement activities among children
and their immigrant families. American Behavioral Scientist (42)4: 720-742.

Valenzuela, Angela. (1999). Subtractive schooling: U.S.-Mexican youth and the
politics of caring. New York: State University of New York Press.

Vygotsky, L. (1978). Mind in society: The development of higher psychological
processes. Cambridge: Harvard University Press.

Waldinger, R. & Lichter, M. (2003). How the other half works. Berkeley:
University of California Press.

Waldinger, R. & Der-Martirosian, C. (2001). The immigrant niche: Pervasive,
persistent, diverse. In R. Waldinger (Ed.), Strangers at the gates: New
immigrants in urban America (pp. 228-271). Berkeley: University of
California Press.

Wall, E., Ferrazzi, G., & Schryer, F. (1998). Getting the goods on social capital.
Rural Sociology 63(2), 300-322.

Watson, B. & Gallois, C. (2004). Emotional expression as a sociolinguistic strategy:
Its importance in medical interactions. In S. Ng, C. Candlin, & C. Chiu
(Eds.). Language matters: Communication, culture and identity. Hong Kong:
City University of Hong Kong Press, 63-84.

265

Weisner, T., Ryan, G., Reese, L., Kroesen, K., Bernheimer, L., & Gallimore, R.
(2001). Behavior sampling and ethnography: Complementary methods for
understanding home-school connections among immigrant Latino families.
Field Methods 13(1), 20-46.

Weisskirch, R.S. & Alva, S.A. (2002). Language brokering and the acculturation of
Latino children. Hispanic Journal of Behavioral Sciences 24, 369-378.

Wellman, B. & Gulia, M. (1999). The network basis of social support: A network is
more than the sum of its ties. In B. Wellman (Ed.), Networks in the global
village: Life in contemporary communities (pp. 83-118). Boulder, CO:
Westview.

Weston, K. (1991). Families we choose. New York: Columbia University Press.

Wilkin, H. A. (2005) Diagnosing communication connections: Reaching
underserved communities through existing communication ecologies,
Dissertation, Annenberg School for Communication, University of Southern
California, Los Angeles, CA.

Wilkin, H. A., Cheong, P. H., & Ball-Rokeach, S. J. (2004, Nov.). Community
change begins at home: The role of family interaction on civic engagement
outcomes. Paper presented at the 90
th
National Communication Association
Conference, Chicago, Ill.

Wilkin, H. & González, C. (June 2006). Are Spanish-language television shows
connecting Latino residents in Los Angeles to their health storytelling
networks? Presented at the International Communication Association Annual
Conference (Dresden, Germany).

Wilkin, H., Katz, V., & Ball-Rokeach, S.J. (in preparation). How family talk affects
community storytelling: The role of family interaction in new immigrant
civic engagement.

Wilson, M. (2002) Community organizations, communication infrastructure, and
community. Paper presented to the National Communication Association
Conference. New Orleans, LA: November 21-24, 2002.

Zelizer, V. (1985). Pricing the priceless child: The changing social value of children.
Princeton: Princeton University Press.

Zentella, A.C. (1997). Growing up bilingual. Oxford: Blackwell Publishing.
266

Zhou, M. & Bankston III, C. (1998). Growing up American: How Vietnamese
children adapt to life in the United States. New York: Russell Sage.

Zhou, M. (1997). Growing up American: The challenge confronting immigrant
children and children of immigrants. Annual Review of Sociology, 23, 63-95.

Zu ňiga, V. & Hamann, E.T. (2006). Going Home?: Schooling in Mexico of
transnational children. Confines 2(4), 41-57.

267

APPENDIX 1:
COMMUNICATION ACTION CONTEXT FEATURES
The communication action context can be understood as the social architecture of
the community as its features can constrain or enable communication between or
within the nodes of the storytelling network. The communication action context is a
set of related feature types (in Ball-Rokeach, Kim & Matei, 2001):
The physical makeup of the urban grid (e.g., streets and freeways)
The relative presence of places that bring people together (e.g., parks, quality
grocery stores, and libraries).
Psychological features, meaning the degree to which people feel free to
engage each other (e.g., perceptions of safety).
Sociocultural characteristics that facilitate (or hinder) communication, such
as racial/ethnic diversity, and inclinations toward individualism or
collectivism.
The nature of Institutions in the community, including schools, health care,
and social services.
Economic features, such as the time and resources available to engage in
conversation with others in the community.

268

Technological features, including access to communication technologies
(e.g., Internet connections, types of new and old local media available to
residents of an area), as well as characteristics of available transportation
systems (e.g., individual cars, bus system, and subway system), which can
also affect how much daily contact individuals have with other residents in
their communities.
269

APPENDIX 2:
CULTURAL VS. SOCIAL CAPITAL
Social capital is the ability to convert resources embedded in informal
networks successful into other kinds of capital, including cultural capital. Cultural
capital refers to the knowledges, practices and artifacts that are valued in a particular
society (Bourdieu, 1986; Monkman, Ronald & Théramène, 2005). Cultural capital
has been applied to studies of immigrants and education in particular, to identify the
factors (which are treated more like facts) held up as evidence that immigrants’
constraints as newcomers explain why school achievement remains elusive for their
children. The concept of culture is both an important descriptive and highly
problematic. On one hand, the term draws attention to school as a ‘culture’ that
values some forms of knowledge and practices more than others. On the other hand,
‘culture’ is not a commodity equally possessed by all members of a group, nor do
cultural differences have the same effects on all individuals and families. Cultural
capital as a concept obscures individual-level differences and therefore denies human
agency of immigrant parents and children in their family networks—and the agency
of institutional agents who may accept or reject those individuals’ efforts to make
institutional connections (Lareau & Horvat, 1999).
The storytelling network of communication infrastructure focuses attention
on the social links between individuals as well as the links between and within
community organizations and local media outlets. Individuals working in
270

institutions have social networks of colleagues within and across institutions. While
two institutions may have formal linkages, the relationships between individuals
working within those institutions determine the (un)successful sharing of resources
between the institutions. Immigrant parents and children also make connections with
individual agents, before they make a connection (if they ever do) to the institution
more generally. Therefore, connections forged between immigrant families and
institutions are actually the result of successfully converted social capital from the
immigrants’ informal networks, into relationships that give access to that
institutional agents’ formal networks within that institution, and across institutions as
well.
Therefore, in this investigation, cultural capital is a less utile concept than
social capital, conceived as embedded in immigrant families’ informal (family and
peer) networks, and the social capital embedded in the connections between
individuals working within and between community institutions. However, the three
types of cultural capital that Bourdieu originally identified can be fruitfully adapted
to this investigation of institutions: institutional capital, embodied capital, and
objectified capital.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Translating race, class, and immigrant lives: the family work of children language brokers

PDF

Doing family on Facebook: connecting communicative affordances of mobile media, parental practices, and social capital

Asset Metadata

Creator Katz, Vikki Sara (author)

Core Title From conversation to conversion: children's efforts to translate their immigrant families' social networks into community connections

School Annenberg School for Communication

Degree Doctor of Philosophy

Degree Program Communication

Publication Date 07/24/2007

Defense Date 06/28/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag brokering,community institutions,family,immigration,Latinos,OAI-PMH Harvest,translating

Language English

Advisor Ball-Rokeach, Sandra (committee chair), Cody, Michael J. (committee member), McLaughlin, Margaret (committee member), Rueda, Robert S. (committee member)

Creator Email vkatz@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m650

Unique identifier UC1169754

Identifier etd-Katz-20070724 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-530964 (legacy record id),usctheses-m650 (legacy record id)

Legacy Identifier etd-Katz-20070724.pdf

Dmrecord 530964

Document Type Dissertation

Rights Katz, Vikki Sara

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu