 

 

gill"

mamsgéﬁéﬁwx...

 

. I! 0411:-

 

gift... .22.!

gﬁdﬁwwﬁm

“r... 5 (I.

. .x, . .1 . .2. , .
gang“: . ...;._._szh.
5 r. 3.422....“ 1.. .10.. 1’; Q . i It: a
V V : . . .351”! . .4902: it
. J. r
235,544...
{Eifix 3 11

4 O ‘ ‘3
, nanny» {a
4 F695»: n.4, .3“
. . i. .15.. a.
. . 3......5? .5... Ex.
..V V . ‘ um .. er. usuypﬁi i.
k n F v .‘k ‘4;
.r . ”n. r a; :3. .— .9-1..'.'
:-’V ”Kugul’AvCo-vli‘oi‘ ’ 4 .IIV.;IIK§. ‘U‘V
a... his? ._ in $1? w:24.;i. ..
.2 1.0;: ‘93.“..“0: n... .4... .:~

. .

v .351
‘ whsﬁm.
ﬁnﬁﬁﬁ . gmﬁaﬁ.

1%,» , . an}
. _ . . .bmmf?

i‘l.
;. .
is 3.2.2:; 3,. .3, sﬁgxumwm . ,
Y. 11. 1... i113}... , 1': ‘QIWJS 117:. .
.3 hi» Lﬁ
50.15 101’.‘
,i. :-

..X.:m.:\tw!wa nu ..
3661‘... 13.3. «1‘1... .
ufll $29.. 7 Ci v
Inﬁ-‘W‘; "2.1“.
a!itat 7:14.!“ .0.“
2.2:
31v v
, ‘ t... ‘
at}. l: 2.3.. :2: ,
. .3 .r {51.3.1.1
:5‘13‘ 3.7

3 :5. 2
. 3.1.5: sh. L.
, {tiﬂzrtiz

:1
Italﬂvxuu
‘ .223)
.1. .
32:113.;1. .11 1... .
Eit?‘ ital:
.31 .51 5.. .

. v .
.NDLV
9|

V .l a.)
Ind; (Q 5‘ ’

1.921.:
:21. . ., all.»
vi a"! D]

a ~_,u\1!, 1.: h

; : . . .e I.

.3 ..\~\.{..uo.i: 3 5.1.50.
6. \y .1 {Ex}: 1‘ p
4)

 

use!

This is to ceitify that the
dissertation entitled

Predicting Latino Youth Academic Success from a

Normative Cultural-ecological Perspective

presented by

Cidhinnia M. Torres Campos

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Psycholgqy

 

 

(“W/M 9 gut/~57"

v/ Major Professor’s ﬁfénature

05/22/06

 

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

-.—--.-u-o--.-n-o----.—.-.-.-.-.-.- —AA

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

Dwfﬁy®s

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/ClRC/DateDue.indd-p.1

PREDICTING LATINO YOUTH ACADEMIC SUCCESS FROM A NORMATIVE
CULTURAL-ECOLOGICAL PERSPECTIVE

By

Cidhinnia M. Torres Campos

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements for the degree
DOCTOR of PHILOSOPHY
Department of Psychology

2006

Abstract

PREDICTING LATINO YOUTH ACADEMIC SUCCESS FROM A NORMATIVE
CULTURAL-ECOLOGICAL PERSPECTIVE

By

Cidhinnia M. Torres Campos

Despite their growing numbers in the United States, research focused on Latino
youth development is sparse (Chapa & Valencia, 1993; Ramos, 2002; Rodriguez &
Morrobel, 2002), while the negative academic issues they face have been well
documented (Dryfoos, 1998; Meir & Stewart, Jr., 1991; Romo & Falbo, 1996; US.
Department of Education, 1992). This study seeks to address whether we have enough
foundational knowledge to develop a core of information about the Latino community
and the characteristics of its children who succeed academically ( President’s Advisory
Commission on Educational Excellence for Hispanic Americans, 2000). In contrast to
deﬁcit-oriented research, this study adapted, augmented, and empirically tested Garcia
C011 and her colleagues’ (1996) model, focusing on normative Latino academic
development and outcomes, with data ﬁ'om the National Longitudinal Study of
Adolescent Health (Add Health) using structural equation modeling (SEM). Although the
model as a whole was not accepted, the exploration of the relationships between proximal
and distal predictors of development and behavioral outcomes offers a foundation for
continued research. This research also challenges educators, researchers, and policy
makers to rise above the historical trend of explanatory models and solutions based on a
deﬁcits perspective (Bartolomé & Balderrama, 2001; Diaz & Flores, 2001; Reyes &

Halcén, 2001; M011, 2001).

© Cidhinnia M. Torres Campos, 2006

Dedication

A mi querida Abuelita Leonor.

Aunque no pudiste ver cl trabajo ﬁnal en vida,

sé que siempre me tienes la velita prendida.

iv

Acknowledgements
There are a great many people who contributed to and supported me through this

process. I cannot even begin to mention everyone, but I’d like to recognize a few of the
many who were central to my ability to successfully complete this process. To all who in
some way supported, assisted, and mentored me, I offer my heartfelt thanks.

I’d like to acknowledge all of the assistance and guidance I received from my
committee chair, Hiram Fitzgerald, Ph.D., and all of my committee members, Richard
Deshon, Ph.D., Francisco Villarruel, Ph.D., and Deborah Johnson, Ph.D. Thank you for
your advice, patience, and encouragement throughout the dissertation process. I will
never forget your dedication to my educational success. To Wm. Thomas Summerfelt,
my mentor, advisor, and friend, my deepest and warmest gratitude for the enduring
support and guidance you have provided over the last 7 years.

To all of my family and friends, all I can say is, iGracias! Meche, a sido mi
privilegio haberte conocido y llarnarte mi amiga. Gracias por siempre apoyarme y sobre
todo en mis momentos malos y de gran felicidad. Mami, gracias por tu dedicacidn
perpétua a mi éxito académico y mi desarrollo como ser humano. Te admiro y aspiro
tener tu fortaleza. A Tristan, tu originalidad y tu sonrisa me ayudaron a siempre mantener
todo en perspectiva y nunca olvidarrne de lo que realmente importa. I am most
appreciative of my beloved Joshua who continued to demonstrate his love and
commitment even when at times it seemed like it would never end. “I know that there’s
something between me and nothing. You are that something that pulls me through. So,

I’m giving it all to you.”

Finally, this research uses data from Add Health, a program project designed by J.
Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and ﬁrnded by a grant P01-
HD31921 from the National Institute of Child Health and Human Development, with
cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R.
Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in
obtaining data ﬁles from Add Health should contact Add Health, Carolina Population

Center, 123 W. Franklin Street, Chapel Hill, NC 27516-2524 (addhealth@unc.edu).

vi

Table of Contents

List of Tables ..................................................................................................................... ix
List of Figures ..................................................................................................................... x
Introduction .......................................................................................................................... 1
Latino Youth in the United States ........................................................................................ 4
History of Latinos in US Education ............................................................................... 6
Current Educational Status of Latinos .......................................................................... 8
Theories of Development and Educational Models of Academic Success .......................... 9
Effective Academic Programs for Latinos ......................................................................... 23
Conclusions ........................................................................................................................ 26
Pathways to Latino Academic Success ........................................................................ 33
Social Position Variables ....................................................................................... 36

Social Stratification Mechanisms .......................................................................... 37
Segregation. ........................................................................................................... 38
Promoting/Inhibiting Environments-School .......................................................... 39
Promoting/Inhibiting Environments-Neighborhood .............................................. 41
Adaptive Culture .................................................................................................... 42

Child Characteristics .............................................................................................. 43
Family .................................................................................................................... 44

Peers ....................................................................................................................... 45
Developmental Competencies ............................................................................... 46
Academic Success .................................................................................................. 50

Study Aim and Hypothesis ................................................................................................ 50
Social Position Variables ............................................................................................. 51
Social Stratification Mechanisms ................................................................................ 52
Segregation .................................................................................................................. 52
Promoting/Inhibiting Environments-School ................................................................ 52
Promoting/Inhibiting Environments-Neighborhood .................................................... 53
Adaptive Culture .......................................................................................................... 53
Child Characteristics ................................................................................................... 54
Family .......................................................................................................................... 54
Peers ............................................................................................................................ 55
Developmental Competencies ...................................................................................... 55
Academic Success ........................................................................................................ 56
Methods .............................................................................................................................. 60
Data Collection ............................................................................................................ 60
Sample .......................................................................................................................... 61

vii

Sample Characteristics ........................................................................................... 62

Instrumentation- Wave I ............................................................................................... 67
The In-School Questionnaire ................................................................................. 67
In-Home Interview: Wave I ................................................................................... 68
School Administrators Questionnaire .................................................................... 68
Parent Questionnaire .............................................................................................. 69

Instrumentation- Wave [1 ............................................................................................. 69
In-Home Interview Wave II ................................................................................... 69
School Administrator Telephone Interviews ......................................................... 69

Instrumentation —Wave III ........................................................................................... 70
In-Home Interview: Wave III ................................................................................ 71

Data Analyses .............................................................................................................. 71
Exogenous Variables ............................................................................................. 73
Endogenous Variables. .......................................................................................... 73
Model Fit ................................................................................................................ 74

Results ................................................................................................................................ 74

Conﬁrmatory Factor Analyses ..................................................................................... 74
Social Position Variables ....................................................................................... 75
Perceived Prejudice ................................................................................................ 75
Segregation ............................................................................................................ 76
Promoting/Inhibiting Environments-School .......................................................... 76
Promoting/Inhibiting Environments-Neighborhood .............................................. 79
Adaptive Culture .................................................................................................... 80
Child Characteristics .............................................................................................. 82
Family .................................................................................................................... 84
Peer Support ........................................................................................................... 86
Developmental Competencies ............................................................................... 88
Academic Success .................................................................................................. 90

Structural Equation Modeling ..................................................................................... 91
Hypothesized Mode] .............................................................................................. 91
Direct Relationships ............................................................................................... 93
Mediating Relationships ........................................................................................ 98

Alternate Models ........................................................................................................ 101

Discussion ........................................................................................................................ 106

Major Findings .......................................................................................................... 107
Overall Hypothesized Model ............................................................................... 107
Predicted Factor Relationships ............................................................................ 110
Alternative Models ............................................................................................... 127

Conclusions ................................................................................................................ 128

Appendix A ...................................................................................................................... 133
References ........................................................................................................................ 139

viii

List of Tables

Table 1 ............................................................................................................................... 63
Table 2 ............................................................................................................................... 63
Table 3 ............................................................................................................................... 64
Table 4 ............................................................................................................................... 65
Table 5 ............................................................................................................................... 66
Table 6 ............................................................................................................................... 93
Table 7 ............................................................................................................................... 97
Table 8 ............................................................................................................................. 100
Table 9 ............................................................................................................................. 102
Table 10 ........................................................................................................................... 102
Table 11 ........................................................................................................................... 104
Table 12 ........................................................................................................................... 106

ix

List of Figures

Figure I. Integrative model of development competencies in minority children .............. 23

Figure 2. Adapted, augmented model of Latino academic developmental competencies 34

Figure 3. Manifest variable model ..................................................................................... 57
Figure 4. School latent construct with estimated parameters ............................................ 79
Figure 5. Neighborhood latent construct with estimated parameters ................................ 80
Figure 6. Adaptive culture latent construct with estimated parameters ........................... 82
Figure 7. Child characteristics latent construct with estimated parameters ...................... 83
Figure 8. Family processes latent construct with estimated parameters ........................... 86
Figure 9. Peer processes latent construct with estimated parameters ................................ 87

Figure 10. Developmental competencies latent construct with estimated parameters ...... 89

Figure 11. Academic success latent construct with estimated parameters ........................ 91
Figure 12. Model altered according to modiﬁcation indices ........................................... 104
Figure 13. Perceived prejudice and segregation direct relationships model ................... 105

Introduction

Existing research has explored factors most often related to the academic failure
of Latino youth, which has lead to policy, educational reform, and programming that
have not signiﬁcantly positively altered the trajectory of these students. These failure
approaches have not lead to effective intervention for the academic success of Latino
youth. A success oriented model able to predict the academic success of Latino youth,
despite growing odds against them, may inform future efforts aimed at improving Latino
student outcomes. Currently, there is a dearth of empirically tested models that have been
speciﬁcally examined in terms of their ability to explain and predict the academic success
of Latino youth. This study seeks to empirically assess the effectiveness of an adapted
and augmented version of Garcia C011 and colleagues’ (1996) integrative model of
minority youth developmental competencies in predicting Latino youth academic
success.

The academic success of adolescents has ramiﬁcations throughout their lives,
particularly given that education has been closely tied to opportunities for employment
and economic development, as well as physical and mental health problems. Students
who underachieve academically are more likely to need social services and contribute
less to the economy than their academically successful peers (Carnegie Council on
Adolescent Development, 1989). The costs to society and to individuals are high. The
US. pays not only through increased social costs, but also in an estimated $260 billion in
lost earnings and tax payments (Carnegie Council on Adolescent Development, 1989;
President’s Advisory Commission On Educational Excellence for Hispanic Americans,
2003; President’s Advisory Commission On Educational Excellence for Hispanic

Americans, 2000).

Although not all Latino students are impoverished, Latino students in general are
viewed as academically “at-risk” (Gutierrez, 2002). One family background characteristic
commonly used to measure risks to future academic outcomes is whether students’
parents primary language is English (Zill & West, 2000). Latino children are the largest
racial/ethnic population among children in the United States. Given the criteria used to
identify future academic risk, Latinos are over represented in “at-risk” categories. Living
in poverty ampliﬁes the barriers and difﬁculties faced by Latino students in school. In
fact, about 40 percent of Latino children live in families with incomes below the poverty
line (Federal Interagency Forum on Child and Family Statistics, 1998). Among the
general Latino population(s) not all students have similar backgrounds, motivations and
perceptions towards school (Reyes & Valencia, 1993). But as a group, Latinos do indeed
have lower levels of educational attainment as compared to their African American and
non-Hispanic Caucasian peers (Meir & Stewart, Jr., 1991; US. Department of Education,
1992). Signiﬁcant gaps between high school graduation rates of Latino and non-Hispanic
students have been found even after holding students’ social class, English-language
proﬁciency, and immigrant status constant (Weiner, Leighton, & Funkhouser, 2000).

Latino youth have been grossly underrepresented in research literature, and what
literature they are included in, is often plagued with methodological and conceptual
issues that misinforrn our understanding of their developmental pathways (Villarruel,
Dunbar, Montero Seiburth, & Outley, 2005). There is even less on subgroups of Latino
youth, such as those from different countries in Latin America and the Caribbean,
different socioeconomic strata, and different generations and levels of acculturation,

despite some data highlighting the within group differences that exist with regard to

educational attainment and SES among Latinos (Boswell, 2002). Moreover, the majority
of the information that does exist oﬁen compares Latino and non-Latino youth, helping to
reinforce perceptions that Latinos under-perform in an array of areas. The unfortunate
consequence of this focus is that the literature tends to emphasize what is wrong, rather
than what might be done to enhance the developmental outcomes for Latino youth. We
know far more about the failure of Latino students and very little about their success.
Researchers have focused on deviant behaviors rather than looking for normative
behavior patterns for Latinos (Garcia Coll, Meyer, & Brillon, 1995).

The knowledge have of why Latino youth fail has not produced policy,
programming or educational reform which have been able to positively alter the
trajectory of these youth (President's Advisory Commission on Educational Excellence
for Hispanic Americans, 2000). Development of a success paradigm, examining Latino
students who do well in school and considering how they differ from less successful
Latino students, can greatly improve our understanding of the developmental pathways of
Latino youth and support practical efforts that increase their opportunities for success. In
order to expand our grasp of normative Latino development and to create a thriving
Latino community, it is important to know why these resilient students succeed while
other Latino students (i.e. nonresilient students) from equally stressful environments
experience greater failure and despair in school. The investigation of resilient students,
youth who succeed academically despite the presence of adverse conditions, has
important implications for the educational improvement of Latino students (Gordon &
Song, 1994; Masten, 1994; McMillan & Reed, 1994; Wang & Gordon, 1994; Winﬁeld,

1991). Masten (1994, p. 46) articulates this point succinctly:

The rationale for examining resilience phenomena rests on the fundamental
assumption that understanding how individuals overcome challenges to
development and recover from trauma will reveal processes of adaptation that can
guide intervention efforts with others at risk.

Academic success has been deﬁned throughout the literature in a variety of ways
and is a concept that has yet to be readily explored from a Latino perspective. The
following review begins with an historic overview of the experiences of Latinos within
the US. education system. Following the historic overview, theories from the ﬁeld of
education and human development that have contributed to our understanding of
academic development of Latino youth are brieﬂy examined. An adaptation and
augmentation of an existing model for understanding the academic development of
Latino youth is then presented with supporting literature, research, and examples from
successﬁil programs and interventions on the underlying factors associated with academic
outcomes (e. g. grades, GPA, graduation, attainment, test scores, etc.). Included in this
review is research regarding the ways in which Latino youth resemble all other youth, as

well as research that highlights their unique strengths.

Latino Youth in the United States

Latinos represent the fastest growing ethnic minority population in the United
States according to the most current US. Census Bureau. From 1990 to 2000, the Latino
population increased by 57.9% (Guzman, 2001). The most recent US. Census (Greico &
Cassidy, 2001) found that Latinos comprise 13% of the total national population.
According to the US. Census Bureau, in October of 1996, for the ﬁrst time in the history

of the United States, there were more youth of Latino origin than any other ethnic or

racial group except non-Hispanic White youth (V illarruel & Montero-Sieburth, 2000).
The Latino youth population encompasses a variety of cultures, histories, and
experiences. Latino is a term used in the US to identify persons of Spanish-speaking
origin or descent who designate themselves as Mexican American, Chicano, Puerto
Rican, Cuban, or of some other Hispanic Origin (Carrasquillo, 1991). This ethnic group
is comprised of individuals from diverse racial backgrounds and diverse countries of
origin, including South America, Central America, and the Caribbean; some who have
immigrated here and some whose families have been here for. As a whole, “Latinos
represent the fastest growing segment of the US. population under the age of 21”
(Perkins & Villarruel, 2000, p. 83). However, it should be noted that differences between
Mexican Americans, Puerto Ricans, and Cuban Americans (the largest Latino subgroups)
exist. Whenever possible, this review reports research ﬁndings according to the
nationality of the youth involved.

Studies have found that Latino students are less likely to graduate from high
school than other students (Dryfoos, 1998; Meir & Stewart, Jr., 1991; US. Department of
Education, 1992). In fact, Latino students are more likely than White students to possess
one or more of the characteristics, such as living in poverty, living in single-parent
families, parents who have lower educational levels, and attending lower quality schools,
that increases their risk of academic failure (Davalos, Chavez, & Guardiola, 1999;
National Commission on Children, 1991). Additionally, Latinos across the country have
been found to have lower levels of educational attainment as compared to other groups
(Meir & Stewart, Jr., 1991; US. Department of Education, 1992). Moreover, Latino

students are often viewed as underachievers, illiterates, and dropouts regardless of their

socioeconomic background (Light, 11. d.; Mercado, 2001). While this image of Latino
students persists, some are successful academically despite adverse conditions and risky
environments. Many Latino youth are doing well, and succeeding despite the fact that a
disproportionate number of their families must contend with poverty, segregation, racism,
discrimination, culture clash, and other stressors. Our understanding of the diverse
backgrounds, motivations, and perceptions towards school is limited due to the lack of
research on, about, and including Latino youth.

History of Latinos in US Education

When it comes to the educational problems faced by Latino students in the United
States, educators, researchers, and policy makers have offered all manner of solutions.
But these movements have rarely considered the needs of Latino students (Reyes &
Halcon, 2001). As Diaz and Flores (2001) put it, “on the whole, school has not been a
very successﬁrl place for these students despite extensive ‘remedial’ efforts by
educators...the failure of these efforts is due to the deﬁcit perspective inherent in them”
(p. 29). Educators have often appropriated the views of policy makers and ascribed the
poor academic performance of Latino students to their cultural, socioeconomic, and
linguistic differences (Diaz & Flores, 2001).

There has also been a tradition in “. . .our nation’s history of negating the political
nature of education” (Bartolomé & Balderrama, 2001, p. 51). In the 1920s and 1930s, for
example, there was widespread use of IQ testing of Spanish-speaking students, in
English, which served as a means of placing a disproportionate number of Mexican
children in remedial education programs (Halcon, 2001). Historically, the majority of

Latino children (many of whom are American born) not only looked different from

mainstream children, but also spoke a language different from that used in schools.
Furthermore, their parents adamantly persisted in maintaining their culture and language
in spite of serious attempts by the schools to assimilate and Americanize the children
(Donato, 1997). Latinos have been viewed as irresponsible, dependent on others, dirty,
stupid, lawless, and Spreaders or disease, and thus unsuited to attend schools with Anglo
children (Donato, 1997). The popular and common solution to what at the time was seen
as the “Mexican problem” was to segregate the children into “Mexican” classrooms and
schools that provided a strict diet of English-language instruction, a rigid mainstream
curriculum, and corporal punishment for speaking Spanish (Carter, 1970; Carter &
Segura, 1979). The goal was to impose English at all costs, even if it meant forcing the
children to lose their language and culture (San Miguel & Valencia, 1998). These
historical trends persist in adapted ways in today’s educational systems, and certainly
have affected Latinos perceptions and beliefs about the US. educational system as a
whole.

Today Latinos continue to be deﬁned as deviating from the norm, thereby
justifying educational policies that place them at the margins of mainstream school
curriculum (Halcon, 2001). As Diaz and Flores (2001) state,

In schools there is often the view that these students are deﬁcient because they

come ﬁ'om poor or working-class backgrounds or have a different language and

culture. Viewed from this perspective it is easy to see why school does not enrich
them. Rather, it creates an artiﬁcial academic “deﬁciency” that is a function of the

attitudes and practices embedded in the social interactions of classroom lessons

(p.45).

Latino students are thus labeled at-risk or high-risk regardless or socioeconomic level or
acculturation status (Gutierrez, 2002). Everyday these students face institutional
practices, “such as devaluing of their home language and culture, or outright hostility and
denigration by society, which must be addressed in order to create favorable educational
circumstances” (Moll, 2001, p. 23).
Current Educational Status of Latinos

Concerns about the performance of Latino students in public schools are well
warranted. Today, Latinos comprise 15% of the elementary school-age population (5-13)
and 13% of students in secondary education (U .8. Census Bureau, 1999). Additionally,
by 2020, Latinos are expected to make up 20 percent of all US. children (NCES, 1998).
Young Latinos and Latinas have the highest dropout rates as compared to Aﬁican
Americans and non-Hispanic Caucasians, (NCES, 1999; Provitera Mcglynn, 2001).
Overall, Latino students perform below the national average in the National Assessment
of Educational Progress (NEAP) (NCES, n. d.). Some studies have found that gaps
between Latino students and their peers emerge as early as the third grade, and continue
to widen in subsequent years (Gandara, 1996). Additionally, the low high school
completion rate for Latinos has not changed substantially in several years. In 1998, the
high school completion rate for Hispanics was only 63% (NCES, n. d.). Meeting the
needs of this growing population continues to be a major challenge for today’s teachers,
who are primarily non-Hispanic Caucasians, middle-class, monolingual English speakers
(Berzins & Lopez, 2001). The lack of academic success for Latino youth is particularly
alarming given that education has been closely tied to long-term outcomes such as

employment, welfare use, and health status. Further, there is a dearth of information

available on empirically tested factors that inﬂuence the academic success of Latino
youth. However, there are a wide variety of theoretical and conceptual models in
education and human development that purport to explain the academic success of youth

in general.

Theories of Development and Educational Models of Academic Success

In order to support the model to be tested in this study, theories of development
and education are reviewed. Theories of development and educational models have
identiﬁed factors that should be attended to and have been found to impact students’
academic success. Theories and models reviewed here include meritocratic ideology,
political explanatory models, bilingual education model, economic and cultural capital
models, theory of resistance, academic failure and success factor explanations,
psychoanalytic theories, social learning theory, cultural
determinism/relativism/conditioning theories, ecological and contextualist theories.

Meritocratic ideology and economic model explanations for achievement reﬂect
the belief that in school, the talented are chosen and moved ahead on the basis of their
achievement alone. What is largely ignored is that those deemed “talented,” for the most
part, are members of the dominant culture whose values comprise the very foundations
that inform the knowledge and skills students must possess or achieve to be designated as
meriting reward (Darder, 1991). Similarly, the dominant approach to human
developmental models of competence have been built upon assumptions, including that a
child’s “school success and perhaps success in adult life depend on the acquisition of
white middle-class competencies through white middle-class child-rearing practices”

(Ogbu, 1981 , p. 414) and presumed universal laws of optimal development (Garcia Coll,

Meyer, & Brillon, 1995). Public schools persistently legitimize the myth or meritocracy,
knowingly or otherwise, and in that way successful participation in the educational
system becomes the only visible means by which individuals are allocated or rewarded
higher social status. Although there are many talented Latino students, they are not
bestowed the label nor perceived as having the attributes necessary to succeed.
Meritocratic ideology and economic model explanations continue to shape and support
many educators’, parents’, and society’s view of why Latino youth have lower rates of
academic success than their peers. It creates a belief system that can dismiss Latino
youth’s lack of academic success as an individual predicament, a lack of ability or
appropriate experiences, and not a systemic problem.

Several theories of human and personality development focus on key
characteristics that may be important for understanding the development of youth in
general, including Latinos. Perkin’s (1981) snowﬂake model of creativity for example,
focuses on the individual traits including a strong commitment to a personal aesthetic, the
ability to excel in ﬁnding problems, mental mobility, a willingness to take risks,
objectivity, and inner motivation, which creative individuals’ posses. Psychoanalytic
theories, such as Erikson’s (1950), have also detailed the intemal/individual factors of
youth that affect behavior in school and academic success. The social and behavioral
problems that adolescents encounter, such as acting out behavior and dropping out of
school, are seen as reﬂective of difﬁculties in resolving earlier life stages and an
adolescent’s experiencing role diffusion/confusion, including a desire for independence, a
need to frnd a job, a search without a goal, and an expression of inner discontent and

restlessness (Erikson, 1950; Muuss, 1962; Muuss, 1996). The task of searching for a

10

personal identity is more difﬁcult in a historical period of rapid social change, such as the
one we are currently experiencing, where the older generation no longer provides
effective role models, and the peer group becomes of central importance (Erikson, 1950).
Although Erikson (1950) never speciﬁcally addressed the identity development of Latino
youth, their search for a sense of identity and desire for independence can be seen as
affected not only by their changing roles within the US culture but also by values and
cultural roles within their native culture that may clash with mainstream US culture. They
must create a sense of identity while facing depreciation of their culture, racism, and
prejudice.

Phinney (1989) has expanded Erikson’s (1950) identity development constructs
into the area of ethnic identity development. Ethnic identity is “a secure commitment to
one’s group, based on knowledge and understanding obtained through an active
exploration of one’s cultural background” (Phinney & Chavira, 1992, p. 272). Phinney
and Alpuria’s (1990) study found that achieving a positive sense of ethnic identity was
related to self-esteem for Mexican American adolescents. Bemal, Saenz, and Knight
(1991) found that a positive sense of ethnic identity may promote achievement by
buffering the psychological stressors experienced by Mexican American youth in school
settings. Behavior that promotes academic success is likely even in instances in which
cultural disparities exist between home and school if academic success is embedded
within students’ ethnic identity. They also suggest that if academic success is not valued
within students’ ethnic identity, then cultural inversion, the tendency of a minority group
to see behaviors, events, and meanings of the dominant group as irrelevant to them, may

occur (Bemal, Saenz, & Knight, 1991).

ll

Bandura’s (1977) social learning theory puts forward that adolescents learn
complex skills most effectively by imitating the behavior of their parents, teachers, and
peers. Teachers serve the model function and, in addition to the inﬂuence they exert
through cognitive, instructional academic curriculum, they have an indirect but potent
inﬂuence in shaping values and attitudes. An individual student may feel respect,
admiration, infatuation, and adoration of an individual teacher and imitate a teacher’s
behavior, even when a more general identiﬁcation with the teacher as a person is lacking
(Muuss, 1996). Additionally, when students are in academic settings where peers value
learning, take school seriously, and aspire to academic success, their own attitudes and
behaviors will be similar to the extent that they accept and imitate this academic striving.
The demand that Latino students be taught by Latino teachers, because they tend to
identify better with them, receives some implicit support from social learning theory’s
concept of imitation of esteemed models (Elizondo, 2005; Hernandez, 2000; Monzo &
Rueda, 2001). Bandura’s (1977) theory also suggests that Latino youth are more likely to
be academically successful when attending a school with Latino teaching faculty and
administrators, as well as other Latino students who value schooling and demonstrate
behaviors related to academic success.

Ecological and contextualist theories go beyond individual characteristics and
focus on the necessity of considering behavior and development as occurring within a
context and on understanding the interdependency of individual and contextual factors.
Ecological theories of human development, such as Bronfenbrenner’s (1979, 1983),
Lemer’s (1983, 1986), and Lewin’s (1935, 1942) have at their center the individual who

develops within varying contexts. Within these models, culture, ethnicity, and race are

12

viewed as critical dimensions of growth and development, underlying the development of
identity, belief, cognition, and social interactions. Bronfenbrenner (1993) has expressed
concern regarding the growing isolation between the family and school systems, such as
the decreased likelith of teachers and parents knowing each other. Another important
factor within Bronfenbrenner’s (1979, 1993) model is the educational system (i.e. school
board or school committee), which sets school policy, directly affecting the resources
within, and the types of interactions that occur inside of school.

Lemer’s (1986) developmental contextualism purports that relevant variables
concerning academic success and other issues within educational situations include more
than just individual characteristics, such as age, gender, past experiences within
educational settings, academic self-concept, and temperament. Developmental
contextualism also highlights the goodness-of-ﬁt between adolescent and the task (i.e.
homework, projects, or tests) or between adolescent and parents and/or educator.
Goodness-of-ﬁt refers to the circumstances that enhance development and adjustment or
impair growth and well-being. Goodness-of-fit takes into consideration the relationship
between an individual’s personality and the corresponding characteristics of the
signiﬁcant other peOple who constitute the social context within which behavior occurs
(Lerner, 1983; Muuss, 1996). The developmental outcome of an interaction is mostly
dependent on the match between individual and context.

Kurt Lewin’s (1935, 1942) ﬁeld theory emphasizes for successful teaching the
social atmosphere, the amount of security, the interest and meaningfulness of the
material, and the atmosphere in which the material is taught. Additionally, students’

perceptions of reality are of key importance. Two aspects, according to ﬁeld theory, that

13

will be relatively stable within a given culture but will diﬁ’er greatly between cultures are:
(l) the ideologies, attitudes, and values that are recognized and emphasized and (2) the
way in which different activities are seen as related or unrelated (Lewin, 1942). Lewin’s
(1942) theory also states that the varying length of the adolescent period between cultures
and between social classes within a culture may account for cultural differences in
adolescent behavior, including academic performance.

Ecological theories imply a complex set of relationships that affect Latino
students. As suggested by ecological theories, peers valuing education and academic
success, more so than parents valuing of education and academic success, has been found
to be predictive of whether Latino adolescents also feel and behave in a manner
consistent with those beliefs (Azmitia & Cooper, 2001; Romo & Falbo, 1996; Steinberg
& Darling 1993). A disconnect between Latino parents and their children’s teachers may
exist due to cultural, linguistic, and socioeconomic differences. The deﬁnitions of
adolescence and the time span of adolescence may be different for the native culture of
Latinos and the US main culture and may be different for members of Latino youth’s
social class and the social class of their educators. A limited amount of research on
goodness-of-ﬁt in classroom settings and teaching styles has found cooperative classroom
cultures, which reﬂect Latino student’s home culture more than traditional classroom
cultures, increase individual learning, improve student attitudes toward school, and
decrease racial tension (Chiu, 1996; Gutierrez, Baquedano-Lopez, & Alvarez, 2001).
Latinos’ involvement at the larger educational systems level is often inﬂuenced by
racism, discrimination, social stratiﬁcation, and politics, thereby decreasing the

likelihood of their having meaningful impact on the system. Additionally, Latino

l4

adolescents who transition from one set of core values from their native country society
to another set of core values in the dominant society of the United States may face
conﬂicting societal values, including gender roles and collectivist versus individualistic
values. Similarly, a disconnect between values and beliefs held by family members,
peers, and the larger societies may cause difﬁculties for Latino adolescents trying to
balance conﬂicting messages or who decide to take part in behaviors (e. g. gang
involvement) considered to be antisocial by the overall culture. Being in possession of
two or more diverse ways of perceiving reality, acting, and conveying meaning, Latino
students have to make choices more frequently (Lima & Lima, 1998).

Similar to other ecological and contextualist models, Valsiner’s (1987, 1988)
cultural-historical and Vygotsky’s (1962, 1978) socio-historical life course theories place
an emphasis on the previous experiences of the developing child and those they interact
closely with, such as parents, teachers, and peers, along with the current societal patterns
and phases that inﬂuence development over time. The child’s life course is seen as the
continual reconstruction of the self or personality as the result of interactions among the
child’s state of biological maturation, existing knowledge and skills, and new encounters
with the environment (Elder, 1996; Elder, Modell, & Parke, 1993). Important among
these encounters are mediators’ transmitting to the child signiﬁcant societal conditions
(Thomas, 2000). Schools are also conceptualized as “cultural settings,” with special
routines and forms of discourse, where adults help children acquire important mediational
means (e. g. literacy and mathematics) of a culture, systems of communicating and

representing knowledge, extending and restructuring the children’s communicative and

15

cognitive abilities (M011, 2001). Thus school characteristics are seen as playing a large
role in learning and academic success.

These theories suggest that the values, bodies of knowledge, and customs of both
the US mainland and native culture of Latino youth, parents/caregiver’s previous
experiences within educational settings and level of educational attainment, family
culture, and Latino youth’s accumulated experiences within educational settings will
deﬁne their perceptions of the importance of academic success and the value of
education. Expectations of and the historical perception of Latino students as poor
learners, and the possibility of racism and discrimination playing a role within teacher-
student relationships, may mean that teachers as mediators may be sending either mixed
or incomplete messages to Latino students in regards to acquiring academic skills.

At-risk explanations and success factors explanations focus on student
vulnerability and resiliency. Educational resilience has been deﬁned as “the heightened
likelihood of success in school and other life accomplishments despite environmental
adversities brought about by early traits, conditions, and experiences” (Wang, et a1. 1999,
p. 46). Alva (1991) used the concept of academic invulnerability to describe students who
“sustain high levels of achievement motivation and performance, despite the presence of
stressful events and conditions that place them at risk of doing poorly in school and
ultimately dropping out of school” (p. 19). From this perspective, it is the balance of
vulnerability, or risk factors, and resiliency, or protective factors in Latino students lives

that deﬁnes how academically successful they will be.

F reire’s (1972) approach to education focuses on the importance of “naming the

world” for those educators who work with students who do not have a voice and are

16

oppressed, and on children learning best when their native language and culture are used
as the springboard for learning educational activities (F reire, 1972; Freire & Macedo,
1987). Additionally, teaching is seen as a political act (F reire, 1972; Freire & Macedo,
1987). Similarly, a major underpinning of Meier and Stewart’s (1991) model of second-
generation discrimination is that education is a political process. Latinos, as political
minorities in the US. face limited opportunities for educational equity. The educational
system often does not support students’ language and culture, but has instead tried to
“Americanize” them (Meier & Stewart, 1991, p. 83). In their research, Meier and Stewart
(1991) found that school districts with greater Latino representation on the school board
and among teaching faculty had signiﬁcantly less second-generation discrimination
against Latino students. Educational policies and reform efforts such as bilingual
education, grouping, and differential assignment can serve to limit Latinos access to
educational opportunity. Given that Latino students come to school with a history of
oppression and having been socialized to a cultural system that is not entirely that of the
dominant culture, indeed is often dismissed or devalued by the dominant culture, they

will be less likely to “read the world” without the support of educators.

Cultural capital, cultural deﬁcit, cultural difference, and culture determinism
theories all center cultural factors in explaining the development of academic
competencies. Bourdieu’s (1997; Bourdieu & Passeron, 1977) cultural capital theory
explains the cultural differences that reproduce social class division. It emphasizes how
children socialized into the dominant culture have an advantage over children not
socialized into this culture because schools are one of the main transmitters of culture and

tend to reward the cultural capital of the dominant classes and devalue that of the lower

17

classes. Latino parents seldom share the same knowledge and disposition that constitute
cultural capital attributes. Cultural deﬁcit theory posits that Latino students’ lack of
academic success is due to problems in their culture, or the lack of the proper experiences
within their cultural context (Montero-Sieburth & Batt, 2001; Valencia & Black, 2002).
This perspective states that Latino students do not achieve because they are not reared in
cognitively stimulating environments and are “socially disadvantaged” or “culturally
deprived,” and allows educators and policymakers to place the responsibility for school
success and especially failure outside the school (Foley, 1997; Montero-Sieburth & Batt,
2001; Valencia & Black, 2002). Often students previously called “culturally deprived” or
“culturally disadvantaged” are today labeled hi gh-risk or at-risk (Gutierrez, 2002).
Cultural difference theory or cultural relativism blames neither the teacher nor the
student, but instead places the responsibility for failure on cultural differences in
communication styles between teachers and students. Latino students and their families
are not blamed for the cultural differences, but it has done little to change the

circumstances within which these differences occur.

Cultural difference model as applied to bilingual education exposes much of the
current ambivalence that many Latino students face in school (Lopez, 1995). Bilingual
education is largely concerned with teaching students English, so that student and teacher
communication styles are similar, and has focused on what works rather than what makes
sense for students and families. Parents may promote language retention while schools
promote language assimilation, and Latino students are caught in the middle. In this
respect, the cultural difference explanation as applied to the bilingual education model

has only created greater discontinuity in linguistic experiences at home and at school, and

18

coupled with other cultural differences, serves to confound Latino students and diminish

their academic aspirations (Montero-Sieburth & Batt, 2001).

Cultural determinism, also known as cultural relativism or cultural conditioning
(Benedict, 1938 as cited in Muuss, 1980), emphasizes that advancing from one grade or
group to another introduces anxiety and insecurity. Moving from one type of school or
school system to another creates transitions that lack continuity and coherence. These
disjointed transitions, often the norm for Western societies, require that the attitudes,
values, and skills children have learned must be unleamed when they become adults. The
development of Latino adolescents, like that of all adolescents is inﬂuenced by the
biological changes occurring in puberty, but they face the additional challenge of
discontinuities found not only in the dominant culture and their native culture, but the

clashes between these as well.

Ogbu’s (1981, 1999) model of human development stresses a cultural contextual
component. Unlike other models, it examines competence in the context of cultural
necessities or constraints in a given population. Competence is the ability to perform a
culturally speciﬁed task, and Ogbu’s (1981, 1999) model focuses on the roles and tasks
of making a living. Children are taught these roles and tasks not only by their parents,
but also in school and in their communities. Differences in societal needs and cultural
tasks exist for populations within the United States, such as middle-class whites and other
minority populations. Ogbu (1981, 1999) posits that voluntary and involuntary minorities
are populations within the US that may have differing cultural tasks as related to
subsistence demands. Voluntary minorities are immigrant groups, such as Cuban

Americans, South and Central Americans, who have historically moved to the United

19

States of their own free will often for economic, social, or political reasons. Voluntary
minorities, despite possibly facing subordination and exploitation, perceive and react to
schooling positively because they regard their current situation in the US. as better than
their situation in their country of origin. In contrast, involuntary minorities, such as
Mexican Americans and Puerto Ricans, are groups who have historically been
involuntarily and permanently incorporated into US. society through slavery, conquest,
or colonization. The model suggests, similar to Paul Willis’s (1972) theory of resistance,
that involuntary minorities are unlikely to work hard in school because they do not wish
to assimilate and because they recognize that, relative to whites, they have limited
opportunities for beneﬁting from education.

While Latinos and white middle-class youth may have similar goals, they may
differ in their belief that the appropriate school credentials will result in these desired
goals. Rules of behaviors for achievement and related competencies for many Latinos
have been developed historically as alternatives to those of the school that represent
white middle-class cultural ways whose racial policies and practices have prevented
generations of Latinos from using these same rules of behavior and competencies to
obtain the desirable adult tasks open to whites. Latino youth, especially those from
involuntary minority groups, may consider “acting white” in school and community
unacceptable, and may form coethnic peer communities that perpetuate this perception
for second generation Latino students from Latino ethnic groups not traditionally
considered involuntary minorities (Portes & Rumbaut, 1996; Portes & Zhou, 1993).
Additionally, involuntary minorities do not have a dual frame of reference that is

available to voluntary minorities, which allows them to view their situation and

20

educational experiences in contrast to the obstacles that were faced “back home” (Suarez-
Orozco, 1987). Ogbu’s (1981) conclusion is that “minority groups experience a
continuing disproportion of school failure mainly when their historical and structural
relationship with the dominant group has led to evolution of alternative competencies” (p.
426)

Recently, some of Ogbu’s (1981, 1987) ideas regarding the patterns of
assimilation and acceptance of US mainstream culture by Latino immigrants has been
called into question. Segmented assimilation theory in particular presents an interesting
model that counters Ogbu’s (1981) original contentions regarding immigrant groups
(Portes & Zhou 1993). For example, a recent study by Bleakley and Chin (2004) found
that children of immigrant Hispanic parents with lower English-speaking proﬁciency are
more likely to drop out of high school or below their age-appropriate grade. Hispanic
immigrants, despite being what Ogbu (1981, 1987) would consider voluntary minorities,
and their descendents do not converge to native levels of education as quickly as non-
Hispanic immigrants and their descendents (Card, DiNardo & Estes, 2000; Grogger &
Trejo, 2002; Smith, 2003). In fact, Bleakley and Chin’s (2004) study found that parental
English-language skills can account for 60% of the difference in dropout rate between
non-Hispanic whites and U.S.-bom Hispanic children of immigrants. The Census 2000
Supplemental Survey found that 65% of U.S.-bom Hispanics retain a non-English
heritage language in the home. This statistics along with research on English proﬁciency
is a striking example of the persistence of heritage language among Latinos. It counters
Ogbu’s (1981, 1999) assertions that voluntary minorities will more quickly assimilate and

adopt U.S. mainstream values and behaviors.

21

Building and expanding on previous theory, Cynthia Garcia C011 and her
associates (Garcia Coll et al., 1996) created a model of human development that is
“anchored within social stratiﬁcation theory and emphasizes the importance of racism,
prejudice, discrimination, oppression, and segregation to the development of minority
children and families” (p. 1892). This model has guided previous research and further
conceptually developed as a framework to examine various areas of child and youth
development (Erkut, Szalacha, & Garcia Coll, 2005; Johnson, J aeger, Randolph, Cauce,
& Ward, 2003). Figure 1 below presents this model.

The model consists of eight major components connected by pathways of
inﬂuence. These eight major components are 1) Social position variables; 2) Racism; and
3) Segregation, which form the social stratiﬁcation structure; these affect 4)
Promoting/Inhibiting Environments that are the intermediate societal institutions; these
environments inﬂuence and interact with 5) Adaptive Culture; and both of these interact
with the 6) Child’s Characteristics; all three of these components inﬂuence and interact
with the 7) Family; it is these last three components that directly inﬂuence children’s 8)
Developmental Competencies. This model challenges researchers to take into
consideration issues of social stratiﬁcation, including racism and oppression, as well as to
identify the alternative competencies in Latino youth not measured by traditional
assessment tools, not only in areas of established developmental competencies but in

areas of bicultural adaptation and coping with racism as well.

22

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#6
Child Characteristics
Age
#2 Temperament
Racism Health status
Biological factors
Prejudice Physical characteristics
Discrimination
Oppression \
/ \ #5
#1 #4 Adaptive Culture #8
Social Position Promoting/Inhibiting Traditions and legacies Developmental
"mam“ E"”"°””'e"“ Economic/political 4 commune”
Race 1 histories C0 itive
Socialmlty Earth 5 “““l‘m‘i” Emotional
Gen: “8 cm C‘m'“ ”mam" Linguistic
er / demands
a
#3
Segregation #7
Residential Family
155062;):3 Structure and roles
s cholo 'cal Values, beliefs, goals
p y 3’ Racial socialization
Socioeconomic status

 

 

 

Figure I. Integrative model of development competencies in minority children (Garcia

Coll, et.al, 1996).

Effective Academic Programs for Latinos

Evaluations of programs aimed at increasing the academic success of Latino
youth may provide insight into the development and empirical testing of a comprehensive
model of Latino academic success. By examining the components of programming that
have been effective, their impact, and the populations with which they have produced the
greatest results we may gain an increased understanding of how to aim future efforts.

Successful programs aimed at Latino students offer insights into multiple factors

at different levels of school systems that are important for their academic success.

23

NCES’s (2001) review of programs for increasing college-going rates of Latino and other
minority students, found that the most effective programs provide a key person, including
guidance counselors, who monitors and guides students over a long period of time. The
most effective programs also provided high-quality instruction through access to the most
challenging courses offered by the school (“untracking”), through special coursework
that supports and augments the regular curricular offerings, or by revamping the
curriculum to better address the learning needs of the students. Paying attention to the
cultural backgrounds of students was also a key component. These early intervention
programs show evidence of doubling the college-going rate of their participants, but do
not appreciably alter their academic achievement because they augment and supplement
what schools do without fundamentally changing the ways schools interact with students
(NCES, 2001).

Weiner, Leighton, and Funkhouser (2000) in their review of programs aimed at
helping Latino students reach high academic standards found that successful programs
share some key features: 1) they provide curriculum instruction that lead to mastery of
standards set for all students in forms that accommodate the particular resources and
needs of Latino students; 2) they offer special support for Latino students who are
English-language learners; and 3) programs serving migrant students tailor their services
to enhance continuity and progress in the educational experience. Chavkin and
Gonzalez’s (2000) review of research and programs addressing Mexican immigrant youth
and resiliency, also found that relationships with caring adults were associated with

resiliency.

24

Researchers at the Center for Research on Education, Diversity, and Excellence at
the University of California, Santa Cruz have similarly identiﬁed ﬁve principals to
govern programs intended to assist Latino students achieve high standards (Rueda, 1998;
Tharp & Gallimore, 1988): 1) joint productive activity; 2) reading and language
development is embedded in the curriculum; 3) connections to everyday life; 4)
challenging expectations; and 5) instructional conversations.

Research on family and community involvement in children’s education suggests
that three key strategies may be particularly effective in promoting Latino families’
participation in school-related activities: 1) bridging language and cultural differences
between school and home; 2) moving beyond traditional school-family activities; and 3)
providing training of parents and staff for effective partnerships (U .S. Department of
Education, 1997).

Although there are a myriad of programs that aim to improve the academic
success of Latino youth, evaluating which components contribute to their effectiveness is
hampered by several factors. The connections between research and practice are not
strong, in large part because of the lack of empirical research involving Latino youth and,
in the existing limited research, absence of theory. The lack of evaluation regarding
program impact means that much of what we know of Latino youth programming is
anecdotal and relatively untested. Additionally, understanding how these programs may
affect Latino students of different nationalities, acculturation level, or immigrant status is
virtually impossible as most studies do not specify these variables. For those programs

that do serve Latino youth, in the context of diverse communities, we know little about

25

what makes them successful, if in fact they are successful with Latino youth (Rodriguez

& Morrobel, 2002).

Conclusions

Although it is well known that Latinos account for most of the grth in the US
youth population (Chapa & Valencia, 1993; Ramos, 2002), research focused on their
development is sparse (Rodriguez & Morrobel, 2002). The academic success of Latino
youth, in particular, has gained national focus. As the largest youth group in the United
States, the academic success of these youth will impact the entire nation. The negative
academic issues facing Latino students have been well documented: low academic
achievement, low levels of high school graduation, and low educational attainment
(Dryfoos, 1998; Meir & Stewart, Jr., 1991; Romo & Falbo, 1996; US. Department of
Education, 1992). Although we may know how many Latinos complete high school or
college, or how many are located in various levels of poverty concentration, we are no
closer to understanding the academic development of Latino youth because of it
(Rodriguez & Morrobel, 2002; Villarruel, Dunbar, Montero Seiburth, & Outley, 2005).

The tradition of focusing on and describing the problems and deﬁcits that
characterize Latino youth is long and deep (Garcia Coll, Meyer, & Brillon, 1995;
Gutierrez, 2002; Light, n. d.; Mercado, 2001; Rodriguez & Morrobel, 2002). Research
and theory focused on negative aspects of Latino youth development, and Latino
academic success in particular, has been based on a relatively unchallenged assumption
that there are baniers that must be overcome to achieve successful youth development
(Diaz & Flores, 2001; Ogbu, 1981; Rodriguez & Morrobel, 2002; Zill & West, 2000).

Often these barriers, sometimes referred to as cultural, socioeconomic, and linguistic

26

differences, are seen as naturally occurring or pre-existing in the lives of Latino youth
(Diaz & Flores, 2001; Rodriguez & Morrobel, 2002). This has led us to another historical
tradition and unfortunate orientation to promote intervention and prevention efforts
(Carter, 1970; Carter & Segura, 1979; Halcon, 2001). Such an orientation is unfortunate
because intervention and prevention presumes that there are universal laws of optimal
academic development and the existence of negative, harmful, or life-threatening
behaviors and conditions that require intervention or the development of prevention
strategies (Garcia Coll, Meyer, & Brillon, 1995; Rodriguez & Morrobel, 2002). Efforts
based on this orientation have already been proven ineffective in positively altering the
acadenric developmental trajectory of Latino youth (President's Advisory Commission on
Educational Excellence for Hispanic Americans, 2000; President’s Advisory Commission
on Educational Excellence for Hispanic Americans, 2003).

The need to develop coherent, contextual, and culturally relevant empirical theory
regarding Latino youth development and academic success drives this work. The lack of
research guided by a normative model of development of competencies as related to
academic outcomes is a glaring omission in the literature on Latino youth (Torres
Campos, 2003). In her review, McLoyd (1998a) also found that empirical studies were
rarely guided by a conceptual or theoretical framework and were primarily exploratory.
She also noted that researchers rarely explained why race or ethnicity should matter. The
research and literature on effective programming for Latino youth has a similar void, in
that it lacks systematic and consistent evaluation and information regarding effectiveness
by nationality, acculturation level, and immigrant or voluntary/involuntary status. Further

research that examines the most effective types of strategies for various groups of Latino

27

children and youth is still necessary. This study promotes an alternate orientation to that
used in the majority of existing research and focuses on youth development as a primary
strategy for studying, intervening, and ultimately positively altering the academic
outcomes of Latino youth.

The theories of human development and education that have been reviewed here
should be both critiqued and built upon in order to establish a model that more accurately
reﬂects the academic development of Latinos. It is important to note that the knowledge
base of child development has generally come from studies of middle-class White
families and based on Euro-American values and standards of behavior (Zayas, 1994;
Zayas & Solari, 1994). Proponents of the dominant model attribute Latino students’
failure in school to developmental “deﬁcits,” and they propose rehabilitation or reform to
correct these deﬁcits and enhance school success. Although meritocratic ideologies and
mainstream theories have dominated education, many theorists and researchers have
moved away from these explanations that use white middle-class values and parenting
patterns as the norm. This is due in part because despite decades of educational reform
Latino students have not seen the expected gains in academic success nor have these
reforms been successful “in permanently inculcating a white middle-class type of school
success” (Ogbu, 1981, p. 415).

Theorists such as Perkins (1981) and Erikson (1950) have identiﬁed several key
characteristics of individuals that are important for youth’s academic success. These
include characteristics such as intrinsic motivation and identity development. But these
theories do not fully explore how these characteristics develop differently or how they

may interact differently with environmental factors in the lives of Latino youth. Ethnic

28

identity is one area of adolescent development that has received signiﬁcant attention and
has been viewed as an aspect of personal identity (Phinney, 1989; Phinney & Alipuria,
1990). But this factor alone does not explain the development of academic competencies
for Latino youth and is highly related to other cultural, contextual, and familial factors in
Latino youth’s lives (Marin & Marin, 1991; Zayas & Solari, 1994).

Theories that view development as occurring within contexts and as
interrelationships between developing individuals and the systems surrounding them,
have expanded our understanding of the environmental factors and interrelationships that
impact the academic success of Latino youth. Theorists such as Bronfenbrenner (1979,
1983), Lerner (1986), Lewin (1935, 1942), Valsiner (1987, 1988), and Vygotsky (1962,
1978) make clear the importance of contextual factors on the development of academic
competencies in youth. These theories and models have generally included a dimension
of “culture” or “ethnicity” in their models of human development. But these theories have
fallen short of producing research or policy that fully takes into account the cultural
factors that affect academic development for Latino youth and often place these forces at
a more distal level in affecting development. Researchers have generally avoided
examining the impact of ethnicity, race, and to some degree, culture on human
development within their investigations. There are a variety of reasons offered for why
ethnicity and other such variables are omitted from research, including but not limited to
difﬁculties in measuring ethnicity and the general question of whether this is a primary or
secondary variable inﬂuencing development. But there is growing recognition within the

ﬁelds of human development that ethnicity must be central to future research endeavors if

29

responsive programs and policies are to be developed (Fisher, Jackson, & Villarruel,
1997)

Educational resilience perspectives, while focusing on the assets or positive
factors in Latino youth’s lives, have not yielded a clear working model or set of key
factors that explain Latino students’ academic success. Additionally, most of the research
using this perspective has not addressed the cultural factors that may impact the academic
development of Latino students. Nor is there a clear sense of how much weight individual
protective/risk factors have in youth’s lives with regards to acadenric outcomes. The
author, in previous research testing the utility of this model, found it was not capable of
distinguishing between academically hi gh-achieving and low-achieving Latino students
(Torres Campos, 2003).

Political explanatory models such as Freire’s (1972) and Meier and Stewart’s
(1991), expand the vision presented in both ecological contextualist theories as well as
educational resilience perspectives to highlight the importance of education as a political
act. These theories also focus our attention on understanding the political connotations of
not only the educational activities that take place within the classroom, but also the
political factors at various levels of development that impact Latino youth’s academic
success. Both of these theories have been unable to create a large empirical or political
movement within the US. to substantiate their effectiveness with Latino populations. In
conjunction with the factors identiﬁed in other theories and models, this political aspect
of education may be critical to understanding the normative development of Latino

academic competencies.

30

Theories which place cultural factors at their center, such as cultural capital,
cultural deﬁcit, cultural difference, and culture determinism while highlighting the
importance of cultural factors on development and particularly academic success, have
not yielded a normative understanding of how Latino youth develop. This is due in part to
these theories deﬁning Latino culture only in relationship to Western, mainstream U.S.
culture, while ignoring in-group differences or attending to Latino culture as a whole.
These perspectives have in some ways perpetuated the view of Latinos as exceptions to
the rule, and “at-risk” as a result. These perspectives continue to guide much of the
current research, programming, and educational policy/reform, despite the lack of
improvement for Latinos.

In contrast, Ogbu (1981, 1987) and Garcia C011 and her colleagues’ (1996)
models have integrated much of the previous research concerning the development of
minority, including Latino, youth into frameworks that explore development from the
perspective of these youth themselves and not in comparison to white, middle-class
values or structures. But Ogbu’s (1981, 1987) model cannot always be clearly applied to
Latinos given the complex set of ecologies in which they develop. Garcia C011 and her
colleagues’ (1996) model builds from Ogbu’s (1981, 1987) model as well as drawing
from other theories to provide a perspective that places important contextual and cultural
factors at the center of development.

From this review of the literature on Latino youth development, education, and
academic success, this study argues that ethical and practical considerations demand a
reorientation. Despite advances in scientiﬁc methodologies, the predominant trend in

social science research has involved the inclusion of race and ethnic identity differences

31

among less dominant racial and ethnic groups (Rodriguez & Morrobel, 2002). This
approach has served to foster and justify supposed differences, particularly, with respect
to the problematic aspects of adolescence (e.g., lower levels of attainment, academic
failure, and high dropout rates) and a tendency to ignore intragroup variability (McLoyd,
1998b; Rodriguez & Morrobel, 2002). As Vonnie McLoyd (1998b) so eloquently put it
“The relative absence of systematic research on normative development among ethnic
minority youth is a bit like the weather: Everyone complains about it, but no one ever
does anything” (p. vii). Consequently, relatively little is known about individual
differences among Latino youth or the sources of deviation from the norms of
development within the Latino population. The importance of attending to intragroup
variation in minority groups is underscored by research demonstrating that variables
which best explain differences within certain minority groups in academic achievement
are sometimes different from those that best explain intergroup differences (Howard &
Scott, 1981). Additionally, race-comparative studies that view Latino youth as abnormal,
incompetent, and change-worthy, draw attention away from the structural forces that
undermine their development.
Because they emphasize the race[Iethnicity] of the subjects (which often is
confounded with social class) or personal characteristics associated with
race[/ethnicity], race-comparative studies often promote person-blarne
interpretations of social problems rather than thoughtful analyses of the roles of
situational and systemic factors (McLoyd, 1990, p. 264).
In response to the overwhelming amount of deﬁcit oriented and exploratory research

conducted on Latino youth (Rodriguez & Morrobel, 2002), this study seeks to conﬁrm

32

and build a model of normative academic development and to challenge educators,
researchers and policy makers to turn historic trends and consider a more balanced and
strengths based approach. Speciﬁcally, Garcia C011 and her colleagues’ (Garcia Coll et
al., 1996) model has been adapted and augmented and was empirically tested focusing on
the development of competencies related to academic outcomes for Latinos. This
resilience research may serve to ﬁll the gaps in our current knowledge of Latino youth
development and as a foundation for future research, intervention, and educational
policy/reform.

Pathways to Latino Academic Success

The adapted and augmented model presented below (see Figure 2) follows the
original closely in terms of its major components. The components of the integrative
model have been adapted to include factors, described in detail below, that are
particularly important to the academic success of Latino students. An augmentation of the
model includes an outcomes factor related to the various ways in which academic success
is deﬁned.

This adapted, augmented model recognizes that Latino students’ development of
competencies related to positive academic outcomes will be greatly affected by the
context in which these youth develop. The key to understanding this model is that the
competencies that youth develop will be adaptive and normative based on their context.
Garcia Coll and her colleagues (1996) state that studies of Latino children need to move
away from conceptualizing developmental outcomes as either negative or positive to a
more balanced conceptualization that reﬂects both the strengths and weaknesses. This

adapted, augmented model provides a framework to empirically examine the

33

developmental processes and competencies of academic success for Latino youth in a

more balanced approach. Latino students are generally considered academically

successful if the normative developmental competencies that result from their contexts

match the competencies expected to contribute to success as deﬁned by mainstream US.

society. In order to positively alter the current disparate trends in academic success, it is

important to know not only what factors are related to the academic success of Latino

youth, but also how these factors interact to produce the competencies that lead to

academic success for Latinos. What follows is a brief overview of the model as a

framework for understanding the relations among the 9 components.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Academic
Success

GPA
High school
graduation
Achievernent/
attainment
Achievement
test scores

 

 

 

 

Socioeconomic status
Support

 

#2 #6
Social Stratiﬁcation Child Characteristics
Mechanisms Age
Perceived and Ethnic identity
experienced: Racism, English proﬁciency
Prejudice, Discrimination, Perceptions towards
Oppression education
I #4 #5 \
#1 Promoting/Inhibiting Adaptive #8
Social Environments Culture Developmental
Position School; (i.e. support, TM‘F‘O‘“ ”‘9 Competencies .
Variables oppoﬂuniti“ for legacies. Culture spoolﬁc and br-
Race positive use of time, _. Economlc/ cultural competencies
Ethnici climate, school Political Academic motivation
S 0C] a] ty engagement) hiStOfies Efforts towards
class/parent Neighborhoods; (i.e. Mimuonand academic mastery goals
educational support, opportunities acculturation Academic self-efﬁcacy
level for positive use of currenttual Slmlll’oeifnon
time, climate/safe ) contex C orrnance
Gender ty demands Attendance
l’ / \ l-
#3 #7
re ation
Residing] F amily/Peers
Economic Perceptions, values,
- . beliefs and goals
Soc all h 1 a1 '
‘ psyc o ogrc towards education

 

 

Figure 2. Adapted, augmented model of Latino academic developmental competencies

34

 

The ﬁgure above depicts 9 major constructs hypothesized to inﬂuence the
academic developmental processes for Latino children (see Figure 2). As with the
original integrative model, social position factors are preeminent. These are attributes or
characteristics of individuals used by society to stratify or place individuals in a social
hierarchy that pertain to Latino children. These factors include race, ethnicity, social
class, and gender. These social position factors have the potential to interact in ways that
magnify or diminish the importance of other factors in the model (Bronfenbrenner &
Crouter, 1983).

The effect of social position is mediated through pervasive social mechanisms of
racism, prejudice, discrimination, and oppression. The segregated environments that
Latino children and their families are subjected to are created by the inﬂuence of social
position variables through these social mechanisms. It is hypothesized that the effects of
racism on Latinos operates through the creation of segregated contexts (Garcia Coll etal.,
1996). The interplay of these factors creates the unique, “nonshared” conditions faced by
Latinos and affects the nature of the developmental processes that operate and the
eventual competencies that result. Segregation directly inﬂuences the inhibiting and
promoting environments experienced by Latino youth.

Racism, discrimination, and prejudice both indirectly, through segregation, and
directly affect Latinos experience through social interactions in inhibiting and promoting
environments. These environments, in turn, directly inﬂuence the adaptive cultures that
are created in response to experiences within these environments. “An adaptive culture

involves a social system deﬁned by sets of goals, values, and attitudes that differs from

35

the dominant culture” (Garcia Coll et al., 1996, p. 1896). Inhibiting/promoting
environments and adaptive culture directly inﬂuence family and peer processes, and
interact with Latino children’s individual characteristics. Latino children inﬂuence their
family and peer processes and contribute to their own socialization. Latino youth’s
developmental competencies result from the direct contributions of adaptive culture,
family processes, peer process, and personal characteristics as individual factors, as well
as from the interactions among these factors.

In terms of Latinos academic success, key developmental competencies include
culture speciﬁc and bicultural competencies, academic motivation, efforts towards
academic mastery goals, academic self-efﬁcacy, time allocation, and academic behaviors.
Behaviors related to academic success include performance on homework, grades, test
scores, and the like, as well as both overall attendance and class/course speciﬁc
attendance. Academic success emerges from the academic competencies developed by
youth. Included below is research supporting the inclusion of each model component and
component factors as related to academic outcomes.

Social Position Variables. Social position variables such as race, ethnicity, social
class, and gender, as described previously in the overview of the history and current
educational status of Latinos in US. education, indirectly affect Latino students’
development of competencies related to academic outcomes. In the United States these
are the most important attributes on which society is stratiﬁed, are overlapping, and result
in additive and multiplicative effects, depending on the degree to which an individual
occupies speciﬁc combinations of these social positions (Gracia Coll et al., 1996;

Krieger, Rowley, Herman, Avery, & Phillips, 1993). For Latino children, their expected

36

gender roles also inﬂuence access to resources, social interactions, and expectations that
will consequently inﬂuence their academic developmental outcomes (Gracia Coll et al.,
1996). The socioeconomic resources available to Latino families, due to social class, and
how they are used are also important inﬂuences on the developmental competencies of
Latino children (Huston, McLoyd, & Garcia C011, 1994). Race, ethnicity social class, and
gender also play a part in teachers’ perceptions of students, as well as the amount of
racism, prejudice, discrimination, and oppression students perceive and experience.

Parent educational level has been found by several researchers to be related to
academic outcomes (Baker, McGee, Mitchell, & Stiff, 2000; Educational Testing
Service, 1996; NCES, 1995; Shumow & Miller, 2001; Torres Campos, 2003; Voydanoff
& Donnelly, 1999). But there have been some inconsistent results on whether parent
educational level is a useful factor for explaining Latino academic success (e. g. Cooper,
Cooper, Azimitia, Chavira & Gullat, 2002; Torres Campos, 2003). The relationship
between parent educational level and other predictors of academic success, such as
parental involvement in schooling, may hold the key to understanding its effect on Latino
children’s academic outcomes (Lareau, 1996; Shumow & Miller, 2001).

Social Stratiﬁcation Mechanisms. Racism, both perceived and experienced,
contributes to Latino students’ development of competencies (Korzenny & Schiff, 1987).
Minority children currently face subtle forms of educational racism that include low
teacher expectations and attitudes, clinical deﬁnitions of academic problems, testing and
tracking, biased curriculum and textbooks, and socializing into lower expectations and
inferior jobs (Fisher, Wallace, & Fenton, 2000; Ogbu, 1991b, 1992). Latinos may be

discriminated against for multiple reasons, including skin color, English language

37

proﬁciency, and poverty (F ennelly, Mulkeen & Giusti, 1998). In one study of a Mexican-
American community, Greenberg, Burgoon, Burgoon, and Korzenny (1983) found that
pervasive racism and discrimination, particularly in the media, were identiﬁed as
contributing to Mexican-American youth’s lack of academic success. Fennelly, Mulkeen
and Giusti’s (1998) study, found that Puerto Rican adolescents and their mothers perceive
discrimination to lead to several negative academic outcomes. These negative outcomes
included a good education not necessarily translating into success for Puerto Ricans,
Puerto Ricans being kept ﬁom getting ahead, and Puerto Rican students being held down
in school despite extensive efforts (F ennelly, Mulkeen & Giusti, 1998). Indeed several
studies have shown that discrimination persists in hiring practices and therefore
minorities have a harder time than whites securing employment (Coontz, 1992; Shulrnan,
1990; Turner, Fix, & Struyk, 1991). Students’ response to discrimination is important in
that it may mediate the inﬂuence of discrimination on academic success (Fisher, Wallace,
& F enton, 2000; Ogbu, 1981). In Fennelly, Mulkeen and Giusti’s (1998) study responses
to discrimination took two forms: 1) Rejecting majority culture to do well in school and
being inﬂuenced by peer pressure to do poorly; and 2) A desire to “prove them wrong”
by succeeding in school and work. Both of these can be seen as adaptive normative
responses, but only the latter will be viewed by society as contributing to academic
success.

Segregation. Segregation is the systematic separation of groups and individuals
based on attributions made in regard to their social position (Taeuber & Taeuber, 1965).
Residential segregation, the most pervasive form of segregation in the US, determines

many of the elements that will either promote or inhibit children’s development (Blau,

38

1981). Economic segregation affects Latino child development through differential
access to resources. Families with higher economic status have greater access to the
resources that enhance the development of children’s competencies than families of
lower economic status (Garcia Coll et. al., 1996). Given the detrimental effects of poverty
on children’s development, the economic segregation produced by economic
discrimination are particularly troublesome (Duncan, Brooks-Gunn, & Klebanov, 1994;
Huston, McLoyd, & Garcia C011, 1994). Social and psychological segregation occurs
when Latino families and children are not permitted to access important social and
emotional resources as a result of social stratiﬁcation mechanisms (Garcia Coll et. al.,
1996). Latino students may be exposed to a combination of high or low levels of all three
types of segregation.

Promoting/Inhibiting Environments-School. School variables that can inﬂuence
Latino youth behavior operate on three levels: 1) The school district or system (i.e.
organizational and instructional philosophies, policies, procedures, and the political
power of Latinos within the district); 2) The individual schools (including personnel and
resources); 3) and the individual classrooms (including child, teacher, and peer
characteristics, classroom structure, curriculum, and instructional strategies) (Wasik,
1992). Most research on the effects of school on academic success has focused at the
individual classroom level. Other important school factors that need to be attended to
when discussing the academic success of Latinos include the inﬂuence of administrators
and school boards, policies on tracking and stafﬁng, and the extent to which the school

promotes parental involvement.

39

Not surprisingly, the climate and support provided by the school environment
have been associated with better academic performance. Students whose schools foster
caring and supportive relationships have higher math, language, and reading achievement
scores (F elner, et al., 1997). Schools that have high expectations, encourage cooperation,
and have teachers who are supportive, have lower levels of failure and dropouts than
schools that emphasize competition, testing, tracking, and have low expectations (Powell-
Cope & Eggert, 1994; Rumberger & Thomas, 2000; Rutter, 1979). Support from teachers
and friends have been found to be related to Mexican American students’ resiliency and
GPA. (Alva, 1991; Lopez, Ehly, & Garcia-Vézquez, 2002). Kaplan’s (1999) study of
black and Latino students revealed that students gave their teachers credit for
strengthening their academic skills. Similarly, Lucas, Henze, and Donato (1990) noted
that support provided by teachers, speciﬁcally in the form of valuing students’ primary
language and encouraging academic achievement, promoted the academic success of
Latino students.

School engagement refers to the kind and level of student “connectedness” to
school. Academic competence and intrinsic achievement motivation have both been
described as two key features governing students’ engagement in school (Wigﬁeld,
Eccles, & Rodriguez, 1998). Level of school engagement is considered an important
early indicator of scholastic achievement and school completion (Connell, Spencer, &
Aber, 1994; Hawkins, Catalano, Kosterrnan, & Hill, 1999). The majority of research on
school afﬁliation, engagement, or bonding has reported that the greater students’ sense of
connectedness to school the greater their academic performance (Paulson, Marchant, &

Rothlisberg’s, 1998; Scales, et al., 2000). The opposite relationship has also been found

40

to be true; students who feel alienated from school are more likely to fail academically
(Eckstronr, Goertz, Pollack, & Rock, 1986). Research examining successful Latino
students has also found that these students were more satisﬁed and felt more involved at
school than less successful Latino students (Alva, 1991; Reyes & Jason 1993; Waxman,
Huang, & Padron, 1997). Sense of school belonging and engagement have similarly been
found to be related to school success for Latino students in general and language minority
Mexican American students in particular (Goodenow & Grady, 1993; Rumberger &
Larson, 1998). Latino students with low levels of school bonding have also been found to
be more at risk for school failure (Robertson, Harding, & Monison, 1998).

Promoting/Inhibiting Environmems-Neighborhood. Children develop the tasks
necessary to be successful and acquire instrumental competencies based on their context
and surroundings (Ogbu, 1981). Although children who grow up in a poor, Latino
neighborhood may not have the availability or access to ample resources (i.e., health care
or good schools), the community can still encourage and provide them with the
competencies needed to be successful outside of the community (Garcia Coll et. al.,
1996). A middle-class environment can provide Latino children with some type of
economic stability, better schools, and access to other resources, but it may not buffer the
effects of prejudice, racism, and discrimination to which these children might be exposed
within and outside the community (Rodriguez, 1975; Tatum, 1987). The neighborhood
community also inﬂuences Latino students through climate and the opportunities for use
of outside-of-school time that are available.

Support provided by other adults and neighbors has been associated with higher

grades and higher academic achievement in the general youth population (Cochran & Be,

41

1989; Entwisle, Alexander, & Olson, 1994; Wenz-Gross & Siperstein, 1997). Extended
family and non-related adults are known to fulﬁll attachment functions in supporting
personal and academic adjustment, particularly within farrrilies of color (Harrison,
Wilson, Pine, Chan, & Buriel, 1990; Kenny & Perez, 1996). Similarly, a review of
research and programs addressing Mexican immigrant youth found that relationships with
caring adults were associated with resiliency (Chavkin & Gonzalez, 2000). Additionally,
Latino student dropout has also been related to a lack of supportive adult relationships
(Chavkin & Gonzalez, 2000; Nesman, Barobs-Gahr, & Medrano, 2001; Rumberger,
2001)

Adaptive Culture. Latino families and children have developed goals, values,
attitudes, and behaviors that set the apart from the dominant culture. This adaptive culture
is a product of Latinos collective cultural, political, and economic history, and current
contextual demands posed by their environments. At the collective level there are three
sources of historical processes that inﬂuence the development of adaptive culture:
traditions and other cultural legacies, economic and political events, and migration and
acculturation patterns. There are large differences among Latinos of different
nationalities and immigrant status in these three historical processes.

Traditions and cultural legacies are part of the group’s collective history for
generations. For example, Puerto Ricans have internalized and extemalized expressions
of a mixed heritage that incorporates T aino, colonial Spanish, and African inﬂuences, as
well as inﬂuences from the United States (Garcia Coll, et. al., 1996). Economic and
political events that have shaped Latino’s collective history also inﬂuence adaptation. In

the case of Puerto Ricans, the history of colonization from Spain and the US. has been

42

proposed as a major contributor to psychosocial development (Garcia Coll & Vézquez
Garcia, 1995). Migration and acculturation patterns include the initial reasons for
migration (i.e. seasonal employment and completion of higher education), length of the
settlement of the community, and the ease of back and forth, or circular, migration
(Garcia Coll, et. al., 1996).

The acculturation level of Latinos has been linked to almost every possible
outcome of interest, including academic success (e. g. Lopez, Ehly, & Garcia-Vazquez,
2002). Garcia Coll, Meyer, and Brillon (1995) have addressed the importance of cohort
effects when assessing the impact of migration and acculturation as ongoing processes
that can be experienced by recent immigrants as well as by individuals who move from
an ethnic neighborhood to one that is predominantly white. Additionally Ogbu’s (1991a)
distinction between voluntary and involuntary minorities may also inﬂuence these
processes.

Current contextual demands are the more immediate sources of inﬂuence that
contribute to the development of adaptive culture. These include factors such as rates of
unemployment, neighborhood safety, racism, prejudice transmitted through systems of
formal education, the media, and interpersonal interactions.

Child Characteristics. Latino child characteristics speciﬁcally related to academic
success include age, ethnic identity, involuntary vs. voluntary minority status (Ogbu,
1991a), English proﬁciency, and perceptions, attitudes, and beliefs regarding education.
Adolescents who have a positive attitude towards school and who value school are less
likely to fail or drop out than adolescents who hold a negative attitude towards school or

value school less (Powell-Cope & Eggert, 1994; Goldschmidt & Wang, 1999; Rumberger

43

2001). Latino students’ perceptions, attitudes, and beliefs regarding education will
determine the basis of behavior displayed during school interactions. Phinney and
Alpuria’s (1990) study found that achieving a positive sense of ethnic identity is another
characteristics related to self—esteem for Mexican-American adolescents. Bemal, Saenz,
and Knight (1991) also found that a positive sense of ethnic identity promotes
achievement for Mexican American youth. English proﬁciency is an individual
characteristic which greatly affects students’ interactions with teachers and other
educators. Ogbu’s (1991b) work has elucidated how minority status affects the
emergence of developmental competencies particularly as related to academics.

Family. Latino family values, beliefs, and goals towards education, along with
support and involvement will directly inﬂuence the emergence of academic
developmental competencies in Latino children. Research has shown that what the family
does to develop language, motivate students, monitor homework, and limit television
watching is more important to student success than family income or education (de
Kanter, Ginsburg, & Milne, 1987; Henderson & Berla, 1994). It is important to note that
some Latino parents view teachers as educational experts and feel it is inappropriate
interfere with their children’s education (Flores, Cousin, & Diaz, 1991). In general,
research has found that students whose parents are involved in school have relatively
high levels of school performance (Furstenberg & Hughes, 1995; Muller, 1993, 1995;
Paulson, 1996; Steinberg, Lambom, Dombusch, & Darling, 1992; Zill & Nord, 1994).
Studies have found that parental involvement in school is associated with Latino
students’ grades, G.P.A., and overall academic performance (Bogenschneider, 1997;

Jones & Velez, 1997; Lopez, 1996) Romo and F albo (1996) found in their longitudinal

44

qualitative study of Latino youth, that parents of academically successful adolescents
were aggressive in making contacts with the schools. But parental involvement with
children relative to education can take many forms and depends on a wide variety of
factors, such as ﬁnancial and human resources, motivation, time constraints, basic
relationship between parent and child, and the willingness to engage parents in the
process of schooling (Jones & Velez, 1997). The contexts for parental involvement can
be the home, the school, and/or families within their cultural group and the broader
society (Jones & Velez, 1997).

An abundance of research has also reported that family support is associated with
higher grades and higher standardized test scores (Bisnaire, Firestone, & Rynard, 1990;
Cauce, F elner, & Primavera, 1982; Christenson, Rounds, & Gomey, 1992; Eccles, Early,
Fraser, Belansky, & McCarthy, 1997; Feldman & Wentzel, 1990; Glasgow, Dombusch,
Troyer, Steinberg, & Ritter, 1997; Masselam, Marcus, & Stunkard, 1990; Rosenthal &
Feldman, 1991). Lopez (1996) found that Latino students receiving higher grades were
more likely to have greater parental support than their low achieving peers. Some
researchers have also found that the expanded role that family and kin networks play in
the developmental processes for minority children may serve to protect them from
economic hardships and social and psychological sources of oppression they experience
as a result of their relative position in society (Harrison, Wilson, Pine, Chan, & Buriel,
1990; McAdoo, 1982).

Peers. There is an abundance of literature that suggests that, similar to families,
peers have a direct inﬂuence on the emergence of academic developmental competencies.

Academically successful youth often have one or more close and stable friendships on

45

which they rely for ongoing emotional support (Werner & Smith, 1992). Positive peer
inﬂuence has been associated with higher academic achievement (Chen & Stevenson,
1995; Hanson & Ginsberg, 1988), higher math achievement (Hanson & Ginsburg, 1988),
and better grades (Mounts & Steinberg, 1995). Adolescents whose peers have high
expectations and positive attitudes towards school are less likely to fail or drop out of
school (Powell-Cope & Eggert 1994; Wang, Hartel, & Walberg, 1999; Rumberger 2001).

Some research has found that for Latino adolescents peers are relatively more
inﬂuential than are parents for academic success (Steinberg & Darling 1993).
Involvement with peers who are motivated to succeed in school, and who spend time
doing homework, has also been found to positively affect Latino students’ academic
achievement (Romo & Falbo, 1996). Muskal and Chairez (1990) found that high-
achieving Latino students, whose families live in middle- to upper- income
neighborhoods of mixed ethnicity, regardless of parents’ birthplaces, beneﬁted from peer
pressure to succeed in school. In contrast, peer pressure, particularly to participate in
risky behaviors, was found to be a barrier to academic achievement in Vera’s (2000)
study of Chicano urban youth.

Developmental Competencies. Key academic developmental competencies
include, culture speciﬁc and bicultural competencies, academic motivation, efforts
towards academic mastery goals, academic self-efﬁcacy/self-concept, and time
allocation. Beyond traditional areas of concern, developmental competencies such as
Latino children’s ability to function in two or more different cultures, cope with racism,
subtle and overt discrimination, and social and psychological segregation must be

considered in relation to academic success. Culture-speciﬁc and bi-cultural competencies

46

are needed if Latino children are to learn the codes that are appropriate to both cultures
and master the activities that are called upon in each (LaFromboise, Coleman, & Gerton,
1993). Research has found that Cuban children and Latino children with bicultural
abilities are less likely to experience family and school conﬂict or to become involved in
illegal drug use (Szapocznik, Kurtines, & Fernandez, 1980; Szapocznik, Santisteban,
Kurtines, Perez-Vidal, & Hervis, 1984). Hurtado, Figueroa, and Garcia (1996) found that
young adults who have learned to be bicultural themselves are able to pass on their
understanding of how to retain community traditions while entering and succeeding in
school and beyond.

Children’s motivation to succeed is considered a key feature of educational
competence (Harter, Runbaugh, Whitesell, & Kowalski, 1992). The achievement
motivation of students has been found to be a predictor of the type of coursework they
select, the effort that put into their work, and their overall levels of educational attainment
and performance across racial/ethnic groups (Anderson & Keith, 1997; Harter et al.,
1992; J essor, Van Den Bos, Vanderryn, Costa, & Turbin, 1995; Keith & Perkins, 1995;
Paulson, Coombs, & Richardson, 1990; Scales, Benson, Leffert, & Blyth, 2000; Wentzel,
1993; Wentzel,l996; Whitehead, 1984). Waxman, Huang, and Padron (1997) found, in
their study of Latino middle school students, that academically successful students are
more motivated than their less successﬁll classmates. Shultz’s (1993) study found that
achievement motivation was a signiﬁcant mediator of academic performance in Hispanic
children, independent of intellectual ability. A study of Latino students attending a

predominantly Latino (80%) high school in California found that academic self efﬁcacy,

47

including acaderrric motivation, was the strongest predictor of academic performance
(Buriel, Perez, De Ment, Chavez, & Moran, 1998).)

Academic self-concept refers to students’ perceptions of their competence or level
of ability within the academic realm and is linked to motivation to succeed in school, as
well as academic performance. Students with more positive academic self-concepts tend
to have higher school grades and test scores (Marsh & Byme, 1999). Students’ academic
self-concept also inﬂuences the courses they choose to take in school, with students being
more likely to take courses in a subject in which they have a more positive self-concept
(Marsh & Yeung, 1997).

Time allocation pertains to students’ time management skills during school and
for academic activities, as well as use of outside of school time. There is an abundance of
literature that has found that positive and creative use of outside of school time by
students is related to greater academic success (Alva, 1991; Barber & Eccles, 1997 ;
Davalos, et al., 1999; Eccles & Barber, 1999; Hanks & Eckland, 1976; Larson, 1994;
McMillan & Reed, 1994; Posner & Vandell, 1994; Quinn, 1995; Scales et al., 2000;
Voydanoff & Donnelly, 1999; Waxman, Huang, & Padron, 1997). Several researchers
have noted that the amount of time an individual devotes to schoolwork is related to his
or her individual achievement (F uligrri, 1997 ; F uligrri & Stevenson, 1995; Leone &
Richards, 1989; Steinberg & Darling, 1993; Walrlberg & Fredrick, 1982). Some studies
of Latino students have found that successful students report higher levels of involvement
in high school activities than unsuccessful students (Alva, 1991; Waxman, Huang, &
Padron, 1997; Davalos, Chavez, & Guardiola, 1999). Davalos and her colleagues (1999),

in their study including Mexican American adolescents found that involvement in

48

extracurricular activities had a signiﬁcant positive effect on whether they were in good
standing at school. Waxman (1997) and her colleagues, in a study of Latino middle
school students, found successful students reported they spent signiﬁcantly more time
doing homework each week and more time on additional reading than unsuccessful
students. Fuligni (1997) in his study of immigrant youth including Latinos, found that the
amount of time students reported they spent studying was a predictor of their academic
achievement as measured by grades.

Behaviors related to academic success are developmental competencies that
directly affect students’ academic success. Students who attend school and class regularly
and who use study skills to increase positive school performance will be more
academically successful. Note taking and test preparation are two study skills Latino
students may use to increase academic performance and thereby increase academic
success (Redd, Cochran, Hair, & Moore, 2002). Most educators and parents consider the
performance of homework a key component of students’ academic responsibilities (US.
Department of Education, n. d.). Students who fail to complete their homework or who
complete it but fail to meet the teacher’s standards are more likely to receive lower
grades in class (US. Department of Education, 11. d.). Research suggests that, for high
school students, greater hours spent on homework are related to higher levels of student
achievement (C00per, 1989; Cooper, Lindsay, Nye, & Greathouse, 1998). Rumberger
and Larson’s (1998) study of Mexican-American Language Minority students in
California also found that how often students were absent, their classroom work habits,

and their classroom social behavior were predictive of students’ GPAs.

49

Academic Success. Academic success is the ﬁnal outcome of this adapted,
augmented integrative model. Academic success involves a wide range and multiple
levels of outcomes that are intertwined and cannot be deﬁned by any one single index,
indicator, or measure. Although academic success has been deﬁned in many ways (i.e.
GPA, persistence, achievement, attainment, etc.) it has yet to be thoroughly explored
from a Latino perspective. Grades have often been used to examine academic success,
but may present only a limited view of what academic success means (Lopez, et al.,
2002). There is a body of research that examines students who persist in school as
examples of success in relation to those who drop out. Other researchers have examined
students who go on to higher education to gauge academic success. Standardized test
scores are also becoming one of the most common ways to rate students in public
schools. Several schools are also using portfolios as measures of student progress. Given
the lack of research on Latino deﬁnitions of academic success, using multiple deﬁnitions
and operationalizations will be important when examining the processes associated with
developing academic competencies. The academic success of Latino students at any
given point in time is in turn an important inﬂuence on the successive ecological

processes that will continue to affect their developmental course.

Study Aim and Hypothesis

A coherent picture of development among ethnic minority youth is missing in
theories of normal adolescence. Research examining the academic development of Latino
youth is rarely guided by a culturally speciﬁc theory or framework. In order to change
this trend in developmental research, this study seeks to conﬁrm and build a model of the

academic development of Latino youth based on Garcia C011 and her colleagues’ (1996)

50

integrative model of minority developmental competencies. The aim of the current study
is to assess pathways through distal and proximal factors that account for social
stratiﬁcation and cultural factors related to developmental competencies and academic
success for Latino youth not found in other ecological models.

Items from the National Longitudinal Study of Adolescent Health (Add Health)
data set used to assess variables and constructs in the model are presented below. Add
Health is a nationally representative study that explores the causes of health-related
behaviors of adolescents in grades 7 through 12 and their outcomes in young adulthood.
Add Health seeks to examine how social contexts (families, friends, peers, schools,
neighborhoods, and communities) inﬂuence adolescents' health and risk behaviors.

The constructs presented in the model below represent latent constructs, which are
multifaceted. Various indicators otherwise known as manifest variables from the Add
Health data set were used to measure each latent constructs or portions of latent
constructs. Whenever possible youth’s self-reported items were used over parent report as
indicators/measured variables of constructs. The indicators below represent the measured
or manifest variables for purposes of structural equation modeling. Appendix A
summarizes the data items used to assess each construct.

Social Position Variables

Social position variables examined in this study include race, ethnicity, gender,
and social class/parent educational level. Race, ethnicity, and gender are collected on all
youth. Social class/parent educational level was measured as parental educational

attainment.

51

Social Stratiﬁcation Mechanisms

For purposes of this study only perceived racism was used as construct of social
stratiﬁcation mechanisms. Youth’s reports of students at school prejudiced were used as
measures of perceived racism.
Segregation

There are three major types of segregation to be considered for purposes of testing
this model. Residential segregation was measured by parents’ report of their reasons for
living in their current neighborhood, including being able to afford better housing,
presence of friends, family, or children, and proximity to work. Economic segregation
was measured using multiple indicators, parents’ report of total family income, ability to
pay bills, employment status, and receipt of public assistance/social welfare. Social
segregation was measured as youth’s level of feeling socially.
Promoting/Inhibiting Environments-School

School as a promoting or inhibiting environment included variables item the
individual school level and the individual classroom level. The items used were drawn
from youth’s perceptions, parents’ perceptions, and school administrators’ reports. Youth
measures included reports of getting along with teachers, perceptions of teacher fairness,
teachers care about them, taking a gun/weapon to school, feeling close to people at
school, feeling part of the school, feeling happy to be at school, and feeling safe at
school. School administrators’ reports of problems at school related to smoking or
tobacco use, drug use, alcohol use, gang violence, teenage pregnancy,
vandalism/thieving, eating disorders, and stress or pressure, school type, average daily

attendance, full-time classroom teachers, average class size, race and ethnicity of full

52

time teachers, gender of full time teachers, presence of parent organization, dropout rates,
retention rates, achievement score rates, percentage of 12th graders in academic or college
preparatory, vocational or occupational preparatory, general pro gram/no divisions, health
related services offered at school, school policies and rules, urbanicity, and size of school
were also included. Parents report of their child’s school being a good one, being a school
that prioritizes learning, and a safe school were also used.
Promoting/Inhibiting Environments-Neighborhood
Youth’s neighborhoods as promoting or inhibiting environments were measured using
both youth and parent report. Measures included youth’s report of feeling safe in the
neighborhood, knowing people in their neighborhood, being able to talk to someone in
their neighborhood, neighbors looking out for one another, feeling happy in their
neighborhood, and whether they would be happy or unhappy if they moved away from
their neighborhood. Parent measures including report of the presence of crime, drugs,
trash, and wanting to leave neighborhood were also used to assess the nature of youth’s
neighborhood environment.
Adaptive Culture

Given the constraints of the data set, acculturation and current contextual
demands were assessed as aspects of adaptive culture. Acculturation was quantiﬁed as
parents’ report of whether child was born in the US. youth, born a US citizen, and for
those not born in the US, parents’ report of when child moved to the US. Youth’s report
of their language use and comfort with different languages were also used as a proxy

measures for acculturation. Current contextual demands were measured using proxy

53

measures including youth’s perceptions that they will live to be 35, be married by 25, will
be killed by 21, and will have a middle class family.
Child Characteristics

Items used to assess characteristics of the youth themselves included English
proﬁciency, self-esteem, learning disabilities, and perceptions towards. Youth’s English
proﬁciency was measured using items regarding the language interviews were conducted,
youth’s report of language used with family members and friends, buying music in
language other than English, and reading, listening, or watching newspaper, radio or TV
in languages other than English. Self—esteem questions included youth’s comparisons of
their intelligence to others, how intelligent they feel they are, feeling as good as others,
feelings of having good qualities, feeling proud, liking themselves as they are, how
conﬁdent they are, and whether they feel life has been a failure. Parents’ and youth’s
report of learning disability and youth’s retrospective report of ADI-[D symptoms were
used as measures of learning disabilities. Perceptions of education were measured by
youth’s feeling they will go college and graduate from college.
Family

Parental support and perceptions of school were measured by parent and youth
report regarding parent involvement in parent teacher organizations, talking with youth
regarding school work or grades, talking with youth about school in general, talking with
teachers, volunteering at school, working with child on school project, being disappointed
if child does not graduate high school, and being disappointed if child does not graduate
college. General parental and family support were measured as youth’s report of parent

caring, closeness, warmth and love, encouraging independence, satisfaction with parental

54

relationship, feeling loved and wanted, understanding, and being paid attention to, along
with parent reports of their relationship with their child.
Peers

Due to constraints of the data set only general peer support and peer substance use
were assessed as part of peer inﬂuence. Support from ﬁiends was measured as how much
time is spent with peers, how often peers are talked to regarding problems, and feeling
that ﬁiends care. Youth report of their peers’ smoking, alcohol, and marijuana use were
also included. One parent report item regarding their perceptions of the type of inﬂuence
their child’s best friend provides was also used.

Developmental Competencies

Culture speciﬁc and bi-cultural competencies are not represented in the Add
Health data set and so were not assessed as part of this study. Other aspects of
developmental competencies assessed in this study are presented below. Youth’s time
allocation was quantiﬁed as youth’s report of time spent in after-school and out-of-school
activities, athletics, physical activity, at religious services or activities and minimization
of time spent watching television or playing video, and/or using the computer for
activities other than schoolwork. Academic motivation was measured as youth’s report of
trying hard at school, paying attention in school, and wanting to go to college.

School performance, attendance, and behavior at school are three additional
developmental competencies assessed in this study. Attendance was measured as youth’s
report of ever skipping school and the number of times they have skipped school. School
performance measures included out of school suspensions, ever being expelled, and

grades skipped and retained. Other behaviors included indicators of ever being high or

55

drunk at school, ever having problems with school because of being drunk or high, and
carrying a weapon to and/or at school as reported by youth.
Academic Success

The effects of the above constructs on academic success were examined for
Latino youth as measured by youth reports of grades, high school graduation, and
attainment. This included grades in English/language arts, mathematics, history, and
science, highest grade completed, high school completion, GED/equivalency attainment,
post-secondary education and graduation, as well as vocational or job training.

The current study seeks to understand the relationships between the above
mentioned constructs and academic success of Latino youth at two time points between
1994 and 1996 when youth are in grades 7 through twelve, and at a ﬁnal time point
between 2001 and 2002 when youth are between the ages of 18 and 26. Changes in
relationships among the constructs may differ across time not only because youth and
their families develop, but many of the youth and families may also have moved and been
in different environments. Appendix A includes a list of the latent constructs with their
underlying measurement variables in simpliﬁed format.

Figure 3 depicts the manifest variable model hypothesized that was tested in the
current study. The effects of social position variables’, in the form of race-ethnicity,
parent educational level, and gender, are hypothesized to be mediated by social
stratiﬁcations mechanisms, in this case youth’s perceived racism, which directly predict
the segregated environments, including residential, economic, and social. The effects of
racism are in turn expected to directly predict youth’s inhibiting and promoting

environments as well as having a mediated effect on promoting/inhibiting environments

56

through youth’s segregated contexts. Segregation is also predicted to directly predict
youth’s promoting/inhibiting environments. Inhibiting/promoting environments and
racism are hypothesized to predict youth and families’ adaptive culture. Both
promoting/inhibiting environments and the adaptive culture are hypothesized to directly
predict family and peer processes. They are also expected to have a mediated or indirect
effect on family/peer processes and developmental competencies through child
characteristics. Developmental competencies are hypothesized to be predicted by
adaptive culture, child characteristics, and family/peer processes. Academic success is

directly predicted by academic developmental competencies.

 

Students at
school

   
 
    
  

 

 

 

 

 

 

 
       
  
 
 

Race/Ethnicity

re'udice Child
{I \\ p J Characteristics
Parent +
educational _
level computed
variable Neighborhood +
Developmental Academic

       

Adaptive

  

 

 

 

 

 

l

computed Culture Competencies Success
1 variable
\ ﬂ

 

 

 

 

 

( Economic Social \
segregation segregation
computed computed
variable variable

Residential
segregation

 

 

computed
k variable J

 

Figure 3. Manifest variable model

Direct Relationships:

0 Social position variables are expected to predict measures of social stratiﬁcation
mechanisms. Identiﬁcation with races other than White/Caucasian,

Latino/Hispanic or Latin American nationality identiﬁcation, female gender,

57

and lower levels of parental educational attainment will predict higher reports of
perceived racism.

Social stratiﬁcation mechanisms are expected to impact segregation. Greater
perceived racism will predict increased residential, economic, and social
segregation

Social stratiﬁcation mechanisms will directly predict promoting/inhibiting
environments. Increased perception of racism will predict more inhibiting
environments.

Social stratiﬁcation mechanisms will impact adaptive culture. Increased
perceptions of racism will predict increased contextual demands and decreased
bicultural orientation.

Promoting/inhibiting enviromnents will predict family/peer processes.
Promoting environments will lead to increased positive family and peer
inﬂuence.

Promoting/inhibiting environments will directly impact adaptive culture.
Promoting environments will predict lower contextual demands, and higher
level of bicultural acculturation.

Adaptive culture will directly impact child characteristics. Low levels of
contextual demands and bicultural orientation will predict positive child
characteristics.

Adaptive culture will directly impact family/peer processes. Low levels of
contextual demands and bicultural orientation will predict positive family/peer

inﬂuence.

58

0 Adaptive culture will directly impact developmental competencies. Low levels
of contextual demands and bicultural orientation will predict positive academic
competencies.

0 Child characteristics will directly inﬂuence family/peer processes. Positive
characteristics will predict positive family/peer inﬂuence.

0 Child characteristics will impact developmental competencies. Positive
characteristics are hypothesized to predict positive academic competencies.

o Family/peer processes are hypothesized to impact developmental competencies.
Positive family and peer inﬂuence will predict positive academic competencies.

0 Developmental competencies are hypothesized to directly impact academic
success. Increased positive academic competencies will predict positive
academic success.

Mediating Relationships:

0 The relationships between social position variables and segregation will be
mediated by social stratiﬁcation mechanisms.

0 Social stratiﬁcation mechanisms will mediate the relationships between social
position variables and promoting/inhibiting environments.

0 Segregation mediates the relationships between social stratiﬁcation mechanisms
and promoting/inhibiting environments.

0 Child characteristics mediate the relationships between promoting/inhibiting
environments and family/peer processes.

0 Family/peer processes mediate the relationships between child characteristics

and developmental competencies.

59

.M
w
.—

Initial analyses will focus on testing the hypothesized model in its entirety. The
entire model may be accepted or rejected despite individual predictions resulting
in signiﬁcant or nonsigniﬁcant relationships. If the overall model reﬂects a poor
ﬁt to the data, additional analyses will examine model components and revisions
to the model in an effort to understand the dynamic relationships suggested by the

full model.

Methods
Data Collection

Data from the Add Health data set will be used in this study. Initiated in 1994
under a grant from the National Institute of Child Health and Human Development
(NICHD) with co-ﬁmding from 17 other federal agencies, Add Health is the largest, most
comprehensive survey of adolescents ever undertaken. Data at the individual, family,
school, and community levels were collected in two waves between 1994 and 1996. In
2001 and 2002, Add Health respondents, 18 to 28 years old, were re-interviewed in a
third wave to investigate the inﬂuence that adolescence has on young adulthood.

Beginning with an in-school questionnaire administered to a nationally
representative sample of students in grades 7 through 12, the study follows up with a
series of in-home interviews of students approximately one, two, and six years later.
Other sources of data used in this study include questionnaires for parents and school
adrrrinistrators.

Waves I and H examine the forces that may inﬂuence adolescents' behavior, in
particular: personal traits, families, ﬁiendships, peer groups, schools, neighborhoods, and

communities. Wave HI explores the transition between adolescence and young adulthood.

60

The clustered sampling design of Add Health is school-based for two reasons:
First, it is the best way to screen for respondents of interest. Second, with the school as a
center, it is relatively easy to access the majority of respondents' peers, whose inﬂuences
are fundamental to the study's hypothesis. Systematic sampling methods and implicit
stratiﬁcation ensured that the 80 high schools selected are representative of US schools
with respect to region of country, urbanicity, size, type, and ethnicity. Eligible high
schools included an 11th grade and enrolled more than 30 students. More than 70 percent
of the originally sampled high schools participated. Each school that declined to
participate was replaced by a school within the stratum.

Participating high schools helped to identify feeder schools—that is, schools that
included a 7th grade and sent at least ﬁve graduates to that high school. From among the
feeder schools, one was selected with probability proportional to the number of students
it contributed to the high school. If the feeder school declined to participate, a
replacement was selected. The recruitment effort resulted in a pair of schools in each of
80 communities (Some high schools spanned grades 7 through 12; for those, a separate
feeder school was not recruited.) There are 132 schools in the core study. Each
participating school provided a student roster. Project staff assigned an identiﬁcation
number to each name and provided copies of the rosters to students for identifying their
friends as they ﬁlled out the questionnaire. Rosters were collected at the end of the class
period and destroyed.

Sample
All students who completed the In-School Questionnaire plus those who did not

complete a questionnaire but were listed on a school roster were eligible for selection into

61

the core in-home sample. This is a nationally representative sample of adolescents in
grades 7 through 12 in the US. in the 1994—95 school year. Students in each school were
stratiﬁed by grade and gender. About 17 students were randomly chosen from each
stratum so that a total of approximately 200 adolescents were selected from each of the
80 pairs of schools. A total core sample of 12,105 adolescents was interviewed.

Based on self-reported data from the In-School Questionnaire, four supplementary
ethnic-group samples were drawn. Following are the numbers of completed cases in these
samples: 1,038 blacks from well-educated families (with a parent with a college degree),
334 Chinese, 450 Cuban, and 437 Puerto Rican. In addition, the main sample contains
more than 1,500 Mexican-Americans and signiﬁcant numbers of Nicaraguans, Japanese,
South Koreans, Filipinos, and Vietnamese.

Sample Characteristics. This study used youth who identiﬁed themselves as being
of Hispanic/Latino background only at the time of the ﬁrst in-home questionnaire. The
overall Latino sample included 3525 respondents. The sample was almost equally divided
by gender, 50.3% male and 49.7% female. At Wave I youth were between 12 and 21
years old with an average age of 16.4. Table 1 presents the frequency distribution for
each Latino subgroup. Of note is over 1800 respondents identiﬁed as being of Mexican
descent, while 363 identiﬁed as either another Latino subgroup or mixed Latino
nationality. Table 2 presents each respondent’s answer when asked to choose a single
race category to identify their racial background. Youth were allowed to choose ﬁ'om all
major racial categories, but were also allowed the option of choosing “Other race.”
Almost half of the respondents (45%) chose “Other” and almost an equal percentage

(44.8%) chose “white.” Almost two-thirds (73.9%) of the youth in the sample were born

62

in the United States and 76.2% were born US. citizens. Only 38.4% of the Latino youth
reported their residential mothers were born in the United States. Only a quarter (25.0%)

of resident fathers were reportedly born in the United States.

 

 

 

 

 

 

 

Table 1
Ethnic subgroup percentages for Latino samples
Percent
Total Sample SEM Sample
Ethnic group (N=3525) (N=1138)
Mexican descent 51.2 47.9
Puerto Rican descent 15.5 16.4
Cuban descent 13.4 16.4
Central American descent 9.6 8.4
Latino/Other/Mixed descent 10.3 10.8
Table 2
Racial background percentages for Latino samples
Percent
Total Sample SEM Sample
.Racial background (N =3525) (N =1 138)
(Missing=29) (Missing=1 8)
White 44.6 45.5
Black or African American 3.3 3.3
American Indian or Native American 4.4 3.9
Asian or Paciﬁc Islander 1.8 2.6
Other 45.0 43.1

 

Youth’s report of their resident parents’ educational levels is presented in Table 3.
Almost identical percentages of resident mothers were reported as having none to an
eighth grade education (21.3%) as were reported to have graduated from high school
(21.4%). A small percentage of resident mothers were reported to have graduated college
(8.4%), while only 2.4% were reported to have had professional training beyond college.
A greater percentage of resident fathers were reported to have none to an eighth grade

education (17%) than were reported to have graduated high school (14.2%). Similar to

63

reports for resident mothers, 6.5% of resident father’s were reported to have completed

college and 3% to have had professional training beyond college.

Table 3

Youth report of residential mother and father educational levels

 

 

 

Percent
Total Sample SEM Sample
Educational Level (N =3525) (N =1 138)
Mother (Missing=497) (Missing=13 1)
Eighth grade or less/none 21.3 20.9
Greater than eighth grade but did not graduate
. 15.4 14.9
hrgh school
Attended vocational/trade/technical school 1 2 1 2
instead of high school ' °
Graduated from high school 21.4 22.5
Completed GED/High school equivalency 3.3 2.7
Attended vocational/trade/technical school after
. 4.0 3.7
hrgh school
Some college/university but did not complete a
8.6 9.8
degree
Graduated from college/university 8.4 9.8
Professional training beyond college 2.4 3.0
Father (Missing=1354) (Missing=364)
Eighth grade or less/none 17.0 17.0
Greater than eighth grade but did not graduate
. 10.0 10.5
hrgh school
Attended vocational/trade/technical school 0 7 0 6
instead of high school ' '
Graduated from high school 14.2 15.4
Completed GED/High school equivalency 1.5 1.5
Attended vocational/trade/technical school after
. 2.6 3.4
hrgh school
Some college/university but did not complete a
6.0 7.2
degree
Graduated from college/university 6.5 8.8
Professional training beyond college 3.0 3.5

 

Of interest in this study are the educational and academic outcomes of the

respondents. Table 4 summarizes the frequencies of various academic outcomes as

reported in Wave ID of the data when respondents were between on average 22.7 years

old (range 18-28 years old). About a quarter of the respondents (25.4%) reported having

64

completed 12th grade, almost a third (29.6%) reported having completed at least one year
of college, and a small portion (0.8%) reported having completed at least one year of
graduate school. About half of the respondents (50.8%) reported having received a high
school diploma. These data are similar to those presented in a recent report which
calculated that the national graduation rate for public high school Latinos in the class of
2001 was 53.2% (Swanson, 2003). Table 5 summarizes the results for degree attainment

by race-ethnicity. Chi-square analyses found no signiﬁcant differences (p = .681) among

 

 

the groups.
Table 4
Academic outcomes for Latino samples
Percent
Total Sample SEM Sample
Items (N=3525) (N=1138)

 

What is the highest grade or year of regular school

you have completed? (MiSSing:1229) (Missing=0)
th

6 0.0 0.0
7th 0.1 0.1
8‘“ 0.2 0.2
9th 1.4 1.6
10th 2.8 3.0
11th 6.5 6.7
12th 25.4 35.6
1 year of college 9.5 15.6
2 years of college 9.7 16.1
3 years of college 4.8 8.9
4 years of college 4.3 8.9
5 of more years of college 1.3 2.5
1 year of graduate school 0.5 0.6
2 years of graduate school 0.2 0.2
3 years of graduate school 0.1 0.0
What degrees or diplomas have you received? (Missing=1227) (Missing=0)
GED/High school equivalency 6.8 9.1
High school diploma 50.8 81.5
Jr. College degree 5.8 9.8
Bachelor’s degree 4.3 7.3
Mater’s degree 0.2 0.4
Doctoral degree 0.0 0.0
Professional degree 0.2 0.3

 

65

Table 5
Academic degrees by Race-ethnicity

 

Degrees
(N=1138)
HS Bachelor’s Master’s Graduate
Race-Ethnicity GED diploma degree degree degree

 

MexicanNative 0 4 17 12 2
MexicanBlack 0 0 2 4 0
MexicanOther 1 33 141 86 25
MexicanAsian 0 l 2 5 l
MexicanWhite 0 19 97 8 1 1 1
PRNative 0 0 1 0 0
PRBlack 0 1 8 8 1
PROther 0 6 36 35 4
PRAsian 0 0 1 4 0
PRWhite 1 6 37 34 4
LatinoNative 0 0 0 3 0
LatinoBlack 0 1 3 0 0
LatinoOther 0 4 26 19 5
LatinoAsian 0 0 4 7 0
LatinoWhite 0 3 19 22 6
CentralNative 0 0 5 1 0
CentralBlack 0 O 0 2 1
CentralOther 0 4 19 22 2
CentralAsian 0 0 4 1 0
C entralWhite 0 5 1 3 1 6 1
CubanNative 0 0 1 0 0
CubanBlack 0 1 2 2 2
CubanOther 0 2 1 l 10 5
CubanAsian 0 0 0 0 0
CubanWhite 0 8 50 65 24

 

The ﬁnal structural equation modeling analyses used 1138 youth from the 3525 in
the overall Latino sample. Youth who did not have any data collected for Wave III were
not included in the overall model analyses. This sample was 47.3% male and 52.7%
female. Table 1 presents the ﬁ'equency distribution for each Latino subgroup for this
SEM sample as compared to the overall Latino sample as a whole. This ﬁnal sample has
similar subgroup distributions as the entire Latino sample, including almost half of the

youth (47.9%) reporting Mexican descent. Table 2 presents each respondent’s single race

66

category choice in comparison to the overall sample. Here again similar frequency
distributions between the SEM sample and the overall sample can be seen. In particular
ahnost equal amounts of youth chose “white” as “other,” 45.5% and 43.1% respectively.
In the SEM sample, over two thirds (78.6%) were born in the United States and 80.8%
were born US. citizens, which represent slightly higher percentages of native US.
citizens that the overall sample. This sample reported 37.9% of their mothers and 27.1%
of their fathers were born in the United States.

Resident parent educational levels are summarized in Table 3. It should be noted
that, due to the high amount of missing data in the overall sample, the SEM sample
represents a slightly more educated sample as compared to the Latino sample as a whole.
Table 4 summarizes the academic outcomes of the SEM sample compared to the Latino
sample overall. As was the case with parent educational level frequencies, due to large
amounts of missing data in the overall Latino sample, the SEM sample represents a more
educated sample of youth as compared to the overall sample. In particular, 81.5% had
completed high school, 9.8% a junior college degree, and 7.3% a bachelor’s degree.
Instrumentation- Wave I

Wave I encompasses all data collection between 1994 and 1995.

The In-School Questionnaire. A self-administered instrument formatted for
optical scanning, was administered to more than 90,000 students in grades 7 through 12
in a 45- to 60-minute class period between September 1994 and April 1995. There was no
"make-up" day for absent students. Parents were informed in advance of the date of the
questionnaire and could direct that their children not participate. The questionnaire

included topics such as social and demographic characteristics of respondents (of interest

67

both as data and as selection criteria for in-home special samples), education and
occupation of parents, household structure, risk behaviors, expectations for the future,
self-esteem, health status, friendships, and school-year extracurricular activities.

In-Home Interview: Wave 1. In-home interviews were conducted between April
and December 1995. All respondents received the same interview, which was one to two
hours long depending on the respondent's age and experiences. The majority of
interviews were conducted in respondents' homes.

To protect conﬁdentiality, no paper questionnaires were used. Instead, all data
were recorded on laptop computers. For less sensitive topics, the interviewer read the
questions aloud and entered the respondent's answers. For more sensitive topics, the
respondent listened through earphones to pre-recorded questions and entered the answers
directly. In addition to maintaining data security, this minimized the potential for
interviewer or parental inﬂuence. The following topics are covered by the In-Home
Interview: health status, health-facility utilization, nutrition, peer networks, decision-
making processes, family composition and dynamics, educational aspirations and
expectations. Care was taken to screen respondents on age and experience so that only
appropriate questions were asked. Additional questions concerning the co-occurrence of
risk behaviors were asked of respondents who indicated multiple behaviors, for example,
ﬁghting while using drugs or drinking while carrying a weapon.

School Administrators Questionnaire. Administrators from participating schools
completed self-administered questionnaires dealing with school policies and procedures,
teacher characteristics, health-service provision or referral, and student body

characteristics.

68

Parent Questionnaire. A parent, preferably the resident mother, of each
adolescent respondent interviewed in Wave 1 was asked to complete an interviewer-
assisted, op—scanned questionnaire covering topics, such as inheritable health conditions,
marriages and marriage-like relationships, neighborhood characteristics, involvement in
volunteer, civic, and school activities, health-affecting behaviors, education and
employment, household income and economic assistance, parent-adolescent
communication and interaction, and parent’s familiarity with the adolescent's ﬁiends and
friends' parents.

Instrumentation- Wave 11

Wave II data collection includes follow-up in-home interviews with adolescents
and follow-up school administrator interviews conducted in 1996. The second wave
surveyed almost 15,000 of the same students one year after Wave 1. The sample for the
Wave II in-home interview comprised the respondents to the Wave I in-home interview,
with a few exceptions.

In-Home Interview Wave 1]. Wave H in-home interviews took place from April
through August 1996. The interview was generally similar to that at Wave 1. Questions
about attributes that should not change, such as ethnic background, were not repeated.

School Administrator Telephone Interviews. In the spring of 1996 school
administrators were contacted by telephone and asked to update information from the
ﬁrst year and add information about speciﬁc dress codes and security procedures on their

campuses.

69

Instrumentation —Wave III

Wave 111 data collection, conducted in 2001 and 2002, includes in-home
interviews with original respondents (now young adults). Data from Wave III allow for
diverse analyses across a spectrum of social, economic, and behaviors.

The in—home Wave III sample consists of Wave I respondents who could be
located and re-interviewed during the ﬁeld-work period, August 2001 to April 2002,
when they were between 18 and 26 years old. Interviews with 15,170 Wave I respondents
were completed at Wave HI. Wave I respondents who were out of the country were
omitted from Wave 111. Every effort was made to re—interview respondents who were
located in correctional facilities. Data collection was conducted nationwide (including
Hawaii and Alaska). To maintain conﬁdentiality, no paper questionnaires were used. As
in earlier waves, data were recorded on laptop computers. For less sensitive material, the
interviewer read the questions and entered the respondent's answers. For more sensitive
material, the respondent entered his or her own answers in privacy. The average length of
a complete interview was 134 minutes. The laptop interview took approximately 90
minutes. Most interviews were conducted in respondents' homes.

Interviews of original Add Health respondents were preloaded with some Wave 1
and Wave II data, including the name, age, and sex of the respondent and identiﬁcations
of parent ﬁgures, friends, and siblings from earlier waves. A monthly Event History
Calendar (EHC) was designed to help respondents remember when events occurred, in a
time continuum relative to pre—loaded public events. Important personal or relationship
events entered by a respondent were automatically displayed in the calendar, which

appeared on screen each time he/she was asked to date an event. The EHC could be

70

accessed at any time during the interview and dates could be corrected after they were
entered.

' Iii-Home Interview: Wave III. The Wave III questionnaire was designed to obtain
relationship, marital, childbearing, and educational histories, and to date key labor force
events. Some questions were unchanged from earlier waves. To enhance longitudinal
measures, new sections focus on topics more relevant to young adults.

Wave III is designed to provide data on the new domains of young adult life,
enabling researchers to model the dynamic processes of change over time. Wave III In-
Home Questionnaire Sections include overview and demographics, household roster and
residence history, parental support and relationships, retrospective ADI-ID, ﬁiends,
education, economics and personal future, tobacco, alcohol, drugs, self-image, and
interviewer's report.

Data Analyses

Before testing the theoretical model in its entirety, conﬁrmatory factor analysis
(CFA) was conducted to obtain estimates of parameters of the model and the residual
error variances of the observed variables as well as assess the ﬁt between the observed
variables and their underlying constructs. Conﬁrmatory factor analysis was conducted
with all latent constructs including, segregation, promoting/inhibiting environments,
adaptive culture, child characteristics, family/peer processes, developmental
competencies, and academic success. Maximum likelihood (ML) estimation was used to
perform the conﬁrmatory factor analyses. In assessing model ﬁt, the independence model

was compared to the hypothesized and re-speciﬁed models. The hypothesized model for

71

this study was based on the theoretical and conceptual model described above. Appendix
A presents the items used in each of the factors tested.

AMOS 5.0, a program for structural equation modeling techniques, was used to
evaluate how well the speciﬁed models adequately described the data. Given the purpose
of this study is to develop an empirical model that can be used for further research and
intervention, it was decided a priori to use model misspeciﬁcation information to revise
and improve models if warranted. It is important to note that AMOS does not allow for
the use of dichotomous categorical variables in analyses due to critical underestimation
errors in factor loadings, factor correlations, and error variances which occur when items
have fewer than three categories (Finch, West, & MacKinnon, 1997). Additionally, the
use of highly correlated items to assess latent constructs will produce poor goodness-of-
ﬁt indices. The use of computed or aggregated variables is preferred when confronted
with a number of correlated items. All statistical analyses were set with an alpha level of
.05 or <.05.

Survey items vary somewhat from survey to survey. As a result, not all items
were asked across all three waves of data collection. This distinction is particularly
noticeable in the case of Wave III. Items from Waves 1 and II and the In-school were
used predorrrinantly as indicators for all latent constructs except academic success. This
was done to account for variations in factors that may inﬂuence development over time.
Wave 111 items were only used in these latent constructs when other measures were not
available or not collected consistently across Waves I and H, for example the
retrospective report of ADI-ID symptoms. Manifest variables/indicators were eliminated

from latent constructs on the basis of the CFA results and previously reviewed literature.

72

Structural equation modeling (SEM) was used to determine whether the model is
a valid representation of Latino academic success. Structural equation modeling was used
to study the complex dependencies among the model constructs. SEM allows for
modeling relationships among multiple predictor and criterion variables, construct
variables (latent variables), model errors in measurements for observed variables, and
statistically test a prion/theoretical and measurement assumptions against empirical data.
This technique allows for multiple measures to be associated with latent constructs, and
to derive unbiased estimates for the relations between latent constructs. Additionally, this
technique can be used to compare multiple groups, such as subgroups of Latino
nationalities, and perform longitudinal studies. The relationships between the theoretical
constructs are represented by regression or path coefﬁcients between the factors. For
purposes of data analyses the hypothesized model to be tested is presented as a path
diagram above (Figure 3). Moreover, alternative more simpliﬁed models were also tested
and assessed for ﬁt. Inclusion or omission of paths within alternative models was based
on theoretical grounds and modiﬁcation indices for model ﬁtting were only used on the
basis of theoretical support (Boosma, 2000; Hoyle & Panter, 1995).

Exogenous Variables. Exogenous variables are those for which the model makes
no attempt to explain. Thus, exogenous variables are independent variables in all
equations in which they appear. In this model only the social position variables construct
is an exogenous variable. The social position construct included parent educational level,
race, gender, and ethnicity.

Endogenous Variables. Endogenous variables are those which the model attempts

to explain. Endogenous constructs are dependent variables in at least one equation,

73

although they may be independent variables in other equations in the system. In the case
of this hypothesized model perceived prejudice, segregation, school, neighborhood,
adaptive culture, child characteristics, family, peers, developmental competencies, and
academic success represent endogenous constructs.

Model Fit. Multiple goodness-of-ﬁt indices, residual error terms, modiﬁcation
indices, and their accompanying expected parameter change (Arbuckle & Wothke, 1999)
were used to assess model ﬁt. The indices used included the x2 to (If ratio, the Goodness
of Fit (GFI) and AGFI (Adjusted GFI) (Joreskog & S6rbom, 1989), the normed ﬁt index
(NF 1), the comparative ﬁt index (CF I), the Tucker-Lewis index (TLI) (Tucker & Lewis,
1973), and the root mean square error of approximation (RMSEA). If a model ﬁts
perfectly, all ﬁt indices except RMSEA should have the value 1. A value of at least .90
was required to accept a model, and a value of at least .95 was required to judge the
model ﬁt as “good.” For RMSEA values less than .05 indicate good ﬁt, values ranging
from .08 to .10 indicate mediocre ﬁt, and those greater than .10 indicate poor ﬁt (Browne

& Cudeck, 1993; MacCallum et al., 1996).

Results

Confirmatory Factor Analyses

Before testing the hypothesized model in its entirety CFA models were conducted
for each of the latent constructs. These were conducted to test the multidimensionality of
the theoretical constructs. Both conﬁrmatory and model generating factor analyses were
used to not only test the proposed models but develop theoretically grounded models that
are also empirically valid. Each of the hypothesized models was tested statistically in a

simultaneous analysis of the entire system of variables to determine the extent to which

74

each is consistent with the data (Byme, 2001). Results for each of the hypothesized latent
constructs are presented below.

Social Position Variables. Variables included as part of the Social Position
construct included ethnic identiﬁcation, racial category identiﬁcation, and parent
educational levels. Gender as a dichotomous variable was ultimately not used in these
analyses do to the limitations of SEM. Items regarding parental educational levels were
highly correlated. In order to account for the relationships between these variables a
single computed variable consisting of average parental educational level, including both
mother and father, was used in analyses. Similarly, youth’s report of their Latino
background and their choice of racial category in Wave 1 were correlated. These variables
were combined in order to create a second computed variable used in analyses. Although
these variables are conceptually linked in the hypothesized model, these constructs are
quite different social constructions and so testing them empirically as a single latent
construct was not appropriate. These computed variables were used as manifest variables
and their error variances calculated with 1138 cases. The error variance for the race-
ethnicity variable was 56.464 and for average parent educational level was 5.243.

Perceived Prejudice. Measures of perceived prejudice are the only indicators of
Social Stratiﬁcation Mechanisms available in the Add Health data set. Youth reported
whether students at their school are/were prejudice in Waves I and 11. These items were
highly correlated and so a computed variable using an average across both waves was
used in analyses. The error variance for this variable was calculated at 1.322 using 1138

cases.

75

Segregation. The Segregation construct included indicators related to residential,
economic, and social segregation. Ten dichotomous parental report items regarding their
reasons for living in their current neighborhood, including being able to afford better
housing, presence of friends, family, or children, and proximity to work, were totaled and
used as an indicator of residential segregation. Indicators of economic segregation
included parent report of total family income, ability to pay bills, and 7 dichotomous
questions regarding receipt of public/economic assistance. These nine questions related to
economic segregation were highly correlated, so a computed variable combining these
items was created as a measure of economic segregation. Social segregation indicators
included youth report of feeling socially accepted, others’ ﬁiendliness towards them, and
others’ liking them from Waves I and II. There were signiﬁcant correlations across these
items and across waves. A computed variable utilizing an average score across items and
waves was used as a measure of social segregation. These three computed manifest
variables were then used in conﬁrmatory factor analyses to test the validity of the
Segregation construct using 1299 cases. The tested model yielded a poor ﬁt to the data
including factor loadings that were not signiﬁcant ,CFI=.000, RMSEA=.091 (CI: .049,
.140). The three computed variables were used as independent manifest variables for
analyses. Error variance results were as follows; economic indicator, 18845.856;
residential indicator, .4.693; and social indicator, .173.

Promoting/Inhibiting Environments-School. The School construct included
indicators from youth, parent, and school administrator reports. Youth reports included
items regarding feeling close to people at school, part of the school, happy to be at

school, and safe at school, getting along with teachers, perceiving teachers to be fair, and

76

feeling teachers care about them from Waves I and H and the In-School survey. Parents’
report of school characteristics included items on whether they felt their child’s school
was a good one, prioritized learning, and was safe. Administrators’ report of school
characteristics included a wide variety of items regarding problems related to smoking or
tobacco use, drug use, alcohol use, gang violence, teenage pregnancy,
vandalism/thieving, eating disorders, stress or pressure, average daily attendance, number
of full-time classroom teachers, average class size, race and ethnicity of full time
teachers, gender of full time teachers, presence of parent organizations, dropout rates,
retention rates, achievement score rates, percentage of 12th graders in academic or
college preparatory, vocational or occupational preparatory, general program/no
divisions, and the availability of health related services, collected in two separate surveys.
Many of the items used to assess the school construct were correlated within and
across waves of data collection. Several computed variables were created in order to
account for the signiﬁcant correlations. The three parent report items were averaged to
create a computed variable representing parents’ perceptions of their child’s school.
Youth report items were averaged across all surveys to create a computed variable
gauging youth’s report of their school sense-of-belonging. Administrator reports of the
percentage of students dropping out in each grade were averaged to create a computed
variable representing school dropout rate. Administrator reports of the percentage of
students retained in each grade were similarly averaged to create a computed variable of
school retention rate. A school climate variable was created using the sum of
administrator report of dichotomous variables regarding problems related to smoking or

tobacco use, drug use, alcohol use, gang violence, teenage pregnancy,

77

vandalism/thieving, eating disorders, stress, or pressure. A variable reﬂecting services
offered was created as the sum of dichotomous items from the second administrator
report regarding the availability of physical and psychosocial services at school.

Several individual items were also included in the conﬁrmatory factor analyses.
These items were included administrator report regarding average daily attendance,
number of full-time classroom teachers, average class size, percent of Hispanic/Spanish
teachers, percent of female teachers, percent of students whose parents belong to the
school’s parent organization, percent of students testing at least one grade level above on
standardized tests, percent of students in academic/college prep courses, and school size.

Conﬁrrnatory factor analyses with 1533 cases resulted in weak indices indicating
the model was a poor ﬁt to the data, CFI=.441, RMSEA=.190 (CI=.185, .194). Using
modiﬁcation indices, several items were removed from the construct. The ﬁnal version of
the construct included 4 indicators, retention rates, average daily attendance, average
class size, and parent perceptions of their child’s school. This model yielded improved ﬁt
indices suggesting a stronger ﬁt to the data, CFI=.993; RMSEA=.032, (.000, .068). The

ﬁnal version of the construct is presented in Figure 4 below.

78

 

 

 

 

ATTEND AVERAGE RETAIN PARENT

 

 

 

 

 

 

 

 

 

 

 

RPRT
1 1 1 1
.19 22.02 12.73 .56

Figure 4. School latent construct with estimated parameters

Promoting/Inhibiting Environments-Neighborhood. The Neighborhood construct
was composed of youth’s report of how happy they are living in the neighborhood, how
happy or unhappy they would be to leave their neighborhood, 4 dichotomous variables
regarding youth knowing most people in the neighborhood, stopping to talk to a neighbor
in the past month, neighbors look out for each other, and feeling safe in their
neighborhood from Waves I and H. Additionally, parent report of their wanting to leave
neighborhood and 3 dichotomous variables regarding the presence of crime, drugs, and
trash in the neighborhood were also included. Computed variables were created that
combined the 4 dichotomous variables from youth reports, one for each wave of data. An
additional computed variable was created using the 3 dichotomous parental report items.
Youth reports of their happiness living in the neighborhood and how happy/unhappy they
would be to leave the neighborhood were used as 4 individual manifest variables.

Conﬁrmatory factor analyses using 2100 cases resulted in mediocre goodness of ﬁt

79

indices (CFI=.811; RMSEA=.119, CI: .111, .126). Upon reviewing the modiﬁcation
indices, two variables, happy/unhappy to move and computed youth report of
neighborhood, both ﬁ‘om Wave 1, were removed from the ﬁnal construct. The resulting
model yielded improved goodness of ﬁt indices, CFI equal to .982 and RMSEA equal to

.049 (CI: .033, .067), representing a good ﬁt to the data. This enhanced model is

 

 

 

presented in Figure 5 below.
.50
Neighborhood
.32 .20 .79 mg 1.00
APPY PARENT leHAaAc APP ‘ MOVE
w1 RPRT W2 W2

 

 

 

 

 

 

 

 

1 1 1

@77 @29 @1,14®.48 @92

Figure 5. Neighborhood latent construct with estimated parameters.

Adaptive Culture. The Adaptive Culture latent construct was tested using
indicators related to acculturation and current contextual demands. Acculturation
indicators included youth report items regarding youth’s US citizenship, country of birth,
language spoken at home, and parental birth country from Waves I and II. Interviewer
report of the language used for administering Wave 1 and H in-home surveys were also
used. Six items from the Wave H in-home interview were also used as indicators of

acculturation. These included language used with family, language used with ﬁiends,

80

purchasing music in a language other than English, and three items regarding the use of
media in non-English languages. Current contextual demands included items ﬁom Waves
I and H, and the In-school survey. These items served as proxy measures gauging youth’s
perceptions that they will have a middle class family, will be killed by age 21, will live to
age 35, and will be married by age 25.

Due to signiﬁcant correlations among adaptive culture construct items and across
waves, various computed variables were created for use in conﬁrmatory factor analyses.
Youth report of parents being born in the US. from Waves I and H along with parental
report of being born in the US were combined into one computed manifest variable.
Another computed variable representing youth’s language preferences consisted of youth
report of language used at home, with family, and with ﬁiends, along with interviewer
reports of language used to conduct surveys for both Waves I and 11. Youth report of their
use of non-English music and media items from Wave HI were averaged into one
indicator. Youth perceptions regarding having a middle class family, being killed by age
21, living to age 35, and being married by 25 from Wave I and the In-school report were
combined into a single manifest variable representing contextual demands. A separate
variable was created using these same items from Wave II. A measure of generational
status was computed using youth report of being born in the United States and born a
US. citizen from Waves 1 and HI.

Conﬁrmatory factor analyses with 1841 cases yielded only mediocre goodness of
ﬁt indices, CFI=.846; RMSEA=.151 (CI: .138, .164). Modiﬁcation indices and a factor
loading that was not signiﬁcant suggested the removal of the Wave II contextual

demands computed variable. The revised version of the Adaptive Culture construct

81

showed greatly improved goodness of ﬁt results, CFI=.990; RMSEA=.049 (CI: .031,

.067). Figure 6 presents this revised model with resulting estimates.

     
   

Adaptive -05

culture

 

usic .’ . Gen.

Media

    

 

 

 

 

 

 

.58 .09 .03 .65 .10

Figure 6. Adaptive culture latent construct with estimated parameters

Child Characteristics. The Child Characteristics latent construct included items
from all three waves of data collection, the In-school survey, and parent report. This
construct included youth’s report of their intelligence, themselves compared to others,
having good qualities, feeling proud of themselves, liking themselves, feeling life was a
failure, trouble paying attention in school, trouble getting homework done, trouble
keeping mind focused, likelihood that they will attend college, likelihood that they will
graduate from college, and whether they consider themselves and/or others consider them
to have a disability ﬁom Waves I and II and the In-school survey. Items used from the
parent report survey included whether their child is mentally retarded, has a learning

disability, has received special education, and is considered to have a disability by parent

82

and/or others. Finally, 18 items from Wave ID that comprised an ADHD retrospective

survey of symptoms were also used.

   
  

1.00
.76

m IISABL LFEST1 LFEST

a“,

Figure 7. Child characteristics latent construct with estimated parameters

 

 

 

 

.02

Several computed variables were created utilizing the above named variables in
order to account for the signiﬁcant correlations among items and across waves. An
indicator of ADHD was computed totaling scores on all 18 ADHD retrospective report
question responses along with youth report items regarding trouble paying attention in
school, trouble getting homework done, and trouble keeping mind focused across Wave I
and H. Items from Wave 1, youth consider themselves and/or others consider them to
have a disability, were averaged to create a computed variable of disability. All of the
parent questions were also combined to create a parental measure of youth
learning/physical disability. Two self-esteem computed variables were created, one for
Wave I and another for Wave H, using youth’s report of their intelligence, themselves
compared to others, having good qualities, feeling proud of themselves, liking

themselves, and feeling life was a failure. Finally, a measure of youth’s attitudes towards

83

education was computed using Wave 1, II, and In-school survey items regarding the
likelihood that they will attend college and likelihood that they will graduate from college
items. Conﬁrmatory factor analyses with 1579 cases showed this model to be a good ﬁt
to the data, CFI=.972 and RMSEA=.045 (CI: .031, .061), despite the youth disability
indicator having a factor loading that was not signiﬁcant. Figure 7 above presents the
Child Characteristics construct with estimated parameters.

Family. The Family construct included indicators regarding parental involvement
in school, parent perceptions towards education, and general parental support. The parent
involvement in school indicators included parent survey items regarding talking with
youth regarding school work or grades, talking with youth about school in general,
talking with youth’s teachers, volunteering at school, and working with youth on school
projects. Parent perceptions towards education included items from Wave I, H, and the
parent survey regarding level of parents’ disappointment if youth did not graduate from
high school and if youth did not graduate from college. Parent support included a series
of dichotomous youth report items regarding going shopping, playing a sport, attending
religious service, discussing life, going to the movies, discussing a personal problem, and
arguing about youth’s behavior with their mother and father, as well as parental caring,
good communication, good relationship, understanding, having fun with family, and
family paying attention from Waves I and 11. Additional items from the parent survey
were also used as indicators of parental support including dichotomous variables
regarding having met their adolescent’s best friend and best friend’s parents, having met
adolescent’s special boyfriend/girlfriend and the special boyfriend/girlfriend’s parents,

talking to the parents of adolescent’s ﬁiends, as well as making decisions together with

84

adolescent, understanding adolescent, trusting adolescent, and to what degree adolescent
interferes with activities.

Computed variables were created to compensate for the large amount of
dichotomous variables and signiﬁcant correlations among items and across waves of data
collection. A measure of parent caring was created using youth report items regarding
mother caring, father caring, and parents caring from Waves I and H. Parental
involvement with adolescent’s ﬁiends was gauged by a computed measure consisting of
parent report of having met adolescent’s best friend and best friend’s parents,
adolescent’s special boyfriend/girlfriend and the special boyfriend/girlfriend’s parents,
and talking to the parents of adolescent’s friends. Parent report of talking with youth
regarding school work or grades, talking with youth about school in general, talking with
youth’s teachers, volunteering at school, and working with youth on school projects were
used to create a computed parental school involvement measure. Parent attitudes towards
education were measured using two computed variables, one for Wave I and another for
Wave 11. Youth report of how disappointed their mother and father would be if they did
not graduate high school and college was used to create each one. An indicator of parent-
child relationship was created using parent report items regarding how well they get
along with adolescent, how often they make decision together, feel they understand
adolescent, can really trust adolescent, and how often adolescent interferes with activities.
Finally, an overall measure of family support was created using the youth report items
from Wave I and H pertaining to family understanding, family having fun together, and
family paying attention to youth, as well as parental caring, good communication, and

good relationship. The family processes construct was tested using 2012 cases and had

85

excellent ﬁt indices. Conﬁrmatory factor analyses results showed CFI to be equal to 1.00,
AGFI equal to .996, and RMSEA equal to .000 (CI: .000, .020). Figure 8 presents this

construct with its estimated parameters.

  

 

 

 

 

 

 

 

 

 

 

 

FAMOVRAL PAR PAR ’ FARCARI rARREL rARFRDl
- DAT1‘ DAT , -

1 1 1 1 1 1

@73 @7. @99 é... ‘31 .7.

Figure 8. Family processes latent construct with estimated parameters.

Peer Processes. The Peer Processes construct included several indicators of peer
inﬂuence. Youth report items from Wave 1, H, and the In-school survey used in this
construct incorporated questions regarding time spent hanging out with friends, a series
of dichotomous questions for male and female friends concerning going over to ﬁiends’
houses, meeting after school, spending time on the weekends, talking about a problem,
and talking on the phone. Additionally, items ﬁ'om Wave I and H regarding ﬁiends
caring, friends smoking, drinking alcohol, and using marijuana were also incorporated. A
single item from the parent survey concerning the inﬂuence of adolescent’s best ﬁiend
was also used

Related peer inﬂuence items were signiﬁcantly correlated within and across
waves of data collection. Computed variables were created to account for these

correlations as well as the dichotomous activities with friends by gender questions. All

86

items from Wave 1 and H with reference to friends drug and alcohol use were combined
to create a substance use inﬂuence variable. Two items concerning time spent with peers
were computed. These items were computed using Wave 1, H, and In-school items by
gender, such that an indicator of time spent with male friends and another of time spent
with female friends were created. Wave I and H items regarding hanging out with friends
were similarly combined to create an overall indicator. Friends caring questions from
Wave I and II were also averaged to create a computed indicator. The item from the

parent survey was used as an independent manifest variable.

.10
1'00 1.95 1.63 --61

 

FRNDSHNGHTIMEBOYS] IMEGRLS] pause]

Figure 9. Peer processes latent construct with estimated parameters.

 

 

Conﬁrmatory factor analyses with 1242 cases showed this construct was a poor ﬁt
to the data, CFI equal to .747 and RMSEA equal to .101 (CI: .086, .117). Modiﬁcation
indices suggested the removal of the parent survey item and the ﬁiend caring computed
indicator. This revised version of the construct had a much better ﬁt to the data. Results
included CFI equal to .989 and RMSEA equal to .037 (CI: .000, .076). Figure 9 above

portrays the revised construct with its estimated parameters.

87

Developmental Competencies. The Developmental Competencies construct
included indicators from Wave 1 and H youth report. These include measure of youth
time allocation such as time spent in after-school and out-of-school activities, athletics,
physical activity, watching television or playing video, using the computer for activities
other than schoolwork, and at religious services or activities. Academic motivation
measures from Wave 1, H, and the In-school included youth’s report of trying hard at
school, paying attention in school, and wanting to go to college. Additionally, a variety of
school behaviors were included in the Developmental Competencies construct. The
measures of school behaviors used were youth’s report of the number of times they have
skipped school, ever being high or drunk at school, and ever having problems with school
because of being drunk or high, and taking/carrying a weapon at/to school from Wave 1,
H, and the In-school survey. Items from Wave I and H regarding grades skipped, grades
retained, suspensions, and expulsions were also incorporated.

Items included as indicators of Developmental Competencies were found to be
signiﬁcantly correlated both within and across waves of data collection. Computed
variables were developed and used in conﬁrmatory factor analyses in order to account for
dichotomous variables and item correlations. A measure of participation in after-school
and out-of-school activities was created incorporating religious activities and services,
clubs, groups, athletics, physical activity, watching television, and playing video games
from Wave I and H items. A computed variable using In—school survey items regarding
club, activity, and athletics participation was also created. A measure of responsible
school behavior was computed using Wave I and II reverse coded items regarding use of

substances at school and carrying a weapon at school from Waves I and H. The average

88

sum of hours spent watching television or videos and/or playing video games was created
using Wave I and 11 items. An academic motivation indicator regarding wanting to go to
college was computed using the youth report items from Wave I and H. A computed
variable combining suspensions, expulsions, and retentions (repeating a grade) was
created using reverse coded items from Waves I and II. Dichotomous items from Wave I
regarding skipping grades were combined into a computed variable. The three items
regarding skipping school from Wave 1, H, and the In-school survey were used as
individual manifest variables for purposes of conﬁrmatory factor analysis. The In-school
survey item regarding working hard at school was also used as a unique manifest

variable.

    

10.53

RETI 1
usnsx-
Nﬁ. RESP 1 1.34
‘—
141.39
1.00 ms 1

Developmental
competencies

    
  

 

 

 

 

 

 

Figure 10. Developmental competencies latent construct with estimated parameters.

89

Conﬁrmatory factor analyses of the Developmental Competencies construct were
conducted using 1533 cases. Results showed this model was a poor ﬁt to the data. Fit
indices results included CFI equal to .609 and RMSEA equal to .102 (CI: .095, .109).
Modiﬁcation indices suggested the removal of three variables, the In-school measure of
trying to do well at school, the computed measure of ever skipping a grade, and the
computed school group measure. The revised construct (see Figure 10 above) had a
stronger ﬁt with the data. CFI was equal to .962 and RMSEA was equal to .036 (CI: .023,
.049).

Academic Success. The Academic Success construct included indicators from
Waves 1, II, and HI of the youth report surveys. Items from Wave 1 and H included youth
report of grades received in English, math, history, and science. Wave IH items
concerning highest grade completed, 8 dichotomous variables regarding degrees
received, and vocational trainings attended were also used. A computed variable of
grades averaged across subjects and waves was created due to signiﬁcant correlations
across items and waves. A computed variable summing all 8 dichotomous degrees
received variables was created as a measure of attainment. The highest grade completed
variable and the vocational training attended variable were used as independent manifest
variables. This construct was tested using 2377 cases and results showed ahnost perfect
ﬁt to the data. CFI was equal to 1.00, AGFI was equal to 1.00, and RMSEA was equal to

.000 (CI: .000, .014). Parameters estimates are shown in Figure 11.

90

     

2.43

  

Academi
success

 

 

 

 

1.00
IGHES EGREE
GRADE
.1 C .1
7.08 .58 .38 .30

Figure 11. Academic success latent construct with estimated parameters.
Structural Equation Modeling

Hypothesized Model. Using the results from the conﬁrmatory factor analyses
presented above, the hypothesized overall adapted augmented model was tested using
structural equation modeling in AMOS 5.0. The observed covariance matrix of the
variables was compared to what would have been expected given the set of relationships
that were proposed. The model included only one dependent factor, academic success,
and thirteen independent factors, average parent educational level variable, race—ethnicity
variable, student prejudice variable, economic segregation variable, residential
segregation variable, social acceptance variable, neighborhood inﬂuence, school
inﬂuence, adaptive culture, child characteristics, family inﬂuence, peer inﬂuence, and
developmental competencies.

The model was constructed based on the factor analyses done previously which
identiﬁed the most unique items for each measure. The revised versions of the constructs
along with their estimated parameters were used to test the model in its entirety. After

imputing missing variables (see Table 5 below) the hypothesized model was tested using

91

1138 cases and resulted in a poor ﬁt to the data. Goodness of ﬁt indices included CF 1
equal to .405, AGFI equal to .774, and RMSEA equal to .065 (CI: .064, .067). Although
the model as a whole was rejected certain expected relationship were found to be

signiﬁcant. The predicted relationships with their standardized estimated parameters and

signiﬁcance levels are speciﬁed below.

92

Table 6

Missing Variable Imputations

 

 

Percent Irnputed

Items (N=1 138)
Contextual demands computed variable 0.2
Parents US born computed variable 0.2
Language preference computed variable 0.2
Non-English music and media use computed variable 0.2
Generational status computed variable 0.2
Youth report of disability computed variable 14.3
Parent report of youth disability computed variable 14.3
Self-esteem Wave I computed variable 14.3
Self-esteem Wave II computed variable 14.3
Youth attitudes towards education 14.3
ADHD retrospective scale score 14.3
Youth ever skip school-Wave I 0.2
Youth ever skip school-Wave II 0.2
Parent involvement with youth friends computed variable 17.1
Parental relationship with youth computed variable 17.1
Youth report of parental/family caring 17.1
Family support computed variable 17.1
Parent attitudes towards education-Wave I computed variable 17.1
Parent attitudes towards education-Wave II computed variable 17.1
Peers’ drug use computed variable 26.9
Friends care computed variable 26.9
Hanging out with ﬁiends computed variable 26.9
Time spent with boys computed variable 26.9
Time spent with girls computed variable 26.9
How happy living in neighborhood-Wave I 14.7
Parent report of neighborhood computed variable 14.7
Neighborhood characteristics-Wave I computed variable 14.7
How happy living in neighborhood-Wave II 14.7
Happy/Unhappy to move-Wave II 14.7
Economic segregation computed variable 28.6
Residential segregation computed variable 28.6
Social segregation computed variable 28.6

 

Direct Relationships. Several direct relationships were predicted according to the
hypothesized adaptive augmented model. Below each of the expected relationships are
presented with their regression results from the SEM analyses. Table 6 summarizes the

direct relationship results below.

93

Social position variables were expected to predict measures of social stratiﬁcation
mechanisms. Social position variables were expected to have a negative relationship to
perceived prejudice. Results showed that neither the race-ethnicity variable ([3 =.049,
p=.099) nor the average parental educational level variable ([3 =-.014, p=.639) were
signiﬁcantly predictive of perceived prejudice.

Social stratiﬁcation mechanisms were expected to impact segregation. Greater
perceived racism should have predicted increased residential, economic, and social
segregation. In this case the constructs should have all been negatively related. SEM
results showed that youth’s perception that students at school were/are prejudice was not
signiﬁcantly related to the economic segregation or residential segregation variables, B:
.022, p=.451 and B: .035, p=.242 respectively. The perceived prejudice factor was
signiﬁcantly predictive of social segregation but not in the direction expected ([3: .113,
p<.001).

Social stratiﬁcation mechanisms were expected to directly predict
promoting/inhibiting environments. Increased perceptions of racism would be predictive
of more inhibiting environments, a negative relationship. The perceptions of prejudice
variable was not signiﬁcantly predictive of either neighborhood environment (B= .062,
p=.062) or school environment ([3= .06, p=.184).

Social stratiﬁcation mechanisms were predicted to impact adaptive culture. Increased
perceptions of prejudice were expected to predict increased contextual demands and
decreased bicultural orientation, a negative relationship. A signiﬁcant relationship was
found between youth’s perceptions of prejudice and adaptive culture factors in the

direction expected, [3: -.078, p=.015.

94

Promoting/inhibiting environments were anticipated to predict family/peer processes.
Promoting environments were reasoned to lead to increased positive family and peer
inﬂuence. School environment was found to be signiﬁcantly and positively related to
family processes, 8: .276, p<.001. School environment was also found to be signiﬁcantly
predictive of peer inﬂuence but not in the direction expected, 8: -.202, p=.005.
Neighborhood environment was also found to be signiﬁcantly and positively predictive
of family processes, B= .291, p<.001. Neighborhood environment was not signiﬁcantly

related to peer inﬂuence (B= .079, p=.125).

Promoting/inhibiting environments were anticipated to directly impact adaptive
culture. Promoting environments were thought to predict lower contextual demands and
higher level of bicultural acculturation or in other words to have positive relationships.
School environment was found to be signiﬁcantly predictive of adaptive culture as
expected, [3: .247, p<.001. Neighborhood environment, on the other hand, was not found

to be signiﬁcantly predictive of adaptive culture (B= .045, p=.214).

Adaptive culture was expected to directly impact child characteristics. Low levels of
contextual demands and bicultural orientation would predict positive child characteristics,
a positive relationship. Results showed that indeed there was a signiﬁcant positive
relationship between adaptive culture and child characteristics, B=.131, p<.001.

Adaptive culture was also anticipated to directly impact family/peer processes. Low
levels of contextual demands and bicultural orientation were expected to predict positive
family processes and positive peer processes. Although adaptive culture was signiﬁcantly

related to family processes it was in a negative rather than a positive direction, 8: -.186,

95

p<.001. Adaptive culture’s relationship to peer inﬂuence was both signiﬁcant and
positive, 8: .256, p<.001

Additionally, adaptive culture was predicted to directly impact developmental
competencies. Low levels of contextual demands and bicultural orientation would be
predictive of positive academic competencies. Results showed that adaptive culture did
not signiﬁcantly predict developmental competencies (B: .083, p=.083)

According to the overall model, child characteristics were presumed to directly
inﬂuence family/peer processes. Positive characteristics were expected to be predictive of
positive family processes and peer processes. Child characteristics was found to be
signiﬁcantly predictive of family processes ([3: .685, p<.001), but not predictive of peer
processes ([3: .018, p=.738).

It was anticipated that child characteristics would impact developmental
competencies, such that positive characteristics would predict positive academic
competencies. Result showed the relationship between child characteristics and
developmental competencies was not signiﬁcant (B= -.102, p=.245).

Family/peer processes were also expected to impact developmental competencies.
Positive family and peer inﬂuences were anticipated to predict positive academic
competencies. Family processes were found to be signiﬁcantly and positively predictive
of developmental competencies, B= .651, p<.001. Peer inﬂuences, on the other hand,

negatively predicted developmental competencies, B: -.343, p<.001.

96

Table 7

Standardized regression weight estimates for predicted direct relationships

 

 

 

Relationship Estimate S.E. P
Race-ethnicity 9 Perceived prejudice -0.014 .005 .099
$163tgigrfaff£el 9 Perceived prejudice (“13 .015 .639
Perceived prejudice 9 Social segregation 0.022 .011 ***
Perceived prejudice 9 Economic segregation 0.035 11.208 .451
Perceived prejudice 9 Residential segregation 0.062 .056 .242
Perceived prejudice 9 Neighborhood 0.06 .021 .062
Perceived prejudice 9 School 0.217 .003 .184
Social segregation 9 Neighborhood 0.086 .059 ***
Economic segregation 9 Neighborhood 0.172 .000 .009
Residential segregation 9 Neighborhood 0.114 .011 ***
Social segregation 9 School 0.177 .008 .010
Economic segregation 9 School -0.047 .000 ***
Residential segregation 9 School -0.078 .001 .288
Perceived prejudice 9 Adaptive culture 0.247 .006 .015
School 9 Adaptive culture 0.045 .159 ***
Neighborhood 9 Adaptive culture 0.131 .01 l .214
Adaptive culture 9 Child characteristics -0. 186 .073 ***
Adaptive culture 9 Family 0.256 .022 ***
Adaptive culture 9 Friends 0.276 .054 ***
School 9 Family -0.202 .090 ***
School 9 Friends 0.291 .235 .005
Neighborhood 9 Family 0.079 .006 ***
Neighborhood 9 Friends 0.685 .016 .125
Child characteristics 9 Family 0.018 .011 **"'
Child characteristics 9 Friends 0.083 .027 .738
Adaptive culture 9 Developmental competencies -0.343 .090 .083
Friends 9 Developmental competencies -0.102 .112 ***
Child characteristics 9 Developmental competencies 0.651 .083 .245
Family 9 Developmental competencies 0.5 .427 ***
Developmental 9 Academic success 0569 .063 ***
competencres

*** p<.001

Developmental competencies were expected to directly impact academic success.

Positive academic developmental competencies were anticipated to predict higher grades

97

and higher levels of attainment. SEM results showed developmental competencies to be
positively and signiﬁcantly predictive of academic success, [3: .500, p<.001.

Mediating Relationships. According to the hypothesized model there were several
mediating relationships between constructs that could be expected. The standardized
indirect effects estimates for all predicted mediating relationships are presented in Table
7 below.

The relationships between social position variables and segregation were expected to
be mediated by social stratiﬁcation mechanisms. This relationship was expected to be a
fully mediated one, such that perceived prejudice would completely mediate the effects
of race-ethnicity and average parent educational level on residential, economic, and
social segregation. The race-ethnicity variable and the average parent educational level
variable were expected to have only indirect effects on the segregation variables. As
previously reported, race-ethnicity and parent educational level had no direct effect on
perceived prejudice and perceived prejudice had no direct effect on either the residential
or economic segregation variables. Results also showed that race-ethnicity had no
signiﬁcant indirect effect on residential segregation (p=.15) or economic segregation
(p=.482). Similarly, average parent educational level showed no signiﬁcant indirect
effects on residential segregation (p=.581) or economic segregation (p=.883). Although
perceived prejudice was shown to have a signiﬁcant direct effect on social segregation,
neither race-ethnicity (p=.06) nor parent educational level (p=.57) had signiﬁcant indirect
effects on social segregation.

It was anticipated that social stratiﬁcation mechanisms would mediate the

relationships between social position variables and promoting/inhibiting environments.

98

Race-ethnicity and parent educational level’s effect on school and neighborhood
environments were expected to be completely mediated by youth’s perceptions of
prejudice. As reported above, race-ethnicity and parent educational level had no direct
effect on perceived prejudice and perceived prejudice had no direct effect on either
school or neighborhood enviromnents. Consequently, results showed no signiﬁcant
indirect effects of race-ethnicity on school environment (p=.23 8) or neighborhood
environment (p=.06). Similarly, no signiﬁcant indirect effects-were found for average
parent educational level on school environment (p=.67) or neighborhood environment
(p=.57).

Segregation was expected to mediate the relationships between social stratiﬁcation
mechanisms and promoting/inhibiting environments. The relationship between perceived
prejudice and school and neighborhood environments was expected to be partially
mediated by residential, economic, and social segregation. Social segregation and
economic segregation were found to have a direct effect on school environment, B= .114,
p=.01 and B= .177, p<.001 respectively. No direct effects were found for either
residential segregation (p=.456) or perceived prejudice (p=.276) on school environment.
No indirect effect was found for perceived prejudice on school environment (p=. 14).

Signiﬁcant direct effects were found for residential segregation, economic
segregation, and social segregation on neighborhood environment, B: .172, p<.001, B=
.086, p=.009, and B= .217, p<.001 respectively. Only an indirect effect, but no direct
effect (p=.092), of perceived prejudice on neighborhood environment was found to be

statistically signiﬁcant ([3= .032, p=.01).

99

Child characteristics were anticipated to mediate promoting/inhibiting environments’

relationship to family/peer processes. School and neighborhood environments were

expected to have both a direct effect and an indirect effect, through child characteristics,

on farmly and peer processes. As reported above, school environment had direct effects

on both family processes and peer inﬂuence, positively and negatively respectively.

Neighborhood environment was also found to have a direct effect on family processes,

but not on peer inﬂuence. Child characteristics were found to have a direct effect on

farrrily processes, but not on peer inﬂuence. Results showed school environment had an

indirect effect on peer inﬂuence ([3: .064, p=.01), but not on family processes (p=.082).

Neighborhood did not have an indirect effect on either peers (p=.31) or family (p=.367).

 

 

 

Table 8

Standardized indirect eﬂects estimates for mediated relationships

Relationship Estimate
Race-ethnicity 9 Residential segregation 0.002
Race-ethnicity 9 Economic segregation 0.001
Race-ethnicity 9 Social segregation 0.006
Race-ethnicity 9 School 0.004
Race-ethnicity 9 Neighborhood 0.005
Average parent educational level 9 Residential segregation 0
Average parent educational level 9 Economic segregation 0
Average parent educational level 9 Social segregation 0002
Average parent educational level 9 School -0.001
Average parent educational level 9 Neighborhood -0.001
Perceived prejudice 9 School 0.015
Perceived prejudice 9 Neighborhood 0.032“
School 9 Friends 0.064"
School 9 Family 0024
Neighborhood 9 Friends 0.012
Neighborhood 9 Family -0.004
Child characteristics 9 Developmental competencies 0.44"

"p=.01

100

Finally, according to the hypothesized model family/peer processes were expected
mediate the relationships between child characteristics and developmental competencies.
Child characteristics were expected to have both a direct and an indirect effect on
developmental competencies. As was reported above, no direct effects were found for
child characteristics on developmental competencies. Altemately, signiﬁcant direct
effects were found for both family and peer processes on developmental competencies. A
signiﬁcant indirect effect was found for child characteristics on developmental
competencies, B: .44, p=.01.

Alternate Models

The hypothesized model as a whole was also tested for its predictive ability with
one Latino subgroup, those of Mexican descent. Structural equation modeling results
with 545 cases yielded poor ﬁt indices for this sub-sample, (CFI=.000, AGFI=.683,
RMSEA=.091). Given the poor ﬁt for the overall sample and this subgroup model
invariance analyses were not undertaken.

Given poor overall model ﬁt results, revised models and portions of the model
were tested using the entire Latino sample to further examine relationships between the
model latent constructs. The ﬁrst alternate model tested removed several items and
constructs that were limited in their predictive ability due to limitations of the Add Health
data set. These included social position factors (race-ethnicity and parent educational
level), the perceptions of prejudice factor, and all three segregation factors (residential,
economic, and social). The rest of the model was tested in it’s entirety as presented
above. Fit indices indicate that this revised model was not a good ﬁt to the data. Model ﬁt

results were CFI equal to .490 and RMSEA equal to .063 (CI: .061, .065). As was found

101

in the initial results for the overall model, there were no signiﬁcant relationships between
neighborhood environment and adaptive culture, neighborhood and peer inﬂuence, child
characteristics and friends, adaptive culture and developmental competencies, or child
characteristics and developmental competencies (Table 8). Tables 8 and 9 summarize the

estimates for all predicted direct and indirect effects in the model.

 

 

 

   
  
  

 

Table 9
Standardized regression weight estimates for 8 component alternate model
Relationship Estimate S.E. P
School 9 Adaptive culture 0.038 .155 ***
Neighborhood 9 Adaptive culture 0.13 .012 .303
Adaptive culture 9 Child characteristics -0.162 .074 ***
Adaptive culture 9 Family 0.245 .022 ***
Adaptive culture 9 Friends 0.237 .053 ***
School 9 Family 0211 .088 ***
School 9 Friends 0.27 .229 .003
eighborhood 9 Family 0.067 .007 **"‘
eighborhood 9 Friends 0.697 .017 .199
'1d characteristics 9 Family 0.015 .012 ***
'ld characteristics 9 Friends 0.078 .027 .774
daptive culture 9 Developmental competencie ~0.344 .090 .098
riends 9 Developmental competencie -0.076 .112 ***
hild characteristics 9 Developmental competencie 0.627 .086 .402
arnily 9 Developmental competencie 0.501 .440 **"‘

 

 

 

 

 

velopmental competencies 9 Academic success 0.038 .063 ***

#*#p<.001

Table 10

Standardized indirect eﬁects estimates for 8 component alternate model

Relationship Estimate
School 9 Friends 0.045”
School 9 Family -0.013
Neighborhood 9 Friends 0.009
Neighborhood 9 Family -0.003
Child characteristics 9 Developmental competencies 0.432"

”p=.01

102

The second alternate model tested was a version of the overall model adapted
according to theory and modiﬁcation indices results. The race-ethnicity and parent
educational level social position factors, the residential segregation factor, and the peer
inﬂuence latent construct were eliminated from this structural equation model. Other
modiﬁcations included the deletion of predicted relationships between perceived
prejudice and economic segregation, neighborhood environment, and school
environment. Finally the relationship between child characteristics and developmental
competencies was also eliminated. Modiﬁcation indices also suggested the addition of
several direct relationships between constructs that were included in this model. These
additional direct relationships were between economic segregation and adaptive culture,
social segregation and child characteristics, social segregation and child characteristics,
school environment and child characteristics, school environment and developmental
competencies, child characteristics and academic success, and ﬁnally social segregation
and academic success. A simpliﬁed latent construct version of this alternate model is
presented in Figure 12 below. All of the expected direct relationships were found to be
signiﬁcant (see Table 10), but the model as a whole remained a poor ﬁt to the data,

CFI=.551 and RMSEA=.059 (CI: .058, .061).

103

 

 

 

 

 

 

 

Economic (
segregation
Neighborhood
Perceived

 

 

 

 

     
    
 
 
 
   

Adaptive
culture

Developmental
competencies

 

 

 

 

prejudice r
* characteristics Academic
success
Social
segregation

Figure 12. Model altered according to modiﬁcation indices

Table 11

Standardized regression weights for simpliﬁed model

Relationship Estimate S.E. P
Economic segregation 9 School 0.157 .000 ***
School 9 Adaptive culture 0.113 .154 .001
Perceived prejudice 9 Social segregation -0.064 .011 "*
Perceived prejudice 9 Adaptive culture 0.204 .006 .036
Residential segregation 9 Adaptive culture 0.231 .000 ***
Social segregation 9 Neighborhood 0.093 .058 ***
Economic segregation 9 Neighborhood 0.636 .000 .005
Social segregation 9 Child characteristics 0.115 .032 "*
School 9 Child characteristics 0.078 .287 .009
Adaptive culture 9 Child characteristics -0.109 .066 .013
Adaptive culture 9 Family 0.252 .021 .011
Neighborhood 9 Family 0.682 .007 ***
Child characteristics 9 Family 0.457 .011 "*
Family 9 Developmental competencies 0.36 .194 ***
School 9 Developmental competencies -0.088 .365 ***
Adaptive culture 9 Developmental competencies 0.366 .083 .041
Developmental competencies 9 Academic success 0.374 .068 ***
Child characteristics 9 Academic success -0.146 .076 ***
Social segregation 9 Academic success 0.157 .070 .003

‘**p<.001

104

A ﬁnal alternate model was tested to explore the possibility of more direct
relationships between, segregation, neighborhood, family, developmental competencies,
and academic success. A simpliﬁed SEM version of the tested model is presented below
in Figure 13. Of note are the predicted direct effects of perceived prejudice on
developmental competencies and social segregation on family processes. Table 11
presents the SEM regression weights results for this model. Results show all the
anticipated relationships were statically signiﬁcant. The model ﬁt indices indicate that the
model as a whole was only a mediocre ﬁt to the data. CFI was equal to .643 and RMSEA
was equal to .053 (CI: .050, .056). Results of the overall and alternate models are

discussed below.

 

Economic
segregation

     
   
   
  
  
    

 

 

 

 

Academic
success

Perceived
prejudice‘

NeighborhOo .

    

 

 

 

 

 

 

 
 

 

 

Residential
egregation
lovelo emel‘etgi
, c m nc
Social pe
egregation

 

 

Figure 13. Perceived prejudice and segregation direct relationships model

105

Table 12

Standardized regression weights for perceived prejudice and segregation direct
relationships model

 

 

 

 

 

 

Estimate S.E. P
erceived prejudice 9 Social segregation 0.219 .011 ***
Social segregation 9 Neighborhood 0.085 .058 ***
onomic segregation 9 Neighborhood 0.174 .000 .010
esidential segregation 9 Neighborhood 0.296 .011 ***
Neighborhood 9 Family 0.452 .007 ***
Social segregation 9 Family 0.518 .011 ***
Family 9 Developmental competencies 0.102 .199 ***
Perceived prejudice 9 Developmental competencies 0.496 .013 .004
Developmental competencies 9 Academic success 0.219 .062 ***
***p<.001
Discussion

The present study examined the empirical validity of a conceptual model of
Latino academic success. Building from Garcia Coll and her colleagues’ (1996) work, an
adapted augmented integrative model of developmental competencies and academic
success was postulated. This conceptual model was broad in scope, centering on social
stratiﬁcation, racism, prejudice, discrimination and oppression, and segregation, and was
tested using the Add Health nationally representative data set. An overarching and
guiding normative model has been largely lacking in previous and current efforts aimed
at improving the academic outcomes of Latinos. The focus of this study was to ﬁll that
gap by identifying an empirical model that could be used to guide further research,
intervention, and policy.

As the largest minority youth population in the United States, Latino youth are
poised to be one of the most inﬂuential groups in the nation. And yet there are gaps in the

levels of academic success Latinos achieve as compared to their black and white peers.

106

Increasing the academic achievement and attainment of Latino youth is a matter of
national importance. But a comprehensive model of Latino academic success has yet to
be elucidated through research. As was presented earlier, the great majority of the
research on Latino youth and their educational outcomes has focused on their failures and
not their successes. These failure approaches have not lead to the kinds of programming
or policy which have been able to reduce the educational gaps between Latinos and their
non-Hispanic peers. This study took a normative approach and used a strengths paradigm
to examine those factors and processes of development that improve Latino youths’
competencies and outcomes.

Results in this study did not support the hypothesized model as whole, but
important progress was made towards the development of an empirically tested
normative model of Latino youth academic development. The implications of the
predicted relationships and overall model results are discussed further. Limitations and
strengths of the study are also addressed in greater detail below. Suggestions for future
areas of research building from these results are discussed. Finally, a challenge is made to
policy makers, intervention designers and staff, and researchers to take on a normative
development frame of reference and be motivated by the results of this study.

Major Findings

Overall Hypothesized Model

The overarching goal of this research was to lay the foundation for the
establishment of a normative empirically tested model of Latino academic outcomes that
could be practically applied to future research, intervention, and policy and educational
reform. Garcia C011 and colleagues’ (1996) integrative model of minority developmental

competencies was used as the basis for developing an empirical model that is centered on

107

normative development and a strengths based approach while recognizing the unique
constellation of distal and proximal factors affecting the lives of Latino youth. Below is a
discussion of the structural equation modeling results for this model and study limitations
affecting these results.

Results from the structural equation modeling analyses showed that the adapted
augmented integrative model of Latino academic success was not a good ﬁt to the
national data set. Analyses included 8 latent constructs, over 45 manifest variables, and
over 25 predicted relationships among model constructs. Given the high level of
complexity present in the model the lack of signiﬁcant ﬁndings is not completely
surprising. Interestingly, although goodness of ﬁt indices overall did not indicate good
model ﬁt, some of the indices, such as the RMSEA (.065), suggested that model with
some modiﬁcations might have better results. The alternative models explored after
initial analyses further examine the relationships suggested by the hypothesized model
and contribute to our knowledge regarding the academic development of Latino youth.

In addition to the high level of complexity in the model, other limitations to this
research may have also contributed to the lack of signiﬁcant ﬁt results for the model. In
particular, although the Add Health data is quite large and includes a myriad of variables,
the data set was designed to capture information more speciﬁcally related to physical
health outcomes. So, for some constructs in the model, particularly social position
variables, social stratiﬁcation mechanisms, and segregation, the data elements available
for use in testing the model were extremely limited. This limitation on the availability of
data elements meant that many aspects of some proposed latent constructs were not

included in analyses. As an example, the limitations in regards to data regarding

108

perceived and experienced prejudice, discrimination, and oppression meant only a limited
aspect of the conceptual social stratiﬁcation mechanism construct was included in the
model. Similarly, neighborhood environment characteristics such as support from
neighbors/other adults and opportunities for positive use of time could not be assessed
using the Add Health data set. Items related to both youth and parent perceptions towards
education were also narrow in scope. Cognitive, social, emotional, and linguistic
developmental competencies as well as academic self-concept and self-efﬁcacy could not
be fully explored using the Add Health data set. These data element limitations also
meant the use of single computed manifest variables in place of full latent constructs for
such components as social position variables, social stratiﬁcation mechanisms, and
segregation.

In a similar fashion, there were limited questions available to assess aspects of
adaptive culture, including acculturation, traditional and cultural legacies, migration
patterns, and current contextual demands, because the Add Health data set was not
designed to examine these constructs in detail. Future research may look to the
neighborhood contextual data ﬁom the Census included in the larger overall Add Health
data set to include factors related to current contextual demands such as rates of
employment and neighborhood safety. The lack of a measure of ethnic identity precluded
the use of this factor in assessing child characteristics. The lack of culturally relevant
variables available in the Add Health data set also meant that aspects of developmental
competencies, in particular bicultural competencies, could not be addressed in this

research.

109

Additional limitations to this study were due to the characteristics of structural
equation modeling as an analysis technique. One example of the limitation of structural
equation modeling was the need to exclude gender as a manifest variable from the social
position construct because it is a dichotomous variable. Future research should consider
the creation of interaction variables that take into account gender when examining the
role of social position variables such as race, ethnicity, and social class on Latino
academic development. Similarly, a manifest variable categorizing youth as either a
voluntary or involuntary minority (Ogbu, 1991a) could not be included as a child
characteristic because it would have been dichotomous in nature. Further exploration into
the concept of voluntary/involuntary minority status (Ogbu, 1991a) and its current
impact, meaning, and operationalization among Latino youth is needed before this
concept can more accurately be incorporated into a model of Latino youth developmental
competencies. Age, as a child characteristic, was also not included in these analyses due
to the use of indicators/manifest variables across time points. Future research could
include age in the analyses by creating a looped model using only Wave I indicators the
ﬁrst time the model is rim and looping the model back ﬁom academic outcomes to
examine Wave II and then Wave IH items.

Predicted Factor Relationships

Garcia Coll and her colleagues’ (1996) model offers a multidimensional and
multilevel perspective of the development of minority children. As part of this study a
critical analysis of the combined effects and relationships between factors known to
contribute to the academic development of Latino youth was conducted. Several

relationships between the factors in the conceptual model were predicted. Although the

110

adapted, augmented, integrative model was not completely supported by the results of
this study, the speciﬁc ﬁndings for each of the expected relationships merit ﬁrrther
examination.

Social position variables. Social position variables were expected to be negatively
related to social stratiﬁcations mechanisms, assessed in this model as perceived
prejudice. Higher scores on the race-ethnicity computed variable and on average parent
education level were predicted to lead to lower levels of perceived prejudice. Results
from the analyses found that neither the race-ethnicity construct nor the parent
educational level construct were signiﬁcantly related to perceived prejudice. Given that
there was only one variable available for use as a perceived prejudice factor, “students at
school are prejudiced,” social stratiﬁcation mechanisms as a construct was not fully
represented as described by Garcia Coll and her colleagues (1996). Similarly, not all
aspects of the variables conceptualized to be part of the social position variables were
assessed by this study and so a full understanding of the effects of social position
variables on social stratiﬁcation mechanisms is not possible in this case. Another
limitation that should be noted is the decision to use youth’s ethnic identiﬁcation and
identiﬁcation of one race from Wave I for analyses. The use of one race for purposes of
identiﬁcation may extremely limit its explanatory power. Additionally, if youth’s
identiﬁcation of race or ethnicity changed over time due to other factors, such as racial
and ethnic identity development or the 2000 Census allowing for identiﬁcation as
multiracial, this was not taken into consideration in these analyses.

There is also conﬂicting research regarding whether parents educational level in

general versus mother’s or father’s level of education are most useful in explaining

111

Latino academic success (Baker, McGee, Mitchell, & Stiff, 2000; Cooper, et al., 2002;
NCES, 1995; Shumov & Miller, 2001; Torres Campos, 2003; Voydanoff & Donnelly,
1999). The use of a construct that included both parents’ educational levels may have
produced a variable that was not as predictive of social stratiﬁcations mechanisms as
separate constructs may have been. Additionally, the possibility of direct effects of parent
educational level on other factors in the model, such as parental involvement in school
and segregation, should be examined directly (Lareau, 1996; Shumov & Miller, 2001).
Future research should also note that ethnicity/ethnic identiﬁcation and perceived
prejudice can have a negative relationship when the prejudice is not seen as directed
towards the self identiﬁed group (McCoy & Major, 2003). Assessing more complete
measures of both ethnicity/etlmic identiﬁcation and perceptions of prejudice will support
continued understanding of how these factors interact to impact youth’s lives.

Perceived prejudice. Youth’s report of perceived prejudice was hypothesized to
be predictive of several factors in the adapted, augmented, integrative model. The ﬁrst of
these were economic segregation, residential segregation, and social segregation. It was
expected that perceived prejudice would be negatively related to all three segregation
factors. Perceived prejudice was only found to be related to one of these three segregation
factors and not in the direction expected. Speciﬁcally, the higher youth scored on
“students at school are prejudice” the higher they scored on a computed measure of
feeling social integrated. This also counters Oliver and Wong’s (2003) recent study that
found a signiﬁcant relationship between negative out-group perceptions and social

isolation.

112

Despite some research that has suggested that there is a relationship between
prejudice and residential segregation (Bobo & Zubrinsky, 1996; Charles, 2000), results of
this study did not ﬁnd a signiﬁcant relationship. A review of the evidence from social
science investigations found that there are multiple causes of racial residential separation
in US. metropolitan areas (Clark, 1986). So, it may be that perceived prejudice as it is
measured in this study is only one of several factors that need to be explored when
predicting residential segregation. Perceptions of prejudice are also not the same as
experiences of oppression and institutional discrimination. Residential segregation would
more likely be a consequence of structural and institutional racism, systemic policies that
inﬂuence and lead to truncated options rather than individual perceptions of prejudice.

Although these ﬁndings are contrary to what would have been expected, they may
be due to the limitations of the data set and the use of structural equation modeling
discussed above. A limitation of examining the relationships between social stratiﬁcation
mechanisms and segregation was the use of single manifest items to represent the
constructs. Further examination of the relationships between perceived and experienced
prejudice/discrimination and segregation is needed before a complete understanding of
how these factors relate to each other and of the pathways that exist to affect youth
development is obtained.

Perceived prejudice as a measure of social stratiﬁcation mechanisms was also
expected to be predictive of promoting/inhibiting environments. In this case, analyses
were conducted that examined perceived prejudice’s relationship to both neighborhood
and school environments. Greater amounts of perceived prejudice were expected to

predict more negative neighborhood and school environments. Previous research has

113

demonstrated a relationship between perceptions of racism and inﬂuences of and
experiences in neighborhood contexts. (Loury, 2005; Quillian & Pager, 2001; Sampson &
Raudenbush, 2005). Previous research has also found a relationship between Latinos
perceived prejudice and perceptions of teachers (Katz, 1999), achievement test score
rates (Osborne, 2001), and dropout rates (Gibson, 1994). The results from this study were
unable to support previous research linking perceptions of racism and prejudice to school
and neighborhood context.

According to the hypothesized model, perceived prejudice was also expected to
be predictive of adaptive culture. Greater reported amounts of perceived prejudice were
expected to be negatively related to adaptive culture. This predicted relationship was
substantiated based on analyses. Researchers have both theorized and examined how
perceptions of prejudice can lead to adaptive behaviors and cultural group histories that
help support youth development (Ogbu, 1981, 1999; Garcia Coll & Vézquez Garcia,
1995; McAdoo, 1981). This ﬁnding offers additional support to the conceptualized
interactions and impacts of perceptions of prejudice on the development of Latino youth.

According to Garcia Coll and colleagues’ (1996) model, social stratiﬁcation
mechanisms, operationalized as perceptions of prejudice for purposes of this study, were
expected to fully mediate the relationship between social position variables and
segregation. As reported above there were no signiﬁcant direct effects for any of the
social position variables used in this study on perceptions of prejudice. Similarly,
perceptions of prejudice had no signiﬁcant direct effect on either the residential or
economic segregation variables. Perceived prejudice was only found to have a direct

effect on social segregation. Structural equation modeling results showed no signiﬁcant

114

indirect effects for either race-ethnicity or average parent educational level on residential
segregation, economic segregation, or social segregation. Social stratiﬁcation
mechanisms were also expected to fully mediate the relationship between social position
variables and promoting/inhibiting environments. Race-ethnicity and average parent
educational level did not have any signiﬁcant direct effects on perceived prejudice.
Perceptions of prejudice also did not have any signiﬁcant direct effects on school or
neighborhood environments. Similarly, no signiﬁcant indirect effects were found
between social position variables and either of the promoting/inhibiting environments.

Although some of the expected relationships between perceived prejudice and
other model constructs were found to be signiﬁcant, it is important to note some
limitations of this construct that should be considered when interpreting these results. The
measure of perceptions of prejudice did not take into account the make up of the school
in terms of racial ethnic background of students. The percentages of Latinos students,
teachers, administrators, and school board members are likely to affect Latino youth’s
perceptions of whether other students are prejudiced Meier & Stewart, 1991). The lack
of variables regarding perceptions of prejudice in both school and neighborhood contexts
similarly may have affected these results. It could also be argued that youth’s perceptions
of students at school being prejudice are more accurately reﬂections of school
context/environment rather than a true measure of perceptions of prejudice. Future
research should consider broader and more inclusive indicators of perceived prejudice
and social stratiﬁcation mechanisms as a whole. Social position variables may also have a
more direct effect on segregation and school and neighborhood that merit ﬁnther

exploration. Moreover, the limited nature of the variables assessed as part of adaptive

115

culture suggests that future research is necessary to completely understand the impact of
social stratiﬁcation mechanisms on the adaptations made by Latino youth and their
families as well.

Segregation. According to Garcia C011 and her colleagues’ (1996) model,
segregation was expected to partially mediate the relationship between social
stratiﬁcation mechanisms and promoting/inhibiting enviromnents. Results did not ﬁnd
any direct effect of perceived prejudice on either school or neighborhood environments.
Signiﬁcant direct effects were found for social segregation and economic segregation but
not residential segregation on school environment. Residential segregation, economic
segregation, and social segregation were all found to have signiﬁcant and direct impact
on neighborhood environments. Perceived prejudice did not have a signiﬁcant indirect
effect on school environment. Signiﬁcant indirect effects were only found for perceived
prejudice on neighborhood environments.

The measurement limitations of both the perceived prejudice construct and the
segregation constructs likely inﬂuenced the lack of signiﬁcant ﬁndings. The signiﬁcant
direct effects found for social and economic segregation support previous work linking
segregation to school contexts (Bankston & Caldas, 1996; Clotfelter, 1999; Moody, 2002;
Orﬁeld, Bachmeier, James, & Eitle, 1997). Similarly, the ﬁnding of direct effects of all
three aspects of segregation on neighborhood environments supports previous research
linking segregation to neighborhood quality and resources (Blau, 1981; Massey, 1990;
Massey & Fong, 1990). But the single manifest variables used for residential segregation,
economic segregation, and social segregation likely do not accurately represent all factors

that contribute to these constructs as expressed in the conceptual model. Another

116

limitation is that the three aspects of segregation were tested separately. Santiago and
Galster (1995) found that economic segregation was predictive of residential segregation
for Puerto Ricans. So it may be that the relationships among the segregation factors are
more complex than were tested in the hypothesized model. Future research should
consider broader measures of segregation in order to assess the nature of the
segregation’s effects on school and neighborhood environments.

Promoting/inhibiting environment-school. The adapted augmented integrative
model hypothesized in this study suggested that school environments would be positively
predictive of family and peer processes. The anticipated relationship between school
environment and family processes was found to be signiﬁcant. This supports the idea that
schools and youth’s interactions in school affect family processes (Cowan, 2001). Future
research should more ﬁnely examine the nature of this interaction. The characteristics
and impact of school on academic success may be due to effects of family sorting into
neighborhoods or family background, so teasing apart the actual causal direction of this
relationship is still necessary (Chase-Lansdale et al., 1997; Ginter et al., 2000; Levine &
Painter, 2000; Sanbonmatsu, et. al., 2006). Additionally, school characteristics, such as
those reported by school administrator’s, may have a more direct effect on some portions
of family processes, such as parental involvement in school/their children’s education and
parental attitudes towards education (Campbell, 1992; Duncan, 1992; Hoover-Dempsey,
et al., 1987; Vaden-Kieman, 1996).

School environment was also found to be a signiﬁcant predictor of peer process,
but in a negative direction. This supports results ﬁ'om other studies that have found

school characteristics to have both positive and negative inﬂuences on peer effects and

117

processes for youth (Crosnoe & Needham, 2004; Osterrnan, 1998). Gibson, Gandara, and
Koyarna (2004) have articulated how schools can both positively and negatively impact
peer relationships. They make the case that concepts such as a sense of school belonging
are measures of relationships between peers and between adults and students (Gibson,
Gandara, & Koyarna, 2004). Some research has found for example that youth who feel
comfortable at school may spend less time with peers, but feel that those peers they do
spend time with care about them (Hamm & Faircloth, 2005; Ostennan, 1998). The causal
relationship between sense of belonging and peer processes may actually function in the
opposite direction, in that peer factors may inﬂuence youth’s school sense of belonging
or bonding (Kester, 1994). School characteristics also may not be enough to increase
access to positively inﬂuencing college-bound schoolmates and peers (Gibson, 2003).
Establishing structures in ways that provide these in-school connections and supports are
also necessary (Stanton-Salazar, Vasquez, & Mehan, 2000). Understanding the
underlying nature of the relationship between school environments and peer inﬂuence on
Latino academic success will require additional research that can examine a wider array
of interactions.

School environment was also expected to positively predict Latino youth’s
adaptive culture. Structural equation modeling results supported school environment’s
positive predictive ability on adaptive culture, supporting previous research that has
found a positive relationship between these factors (Trickett & Birman, 2005). Given
differences in traditions, cultural legacies, economic and political events, and migration
and acculturation patterns among Latinos of different nationalities, the relationship

between school factors and adaptive culture should, in the future, be tested with Latino

118

subgroups. Including measures that describe the presence of Latinos in the school system
and school district, including employment levels and school board positions, would help
to understand whether Meier and Stewart’s (1991) model of second-generation
discrimination also impacts this relationship.

Promoting/inhibiting environments-neighborhood. Garcia C011 and colleagues’
(1996) model suggests that neighborhood environments will affect family processes. This
study did not ﬁnd a signiﬁcant effect for neighborhood contexts on Latino youth family
functioning. This reﬂects the complex interplay between community context and family
inﬂuence on youth development (Brooks-Gunn et al., 1997; Klebanov et a1. 1997;
Caughy & Franzini, 2005; Pebley & Sastry, 2003; Pebley & Vaiana, 2002; Pinderhughes
et a1. 2001; Rodriguez, 1975; Tatum, 1987). Aspects of neighborhood contexts such as
extended families and non-parental adult support which have been found to be related to
general youth and Latino youth academic success (Chavkin & Gonzalez, 2000; Cochran
& Be, 1989; Entwilse, Alexander, & Olson, 1994; Harrison, Wilson, Pine, Chan, &
Buriel, 1990; Kenny & Perez, 1996; Wenz-Gross & Siperstein, 1997) were not included
in this model. Studies have also found that what has been considered neighborhood
effects on academic success may be more correctly categorized as family effects (Chase-
Lansdale et al., 1997; Ginter et al., 2000; Sanbonmatsu, et. al., 2006). Similarly, a recent
longitudinal study of neighborhood effects on child outcomes in Los Angeles found that
effects vary by race (Jackson & Mare, 2005). Further research on which speciﬁc
neighborhood characteristics affect family process and Latino academic success is still

needed.

119

The hypothesized model also suggested neighborhood contexts would positively
predict peer processes. Again, this study did not ﬁnd this expected relationship to be
statistically signiﬁcant. Recent research suggests that the relationship between
neighborhood characteristics and peers is a complex one. For example, a recent study
found negative neighborhood effects were related to an increase in the exposure and time
spent with peers as well as to the characteristics of peer networks (Harding, 2005).
Research considering alternate pathways between neighborhood characteristics and peers
inﬂuence on academic success suggests that how these factors affect Latino youth’s
development needs to be further explored (Gonzales, Cauce, Friedman, & Mason, 1996).

Limitations regarding the measurement of neighborhood environment should be
noted. This study used a composite factor that included indicators from both Wave I and
H, not taking into account family mobility between surveys. Research has shown that
there are varying impacts on youth developmental outcomes for “old” and “new”
neighborhoods that were not examined in this study (Briggs, 1997; Duncan, 1992;
Tienda, 1991; Yonkers Family and Community Project, 1997). Additionally, researchers
examining neighborhood effects have not always agreed upon the best deﬁnition and
measures of neighborhood characteristics. Some take an empirically-based approach in
which numerous measures are submitted to factor analysis, which leads to the creation of
composite scores based on somewhat disparate variables such as housing quality and
density, unemployment, family structure, afﬂuence, racial composition, population
change, and other measures readily available through census data (Duncan & Aber 1997;
Aneshensel & Sucoff 1996; Geis & Ross 1997). In this case, data based On parent and

youth report of neighborhood characteristics was used. The Add Health data set also

120

includes information regarding neighborhood characteristics based on Census data.
Including that data in analyses may yield different results than what was found in this
study. A potential drawback to using Census data is that neighborhoods, as deﬁned by the
Census, are not necessarily the most relevant units from a scientiﬁc perspective.
According to the hypothesized model, the neighborhood construct should have
positively predicted adaptive culture. In this study, neighborhood environment was not
found to signiﬁcantly predict adaptive culture. This counters previous research that has
demonstrated a positive relationship between acculturation and neighborhood
characteristics (Alba & Nee 1997; Rosenbaum, Friedman, Schill, & Buddelrneyer, 1998).
It is important to note that some research examining neighborhood and acculturation has
focused on acculturation’s prediction of neighborhood characteristics (Kwon, Zuiker, &
Bauer, 2004; Rosenbaum, Friedman, Schill, & Buddelrneyer, 1998; Rosenbaum &
Friedman, 2001). So, the causal direction of this relationship is still very much in
question. A recent study on the processes that link neighborhood effects and acculturation
for Latinos found that it is spatial and social integration within neighborhoods that
positively impact acculturation (Bauder, 2001). Spatial and social integration are related
to the concepts of segregation assessed in this study. Future research should include these
components and interactions with neighborhoods effects to further examine the linkages
between neighborhood and acculturation, as well as other aspects of adaptive culture.
Neighborhood resources not included in this study, such as social, human, economic, and
physical capital, also need to be assessed in further research examining the relationship
between neighborhood and adaptive culture (Bankston & Zhou, 1997; Portes & Rumbaut

1996; Portes & Zhou, 1993; Waters, 1994; Zhou, 1997).

121

Adaptive culture. The adapted augmented integrative model tested in this study
projected that adaptive culture would signiﬁcantly and positively impact child
characteristics. Results from this study supported the positive predictive relationship
between adaptive culture and child characteristics. This ﬁnding supports the premise that
it is more than just genes and family characteristics that impact child characteristics
(Bowman, 1994; Garcia Coll et, al., 1996; Harris, 2002). Adaptive culture was also
expected to positively impact family processes. Structural equation modeling results
showed adaptive culture negatively inﬂuenced family processes. This counters the work
of Garcia C011 and her colleagues (1996) and other researchers whom have found
signiﬁcant effects between culture and family processes (Harwood, Schoelmerich,
Ventura Cook, Schulze, & Wilson, 1996). It was similarly anticipated that adaptive
culture would positively impact peer processes. Results of this study supported previous
work suggesting adaptive culture’s signiﬁcance as a predictive factor of peer processes
(Birmarr, Trickett, & Vinokurov, 2002; Garcia Coll, et al., 1996; Wong, 1999). The
hypothesized model additionally conceptualized adaptive culture as having a direct
inﬂuence on developmental competencies. This expected relationship was not supported
by this study’s results.

The mixed ﬁndings regarding adaptive culture’s expected relationships
demonstrates the complicated nature of this construct’s impact on the lives and
development of Latino youth. The limitations previously discussed regarding the
measurement of adaptive culture should be carefully considered when interpreting these
study results. In further research, it is vital to examine the relationship between adaptive

culture and speciﬁc familiar processes rather than family inﬂuence as a whole. Similarly,

122

the separate aspects of adaptive culture need to be further explored to understand which
aspects have the greatest impact on family, peers, and developmental competencies.

Child characteristics. According to Garcia C011 and her colleagues’ (1996) model,
child characteristics were presumed to directly impact family processes. Child
characteristics were indeed found to positively predict family processes, which supports
the notion that children are not just passive recipients but also impact their families’ and
parents’ behavior (Bell,1979; Garcia Coll, et al., 1996; Harris, 1998; Langlois, Ritter,
Casey, & Sawin, 1995; Scarr & McCartney, 1983). The adapted augmented model
suggested child characteristics would also directly impact peer processes. The ﬁndings
from this study did not support this expected relationship. It is important to note that the
majority of research examining the multidimensional relationship between peers and
child characteristics has most often investigated the effects peers have on youth, rather
than how youth impact peers (Bobonis & F inan, 2005; Ladd, Kochenderfer, & Coleman,
1996; Powell-Cope & Eggert 1994; Wang, Hartel, & Walberg, 1999; Rumberger 2001).
Asher and Williams (1993) have discussed the characteristics of children that affect their
acceptance or rejection by peers, but many of the qualities they considered important
were not used in this study. Additionally, only youth’s self report of characteristics and
not peer reports were used in this study. This study did support some research that has
found children who show positive characteristics related to school success spend less
time hanging out with peers in unstructured activities (Posner & Vandell, 1999).

Child characteristics were also expected to be directly predictive of
developmental competencies. This relationship was not supported by the results of this

study. Elliott, Hutton, Hildreth, and Illushin (1999) similarly found that US. students had

123

positive attitudes towards education, an aspect of child characteristics, but had less
commitment to educational endeavors. Another recent study of high school students
using SEM found that level of aspiration mediated that relationship between school
attitudes and achievement (Abu Hilal, 2000). Additionally, the relationship between
developmental competencies, such as time allocation, and child characteristics, such as
attitudes towards school and school adj ustrnent, may be highly interrelated (Heath &
Soep, 1998; Huang etal., 2000; Rodriguez etal., 1999; Roth, Brooks-Gunn, Murray, &
Foster, 1998; Posner & Vandell, 1999). Further research is needed to understand the
relationships between these factors in the lives of Latino youth.

According to the adapted augmented version of the integrative model of minority
developmental competencies (Garcia Coll, et. a1, 1996), child characteristics were
assumed to partially mediate the relationship between promoting/inhibiting environments
and family/peer processes. Indeed, as mentioned above child characteristics were found
to directly affect family processes. But no direct effects were found for child
characteristics on peer processes. Structural equation modeling results found no
signiﬁcant direct effects of neighborhood environment on either family or peer processes.
Altemately, school environment was found to directly impact both family and peer
processes. The only signiﬁcant indirect effect found was between school and peer
processes.

Family processes. Garcia C011 and her colleagues (1996) propose that family
processes directly affect youth’s developmental competencies. This expected relationship
was supported by the structural equation modeling results. This ﬁnding supports vast

research on familial and parental impact on youth academic development (Bisnaire,

124

Fireston, & Rynard, 1990; de Kanter, Ginsburg, & Milne, 1987; Harrison, Wilson, Pine,
Chan, & Buriel, 1990; Henderson & Berla, 1994; Lopez, 1996; Romo & Falbo, 1996).
The family processes latent construct was also expected to mediate the relationship
between child characteristics and developmental competencies. As reported earlier child
characteristics was not directly related to developmental competencies while family
processes were. Results did ﬁnd a signiﬁcant indirect affect for child characteristics on
developmental competencies, supporting the proposed mediation relationship.

The encouraging ﬁndings regarding the inﬂuence and interaction of family
processes and other model constructs provide a foundation to further explore familial
impact on youth development. Several key aspects of familial processes suggested by
Garcia C011 and her colleagues’ (1996) were not assessed in this study. Future research
assessing the utility of this model would do well to incorporate elements of data
regarding structure and roles within the family, as well as parental racial and ethnic
socialization. The mediation relationship suggested by the hypothesized model also needs
to be further explored to understand whether it is a complete or partial mediation. Future
research which examines multiple factors of family inﬂuence will also expand our
knowledge of which family factors are most important for Latino academic
developmental competencies.

Peer processes. According to the hypothesized model, peer processes were
expected to be positively related to youth’s academic developmental competencies.
Although results did ﬁnd that peer processes signiﬁcantly impact developmental
competencies the relationship was a negative one. The results of previous research

suggest that this relationship may be more complicated than expressed in the

125

hypothesized model. The ﬁndings of Posner and Vandell’s (1999) study of low-income
urban youth suggests that time spent in out-of-school activities may also be predictive of
time spent hanging out with peers; so the causal relationship may be from developmental
competencies to peers rather than the other way around. Similarly, other research has
found a correlational relationship between peers and individual behaviors (Rowe, 1994;
Harris, 2000). Kinderrnan’s (1993) study circumvented this problem, controlling for
child’s IQ and parents’ attitudes, and observed the effects of changes in group
membership over the course of the year. When a child switched ﬁ'om a clique of
academic achievers to a clique of non-achievers (or vice versa), the child's attitude toward
schoolwork shifted to match that of the new group. Changes in peer groups and peer
effects over time were not considered in this study. Changes in group membership and
peer inﬂuence over time should be further investigated to understand peer effects on the
developmental competencies of Latino youth.

Developmental Competencies. The hypothesized model held that developmental
competencies should have a direct and positive predictive effect on academic success.
This relationship was supported by the structural equation modeling results. This ﬁnding
provides support for expanding Garcia C011 and her colleagues’ (1996) model to examine
speciﬁc youth outcomes. This ﬁnding also supports research that has found relationships
between academic developmental competencies and academic outcomes (Alva, 1991;
Anderson & Keith, 1997; Buriel, Perez, De Ment, Chavez, & Moran, 1998; Davalos, et
al., 1999; Eccles & Barber, 1999; Halter, et al., 1992; J essor, Van Den Bos, Vanderryn,
Costa, & Turbin, 1995; Keith & Perkins, 1995; Larson, 1994; McMillan & Reed, 1994;

Paulson, Coombs, & Richardson, 1990; Posner & Vandell, 1994; Scales, et al., 2000;

126

Shultz, 1993; Voydanoff & Donnelly, 1999; Waxman, Huang, & Padron, 1997). Future
studies may expand this work by examining “skipping,” cheating, completion of
homework, and self sabotaging behavior (Cowman & Ferrari, 2002; Chrisman, Pieper,
Glance, Holland, & Glickauf-Hughes, 1995; Ferrari, 2005; Ferrari & Thompson, 2006).
Similarly, ﬁrture studies should consider expanded indicators regarding participation in
after-school and out-of-school activities.
Alternative Models

Several alternate models were tested to further explore the relationships between
constructs suggested by the conceptual model and existing literature. While none of the
models were a good ﬁt to the data, the ﬁndings do suggest some interesting relationships
between factors, as well as some areas for ﬁrture research beyond the noted areas and
limitations previously discussed. The poor ﬁt indices for the adapted augmented
integrative model, when tested using the Latino sample as a whole and the Mexican
descent subgroup only, suggests that the model has not fully incorporated all the factors
and pathways through which Latino students reach academic success. Recent data have
demonstrated the high level of variability among Latinos, in particular between Cuban
Americans and other Latino subgroups, on levels of educational attainment, income, and
employment (Boswell, 2002). Future research should further test the model to examine
invariant factor loadings as well as invariant structural relations for Latino subgroups to
better understand the various pathways taken by students from different Latino
backgrounds.

Even after removing those factors for which there was insufﬁcient data to be able

to create latent constructs, the model was not a good ﬁt to the data. This may be

127

suggestive of the importance of these factors in explaining youth development. Indeed,
perceived prejudice and segregation factors were subsequently examined models were
found to have direct relationships to such constructs as adaptive culture, child
characteristics, family, and academic success indicating that these constructs are indeed
important to the academic development of Latino youth. Although longitudinal studies of
the effects of segregation, prejudice, and oppression are complicated and often difﬁcult to
undertake, they will be necessary if we are to gain further clarity on how these pervasive
community factors affect Latino youth development. The overall model as a whole also
provides a wealth of possible relationships to explore in further research and may
ultimately also prove useful as a conceptual framework guiding intervention and policy.
Conclusions

Although the proposed model as a whole wasn’t accepted, several strengths of
this study can be built upon in future research. As previously noted, few studies have
examined the large constellation of factors affecting Latino youth development
considered in this research. It provides much needed knowledge addressing how Latino
youth behave and the underlying processes that explain their academic outcomes. The
exploration of the relationships between proximal and distal predictors of development
and behavioral outcomes contributes to our knowledge of the developmental pathways
taken by Latino youth. This knowledge can support efforts to foster developmentally
appropriate environments that embrace the culturally unique strengths of Latino youth in
ways to enhance their ability to take advantage of the assets they have (Rodriguez &
Morrobel, 2002). This study’s use of a large nationally representative data set allows for

greater generalization of the research results of Latinos across the country. The

128

developmental pathways investigated were able to take into account the various
nationalities, socioeconomic statuses, and regional differences present among Latinos
living in the US.

The relationships which proved to be signiﬁcant predictive links offer a
foundation for continued research geared to further understanding of Latino youth
developmental pathways. In particular, adaptive culture proved to be an important
mediator of the effects of perceptions of prejudice, school environment, and
neighborhood environment on individual youth characteristics, youth’s family and
friends, and youth’s developmental competencies. Further research that broadens our
understanding of the adaptive cultures developed by the Latino community may well
provide the cornerstone to understanding how to intervene in the lives of Latino youth in
order to increase their academic success. This research also supports peers’ role in the
development and academic success of Latino youth and suggests both adaptive culture
and school environment are predictive of the type of inﬂuence peers will have. Similarly,
despite data limitations, perceptions of prejudice and segregation had signiﬁcant direct
and indirect effects on family, developmental competencies, and academic success. These
signiﬁcant effects suggest that the role of social stratiﬁcation mechanisms and
segregation on the development and academic outcomes of Latino youth merit further
investigation. Given that the individuals in this sample were adolescents, it is not entirely
surprising that school environment had such signiﬁcant effects on adaptive culture,
family, friends, developmental competencies, and academic success. But it does suggest

that a greater focus on school contexts and their effects on the academic outcomes of

129

Latino youth are necessary. Examining the environments rather than simply blaming the
students will be necessary in order to understand what truly predicts academic success.

The results of this study demonstrate that the dimensions and concepts included in
Garcia C011 and her colleagues’ (1996) model are critical components for understanding
the developmental competencies of Latino youth. The pathways linking constructs and
the model as a whole may require reconﬁguration or adjustment to accurately capture the
normative development of Latinos. The model in its entirety may be most useful as a
conceptual model that identiﬁes factors in the lives of Latino youth that impact their
development. The model may be built upon and further adapted in order to develop a
framework that may be empirically investigated.

Dialogue and research in youth development has not kept pace with demographic
trends in the United States. Estimates project that between 2010 and 2020, 20% of youth
ages 10 to 20 will be of Latino origin (Bureau of the Census, 2002; NCES, 1998).
Although Latinos are the fastest growing youth group in the United States (Chapa &
Valencia, 1993; Perkins & Villarruel, 2000; Ramos, 2002), research focused on their
development is sparse. The negative academic outcomes of Latino youth on the other
hand have been well documented, including low graduation rates and low educational
attainment (Dryfoos, 1998; Meir & Stewart, Jr., 1991; NCES, 1999; Provitera Mcglynn,
2001; Romo & Falbo, 1996; US. Department of Education, 1992; Weiner, Leighton, &
Funkhouser, 2000). The limited existing research has focused on factors related to the
failure of Latino youth, leading to policy, educational reform, and programming that have
not signiﬁcantly positively altered the trajectory of Latino students (Garcia Coll, Meyer,

& Brillon, 1995; President's Advisory Commission on Educational Excellence for

130

Hispanic Americans, 2000; Villarruel, Dunbar, Montero Seiburth, & Outley, 2005).
Latino students in general continue to be viewed as academically “at-risk,”
underachievers, illiterates, and dropouts (Gutierrez, 2002; Light, 11. d.; Mercado, 2001). If
the current educational trends are not changed the ramiﬁcations will affect not only
Latino youth as individuals but the nation as a whole (Carnegie Council on Adolescent
Development, 1989; President’s Advisory Commission On Educational Excellence for
Hispanic Americans, 2003; President’s Advisory Commission On Educational
Excellence for Hispanic Americans, 2000).

Through the development of an empirically tested model, this study seeks to
inform and begin to ﬁll the gaps in our research regarding the normative development of
Latino youth. Using a success paradigm, examining Latino students who do well in
school and considering how they differ from less successful Latino students, this study
seeks to improve our knowledge regarding the developmental pathways of Latino youth
and to motivate researchers, policymakers, and the community at large to create efforts
that increase their opportunities for success. Researching resilience has important
implications for the educational improvement of Latino youth and provides opportunities
for examining the processes of adaptation that can guide future intervention efforts
(Gordon & Song, 1994; Masten, 1994; McMillan & Reed, 1994; Wang & Gordon, 1994;
Winﬁeld, 1991).

This research also hopes to inspire educators, researchers, and policy makers to
overturn the historical trend of explanatory models and solutions based on a deﬁcits
perspective, which rarely consider the needs and values of Latino students, and negate the

political nature of education thereby continuing to justify placing Latinos at the margins

131

of mainstream school curriculum (Bartolomé & Balderrama, 2001; Diaz & Flores, 2001;
Reyes & Halco'n, 2001; Moll, 2001). This research encourages others to move away from
using a race-comparative paradigm that has impeded the development of a rich,
meaningful, and culturally anchored knowledge base about Latino youth. Rather than
focusing on documenting the ways in which Latino youth do not behave, directing
research onto how they do behave, and the underlying processes or mechanisms that
explain these behavioral outcomes, is the call of this study. Policy making and
educational reform efforts should be informed with relevant youth development evidence
and seek that evidence vigilantly (Rodriguez & Morrobel, 2002).
Educators and policy makers are not released from their responsibility because of
the lack of empirical research on Latino youth development. Ethical standards for
policy and program design, implementation, and evaluation must be maintained,
even in the face of empirical research drought (Rodriguez & Morrobel, 2002, p.
22).
Shifting to a success paradigm in research, education, and policy making will move us
forward in our understanding of what makes Latino youth academically successful. The
integrative model and research presented here provide a template of proximal and distal

components that can inform knowledge seeking and intervention processes.

132

Appendix A

133

Appendix A

List of constructs and corresponding Add Health items

 

Observed Variables for Each Latent Construct

 

Social Position Variables
Race
Ethnicity
Gender
Parent educational attainment
Social Stratiﬁcation Mechanisms
Youth 's Perceived Prejudice
Students at school prejudiced
Segregation
Can afford better housing in that neighborhood as compared to other
neighborhoods
Neighborhood close to where used to work
Neighborhood close to where currently work
Moved to neighborhood because had outgrown previous housing
Less crime in this neighborhood than others
Less drugs and illegal activity by adolescents in this neighborhood
Neighborhood close to friends or relatives
Schools better here than in other neighborhoods
Same age children as your children in neighborhood
Self or spouse/partner born in this neighborhood
Total family income
Ability to pay bills
Receipt of public assistance/social welfare
People were unfriendly
Feel socially accepted
Felt disliked by people
Promoting/Inhibiting Environment
School
Youth gets along with teachers
Youth perceives teachers fair
Youth feels teachers care about them
Youth feels close to people at school
Youth feels part of the school
Youth feels happy to be at school
Youth feels safe at school
School administrators’ reports of problems at school related to smoking or
tobacco use, drug use, alcohol use, gang violence, teenage pregnancy,
vandalism/thieving, eating disorders, stress or pressure
School administrators’ reports school type
(list continues)

134

 

Observed Variables for Each Latent Construct

 

School administrators’ reports average daily attendance
School administrators’ reports full-time classroom teachers
School administrators’ reports average class size
School administrators’ reports race and ethnicity of full time teachers
School administrators’ reports gender of full time teachers
School administrators’ reports presence of parent organization
School administrators’ reports dropout rates
School administrators’ reports retention rates
School administrators’ reports achievement score rates
School administrators’ reports percentage of 12th graders in academic or
college preparatory, vocational or occupational preparatory, general
program/no divisions
School administrators’ reports health related services offered at school
School administrators’ reports school policies and rules
School administrators’ reports urbanicity
School adnrinistrators’ reports size of school
Parents report of the their child’s school being a good one
Parents report of the their child’s school being a school that prioritizes
learning
Parents report of the their child’s school a safe school
Neighborhood
Youth feel safe in the neighborhood
Youth know people in their neighborhood
Youth able to talk to someone in their neighborhood
Youth feels neighbors looking out for one another
Youth feels happy in their neighborhood
Youth would be happy or unhappy if they moved away ﬁom their
neighborhood
Parent report wanting to leave neighborhood
Litter and trash problem
Drug dealers and drug user problem
Adaptive Culture
Acculturation
Parents’ report of whether child was born in the US youth
Parents’ report of whether child was born into neighborhood
Parents’ report of whether child was born a US citizen
Parents’ report of when child moved to the US
Youth’s language use
Youth’s comfort with different languages
Language the interview was conducted in
Language used with family member and friends
Buying music in language other than English
Reading, listening, or watching newspaper, radio or TV in languages other
than English

 

(list continues)

135

 

Observed Variables for Each Latent Construct

 

Current Contextual Demands
Youth’s perceptions that they will live to be 35
Youth’s perceptions that they will be married by 25
Youth’s perceptions that they will be killed by 21
Youth’s perceptions that they will have a middle class family
Child Characteristics
Self-esteem
Feels as intelligent as others
Feels intelligent
Feels as good as others
Feels has good qualities
Feels proud
Like themselves as they are
Conﬁdence
Felt life was a failure
Learning Disability
Parents’ report of learning disability
Parent report of mental retardation/learning disability
Parent report of participation in special education
Youth consider themselves disabled
Youth feel others consider them disabled
Youth’s retrospective report of ADHD symptoms
Perceptions Towards Education
Youth feels will go to college
Youth feels will graduate from college
Family
Parental Involvement In School
Parent involvement in parent teacher organizations
Parent talks with youth regarding school work or grades
Parent talks with youth about school in general
Parent talks with youth’s teachers
Parent volunteers at school
Parent works with child on school project
Parent would be disappointed if child does not graduate high school
Parent would be disappointed if child does not graduate college
Parent report met best friend/best friend’s parents
Parent report met special friend/special friend’s parents
Parent report talk to parents of youth’s friends
Parental activities: shopping, playing a sport, attending religious service,
discussing life, going to the movies, discussing a personal problem, and
arguing about youth’s behavior
Parent caring
Parent closeness
Parent warmth
(list continues)

136

 

Observed Variables for Each Latent Construct

 

Parent love
Parent encourages independence
Youth satisfaction with parental relationship
Youth feel loved and wanted
Parent/family understanding
Youth feel parent pays attention to them
Parental communication
Parent report get along with youth
Parent report understand youth
Parent report feel can really trust youth
Parent report make decision together with youth
Parent report youth interferes with activities
Peers
Time spent hanging out
Frequency of talking to peers regarding problems
Youth feeling that friends care
Go over to friends’ houses
Meet friends after school
Spend time with friends on the weekends
Talk on the phone with ﬁiends
Peers smoking, drinking alcohol, and using marijuana
Parent report of best friend’s inﬂuence
Developmental Competencies
Time Allocation
Participation in clubs/organizations/teams
Time spent in after—school and out-of-school activities
Time spent in athletics
Time spent in physical activity/exercising
Time spent in watching television or playing video
Time spent using the computer for activities other schoolwork
Time spent at religious services or activities
Time spend doing hobbies
Time spent roller-balding, roller-skating, skateboarding, bicycling
Academic Motivation
Youth try hard at school
Youth pay attention in school
Youth want to go to college
Behaviors
Number of times youth have skipped school
Youth get homework done
Out of school suspensions
Youth ever expelled
Grades youth retained
Grades youth skipped

 

(list continues)

137

 

Observed Variables for Each Latent Construct
Youth take a weapon to school
Youth high at school
Youth drunk at school
Academic Success
Grades in English/language arts
Grades in mathematics
Grades in history
Grades in science
GED/equivalency
High school completion
Post-secondary education
Post-secondary graduation
Vocational or job training
Attainment
Highest grade completed

138

References

Abu Hilal, Maher, (2000) A structural model of attitudes toward school subjects,
academic aspiration and achievement. Educational Psychology, 20, 75-84.

Alba, R. & Nee, V. (1997). Rethinking assimilation theory for a new era of
immigration. International Migration Review, 31(4), 826-874.

Alva, S. A. (1991). Academic invulnerability among Mexican-American students: The
importance of protective resources and appraisals. Hispanic Journal of Behavioral
Sciences, 13, 18-34.

Anderson, E. S., & Keith, T. Z. (1997). A longitudinal test of a model of academic
success for at-risk high school students. Journal of Educational Research,
90(5), 259-268.

Aneshensel, C. S. & Sucoff, C. A. (1996). The neighborhood context of adolescent
mental health. Journal of Health and Social Behavior, 37(4), 293-310.

Arbuckle, J. L. & Wotlr, W. (1999). AMOS 4.0 user ’s guide. Chicago: Smallwaters.

Asher, S. R. & Williams, G. A. (1993). Children without friends, Part 2: The reasons for
peer rejection. In Todd, C.M. (Ed.), Day care center connections, 3(1), pp. 3-5.
Urbana-Champaign, E: University of Illinois Cooperative Extension Service.

Aznritia, M., & Cooper, C. (2001). Good or bad? Peer inﬂuences on Latino and European
American adolescents’ pathways through school. Journal of Education for
Students Placed at Risk, 6(1-2), 45-71.

Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.

Bankston, C. & Caldas, S. J. (1996). Majority African American schools and social
injustice: The inﬂuence of de facto segregation on academic achievement. Social
Forces, 75(2), 535-555.

Bankston, C. & Zhou, M. (1997). The social adjustment of Vietnamese American
adolescents: Evidence for a segmented assimilation approach. Social Science
Quarterly, 78, 508-23.

Baker, S. R., McGee, Z. T., Mitchell, W. S., & Stiff H. R. (2000) Structural eﬂ'ects on
academic achievement of adolescents. (Report No. PS-029-115) Hampton
University. (ERIC Documents Reproduction Services No. ED448890).

Barber, B., & Eccles, J. S. (1997, April). Student council, volunteering, basketball, or

marching band: What kind of extracurricular involvement matters? In G. H. Elder
(Chair), Family inﬂuence son Adolescent Involvement in Community Activities.

139

Symposium conducted at the biennial meeting of the Society for Research on
Child Development, Washington DC.

Bartolomé, L. I. & Balderrama, M. (2001). The Need for Educators with Political and
Ideological Clarity: Providing Our Children with "the Best." In M. Reyes & J.
Halcon (Eds) The Best for Our Children: Latina/Latino Views on Literacy. New
York: Teachers College Press (pp. 48-64).

Bauder, H. (2001). You're good with your hands, why don't you become an auto
mechanic: Neighborhood context, institutions and career development.
International Journal of Urban and Regional Research. 25 (3): 593-608.

Bell, R. Q. (1979). Parent, child, and reciprocal inﬂuences. American Psychologist, 34,
821-826.

Benedict, R. (193 8/ 1980). Continuities and discontinuities in cultural conditioning. In R.
E. Muuss (Ed.), Adolescent behavior and society (3rd ed.). New York: Random
House.

Bemal, M. E., Saenz, D. S., & Knight, G. P. (1991). Ethnic identity and adaptation of
Mexican American youths in school settings. Hispanic Journal of Behavioral
Sciences, 13, 135-154.

Berzins, M. E. & Lopez, A. E. (2001). Starting off right: Planting the seeds for biliteracy.
In Maria de la Luz Reyes & John Halcon (Eds.), The Best for Our Children. New
York: Teachers College Press.

Birman, D., Trickett, E. J ., & Vinokurov, A. (2002). Acculturation and adaptation of

Soviet Jewish refugee adolescents: Predictors of adjustment across life domains.
American Journal of Community Psychology, 30(5), 585-607.

Bisnaire L. M., Firestone P., & Rynard, D. (1990). Factors associated with academic
achievement in children following parental separation. American Journal of
Orthopsychiatry, 60(1), 67-76.

Blau, Z. S. (1981). Black children/white children: Competence, socialization and social
stratification in comparative perspective. New York: Free Press.

Bleakley, H. & Chin, A. (2004). What Holds Back the Second Generation? The
Intergenerational Transmission of Language Human Capital among Immigrants.
CCIS Working Papers no.104; Center for Comparative Immigration Studies;
University of California, San Diego.

Bobo, L. & Zubrinsky, C. L. (1996). Attitudes on residential integration: Perceived status

differences, mere in-group preference, or racial prejudice? Social Forces, Vol. 74
(3), pp. 883-909.

140

Bobonis, G. & Finan, F. (2005), Endogenous Social Interaction Eﬂects in School
Participation in Rural Mexico, mimeo, University of California Berkeley.

Bogenschneider, K. (1997). Parental involvement in adolescent schooling: A proximal
process with transcontextual validity. Journal of Marriage and the Family, 5 9,
7 1 8-733.

Boomsa A. (2000). Reporting analyses of covariance structures. Structural Equation
Modeling, 7, 461-483.

Boswell, T. (2002). A demographic profile of Cuban Americans. Miami: The Cuban
American National Council, Inc.

Bourdieu, P. (1997). Capital cultural, escuela y espacio social. Compilacion y
traduccion: Isabel Jimenez. Mexico, D.F.: Siglo XXI.

Bourdieu, P., & Passeron, J. C. (1977). Reproduction in education, society, and culture.
Beverly Hills, CA: Sage.

Bowman, B. T. (1994). Cultural Diversity and Academic Achievement Urban
Monograph Series, Oak Brook, IL: North Central Regional Educational
Laboratory

Briggs, X. S. (1997). Moving up versus moving out: Neighborhood effects in housing
mobility programs. Housing Policy Debate, 8(1), 195-234.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature
and design. Cambridge, MA: Harvard University Press.

Bronfenbrenner, U. (1993). The ecology of cognitive development: Research models and
fugitive ﬁndings. In R. H. Wozniak & K. W. Fisher (Eds), Development in
context: Acting and thinking in specific environments. Hillsdale, NJ: Erlbaum.

Bronfenbrenner, U., & Crouter, A. C. (1983). The evolution of environmental models in
developmental research. In W. Kessen (Ed.), P. H. Mussen (Series Ed.),
Handbook of child psychology: Vol. I. History, theory, and methods, 4th ed. (pp.
357-414). New York: Wiley.

Brooks-Gunn, J ., Duncan, G. J ., and Aber, J. L. (1 997) Neighborhood poverty: Context
and consequences for children, Vol. 1. New York: Russell Sage Foundation.

Browne, M. W. & Cudeck, R. (1993). Alternative ways of assessing model ﬁt. In K. A.
Bollen & J. S. Long (Eds.), Testing structural equation models, pp. 445-455.
Newbury Park, CA: Sage.

141

Buriel, R., Perez, W., De Ment, T. L., Chavez, D., & Moran, V. R. (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.

Byme, B. M. (2001). Structural equation modeling with AMOS: Basic concepts,
applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates.

Campbell, L. (1992, April). Parents and schools working for student success. NASSP
Bulletin, 76(543), 1-4.

Card, D., DiNardo, J. E. & Estes, E. (2000). The More Things Change: Immigrants and
the Children of Immigrants in the 1940s, the 19703, and the 1990s. In G. Borjas

(Ed.), Issues in the Economics of Immigration. University of Chicago Press for
NBER: Chicago.

Carnegie Council on Adolescent Development. (1989). Turning points: Preparing
American youth for the 21'" century. New York: Carnegie Corporation of New
York.

Carrasquillo, A. L. (1991). Hispanic Children and Youth in the United States: A
Resource Guide. Reference Books on Family Issues (V 01. 20). Garland Reference
Library of Social Science (Vol. 608). New York: Garland Publishing, Inc.

Carter, T. P. (1970). The way beyond bilingual education. Civil Rights Digest, 3(4), 14-
21.

Carter, T. P. & Segura, R. D. (1979). Mexican Americans in School: A Decade of
Change. New York: College Entrance Examination Board.

Cauce, A. M., Felner, R. D., & Primavera, J. (1982). Social support in high-risk
adolescents: Structural components and adaptive impact. American Journal of.
Community Psychology, 10(4), 417-428.

Caughy, M. O. & Franzini, L. (2005). Neighborhood correlates of cultural differences in
perceived effectiveness of parental disciplinary actions. Parenting, 5(2), 119-151.

Chapa, J ., & Valencia, R. R. (1993). Latino population growth, demographic
characteristics, and educational stagnation: An examination of recent trends.
Hispanic Journal of Behavioral Sciences, 15(2), 165-187.

Charles, C., (2000). Neighborhood racial-composition preferences: Evidence from a
multiethnic metropolis, Social Problems, 47(3), 379—407.

Chase-Lansdale, P.L., Gordon, R. A., Brooks-Gunn, J., 8: Connell, J. P. (1997)
Neighborhood and Family Inﬂuences on the Intellectual and Behavioral

142

Competence of Preschool and Early School-Age Children. In Brooks-Gunn, J .,
G.J. Duncan, & J .L. Aber, eds., Neighborhood Poverty Vol. I .' Context and
Consequences for Children. New York: Russell Sage Foundation.

Chavkin, N. F ., & Gonzalez, J. (2000). Mexican immigrant youth and resiliency:
Research and promising programs. Charleston, WV: ERIC Clearinghouse on
Rural Education and Small Schools, AEL, Inc. (ERIC Document Reproduction
Service No. ED447990).

Chen, C., & Stevenson, H. W. (1995) Motivation and mathematics achievement: A
comparative study of Asian-Americans, Caucasian-American, and East Asian
high school students. Child Development, 66, 493-506.

Chiu, M. (1996). Building mathematical understanding during collaboration.
Unpublished doctoral dissertation, university of California, Berkeley.

Chrisman, S. M., Pieper, W., Glance, P. R., Holland, C. L., & Glickauf-Hughes, C.
(1995). Validation of the Glance Irnpostor Phenomenon scale. Journal of
Personality Assessment, 65, 456-467.

Christenson, S. L., Rounds, T., & Gomey, D. (1992). Family factors and student
achievement: An avenue to increase students’ success. School Psychology
Quarterly, 7(3), 178-206.

Clark, W.A.V. (1986). Residential segregation in American cities: A review and
interpretation. Population Research and Policy Review (Historical Archive), 5(2),
95.

Clotfelter, C. T. (1999). Public school segregation in metropolitan areas. Land
Economics, 75(4), 487-504.

Cochran, M., & Be, I. (1989). The social networks, family involvement, and pro-and
anti-social behavior of adolescent males in Norway. Journal of Youth and
Adolescence, 18, 377-398.

Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience in
African-American youth: Context, self, action and outcomes in school. Child
Development, 65, 493-506.

Coontz, S. (1992). The way we never were. New York: Basic
Cooper, C. R., Cooper, Jr., R. G., Azimitia, M., Chavira, G., & Gullat, Y. (2002).
Bridging multiple worlds: How African-American and Latino youth in academic

outreach programs navigate math pathways to college. Applied Developmental
Science, 6(2), 73-87.

143

Cooper, H. (1989). Homework New York: Longrna

Cooper, H., Lindsay, J. J ., Nye, B., & Greathouse, S. (1998). Relationships between
attitudes about homework, the amount of homework assigned and completed, and
student achievement. Journal of Educational Psychology. 90, 70-83.

Cowan, RA. (2001). How Family and Peer Relationships Promote Social Competence.
Paper presented at The Pre-Teen And Teenage Years: New Findings, New
Challenges, UC Berkeley Clark Kerr Conference Center, Berkeley, CA.

Cowman, S. & Ferrari, J. R. (2002). "Am I for real?" Predicting irnpostor tendencies from
self handicapping and affective components. Social Behavior and Personality,
30(2), 119-126.

Crosnoe, R. & Needham, B. (2004). Holism, contextual variability, and the study of
ﬁiendships in adolescent development. Child Development, 75(1), 264-279.

Darder, A. (1991). Culture and Power in the Classroom: A Critical Foundation for
Bicultural Education. Critical Studies in Education and Culture. Westport, CT:
Bergin & Garve

Davalos, D. B., Chavez, E. L., & Guardiola, R. J. (1999). The effects of extracurricular
activity, ethnic identiﬁcation, and perception of school on student dropout rates.
Hispanic Journal of Behavioral Sciences, 21(1), 61-77.

de Kanter, A., Ginsburg, A., & Milne, Am. (1987). Parent involvement strategies: A new
emphasis on traditional parent roles. Washington, DC: US. Department of
Education.

Diaz, E. & Flores, B. (2001). Teaching to the Potential: Teacher as Sociocultural and
Sociohistorical Mediator. In Maria de la Luz Reyes 8; John Halcon (Eds), The
Best for Our Children. New York: Teachers College Press.

Donato, R. (1997). The other struggle for equal schools: Mexican Americans during the
Civil Rights era. Albany, NY : State University of New York Press.

Dryfoos, J. G. (1998). Safe passage: Making it through adolescence in a risky society.
What parents, schools, and communities can do. New York: Oxford University
Press.

Duncan, C. (1992, April). Parental support in schools and the changing family structure.
NASSP Bulletin, 76(543), 10-14.

Duncan, G. J. & Aber, J .L. (1997). Neighborhood Poverty Vol. I: Context and
Consequences for Children. New York: Russell Sage Foundation.

144

 

Duncan, J. D., Brooks-Gunn, J ., & Klebanov, P. D. (1994). Economic deprivationa nd
early childhood development. Child Development, 63, 296-318.

Eccles, J. S., & Barber, B. L. (1999). Student council, volunteering, basketball, or
marching band: What kind of extracurricular involvement matters? Journal of
Adolescent Research, 14(1), 10-43.

Eccles, J. 8., Early, D., Frasier, K., Belansky, E., & McCarthy, K. (1997). The relation of
connection, regulation, and support for autonomy in the context of family, school,

and peer group to successful adolescent development. Journal of Adolescent
Research, 12, 263-286.

Eckstrom, R. B., Goertz, M. E., Pollack, J. M., & Rock, D. A. (1986). Who drops out of
high school and why? Findings ﬁom a national study. Teachers College Record,
87(3), 356-373.

Educational Testing Service. (1996). Report in Brief NAEP 1994 Trends in Academic
Progress. Washington, DC: Author.

Elder, G. H., Jr. (1996). Human lives in changing societies: Life course and
developmental insights. In R. B. Cairns, G. H. Elder, Jr., & E. J. Costello,
Developmental science. Cambridge, England: Cambridge University Press.

Elder, G. H., Jr., Modell, J ., & Parke, R. D. (1993). Studying children in a changing
world. In G. H. Elder, Jr, J. Modell, & R. D. Parke. Children in time and place.
Cambridge, England: Cambridge University Press.

Elizondo, FR. (2005). A Latino perspective: Better teachers and administrators. School
Administrator, 62, 22—23.

Elliott, J., Hutton, N., Hildreth, A., & Illushin, L. (1999). Factors inﬂuencing educational
motivation: a study of attitudes, expectations and behaviour of children in

Sunderland, Kentucky and St Petersburg. British Educational Research Journal,
25(1), 75-94.

Entwilse, D. R., Alexander, K. L., & Olson, L. S. (1994). The gender gap in math: Its
possible origins in neighborhood effects. American Sociological Review, 59(6),
822-838.

Erikson, E. H. (1950). Childhood and Society. New York: Norton (2“d ed. 1963).

Erkut, S., Szalacha, L., & Coll, C. (2005). A framework for studying minority youths'

transitions to fatherhood: The case of Puerto Rican adolescents. Adolescence,
40(160), 709-727.

145

Federal Interagency Forum on Child and Family Statistics. (1998). America's children:
Key national indicators of well-being. Washington, DCzAuthor.

Feldman, S. S., 8c Wentzel, K. R. (1990). Relations among family interaction patterns,
classroom self-restraint, and academic achievement in preadolescent boys.
Journal of Educational Psychology, 82, 813-819.

Felner, R. D., Jackson, A. J ., Kasak, D., Mullhall, P., Brand, 8., & Flowers, N. (1997).
The impact of school reform for the middle years. Phi Delta Kappan, 78(7), 528-
532,541-550.

Fennelly, K., Mulkeen, P., & Giusti, C. (1998). C0ping with racism and discrimination:
The experience of young Latino adolescents. In H. McCubbin&E. Thompson
(Eds), Resiliency in Native American and immigrant families (Vol. 2, pp. 367-
383). Thousand Oaks: Sage.

Ferrari, J. R. (2005). Irnpostor tendencies and academic dishonesty: Do they cheat their
way to success? Social Behavior and Personality, 33(1), 11-18.

Ferrari, J. R. & Thompson, T. (2006). Irnpostor fears: Links with self-presentational
concerns and self-handicapping behaviours. Personality and Individual
Diﬂerences, 40(2), 341-352.

Finch, J. F ., West, S. G., & MacKinnon, D. P. (1997). Effects of sample size and
nonnormality on the estimation of mediated effects in latent variable models.
Structural Equation Modeling: A Multidisciplinary Journal, 4, 87-107.

Fisher, C. B., Jackson, J. J ., & Villarruel, F. A. (1997). The study of ethnic minority
children and youth in the United States. In R. M. Lerner (Ed), Theoretical models
of human development (pp. 1145-1207). New York: Wiley Press.

Fisher, C. B., Wallace, S. A., & Fenton, R. E. (2000). Discrimination distress during
adolescence. Journal of Youth and Adolescence, 29(6), pp. 679 — 695.

Flores, B., Cousin, R., & Diaz, E. (1991). Transforming deﬁcit myths about learning,
language, and culture. Language Arts, 68, 369-379.

Foley, D. E. (1997). Deﬁcit thinking models based on culture: The anthropological
protest. In R. R. Valencia (Ed.), The evolution of deﬁcit thinking: Educational
thought and practice (pp.1 13-131). The Stanford Series on Education and Public
Policy. London: F ahner

F reire, P. (1972). Pedagogy of the Oppressed, Harmondsworth: Penguin.

Freire, P., & Macedo, D. (1987). Literacy: Reading the word and the world. South
Hadley, MA: Bergin & Garvey.

146

Fuligrri, A. J. (1997). The academic achievement of adolescents form immigrant families:
The role of family background, attitudes, and behavior. Child Development,
68(2), 351-363

Fuligni, A. J. & Stevenson, H. W. (1995). Time use and mathematics achievement among
American, Chinese, and Japanese high school students. Child Development, 66(3),
830-842.

Furstenberg, Jr., F. F., & Hughes, M. E. (1995). Social capital and successful
development among at-risk youth. Journal of Marriage and the Family, 37(3),
580-591.

Gandara, P. (1996). The challenge of Latino education: Implications for social and
educational policy. In A. Yanez-Chavez (Ed.), Latino politics in California (pp.
51-57). San Diego: Center for US. Mexican Studies, University of California

Garcia Coll, C. T., Lamberty, G., Jenkins, R., McAdoo, H. P., Cmic, K., Wasik, B. H., &
Garcia, H. V. (1996). An integrative model for the study of developmental
competencies in minority children. Child Development, 67, 1891-1914.

Garcia Coll, C. T., Meyer, E. C., & Brillon, L. (1995). Ethnic and minority parenting. In
MH Bomstein (Ed.), Handbook of Parenting. Vol. 2: Biology and Ecology of
Parenting, pp. 189-209. Hillsdale, NJ: Erlbaum.

Garcia Coll, C. T., & Vazquez Garcia, H. A. (1995). Hispanic children and their families:
On a different track from the very beginning. In H. E. Fitzgerald, B. M. Lester, &
B Zuckerrnan (Eds), Children of poverty: Research, health care, and policy
issues @p. 5 7-83). New York: Garland.

Gibson, M. (1994). Additive acculturation as a strategy for school improvement. In R.
Rumbaut & W. Cornelius, Eds. Immigrant children in California schools. San
Diego, CA: Center for US-Mexican Studies, UCSD.

Gibson, M. (2003). Improving Graduation Outcomes for Migrant Students. ERIC Digest,
EDO-RC-03-2. ERIC Clearinghouse on Rural Education and Small Schools.

Gibson, M. A., Gandara, P. & Koyarna, J. P. (2004). How Schools Can Shape Peer
Relations to Promote Achievement Among Mexican-Origin Youth. UC/ACCORD
Policy Briefs, pb-001-0504. University of California All Campus Consortium on
Research for Diversity.

Ginter, D, Haveman, R., & Wolfe, B. (2000). Neighborhood attributes as determinants of

children’s outcomes: How robust are the relationships, Journal of Human
Resources, 35(4), 603-642.

147

Glasgow, K. L., Dombusch, S. M., Troyer, L. Steinberg, L., 8; Ritter, P. L. (1997).
Parenting styles, adolescents' attributions, and educational outcomes in nine
heterogeneous high schools. Child Development, 63(3), 507-529.

Goldschnridt, P., & J. Wang (1999). When can schools affect dropout behavior? A
longitudinal multilevel analysis. American Educational Research Journal, 36,
71 5-73 8.

Gonzales, N. A, Cauce, A. M., Friedman, R. J ., & Mason, C. A. (1996). Family, peer, and
neighborhood inﬂuences on academic achievement among African-American
adolescents: one-year prospective effects. American Journal of Community

Psychology, 24(3), 365-387.

Goodenow, C. & Grady, K. E. (1993). The relationship of school belonging and friends'
values to academic motivation among urban adolescent students. Journal of
Experimental Education, 62(1), 60-71.

Gordon, B. W., & Song, L. D. (1994). Variations in the experience of resilience. In M. C.
Wang & E. W. Gordon (Eds.), Educational resilience in inner-city America:
Challenges and prospects (pp.27-43). Hillsdale, NJ: Lawrence Earlbaum.

Greenberg, B. S., Burgoon, M., Burgoon, J. K., & Korzenny, F. (1983). Mexican
Americans and the mass media. Norwood, NJ: Ablex.

Greico, E. M., & Cassidy, R. C. (2001). Overview of Race and Hispanic Origin (Census
2000 Brief No. C2KBR/01-1). Washington, DC: US. Government Printing
Ofﬁce.

Grogger, J. & Trejo, S. J. (2002). Falling Behind or Moving Up? The Intergenerational
Progress of Mexican Americans. San Francisco: Public Policy Institute of
California.

Gutierrez, K. D., Baquedano-Lopez, P., & Alvarez, H. H. (2001). Literacy as hybridity:
Moving beyond bilingualism in urban classrooms. In M. L. Reyes & J.J. Halcon
(Eds) The Best for Our Children: Critical Perspectives on Literacy for Latino
Students. Teachers College, Columbia University.

Gutierrez, P. S. (2002). In search of bedrock: Organizing for success with diverse needs
children in the classroom. Journal of Latinos and Education, 1(2), pp. 49-64.

Guzman, G. (2001) The Hispanic Population (Census 2000 Brief No. C2KBR/01-3).
Washington, DC: US. Government Printing Ofﬁce.

Halcon, J .J . 2001. Mainstream ideology and literacy instruction for Sparrish-speaking
children. In The Best for Our Children: Critical Perspectives on Literacy for
Latino Students, (pp. 65-77). New York: Teachers College Press.

148

Hamm, J. V. & Fairclotlr, B. S. (2005). Peer context of mathematics classroom belonging
in early adolescence. Journal of Early Adolescence, 25(3), 345-366.

Hanks, M., & Eckland, B. K. (1976). Athletics and social participation in the educational
attainment process. Sociology of Education, 49(4), 271-294.

Hanson, S. L., & Ginsburg, A. L. (1988). Gaining ground: Values and high school
success. American Educational Research Journal, 25(3), 334-365.

Harding, D. J. (2005). Neighborhood violence and the age structure of peer networks:
socialization of adolescent boys in disadvantaged neighborhoods. PSC Research
Report 05-586, Population Studies Center, University of Michigan.

Harris, J. R. (1998). The nurture assumption: Why children turn out the way they do.
New York: Free Press.

Harris, J. R. (2002). Socialization, Personality Development, and the Child's
Environments: Comment on Vandell (2000). Developmental Psychology, 36(6),
71 1-7 23.

Harrison, A. O., Wilson, M. N., Pine, C. J ., Chan, S. Q., & Buriel, R. (1990). Family
ecologies of ethnic minority children. Child Development, 61, 347-362.

Harter, S., Runbaugh, Whitesell, N., & Kowalski, P. (1992). Individual difference in the
effects of educational transitions on young adolescents' perceptions of
competence and motivation orientation.

Harwood, R. L., Schoelmerich, A., Ventura-Cook,E., Schulze, P. A., & Wilson, S. P.
(1996). Culture and class inﬂuences on Anglo and Puerto Rican mothers' beliefs

regarding long-term socialization goals and child behavior. Child Development,
67(5), 2446-2461.

Hawkins, J. D., Catalano, R. F ., Kosterman, R., Abott, R., & Hill, K. G. (1999).
Preventing adolescent health-risk behaviors by strengthening protection during
childhood. Archives of Pediatric and Adolescent Medicine, 8, 187-220

Heath, S. B. & Soep, E. (1998).Youth development and the arts in non-school hours,
Grantmakers in the Arts Newsletter, 9, 9-17).

Henderson, A. (1988) Good news: All ecologically balanced approach to academic
improvement. Educational Horizons, 66, 60-62.

Henderson, A. T., & Berla, N. (Eds). (1994). A new generation of evidence: The family is

critical to student achievement. Washington DC: National Committee for
Citizens in Education. (ERIC Document Reproduction Service No. ED375968).

149

Hernandez, J. C. (2000). Understanding the retention of Latino college students. Journal
of College Student Development, 41, 575—588.

Hoover-Dempsey, K. V., Bassler, O. C., & Brissie, J .S. (1987). Parent involvement:
contributions of teacher efﬁcacy, school socioeconomic status, and other school
characteristics. American Educational Research Journal, 24(3), 417-435.

Howard, A., & Scott, R. A. (1981). The study of minority groups in complex societies. In
R. H. Munroe, R. L. Munroe, & B. Whiting (Eds), Handbook of cross-cultural
human development (pp. 113-152). New York: Garland.

Hoyle, R. H. & Panter, A. T. (1995). Writing about structural equation models. In R. H.
Hoyle (Ed.), Structural Equation Modeling: Concepts, issues, and applications,
pp. 158-176. Thousand Oaks, CA: Sage.

Huang, D., Gribbons, 3., Kim, K. S., Lee, C., & Baker, E. L. (2000). A Decade of
Results: The Impact of the LA ’s Best After School Enrichment Program on
Subsequent Student Achievement and Performance. Los Angeles, CA: UCLA
Center for the Study of Evaluation.

Hurtado, A., Figueroa, R., & Garcia, E. E. (eds.). (1996). Strategic interventions in
education: Expanding the Latina/Latino pipeline. Santa Cruz, CA: Regents of the
University of California.

Huston, A. C., McLoyd, V. C., & Garcia Coll, C. (Guest Eds). (1994). Children and
poverty: Issues in contemporary research. Child Development, 65(2).

Jackson, M. I. & Mare, R. D. (2005). Cross-Sectional and longitudinal measurements of
neighborhood experience and their effects on children. California Center for
Population Research On-Line Working Paper Series, CCPR-Ol 1-05, California
Center for Population Research, University of California, Los Angeles.

J essor, R., Van Den Bos, J ., Vanderryn, J ., Costa, F. M., & Turbin, M. (1995). Protective
factors in adolescent problem behavior: Moderator effects and developmental
change. Developmental Psychology, 31, 923-933.

Johnson, D. J ., Jaeger E., Randolph, S.M., Cauce, A.M., & Ward, J. (2003). Studying the
effects of early child care experiences on the development of children of color in
the United States: Toward a more inclusive research agenda. Child Development,
74(5), 1227-1244.

Jones, T. G., & Velez, W. (1997, March). Eﬂects of Latino Parent Involvement on
Academic Achievement. Paper presented at the Annual Meeting of the American
Educational research Association, Chicago. (ERIC Document Reproduction
Service No. ED410676).

150

J tireskog, K. G. & Sdrbom, D. (1989). Lisrel 7: A guide to the program and applications.
Chicago: SPSS.

Kaplan, E. B. (1999). "It's going good" inner-city Black and Latino adolescents'
perceptions about achieving an education. Urban Education, 34(2), 181-213.

Katz, S. R. (1999). Teaching in tensions: Latino immigrant youth, their teachers, and the
structures of schooling. Teachers College Record, 100(4), 809-840.

Keith, J. G., & Perkins, D. F. (1995). 13,000 adolescents speak: A profile of Michigan
youth. East Lansing, MI: Community Coalitions In Action, Institute for Children,
Youth, and Families, Michigan State University.

Kenny, M. E., & Perez, V. (1996). Applications of attachment theory to adjustment and
identity among ethnically diverse ﬁrst-year college students. Journal of College
Students Development, 37, 527-535.

Kester, V. M. (1994). Factors that affect African-American students' bonding to middle
school. Elementary School Journal, 95(1), 63-73.

Klebanov, P.K., Brooks-Gunn, J ., Chase-Lansdale, P.L., & Gordon, R. A. (1997). Are
neighborhood effects on young children mediated by features of the home
environment? In Brooks-Gunn, J., G.J. Duncan, & J .L. Aber, eds, Neighborhood
Poverty Vol. 1: Context and Consequences for Children. New York: Russell Sage
Foundation.

Korzenny, F. & Schiff, E. (1992). Media effects across cultures: Challenges and
opportunities, in F. Korzenny, S. Ting-Toomey& E. Schiff (Eds), Mass media
eﬂects across cultures, (pp.1-8). London: Sage.

Krieger, N., Rowley, D. L., Herman, A. A., Avery, B., & Phillips, M. T. (1993). Racial
differences in preterm delivery—Developing a new research paradigm. American
Journal of Preventative Medicine, 9(Suppl. 6), 82-122.

Kwon, H. K., Zuiker, V. S., 8:. Bauer, J. B.(2004). Factors associated with the poverty
status of asian immigrant householders by citizenship status. Journal of Family
and Economic Issues, 25(1), 101-120.

Ladd, G. W., Kochenderfer, B. J ., & Coleman, C. C. (1996). Friendship quality as a
predictor of young children’s early school adjustment. Child Development, 6 7,
1003-1118.

LaFromboise, T., Coleman, H. L. K., & Gerton, J. (1993). Psychological impact of
biculturalism: Evidence and theory. Psychological Bulletin, 114(3), 395-412.

151

Langlois, J. H., Ritter, J. M., Casey, R. J .. & Sawin, D. B. (1995). Infant attractiveness
predicts maternal behaviors and attitudes. Developmental Psychology, 31 , 464-
472.

Lareau, A. (1996) Assessing parent involvement in schooling: A critical analysis. In A.
Booth & J. Dunn (Eds), F amily-school links: How do they aﬂect educational
outcomes (pp. 57-64). Mahwah, NJ: Lawrence Erlbaum.

Larson, R. (1994). Youth organizations, hobbies, and sports as developmental contexts.
In R. K. Silberiesen & E. Todt (Eds), Adolescence in context. (pp. 46-65). New
York: Teachers College Press.

Leone, C. M., & Richards, M. H. (1989). Classwork and homework in early adolescence:
The ecology of achievement. Journal of Youth and Adolescence, 18(6), 531-548.

Lerner, R M. (1983). A "goodness of ﬁt" model of person-context interaction. In D.
Magnusson & V. L. Allen (Eds), Human Development: An Interactional
Perspective. New York: Academic Press.

Lerner, R. M. (1986). Concepts and theories of human development (2‘“(1 ed.). New York:
Random House.

Levine, D. I. & Painter, G. (November 20, 2000). Are measured school effects just
sorting? Identifying causality in the National Education Longitudinal Survey.
Institute of Industrial Relations Working Paper Series. Paper iirwps-078-00.

Lewin, K. (1935). A dynamic theory of personality. New York: McGraw-Hill.

Lewin, K. (1942). Field theory and learning. In F orty—first yearbook, National Society for
the Study of Education (Part II). Bloomington, IL: Public School Publishing.

Light, R. (n. d.). A note on myths about language, learning, and minority children.
Retrieved June 14, 2003 from http://www.albany.edu/lap/Papers/mythshtml

Lima, E. S. and Lima M. G. (1998). Identity, cultural diversity, and education. In Y. Zou
and E. T. Trueba (Eds). Ethnic Identity and Power: Cultural Contexts of Political
Action in School and Society. (pp. 321-343). Albany, NY: State University of
New York.

Lopez, A. S. (Ed.). (1995). Latino language and education: Communication and the
dream deferred. New York: Garland.

Lopez, E. J ., Ehly, S., & Garcia-Vazquez, E. (2002) Acculturation, social support and

academic achievement of Mexican and Mexican American high school students:
An exploratory study. Psychology in the School, 39(3), 245-257.

152

Lopez, E. S. (1996). Social capital and the educational performance of Latino and Non-
Latino youth (Research Rep. No. 11). East Lansing, MI: Julian Samora Research
Institute, Michigan State University.

Loury, G. C. (2005). Racial stigma and its consequences, Focus, 24 (1), 1-6.

Lucas, T., Henze, R., & Donato, R. (1990). Promoting the success of Latino language-
minority students: An exploratory study of six high schools. Harvard Educational
Review, 15, 230-257.

MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and
determination of sample size for covariance structure modeling. Psychological
Methods, 1, 130-149.

Marin G., & Marin, B. (1991). Research with Hispanic populations. Newbury Park, CA:
Sage Publications.

Marsh, H. W., & Byme, B. M. (1999). Causal ordering of academic self-concept and
achievement: Reanalysis of a pioneering study and revised recommendations.
Educational Psychologist, 34, 148-155.

Marsh, H. W., & Yeung, A. S. (1997). Coursework selection: Relations to academic self-
concept and achievement. American Educational Research Journal, 34, 691-720.

Masselam, V. S., Marcus, R. F., & Stunkard, C. L. (1990). Parent-adolescent
communication, family functioning, and school performance. Adolescence,
25(99), 725-37.

Massey, D. S. (1990). American apartheid: Segregation and the making of the underclass.
American Journal of Sociology, 96(2), 329-357.

Massey, D. S., & Fong, E. (1990). Segregation and neighborhood quality: Blacks,
Hispanics, and Asians in the San Francisco Metropolitan Area. Social Forces,
69(1),15-32.

Masten, A. S. (1994). Resilience in individual development: Successful adaptation
despite risk and adversity. In M. C. Wang & E. W. Gordon (Eds), Educational

resilience in inner-city America: Challenges and prospects (pp.27-43). Hillsdale,
NJ: Lawrence Earlbaum.

McAdoo, HP. (1981). Upward mobility and parenting in middle-income black families.
Journal of Black Psychology, 8, 122.

McAdoo, H. P. (1982). Stress absorbing systems in black families. Family Relations, 31,
379-188.

153

McCoy, S. K. & Major, B. (2003). Group identiﬁcation moderates emotional responses to
perceived prejudice. Personality and Social Psychology Bulletin, 29(8), 1005-
1017.

McLoyd, V. (1990). Minority children: Introduction to the Special Issue. Child
Development, 61, 263-266.

McLoyd, V. C. (1998a). Socioeconomic disadvantage and child development. American
Psychologist, 53, (2), 185-204.

McLoyd, V. C. (1998b). Changing demographics in the American population:
Implications for research on minority children and adolescents. In V. C. McLoyd
& L. Steinberg (Eds), Studying minority adolescents: Conceptual,
methodological, and theoretical issues (pp. 3-28). Mahwah, NJ: Lawrence
Erlbaum Associates.

McLoyd, V. C., & Steinberg, L. (Eds), (1998). Studying minority adolescents:
Conceptual, methodological, and theoretical issues. Ann Arbor, MI: University of
Michigan.

McMillan, J. H., & Reed, D. F. (1994) At-risk students and resiliency: Factors
contributing to academic success. The Clearing House, 6 7, 137-140.

Meier, K. J ., & Stewart, Joseph, Jr. (1991). The Politics of Hispanic Education. Un Paso
pa 'lante y dos pa ’tras. Albany, NYzState University of New York.

Mercado, C. I. (2001). Reﬂections on the power of Spanish: Confessions from the ﬁeld.
In M. L. Reyes & J .J . Halcon (Eds) The Best for Our Children: Critical
Perspectives on Literacy for Latino Students, New York: Teachers College Press.

Moll, L. C. 2001. The diversity of schooling: A cultural-historical approach. In M. L.
Reyes & J .J . Halcén (Eds) The Best for Our Children: Critical Perspectives on
Literacy for Latino Students, (pp. 13—28). New York: Teachers College Press.

Montero-Sieburth, M., & Batt, M. C. (2001). An overview of the educational models
used to explain the academic achievement of Latino students: Implications for
research and policies into the new millennium. In R. E. Slavin & M. Calderon
(Eds), Eﬂ'ective programs for Latino students. (pp. 331-368). Mahwah, NJ:
Lawrence Erlbaum.

Monzo, L. D. & Rueda, R. S. (2001). Sociocultural Factors in Social Relationships:
Examining Latino T eachers' and Paraeducators' Interactions with Latino
Students, Research Report 9, Center for Research on Education, Diversity &
Excellence, Santa Cruz, CA & Washington, DC.

154

Moody, J. (2002). Race, school integration, and ﬁiendship segregation in America.
American Journal of Sociology, 107, 679-716.

Mounts, N. S. & Steinberg, L. (1995). An ecological analysis of peer inﬂuence on
adolescent grade point average and drug use. Developmental Psychology, 31(6),
91 5-922.

Muller, C. (1993). Parent involvement and academic achievement. In B. Schneider & J.
S. Coleman (Eds), Parents, their children, and school. (pp. 77-113). Boulder,
CO: Westview.

Muller, C. (1995). Maternal employment, parent involvement , and mathematics
achievement among adolescents. Journal of Marriage and the Family, 5 7, 85-100.

Muskal, F. & Chairez, M. (1990). The Mobility Strategies of Successful Hispanic High
School Students. (ERIC Document Reproduction Service No. ED ED369862).

Muuss, R. E. (1962). F irst—aid classroom discipline problems. New York: Holt, Rinehart
& Winston.

Muuss, R. E. (1996). Theories ofAdolescence (6" ed.) New York: McGraw-Hill.

National Center for Educational Statistics (NCES). (n. d.). Latinos in elementary
education (K-8). Retrieved February 10, 2001, from
http://www.ed.gov/ofﬁces/OHA/Hispaniclrr/K-8.htrnl

National Center for Educational Statistics (NCES). (2001). Paving the way to
postsecondary education: K -1 2 intervention programs for underrepresented
youth. Washington, DC: US. Department of Education.

National Center for Educational Statistics (N CES). (1999). Digest of education statistics
1998. Washington, DC: US. Department of Education.

National Center for Educational Statistics (NCES). (1998). Issues in focus: The
educational progress of Hispanic students. Washington, DC: US. Department of
Education.

National Center for Educational Statistics (NCES). (1995). Fathers' and Mothers'
Involvement in Their Children 's Schools by Family Type and Resident Status
(NCES 2001-032) By C. W. Nord and J. West. Washington, DC: Author.

National Commission on Children. (1991). Beyond rhetoric: A new American agenda for
children and families. Final Report. Washington, DC: Author.

155

Nesman, T. M., Barobs-Gahr, B., & Medrano, L. (2001, April). They Are Our Kids:
Signiﬁcant Findings from a 1998 Latino Dropout Study. Paper presented at the
National Migrant Conference, Orlando, FL.

Ogbu, J. U. (1974). The next generation: An ethnography of education in an urban
neighborhood. New York: Academic Press.

Ogbu, J. U. (1981). Origins of human competence: A cultural-ecological perspective.
Child Development, 52, 413-429.

Ogbu, J. U. (1987). Variability in minority school performance: A problem in search of
an explanation. Anthropology & Education Quarterly, 18, 312-334.

Ogbu, J. U. (1991a). Involuntary minorities. In M. A. Gibson & J. U. Ogbu (Eds),
Minority status and schooling (pp. 3-22). New York: Garland.

Ogbu, J. U. (1991b). Low performance as an adaptation: The case of blacks in Stockton,
California. In M. A. Gibson & J. U. Ogbu (Eds), Minority status and schooling
(pp. 249-285). New York: Garland.

Ogbu, J. U. (1992). Understanding cultural diversity and learning. Educational
Researcher, 21(8), 5-14.

Ogbu, J. U. (1999). Cultural context of children’s development. In H. E. Fitzgerald, B. M.
Lester, & BS. Zuckerrnan (Eds) Children of color: research, health, and policy
issues, (pp. 73-92). New York: Garland Press.

Ogbu, J. U., & Matute-Bianchi, M. E. (1986). Understanding sociocultural factors:
Knowledge, identity and school adjustment. In D. D. Holt (Ed), Beyond
language: Social and cultural factors in schooling language minority students
(pp. 73-142). Sacramento: California State Department of Education, Bilingual
Education Ofﬁce.

Oliver, J. E. & Wong, J. (2003). Intergroup prejudice in multiethnic settings. American
Journal of Political Science, 47(4), 567-582.

Orﬁeld, G., Bachmeier, M. D., J arnes, D. R., Eitle, T. (1997). Deepening segregation in
American public schools: A special report from the Harvard Project on School
Desegregation. Equity & Excellence in Education, 30 (2), 5-24.

Osborne, J. W. (2001). Testing stereotype threat: Does anxiety explain race and sex

differences in achievement? Contemporary Educational Psychology, 26, 291—
3 1 0.

156

Osterrnan, K. F. (1998). Student Community within School Context: A Research
Synthesis. Paper presented at the Annual Meeting of the American Educational
Research Association. San Diego, CA.

Paulson M. J ., Coombs R. H., & Richardson, M. A. (1990). School performance,
academic aspirations, and drug use among children and adolescents. Journal of
Drug Education, 20(4), 289-303.

Paulson, S. E. (1996) Maternal employment and adolescent achievement revisted: an
ecological perspective. Family Relations, 45, 201-208.

Paulson, S. E., Marchant, G. J ., & Rothlisberg, B. A. (1998). Early adolescents'
perceptions of patterns of parenting, teaching, and school atmosphere:
Implications for achievement. Journal of Early Adolescence, 18(1), 5-26.

Paulson, S. E., Marchant, G. J ., & Rothlisberg, B. A. (2001). Relations of middle school
students' perceptions of family and school contexts with academic achievement.
Psychology in the Schools, 38(6), 505-519.

Pebley, A. R & Sastry, N. (2003). Concentrated Poverty vs. Concentrated Aﬂluence:
Eﬂects on Neighborhood Social Environments and Children 's Outcomes. RAND
Labor & Population Working Paper Series 03—24.

Pebley, A. R. & Vaiana, M. E. (2002). In our backyard: How 3 LA. neighborhoods aﬂect
kids ’ lives. Santa Monica, CA: RAND.

Perkins, D. F., & Villaruel, F. A. (2000) An Ecological, Risk-Factor Examination of
Latino Adolescents; Engagement in Sexual Activity. In F. A. Villaruel (Ed.).
Making Invisible Latino Adolescents Visible. (pp. 83-106) Newbury Park, CA:
SAGE Publications, Inc.]

Perkins, D. N. (1981). The Mind is Best Work. Cambridge, MA: Harvard University Press

Phinney, J. S. (1989). Stages of ethnic identity in minority group adolescents. Journal of
Early Adolescents, 9, 34-49.

Phinney, J ., & Alpuria, L. (1990). Ethnic identity in older adolescents ﬁom four ethnic
groups. Journal of Youth and Adolescence, 9, 163-173.

Phinney, J. & Chavira, V. (1992). Ethnic identity and self-esteem: An exploratory
longitudinal study. Journal of Adolescence, 15, 271-281.

Pinderhughes, E.E., Nix, R., Foster, E. M., Jones, D., & TCPPRC (2001). Parenting in
context: Impact of neighborhood poverty, residential stability, public services,
social networks, and danger on parental behaviors. Journal of Marriage and
Family, 63, 941-53.

157

Portes, A., & Rumbaut, R. (1996). Immigrant America: A portrait (2“d ed.). Berkeley,
CA: University of California Press.

Portes, A., & Zhou, M. (1993). The new second generation: Segmented assimilation and
its variants. Annals, AAPSS, 530, 75-96.

Posner, J. K., & Vandell, D. L. (1994). Low -income children's after-school care: Are
there beneﬁcial effects of after school programs? Child Development, 65(2), 440-
456.

Posner, J. K., & Vandell, D. L. (1999). After-school activities and the development of
low-income urban children: a longitudinal study. Developmental Psychology,
35(3), 868-879.

Powell-Cope, G. M., & Eggert, L. L. (1994). Psychosocial Risk and protective Factors:
Potential high school dropouts versus typical youth. In R. C. Moss (Ed). Using
what we know about at-risk youth: Lessons ﬁom the ﬁeld. (pp. 23-51). Lancaster,
PA: Technomic.

President’s Advisory Commission on Educational Excellence for Hispanic Americans
(2000). Testing Hispanic students in the United States: Technical and policy
issues. Washington, DC: White House Initiative on Educational Excellence for
Hispanic Americans.

President's Advisory Commission on Educational Excellence for Hispanic Americans,
(2003). From Risk to Opportunity: Fulfilling the Educational Needs of Hispanic
Americans in the 21 " Century. Washington, DC: White House Initiative on
Educational Excellence for Hispanic Americans.

Provitera Mcglynn, A. (2001, October 8). Hispanic girls most likely to drop out and
stay out: Mixed messages and discrepant expectations faulted. The Hispanic
Outlook in Higher Education, p. 30.

Quillian, L. &. Pager, D. (2001). Black neighbors, higher crime? The role of racial
stereotypes in evaluations of neighborhood crime, American Journal of Sociology
1 07: 7 17—67.

Quinn, J. (1995). Positive effects of participation in youth organizations. In M. Rutter
(Ed). Psychological disturbances in young people: Challenge for Prevention. (pp.
274-304). Cambridge: Cambridge University Press.

Ramos, J. (2002). The other face of America: Chronicles of the immigrants shaping our
future. New York: Harper Collins Publishers.

158

Redd, Z., Cochran, S. W., Hair, E., & Moore, K. A. (2002). Academic achievement
programs and development: A synthesis. Washington, DC: Child Trends.

Reyes, M. de la Luz, & Halcon, J. J. (Eds). (2001). Best for our children: Critical
perspectives on literacy for Latino students. New York: Teachers College Press

Reyes, O. &. Jason, L. A. (1993). Pilot study examining factors associated with academic
success for Hispanic high school students. Journal of Youth and Adolescence,
22(1), 57-71.

Reyes, P., & Valencia, R. R. (1993) Educational policy and the growing Latino student
population: Problems and prospects. Hispanic Journal of Behavioral
Sciences,15(2), 258-283.

Robertson, L. M., Harding, M. S., 81. Morrison, G. M. (1998). A comparison of risk and
resilience indicators among Latino/a students: Differences between students

identiﬁed as at-risk, learning disabled, speech impaired and not at-risk. Education
and Treatment of Children, 21(3), 333-353.

Rodriguez, C. (1975). A cost beneﬁt analysis of subjective factors affecting assimilation:
Puerto Ricans. Ethnicity, 2, 66-80.

Rodriguez, E., Hirschl,T. A., Mead, J. P., & Groggin, S. E. (1999). Understanding the
Drﬂ'erence 4-H Clubs Make in the Lives of New York Youth: How 4-H Contributes
to Positive Youth Development. Ithaca, NY: Cornell University.

Rodriguez, M. C. & Morrobel, D. (2002). Latino youth development: a vision of success
in a period of empirical drought. Paper Presented at the annual meeting of the
American Educational Research Association, New Orleans, LA.

Romo, H., & F albo,T. (1996) Latino high school graduation: Defying the odds. Austin,
TX: University of Texas Press.

Rosenbaum, E. & Friedman, S. ( 2001). Differences in the locational attainment of
immigrant and Native-bom households with children in New York City.
Demography, 38(3), 337-348.

Rosenbaum, E., Friedman, S., Schill, M. H., & Buddelrneyer, H. (1998). Nativity
Diﬁ'erences in Neighborhood Quality Among New York City Households, 1996.
Working Paper 98-4, University School of Law Center for Real Estate and Urban
Policy.

Rosenthal, D.A. & Feldman, S. S. (1991). The inﬂuence of perceived family and personal

factors on self-reported school performance of Chinese and Western high school
students. Journal of Research in Adolescence, I , 135-154.

159

Roth, J ., Brooks-Gunn, J ., Murray, L., & Foster,W. (1998). Promoting healthy
adolescents: Synthesis of youth development program evaluations. Journal of
Research on Adolescence, 8, 423-459.

Rowe, D. C. (1994). The limits of family inﬂuence: Genes, experience, and behavior.
New York: Guilford Press.

Rueda, R. (1998). Standards for professional development: A sociocultural perspective.
Santa Cruz, CA: University of California.

Rumberger, R. W. (2001, January). Why Students Drop Out of School and What Can Be
Done. Paper presented at the conference of Drop Outs in America: How Severe is
the Problem? What Do We Know about Intervention and Prevention?, Harvard
University.

Rumberger, R. W., & Larson, K. A. (1998) Toward explaining differences in educational
achievement among Mexican American language-minority students. Sociology of
Education, 71, 69-93.

Rumberger, R. W. & Thomas, S. L. (2000) The distribution or dropout and turnover rates
among urban and suburban high schools. Sociology of Education, 73, 39-67.

Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In
M. W. Kent, & J. E. Rolf (Eds), Primary prevention of pathology: Social
competence in children (W01. 3). Hanover: University of New England.

Sampson, R. J. & Raudenbush, R. W. (2005). Neighborhood stigma and the perception of
disorder. Focus, 24(1), 7-11.

San Miguel G., Jr. & Valencia, R. R. (1998). From the Treaty of Guadalupe Hidalgo to
Hopwood: The educational plight and struggle of Mexican Americans in the
Southwest. Harvard Educational Review, 68(3), 353-412.

Sanbonmatsu, L., Kling, J. R., Duncan, G. J ., & Brooks-Gunn, J. (2006). Neighborhoods
and Academic Achievement: Results ﬁ'om the Moving to Opportunity
Experiment. NBER working paper series, v no. WI 1909.

Santiago, A. M. & Galster, G. (1995). Puerto Rican segregation in the United States:
cause or consequence of economic status? Social Problems, 42(3), 361-3 89.

Scales, P. C., Benson, P. L., Leffert, N., & Blyth, D. A.. (2000). Contribution of

developmental assets to the prediction of thriving among adolescents, Applied
Developmental Science, 4(1), 27-46.

Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory
of genotype -environment effects. Child Development, 54, 424-435.

160

Shultz, G. F., (1993). Socioeconomic advantage and achievement motivation: Important
mediators of academic performance in minority children in urban schools. Urban
Review, 25, 221-232.

Shulrnan, S. (1990). The causes of black poverty: Evidence and interpretation. Journal of
Economic Issues, 24 (4), 995-1016.

Shumow, L., & Miller J. D. (2001). Parents' at-home and at-school academic
involvement with young adolescents. Journal of Early Adolescence, 21(1), 68-91.

Smith, J. P. (2003). Assimilation across the Latino Generations. AEA Papers and
Proceedings, 93(2), 3 15-3 19.

Stanton-Salazar, R. D., Vasquez, O. A., & Mehan, H. (2000). Re-engineering academic
success through institutional support. In S. T. Gregory (Ed.), The academic
achievement of minority students: Perspectives, practices, and prescriptions (pp.
213-305). Lanham, MD: University Press of America, Inc.

Steinberg, L. D. & Darling, N. (1993). The broader context of social inﬂuence in
adolescence. in R. K. Silbereisen & E. Todt (Eds), Adolescence in Context: The
interplay of family, school, peers, and work in adjustment. (pp. 25-45). New
York, Springer-Verlag.

Steinberg, L., Lambom, S. D., Dombusch, S. M., & Darling, N. (1992). Impact of
parenting practices on adolescent achievement: Authoritative parenting, school
involvement, and encouragement to succeed. Child Development, 63, 1266-1281.

Suarez-Orozco, M. (1987). Towards a psychological understanding of Hispanic
adaptation to American schooling. In H. E. Trueba (Ed.), Success or failure,
learning and the language minority student, (pp. 156-168). Cambridge, MA:
Newbury House.

Swanson, C. B. (2003). Who Graduates? Who Doesn 't? A Statistical Portrait of Public
High School Graduation, Class of 2001 .Washington, DC: The Urban Institute.

Szapocznik, J ., Kurtines, W., & Fernandez, T. (1980). Bicultural involvement and
adjustment in Hispanic-American youths. International Journal of Intercultural
Relations, 4, 353-365.

Szapocznik, J ., Santisteban, D., Kurtines, W., Perez-Vidal, A., & Hervis, O. (1984).
Bicultural effectiveness training: A treatment intervention for enhancing
intercultural adjustment in Cuban American families. Hispanic Journal of
Behavioral Sciences, 6, 314-344.

161

Taeuber, K. E. & Taeuber, A. F. (1965). Negroes in cities: Residential segregation and
neighborhood change. Chicago: Aldine.

Tatum, B. (1987). Assimilation blues. Westport, CT: Greenwood.

Tharp, R. G., & Gallimore, R. (1988). Rousing minds to life: Teaching, learning, and
schooling in social context. New York: Cambridge University Press.

Thomas, R. M. (2000). Comparing Theories of Child Development (511’ Ed.).
Belmont,CA: Wadsworth/Thomas Learning.

Tienda, M. (1991). Poor people, poor places: Deciphering neighborhood effects on
poverty outcomes, in J. Huber (Ed.). Macro-Micro Linkages in Sociology, pp 244-
62. Sage Publications.

Torres Campos, C. M. (2003). Individual and environmental diﬂerence between high-
achieving and low-achieving Michigan Latino students. Unpublished master's
thesis, Michigan State University.

Trickett, E. J ., Birman, D. (2005). Acculturation, school context, and school outcomes:
Adaptation of refugee adolescents ﬁom the former Soviet Union. Psychology in
the Schools, 42(1), 27-38.

Tucker, C. & Lewis, C. (1973). A reliability coefﬁcient for maximum likelihood factor
analysis. Psychometrika, 38, 1-10.

Turner, M., Fix, M., & Struyk, R. J. (1991). Opportunities denied, opportunities
diminished: Discrimination in hiring. Washington, DC: Urban Institute Press.

US. Census Bureau (November 1999). Resident population estimates of the United
States by sex, race, and Hispanic origin: April 1, 1990 to October 1, 1999.
Washington D. C.: US. Census Bureau.

US. Department of Education. (n. d.). How important is homework? Retrieved June 1,
2004 ﬁ‘om
http://www.kidsource.com/kidsource/content/how_important__homework.htrnl

US. Department of Education. (1997). Overcoming barriers to family involvement in
Title I schools: Report to Congress. Washington, DC: Ofﬁce of Educational
Research and Improvement.

US. Department of Education. (1992). Dropout rates in the United States, 1991.
Washington, DC: US. Government Printing Ofﬁce.

Vaden-Kieman, N. (1996), Parent's Reports of School Practices to Involve Families.
(NCES 97-327). Washington DC: US. Department of Education.

162

Valencia, R R., and Black, M. S. (2002). “Mexican Americans don’t value education!”—
On the basis of the myth, mythmaking, and debunking. Journal of Latinos and
Education, 1(2), pp. 81-103.

Valsiner, J. (Ed.). (1988). Child development within culturally structured environments
(Vols. 1& 2). Norwood, NJ: Ablex.

Valsiner, J. (1987). Culture and the development of children 's action. Chicester, England:
Wiley.

Vera, E. M. (2000). Aspirations, barriers, and community strengths: A qualitative survey
of urban Chicano youth and families (Occasional Paper No.38). East Lansing, MI:
Michigan State University, Julian Sarnora Research Institute.

Villarruel, F. A., & Montero-Sieburth, M. (2000). Introduction: Latino Youth and
America, Time for Change. In F. A. Villarruel (Ed.). Making Invisible Latino
Adolescents Visible. (pp. xi-xxx) Newbury Park, CA: SAGE Publications, Inc.

Villarruel, F.A., Montero-Sieburth, M., Dunbar, C., & Outlay, C. W. (2005). Dorothy,
there is no yellow brick road: The paradox of community youth development
approaches for Latino and African American urban youth. In J .L. Mahoney, R.W.
Larson & J .S. Eccles (Eds), Organized activities as a context of development:

Extracurricular activities, after-school, and community programs (pp. 111-130).
Mahwah, NJ: Lawrence Erlbaum Associates.

Voydanoff, P., & Donnelly, B. W. (1999). Risk and protective factors for psychological

adjustment and grades among adolescents. Journal of Family Issues, 20(3), 328-
349.

Vygotsky, L. S. (1962). Thought and language. Cambridge, MA: MIT Press.

Vygotsky, L. S. (1978). In M. Cole, V. John-Steiner, S. Scribner, & E. Souberrnan (Eds),
Mind in society. Cambridge, MA: Harvard University Press.

Wahlberg, B., & Fredrick, W. C., (1982). Instructional time and learning. Encyclopedia
of Educational Research, 2, 917-914.

Wang, M. C. & Gordon, B. W. (Eds). (1994). Educational resilience in inner-city
America: Challenges and prospects. Hillsdale, NJ: Lawrence Earlbaum.

Wang, M. C., Hartel, G. D., & Walberg, H. J. (1999). Psychological and Educational
Resilience. In A. J. Reynolds, H. J. Walberg and R. P. Weissberg (Eds).

Promoting positive outcomes: Issues in children 's and families' lives. (pp. 329-
365) Washington, DC: Child Welfare League of America.

163

Wasik, B. H. (1992, January). The context of the classroom and children's learning. In J.
Belsky (Chair), The context of schooling. Symposium conducted at the School
Readiness: Scientiﬁc Perspectives National Conference, Columbia, MD.

Waters, M. (1994). Ethnic and racial Identities of second-generation black immigrants in
New York City. in A. Portes (Ed). The New Second Generation, pp. 171-198.
New York: Russell Sage Foundation.

Waxman, H. C., Huang, S. L., & Padron, Y. N. (1997). Motivation and learning
environment differences between resilient and non-resilient Latino middle school
students. Hispanic Journal of Behavioral Sciences, 19(2), 137-155.

Weiner, L., Lighton, M., & Funkhouser, J. (2000). Helping Hispanic students reach high
academic standards. Washington, DC: US. Department of Education.

Wentzel, K. R. (1996). Social and academic motivation in middle school: Concurrent and
long-term relations to academic effort. Journal of Early Adolescence, 16(4), 390-
406.

Wentzel, K. R. (1989) Adolescent classroom goals, standards for performance, and
academic achievement: An interactionist perspective. Journal of Educational
Psychology, 81(2), 131-142.

Wenz-Gross, M., & Siperstein, G. N. (1997). Importance of social support in the
adjustment of children with learning problems. Exceptional Children, 63, 183-
193.

Werner, E. & Smith, R. (1992). Overcoming the odds: High risk children from birth to
adulthood. Ithaca, NY, Cornell University.

Whitehead, J. (1984). Motives for higher education: A study of intrinsic and extrinsic
motivation in relation to academic attainment. Cambridge Journal of Education,
14, 26-34.

Wigﬁeld, A., Eccles, J. S., & Rodriguez, D. (1998). The development of children’s
motivation in school contexts. In P. D. Pearson & A. Iran-Nejad (Eds), Review of
research in education (pp.79-118). Washington, DC: American Educational
Research Association.

Willis, P. (1972). Learning to labour: How working class kids get working class jobs.
Westrnead, Famborough, Hants. England: Saxon House.

Wong, S. K. (1999). Acculturation, peer relations, and delinquent behavior of Chinese-
Canadian youth. Adolescence, 34(133), 107-1 19.

Zayas, L. H. (1994). Hispanic family ecology and early childhood socialization: Health
care implications. Family Systems Medicine, 12, 315-325.

164

Zayas, L. H., & Solari, F. (1994). Early childhood socialization in Hispanic families:
Context, culture, and practice implications. Professional Psychology: Research
and Practice, 25, 200-206.

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

Zill, N. & Nord, C. W. (1994). Running in Place: How American Families are Faring in
a Changing Economy and an Individualistic Society. Washington, DC: Child
Trends, Inc.

Zill, N. & West, J. (2000). Entering kindergarten: A portrait of American children when

they begin school, in US. Department of Education, NCES, The Condition of
Education, Washington, DC: US. Departrrrent of Education.

165

EUN SITVL ES1

111111111111111111411511111)1

 

 

’4 4A ./-—'.‘T~——..