help turn paper
ani1010
ContentServer.pdf
E M P I R I C A L R E S E A R C H
Trajectories of Mexican American and Mainstream Cultural Values Among Mexican American Adolescents
George P. Knight • Camille D. Basilio •
Heining Cham • Nancy A. Gonzales •
Yu Liu • Adriana J. Umaña-Taylor
Received: 9 April 2013 / Accepted: 10 July 2013 / Published online: 23 July 2013
� Springer Science+Business Media New York 2013
Abstract Mexican Americans are one of the largest and
fastest growing ethnic groups in the United States, yet we have
limited knowledge regarding changes (i.e., developmental
trajectories) in cultural orientation based upon their exposure
to the Mexican American and mainstream cultures. We
examined the parallel trajectories of Mexican American and
mainstream cultural values in a sample of 749 Mexican
American adolescents (49 % female) across assessments
during the fifth grade (approximately 11 years of age), the
seventh grade (approximately 13 years of age) and the tenth
grade (approximately 16 years of age). We expected that these
values would change over this developmental period and this
longitudinal approach is more appropriate than the often used
median split classification to identify distinct types of accul-
turation. We found four distinct acculturation trajectory
groups: two trajectory groups that were increasing slightly
with age in the endorsement of mainstream cultural values,
one of which was relatively stable in Mexican American
cultural values while the other was declining in their
endorsement of these values; and two trajectory groups that
were declining substantially with age in their endorsement of
mainstream cultural values, one of which was also declining in
Mexican American cultural values and the other which was
stable in these values. These four trajectory groups differed in
expected ways on a number of theoretically related cultural
variables, but were not highly consistent with the median split
classifications. The findings highlight the need to utilize lon-
gitudinal data to examine the developmental changes of
Mexican American individual’s adaptation to the ethnic and
mainstream culture in order to understand more fully the
processes of acculturation and enculturation.
Keywords Cultural values � Mexican American culture � Longitudinal trajectories � Acculturation and enculturation
Introduction
The rapidly increasing absolute and relative size of several
ethnic minority populations in the United States (particu-
larly Mexican Americans: US Census Bureau 2012) is
noteworthy given the empirical evidence that increased
exposure to the mainstream culture of the United States
may be associated with increased negative behavioral and
mental health outcomes for some ethnic minority youth
(e.g., Rogler et al. 1991; Samaniego and Gonzalez 1999).
Some ethnic minority youth may be at risk for negative
outcomes (e.g., internalizing problems, externalizing
problems, academic failure, and drug and alcohol abuse)
because the demands to adapt to both the mainstream and
G. P. Knight (&) � C. D. Basilio � H. Cham � N. A. Gonzales � Y. Liu Department of Psychology, Arizona State University,
950 S McAllister Ave, Tempe, AZ 85287-1104, USA
e-mail: [email protected]
C. D. Basilio
e-mail: [email protected]
H. Cham
e-mail: [email protected]
N. A. Gonzales
e-mail: [email protected]
Y. Liu
e-mail: [email protected]
A. J. Umaña-Taylor
T. Denny Sanford School of Social and Family Dynamics,
Arizona State University, P.O. Box 873701, Tempe,
AZ 85287-3701, USA
e-mail: [email protected]
123
J Youth Adolescence (2014) 43:2012–2027
DOI 10.1007/s10964-013-9983-8
ethnic cultures require adherence to the behavioral expec-
tations and values of the ethnic culture and adherence to
the behavioral expectations and values of the mainstream
culture (e.g., Gonzales et al. 2002; Szapocznik and Kur-
tines 1993). However, some authors have suggested that a
strong connection to the ethnic culture may be protective
and suppress these negative effects because these youth
may be less differentiated from their parents, less likely to
experience family conflict, and more likely to receive
strong social support from the family (e.g., Atzaba-Poria
and Pike 2007; Schwartz et al. 2006; Szapocznik and
Kurtines 1993; Vega and Gil 1999). Hence, understanding
the nature of this dual cultural adaptation process is critical
to understanding the potential risks, and protections, that
may be associated with these demands.
Although some (e.g., Berry 2006) describe this dual cul-
tural adaptation under the rubric of acculturation, we describe
this dual cultural adaptation as occurring through the pro-
cesses of acculturation and enculturation (e.g., Gonzales et al.
2002) to differentiate those forces promoting mainstream
adaptation from those promoting ethnic adaptations. Accul-
turation is the process of the adaptation to the mainstream
culture, while enculturation is the process of adaptation to the
ethnic culture. Although acculturation and enculturation pro-
cesses are separable, they are not independent or orthogonal,
and they lead to outcomes in which an individual may achieve
any combination of levels along each dimension. These dual-
axis processes of adaptation lead to change over time in a wide
array of psychosocial dimensions including cultural knowl-
edge, behaviors, beliefs, attitudes, and values (e.g., Berry
2006; Cuellar et al. 1995; Felix-Ortiz et al. 1994; Gonzales
et al. 2002; LaFramboise et al. 1993; Tsai et al. 2002). Fur-
thermore, these adaptations occur through normal develop-
mental and socialization processes (broadly defined) that
unfold throughout the lifespan of ethnic minorities who have
immigrated recently, as well as those that have been in the
United States for several generations. The resulting develop-
mental changes in culturally related knowledge, behaviors,
beliefs, attitudes, and values become an integral part of the
ethnic minority individuals’ ethnic and mainstream social
identities depending on the culture with which they are asso-
ciated. Much, but not all, ethnic socialization occurs in the
family and ethnic community. Much, but not all, mainstream
socialization occurs in schools, mainstream community, and
media. Because of the variability in the cultural contexts in
which ethnic minority youth live (e.g., school and neighbor-
hood ethnic composition, language spoken in the home, the
availability of ethnically related services and products), there
is considerable variability in the degree of ethnic and main-
stream socialization pressures they experience and in their
connection to the ethnic and mainstream cultures. This vari-
ability has led many theorists to believe that there are quali-
tatively different types of dual cultural adaptations and
associated developmental trajectories experienced by ethnic
minority youths (e.g., Szapocznik and Kurtines 1993; Tadmor
and Tetlock 2006).
Unfortunately, the research literature examining these dual
cultural adaptation processes has been limited in several ways.
First, although these processes often are described as resulting
in changes in response to exposure to both the mainstream and
ethic cultures, most of the studies on this topic have been
conspicuously cross-sectional and have relied on single point
in time assessments. Hence, most of these studies do not
directly assess changes in cultural orientation associated with
age related changes in exposure to the ethnic culture and the
mainstream culture. The few examinations of the longitudinal
trajectories associated with these types of adaptations by
ethnic minority individuals generally have investigated only
one of these processes (often enculturative changes) and/or
examined these changes in very select samples (e.g., clinical
patients, college students or juvenile offenders, etc.: Altschul
et al. 2006; French et al. 2006; Kiang et al. 2010; Knight et al.
2012; Pahl and Way 2006; Schwartz et al. 2013; Syed and
Azmitia 2009). The primary purpose of the present study was
to examine the longitudinal trajectories of developmental
change associated with acculturative and enculturative adap-
tation processes in a relatively representative sample of
Mexican American adolescents.
Second, much of this research has focused on a limited
range of behavioral indicators, such as language use and
affiliation patterns, that may be somewhat, if not strongly,
determined by the adults (i.e., parents and teachers) in the
adolescent’s life. Therefore, we examine the parallel tra-
jectories of change in the endorsement of values relatively
more often associated with the Mexican American culture
(i.e., referred to as Mexican American cultural values) and
values relatively more often associated with the main-
stream culture of the United States (i.e., referred to as
mainstream cultural values) among Mexican American
adolescents across three assessments over a 6-year time
period. The specific Mexican American cultural values
(i.e., Familism-Support, Familism-Obligations, Familism-
Referents, Respect, and Religiosity) and mainstream cul-
tural values (i.e., Material Success, Independence and Self-
Reliance, and Competition and Personal Achievement)
examined in this study were identified as associated with
the respective culture by a sequence of focus groups of
Mexican American adolescents, mothers, and fathers
(Knight et al. 2010). We decided to examine the trajecto-
ries of change in culturally related values, and across this
age span, because this is a time period during which there
are substantial changes in their exposure to the mainstream
culture (e.g., Daniel et al. 2012). Although there is con-
siderable variability during the elementary school years,
Mexican American children are more likely to attend
neighborhood schools that have a relatively higher
J Youth Adolescence (2014) 43:2012–2027 2013
123
proportion of Mexican American students, teachers, and
staff. As they transition into middle schools and eventually
high school they are likely to experience an increase in the
proportion of students and adults from a more mainstream
background. These transitions represent a developmental
time period during which adolescents are increasingly
being influenced by peers and during which familial
influence may decease somewhat (Brown 1990; Knight
et al. 2009), and they increasingly experience a wider
variety of immediate contexts to which they must adapt.
The internalization of such values ultimately becomes an
important guide to behavior in a wide range of contexts
(e.g., Schwartz 1999). Because of the great variability, and
sometime novelty, in the contexts to which they are
exposed, it is likely that they increasingly rely on the
values they have internalized in their ethnic and main-
stream identities to guide them to appropriate behavioral
responses in these contexts (e.g., Daniel et al. 2012). Sec-
ond, the internalization of such values very accurately
reflects the psychological state of the adolescent experi-
encing the dual-cultural adaptation demands associated
with the processes of acculturation and enculturation.
Furthermore, adolescence is a developmental period during
which the normative processes of identity development
(Erikson 1968) lead minority adolescents to become more
aware of the importance of ethnic group membership and
to explore their ethnic origins and commit to that ethnic
group (e.g., Phinney 1990; Umaña-Taylor et al. 2009).
During these developmental transition periods, social
identity and self-categorization processes (e.g., Tajfel and
Turner 1986; Turner et al. 1987) are likely to lead Mexican
American youth to internalize values associated with
Mexican American culture in part because of their
increasing commitment to, and exploration of, their ethnic
group membership. While younger children may behave in
accordance with the cultural values of the parents and other
socialization agents often because of the sanctions (both
positive and negative) associated with behaving accord-
ingly; adolescents may be abstracting value rules from their
socialization experiences and internalizing these values
into their system of social identities so that they become
self-chosen guides for behavior (e.g., Knight et al. 2011).
Indeed, there is evidence that late childhood or early ado-
lescence is a time at which important changes in reasoning
and values begin to emerge (e.g., Daniel et al. 2012;
Eisenberg et al. 1991).
A third limitation often associated with the preponder-
ance of the studies that rely on single-point-in-time
assessments is the frequent use of the median-split proce-
dure to classify individuals into different types of accul-
turation/enculturation groups. The use of sample specific
median-splits can be very problematic because the inherent
positive skew of many of the indicators of cultural
orientation (i.e., measures of acculturation, ethnic identity,
ethnic pride, cultural values, etc.) often have median scores
that are relatively high (i.e., often nearly a 4.0 or above on
a five-point response scale) and the respondents that are
below this median are classified as though they are low
even though very few may actually score low on these
actual response scale (Coatsworth et al. 2005; Knight et al.
2009). In contrast, we rely on person-centered analytical
methods to identify individuals’ longitudinal changes in
these culturally related values as they transition from the
elementary school years through middle schools and high
school years. We also identify groups of Mexican Ameri-
can adolescents who are experiencing similar trajectories
of change, and we do so without imposing an arbitrary
dichotomization like a median-split. We also examine the
association of the resulting trajectory groups to a number of
variables reflecting adolescents’ cultural adaptations (i.e.,
the adolescents reported ethnic pride, ethnic identity, per-
ceptions of ethnic discrimination, biculturalism, and use of
the Spanish and English languages) and their gender and
nativity.
A secondary purpose of this study is to examine the
relationship of these longitudinal trajectories to the single-
point-in-time median-split classifications (e.g., Berry 2007;
Szapocznik et al. 1980) in which ethnic minority partici-
pants have been classified as ‘‘bicultural’’ if they score
above the median on a focal assessment of ethnic cultural
orientation and above the median on an assessment of
mainstream cultural orientation. The other three quadrants
created by a median split on ethnic orientation and main-
stream orientation have been labeled ‘‘assimilated’’ (high
mainstream, low ethnic); ‘‘separated’’ or ‘‘alienated’’ (high
ethnic, low mainstream); and ‘‘marginalized’’ (low on
both). Although there has been some debate regarding
whether a mid-scale split might be more conceptually
defensible (e.g., Arends-Tóth and van de Vijver 2006), the
empirical evidence indicates that the dichotomization of a
naturally continuous variable results in: lower measure-
ment reliability and the associated loss of effect size and
power, spurious relations and the overestimation of effect
sizes in analyses with multiple independent variables,
overlooking non-linear relations (MacCallum et al. 2002).
Indeed, analyses that dichotomized continuous variables
perform well only when the: underlying distribution of the
variable is strongly categorical, the proportion of partici-
pants assigned to each dichotomized categories matches
the proportion found in latent variable analyses, and the
continuous measure is highly reliable (DeCosta et al.
2009). These problems may well be why recent studies that
relied on a person-centered approach (Coatsworth et al.
2005; Knight et al. 2009), one of which was also longitu-
dinal (Knight et al. 2009), have not found strong evidence
of these four acculturation/enculturation types that result
2014 J Youth Adolescence (2014) 43:2012–2027
123
from the median-split procedure. However, these person-
centered studies also relied on select samples of adoles-
cents, either youth in an intervention program or juvenile
delinquents.
The Current Study
Researchers have suggested that the dual cultural adapta-
tions that ethnic minority youth experience during their
routine contact with the ethnic and mainstream cultures
may lead to changes in cultural knowledge, behaviors,
beliefs, attitudes, and values over time; and that these
changes may create either risk factors for, or protective
factors against, negative outcomes. Hence, the present
study was designed to identify groups of Mexican Ameri-
can youth demonstrating distinct trajectories of change in
Mexican American and mainstream cultural values across a
7 year developmental time frame. Because the limited
longitudinal research on these dual cultural adaptation
processes have not supported the four-fold typologies
identified by the median-split methodologies, we expected
to find a number of distinct trajectory groups but did not
have a priori expectations regarding the exact nature of
these groups. However, these trajectories may differ in
their intercepts and slope of change with age. We also
expected that trajectory group membership would be rela-
ted systematically to a variety of other indicators of cul-
tural orientation (i.e., generation of immigration, ethnic
pride, ethnic identity, perceived ethnic discrimination,
biculturalism, and language use) in ways that make sense
given the nature of the dual cultural adaptations occurring
in each trajectory group. For example, the adolescents in
trajectory groups that indicate a high endorsement (either
initially or increasing over age) of Mexican American
cultural values should score high in ethnic pride, ethnic
identity, and Spanish language use. Adolescents in trajec-
tory groups that are high in both Mexican American and
mainstream cultural values should be high in these same
culturally related variables and also be high in bicultural-
ism and low in perceived discrimination. We also expected
that the dual cultural adaptations represented by the distinct
trajectory groups would not be highly consistent with the
groups resulting from single-point-in-time median-split
analyses.
Method
Participants
Data for this study come from the first, second, and third
assessments (between 2004 and 2011) of an ongoing
longitudinal study investigating the role of culture and
context in the lives of Mexican American families (Roosa
et al. 2008). Participants were 749 Mexican American
adolescents (49 % female) selected from the rosters of
schools that served ethnically and linguistically diverse
communities in the Phoenix metropolitan area. To recruit a
representative sample of Mexican American families, a
multiple step process was implemented that included: a
stratified random sampling strategy to select neighborhoods
diverse in cultural and economic qualities, recruitment
through 47 schools across 35 neighborhoods, the use of
culturally sensitive recruitment and data collection pro-
cesses, conducting interviews in participants’ homes in
English or in Spanish according to the participants’ pref-
erence, and a financial incentive (Roosa et al. 2008). We
originally identified 237 potential communities for inclu-
sion by identifying all public schools (including Catholic
schools and Charter schools) in the metropolitan area with
at least 20 Mexican American students in the fifth grade.
We then scored the degree to which these communities
provided support for parental enculturation efforts using
multiple indicators (i.e., Mexican American population
density, percentage of elected and appointed Latino office
holders, the number of Churches holding service in Span-
ish, the number of locally owned stores selling traditional
Latino foods/medications/household items, and the pres-
ence of traditional Mexican-style stores). We then selected
the five communities that were scored as most ‘‘Mexican
American’’ (because they represented Mexican ethnic
enclaves). We ensured that we had represented the entire
spectrum of community context by choosing a random
starting point from within the 10 lowest scoring commu-
nities and then selecting every 9th school thereafter. A
description of the breadth of these communities is provided
in Roosa et al. (2008).
After receiving consent to contact families through a
letter to the home, we attempted to contact 1,982 families.
Of these, 12 (0.6 %) families could not be contacted, 55
(2.8 %) declined to participate before being screened, and
1,970 were screened to determine if they met the eligibility
requirements. Of the 1,085 who met the eligibility
requirements 749 (69.0 %) completed the initial interview,
270 (24.9 %) declined to participate, 4 (0.3 %) began the
interview but were unable to complete it, and 61 (5.6 %)
were not asked to participate because we had reached our
recruitment goal. Of those who were ineligible: 56 (2.8 %)
no longer attended the participating school, 99 (5.0 %) and
243 (12.3 %) did not have a biological mother or father
(respectively) in the home, 298 (15.1 %) and 106 (5.4 %)
did not have a Mexican American biological mother or
father (respectively), 16 (0.8) were severely learning dis-
abled, 3 (.01 %) could not speak either English or Spanish,
and 9 (.04 %) were participating in another research
J Youth Adolescence (2014) 43:2012–2027 2015
123
project. We included only families with both a Mexican
American biological mother and father to limit the poten-
tial of another cultural influence within the home. Hence,
the overall recruitment success was 73.2 % of those who
were eligible and asked to participate. This sample of
Mexican American families was diverse with respect to
both SES and language (Roosa et al. 2008). Family
incomes ranged from less than $5,000 to more than
$95,000, with the average family reporting an income of
$30,000–$35,000. In terms of language, 30.2 % of moth-
ers, 23.2 % of fathers, and 82.5 % of adolescents were
interviewed in English. At Time 1, the mean age of
mothers in our study was 35.9 (SD = 5.81) and mothers
reported an average of 10.3 (SD = 3.67) years of educa-
tion. At Time 2, the mean age of fathers was 38.1
(SD = 6.26) and fathers reported an average of 10.1
(SD = 3.94) years of education. The adolescents (48.7 %
female) ranged in age from 9 to 12 with a mean of 10.42
(SD = .55; with 97.6 % being 10 or 11 years old) at Time
1. A majority of mothers and fathers were born in Mexico
(74.3, 79.9 %, respectively), and a majority of adolescents
were born in the United States (70.3 %).
At Time 2, approximately 2 years after Time 1 data
collection, most students were in the 7th grade. Of the 39
(5.2 %) families who did not participate at Time 2, 16
(2.1 %) refused to participate. Families who participated at
Time 2 were compared to families who did not participate
at Time 2 on several demographic variables and no dif-
ferences emerged among adolescent (i.e., gender, age,
generational status, language of interview), mother (i.e.,
marital status, age, generational status), or father charac-
teristics (i.e., age, generational status). At Time 3,
approximately 3 years after Time 2 data collection, most
students were in the 10th grade. Of the 109 (14.6 %)
families who did not participate at Time 3, 37 (4.9 %)
refused to participate. Families who participated at Time 3
were compared to families who did not participate at Time
3 on several demographic variables and no differences
emerged among adolescent (i.e., gender, age, generational
status, language of interview), mother (i.e., marital status,
age, generational status), or father characteristics (i.e., age,
generational status).
Procedure
Adolescents completed computer assisted personal inter-
views at their home, scheduled at the family’s convenience,
that were about 2.5 h long. The interviewers were:
80–90 % female (depending upon the assessment year),
between 23 and 60 years of age (with the exception of a
19 year old civil rights activist), fluent in both English and
Spanish, recipients of a master’s or bachelor’s degree (or
the combination of education and a least 2 years of
professional experience in a social service agency), strong
in communication and organizational skills, and knowl-
edgeable about computers. Each interviewer received at
least 40 h of training that included information on the
project’s goals, characteristics of the target population, the
importance of professional conduct when visiting partici-
pants’ homes as well as throughout the process, and the
critical role they would play in collecting the data. Inter-
viewers read each survey question and possible responses
aloud in participants’ preferred language to reduce prob-
lems related to variations in literacy levels. Participating
adolescents were compensated $45 at Time 1, $50 Time 2,
and $55 at Time 3.
Measures
Generation
Participants’ generation of immigration to the United
States was determined based on their country of birth and
the number of parents and grandparents that were born in
Mexico. For example, a participant was coded as a first
generation immigrant if they and their parents and grand-
parents were born in Mexico. A participant was coded as a
second generation immigrant if they were born in the
United States and their parents and grandparents were born
in Mexico. A participant was coded as a third generation
immigrant if they and their parents were born in the United
States and their grandparents were born in Mexico. A
participant was coded as a fourth generation immigrant
(technically a fourth or beyond generation immigrant) if
they, their parents, and their grandparents, were born in the
United States. Participants whose background deviated
slightly from these coding rules (i.e., when a grandparent
was born in another Latin American country) were
assigned the nearest appropriate code.
Mexican American and Mainstream Cultural Values
Adolescents completed the Mexican American Cultural
Values Scale (MACVS: Knight et al. 2010) to assess
Mexican American values and mainstream values. The
development of the MACVS was based on the values that
Mexican American mother, father, and adolescent focus
groups identified as associated with the Mexican American
and mainstream American cultures. The Mexican Ameri-
can values scale consists of 5 correlated subscales from
MACVS: Familism-Support (6 items, e.g., ‘‘parents should
teach their children that the family always comes first’’);
Familism-Obligation (5 items, e.g., ‘‘if a relative is having
a hard time financially, one should help them out if pos-
sible’’); Familism-Referents (5 items, e.g., ‘‘a person
should always think about their family when making
2016 J Youth Adolescence (2014) 43:2012–2027
123
important decisions’’); Respect (8 items, e.g., ‘‘children
should always be polite when speaking to any adult’’); and
Religiosity (7 items, e.g., ‘‘one’s belief in God gives inner
strength and meaning to life’’). The mainstream values
scale consists of 3 substantially correlated subscales from
the MACVS: Material Success (5 items, e.g., ‘‘the best way
for a person to feel good about himself/herself is to have a
lot of money’’); Independence and Self-Reliance (5 items,
e.g., ‘‘as children get older their parents should allow them
to make their own decisions’’); and Competition and Per-
sonal Achievement (4 items, e.g., ‘‘one must be ready to
compete with others to get ahead’’). Adolescents indicated
their endorsement of each item by responding with a five-
point Likert-type scale ranging from (1) not at all to (5)
very much. Confirmatory factor analyses indicated that the
items for each subscale fit best on the respective subscale;
that these subscales also loaded on these two higher order
factors; and that there was reasonable measurement
invariance between 5th grade adolescent and their mothers
and fathers (Knight et al. 2010). Knight et al. (2010) also
provide evidence of the validity of the MACVS subscale
scores. The Cronbach’s a for the Mexican American values scale was .85, .90, and .92 at T1, T2, and T3, respectively.
The Cronbach’s a for the mainstream values scale was .84, .84, and .81 at T1, T2, and T3, respectively.
Language Use
Adolescents completed the four English language use and
four Spanish language use items from the Acculturation
Rating Scale for Mexican Americans-II (Cuellar et al.
1995) at the first assessment. These items best represented
the different contexts in which language use may be rep-
resented that assess both English (a = .69) and Spanish (a = .83) language use. The subset of language use items consists of an English use and Spanish use subscale. The
subscales evaluate frequency of language use (e.g., ‘‘How
often do you speak Spanish?’’). The response scale ranged
from 1 (almost never) to 5 (almost always) and a mean
score was computed for each language, with higher scores
indicating greater language use.
Ethnic Pride
Adolescents completed the Mexican American Ethnic
Pride Scale (Thayer et al. 2002) at the first assessment. The
four-item scale (a = .78) assessed ethnic pride for in Mexican Americans (e.g., ‘‘You feel proud to see Latino
actors, musicians, and artists being successful’’). The
response scale ranged from 1 (not at all true) to 5 (very
true) and a mean score was computed with higher scores
indicating higher levels of ethnic pride.
Ethnic Identity
Adolescents completed the Ethnic Identity Scale (Umaña-
Taylor et al. 2004) at the second assessment. The 17-item
scale (a = .84) assessed ethnic exploration (e.g., ‘‘You have attended events that have helped you learn more about
you ethnic background’’), resolution (e.g., ‘‘You are clear
about what your ethnic background means to you’’), and
affirmation (e.g., ‘‘You feel negatively about your ethnic
background’’; reverse coded). The response scale ranged
from 1 (not at all true) to 5 (very true) and an overall mean
score was computed with higher scores indicating higher
levels of ethnic identification.
Perceived Ethnic Discrimination
Adolescents completed nine items designed to assess dis-
crimination experiences from peers and teachers at the first
assessment. Because we were aware of no well-developed
measures of perceived ethnic discrimination for use with
Mexican Americans at the outset of the study, items were
selected and adapted from three measures that had been
validated for other groups [Hughes and Dodge (1997):
Racism in the Workplace Scale; Landrine and Klonoff
(1996): Schedule of Racist Events; and Klonoff and
Landrine (1995): Schedule of Sexist Events]. There were
4 peer items (e.g., ‘‘How often have kids at school called
you names because you are Mexican American?’’) and 5
teacher items (e.g., ‘‘How often have you had to work
harder in school than White kids to get the same praise or
the same grades from your teachers because you are
Mexican American’’) had a Cronbach’s a was .78. The response scale ranged from (1) Not at all true to (5)
Very true and an overall mean score was computed with
higher scores indicating higher levels of perceived
discrimination.
Biculturalism
Adolescents completed the Mexican American Bicultural-
ism Scale (Basilio et al. 2013) at the third assessment. This
measure had been developed shortly before the beginning
of the third assessment of the second cohort for this sample
and was only available for this half of the sample. This
measure has three subscales: Bicultural comfort (9 items),
bicultural ease (9 items), and bicultural advantage (9
items), which represent the affective, behavioral, and
cognitive components of biculturalism respectively. The
response scale for bicultural comfort ranged from 1 (e.g.,
‘‘I am only comfortable when I need to interact with other
Mexican/Mexican Americans’’ or ‘‘I am only comfortable
when I need to interact with Whites (Gringos).’’) to 5 (e.g.,
‘‘I am always comfortable in both of these situations.’’).
J Youth Adolescence (2014) 43:2012–2027 2017
123
The response scale for bicultural ease (e.g., ‘‘Needing to
work with Mexicans/Mexican Americans sometimes, and
with Whites (Gringos) other times is’’) ranged from 1 (very
difficult) to 5 (very easy). The response scale for bicultural
advantage (e.g., ‘‘For me, being able to interact with other
Mexicans/Mexican Americans sometimes, and being able
to interact with Whites (Gringos) other times has’’) ranged
from 1 (many disadvantages) to 5 (many advantages). An
overall mean biculturalism score was computed using the
items from all three subscales (a = .88) with higher scores indicating higher levels of biculturalism.
Results
Data Analysis Plan
Even though we expected qualitatively different types of
developmental trajectories we conducted preliminary lon-
gitudinal growth models (not presented here) separately,
and jointly, for the trajectories of Mexican American cul-
tural values and mainstream cultural values using Mplus
6.1 (Muthén and Muthén 2010) to confirm that there was
indeed significant variability in the intercepts and slopes of
the developmental trajectories. Based on the hypothesis of
growth across school transitions, the time metric used in
the longitudinal growth curve models was grade (grades 5,
7, and 10; approximately 11, 13, and 16 years of age). The
standard deviations of participants age at each assessment
was relatively small (SD = 0.55) indicating that using age
would provide little or no information gain in growth curve
modeling, but would entail much greater modeling
demands. Furthermore, preliminary analyses revealed that
using age and grade produced virtually identical log-like-
lihood estimates, parameter estimates, and estimated stan-
dard errors. Based on the evidence of significant variability
in the intercepts and slopes of parallel process longitudinal
growth curve modeling for the Mexican American cultural
values and the mainstream cultural values, we conducted
parallel latent class growth models centered at fifth grade
(approximately 11 years of age). We examined latent class
growth models for 1–5 classes, and examined the hetero-
geneity of residual variances across time and latent classes,
and that of the intercept variance across latent classes, in
order to minimize chances of over-extracting latent classes
(Muthén 2004) and producing biased results (Enders and
Tofighi 2008). To determine the optimal number of tra-
jectory groups we relied on several criteria including: the
Bayesian Information Criterion and the sample size
adjusted Bayesian Information Criterion (BIC and saBIC),
the adjusted Lo–Mendell–Rubin test (LMR), all trajectory
groups consisting of at least 5 % of the sample; the prob-
abilities of highest trajectory group membership being near
.80, and the interpretability of the trajectories (Nylund et al.
2007).
After identifying the trajectory groups, we examined the
association of these trajectory groups with several variables
representing the ethnic adaptation of adolescents (adoles-
cents English and Spanish use, ethnic pride, ethnic identity,
perceptions of ethnic discrimination, and biculturalism)
and adolescents’ gender and nativity. Preliminary analyses
revealed that the trajectory groups differed on the inter-
cepts of the trajectories of these culturally related variables
that could change over time, but not the changes over time;
so we examined the differences between the trajectory
groups on these variables measured at the first assessment
(with the exception of ethnic identity which was first
measured at the second assessment and biculturalism
which was first measured at the third assessment). To
examine the relationship between the trajectory groups and
each categorical variable (i.e., gender and generation) we
conducted separate v2 tests of association. To examine the relationship between the trajectory groups and each
remaining continuous variable we conducted a series of
one-way ANOVAs using the trajectory groups as the
independent variable. Each significant omnibus F was
followed by pairwise comparisons. Most of the missing
data resulted from individual participants’ being unavail-
able after the initial data collection (represented by the
noted retention rates) or the measure of biculturalism being
administered to only half of the sample. Beyond this, less
than 1 % of participants had missing data for any variable
included in this report at any assessment.
Finally, to compare our trajectory findings based upon
latent class longitudinal growth curve modeling to the
acculturation categorization approach used in earlier
research, we compared the obtained trajectory groups to a
median split classification (i.e., using those at or above the
median versus those below the median) of the Mexican
American cultural values and the mainstream cultural
values at the first and third assessment. In addition, we
compared the median split classification results at the first
assessment with those at the third assessment.
Parallel Latent Class Growth Models for Mexican
American and Mainstream Values
Based on the fit criteria, we selected a four-class model in
which the log-likelihood estimate = -2,473.54; the
BIC = 5,211.83; the saBIC = 5,084.81; the adjusted LMR
p = 0.17; the percentage of cases in each trajectory group
ranged from 14.3 to 33.8 %; and average probabilities of
highest trajectory group membership ranged from 0.73 to
0.86 across the four groups. In addition, the average
probability of any case being in a different trajectory group
ranged from 0.00 to 0.13, indicating that the degree of
2018 J Youth Adolescence (2014) 43:2012–2027
123
classification error was not serious. The mean trajectories
of the four classes for the Mexican American cultural
values and the mainstream cultural values are displayed in
Fig. 1 with the intercepts and slopes for each class reported
in Table 1. When the latent class growth model indicated
the need for both a linear and quadratic slope but neither of
the mean slopes was significant for a given class, we
conducted a subsequent latent basis model (see Grimm
et al. 2011) to determine if there is a significant mean
change over time (i.e., from the first to the third assess-
ment) represented by the combined mean linear and qua-
dratic slopes. The mean intercepts for all four trajectory
groups were significantly greater than zero.
The first trajectory group (Group 1) includes 241
(32.2 %) adolescents. These adolescents endorsed the
Mexican American cultural values at a very high level and
were relatively stable in the endorsement of these values
over time. These adolescents also were moderate in their
endorsement of mainstream cultural values (see Table 1).
Neither the mean linear or quadratic slopes in-and-of-
themselves indicated significant changes in mainstream
cultural values over time; however, the latent basis model
analyses indicated that this trajectory group on average did
increase significantly (mean latent basis slope = 0.345,
p \ .001) from the first to the third assessment (see Fig. 1). Although these adolescents endorse the Mexican American
cultural values at very high levels and the mainstream
cultural values at moderate levels, they appear to be
increasingly accepting the mainstream cultural values more
than any other trajectory group. Hence, these adolescents
are high in Mexican American values and moderate but
increasing in mainstream values and may be experiencing
the demands of the dual cultural adaptation by becoming
connected to both the ethnic and mainstream cultures.
The second trajectory group (Group 2) included 253
(33.8 %) adolescents. These adolescents were initially the
lowest, but still quite high, in the endorsement of Mexican
American cultural values (i.e., their mean is 4.277 on a
5-point scale); but their endorsement of Mexican American
cultural values declined over time. These adolescents were
also moderate in their endorsement of mainstream cultural
values (see Table 1). Again, although neither the mean
linear or quadratic slopes were statistically significant, the
subsequent latent basis model analyses indicated that this
trajectory group on average did increase somewhat in
mainstream cultural values (mean latent basis slope =
0.234, p \ .001) from the first to the third assessment (see Fig. 1). The adolescents in this trajectory group were
becoming more similar in their level of endorsement of
Mexican American and mainstream values over time;
however, this is mostly because of their waning endorse-
ment of Mexican American cultural values rather than their
very modest increase in their endorsement of mainstream
cultural values.
The third trajectory group (Group 3) included 107
(14.3 %) adolescents. These adolescents were initially the
highest in the endorsement of Mexican American cultural
values, and this endorsement remained at a very high level
Fig. 1 The parallel process longitudinal trajectory groups
for Mexican American and
mainstream cultural values
J Youth Adolescence (2014) 43:2012–2027 2019
123
over time. These adolescents were also initially moderately
high in the endorsement of mainstream cultural values, but
this endorsement declined considerably over time. How-
ever, the rate of decline in their endorsement of mainstream
cultural values was also decreasing somewhat over time
(see Fig. 1). These adolescents are maintaining their high
endorsement of the Mexican American cultural values, but
over time their endorsement of the mainstream cultural
values was waning. These adolescents were high in Mex-
ican American Values and moderately high but declining in
mainstream values. Hence, these adolescents may be
experiencing the demands of the dual cultural adaptation
by maintaining their connection to the ethnic culture while
distancing themselves from the mainstream culture.
The fourth trajectory group (Group 4) included 148
(19.8 %) adolescents. These adolescents initially endorsed
the Mexican American cultural values at a very high level,
but this endorsement declined considerably over time. This
group was also initially moderately high in their endorse-
ment of mainstream cultural values (i.e., 3.437 on a 5-point
scale), but their endorsement declined considerably over
time. However, the rate of decline in their endorsement of
mainstream cultural values was also decreasing somewhat
over time (see Fig. 1). These adolescents were high in
Mexican American values and moderately high in main-
stream values but declining substantially in both set of
values. Hence, these adolescents may be experiencing the
demands of dual cultural adaptation by distancing them-
selves from both the Mexican American and the main-
stream cultures. However, it is important to note that even
after these substantial declines these adolescents were still
endorsing Mexican American cultural values at moderately
high levels and mainstream cultural values at moderate
levels (see Fig. 1).
Relation of Trajectory Groups to Cultural Variables
Table 2 presents the relationships of the trajectory groups
to the adolescent’s generational status. Compared to the
adolescents in the third and fourth trajectory groups, the
adolescents in the first trajectory group, and to some extent
those in the second trajectory group, tend to more often
have parents who were born in the United States (i.e., be
3rd and 4th generation immigrants). However, the gener-
ational differences across the four trajectory groups were
quite small. Gender was not significantly associated with
the trajectory groups [v2 (3) = 3.43]. The series of one-way ANOVAs indicate that the four
trajectory groups differed significantly in Ethnic Pride [F(3,
745) = 23.19, p \ .001], Ethnic Identity [F(3, 706) = 9.96, p \ .001], Perceived Ethnic Discrimination [F(3, 745) = 6.38, p \ .001], Biculturalism [F(3, 301) = 4.71, p \ .01], and Spanish Use [F(3, 745) = 12.25, p \ .001], but not English Use [F(3, 745) = 0.96, ns]. The means of each
culturally related variable for each trajectory group are
presented in Fig. 2. We performed all pairwise comparisons
between the four trajectory groups using Tukey’s HSD test,
and the superscripts above each bar indicate which groups
differed significantly (p B .05) The trajectory groups that
maintained their high endorsement of Mexican American
cultural values over time but declined in their endorsement of
mainstream cultural values (Group 3) were the highest in
ethnic pride while the group declining in Mexican American
cultural values and slightly increasing in mainstream cultural
values (Group 2) was the lowest in ethnic pride. Trajectory
Group 1 (high and stable Mexican American values and
increasing in mainstream cultural values) and Group 4 (high
but declining in Mexican American cultural values and
declining in mainstream values) reported similar levels of
ethnic pride and were between the other two groups. The two
trajectory groups that maintained their high level of
endorsement of Mexican American cultural values over time
(Groups 1 and 3) were higher in ethnic identity than either of
the two groups that were declining in their endorsement of
Mexican American cultural values (Groups 2 and 4). The two
trajectory groups that were increasing somewhat in their
endorsement of mainstream cultural values over time
(Groups 1 and 2) reported less perceived ethnic discrimina-
tion than either of the two groups that were declining in their
endorsement of mainstream cultural values (Groups 3 and 4).
Trajectory Group 1 (high and stable Mexican American
values and increasing in mainstream cultural values) scored
highest in biculturalism, and was significantly higher in
biculturalism than either of the groups that were declining in
their endorsement of Mexican American cultural values
Table 1 Intercepts, slopes, and standard errors (in parentheses) of the Latent Class Longitudinal Trajectory Groups for Mexican American
and Mainstream cultural values
Group Parameter Mexican American
cultural values
Mainstream
cultural values
Group 1
(n = 241)
Intercept 4.544 (0.038)** 2.578 (0.069)**
Linear -0.018 (0.014) 0.017 (0.052)
Quadratic 0.010 (0.010)
Group 2
(n = 253)
Intercept 4.277 (0.031)** 2.571 (0.059)**
Linear -0.091 (0.013)** 0.010 (0.049)
Quadratic 0.006 (0.008)
Group 3
(n = 107)
Intercept 4.776 (0.028)** 3.560 (0.271)**
Linear -0.027 (0.018) -0.333 (0.102)**
Quadratic 0.041 (0.012)**
Group 4
(n = 148)
Intercept 4.618 (0.046)** 3.437 (0.189)**
Linear -0.119 (0.013)** -0.399 (0.093)**
Quadratic 0.051 (0.013)**
Intercept centered at 11 years of age
** p \ .01
2020 J Youth Adolescence (2014) 43:2012–2027
123
(Groups 2 and 4). Trajectory Group 3 (high and stable
Mexican American values and decreasing in mainstream
cultural values) was not significantly different from any of
the other groups in biculturalism. Trajectory Group 3 (high
and stable Mexican American values and decreasing in
mainstream cultural values) also reported the most Spanish
use compared to any of the other trajectory groups; while the
two groups who were increasing somewhat in mainstream
cultural values (Groups 1 and 2) were similarly lower in
Spanish use. Finally, among the two trajectory groups who
were declining in their endorsement of Mexican American
cultural values (Groups 2 and 4), the group declining in
mainstream cultural values (Group 4) reported higher
Spanish use than the group increasing somewhat in main-
stream cultural values (Group 2).
Relation among the Trajectory Groups and the Median
Split Classifications
Table 3 presents the cross-tabulation of the observed tra-
jectory groups with the median split classifications at the first
and third assessment. Although the associations are statisti-
cally significant at each assessment (Time 1: v2 (9) = 459.52, p \ .001; Time 3: v2 (9) = 345.61, p \ .001), the
a a
a
a
ab
a
b b
a
b
a
a
c
a
b
ab
c
a a
b
b
b
b
a
1
2
3
4
5
Ethnic Pride Ethnic Identity Total Score (T2)
Perceived Ethnic Discrimination
Biculturalism (T3) Spanish Use English Use
Group 1 Group 2 Group 3 Group 4
Fig. 2 Comparisons of each trajectory group on each culturally related variable. [Groups with a different superscript differ significantly (p B .05).]
Table 2 The percentage of each trajectory group by generation status
Trajectory group v2 (df)
Group 1
(n = 239) (%)
Group 2
(n = 252) (%)
Group 3
(n = 107) (%)
Group 4
(n = 148) (%)
Generation
First 21.34 27.78 39.25 37.84 v2 (9) = 39.69***
Second 40.59 47.62 40.19 40.54
Third 16.74 5.56 4.68 10.81
Fourth 21.34 19.05 15.89 10.81
Only 746 out of 749 adolescents reported their generation status
*** p \ .001
J Youth Adolescence (2014) 43:2012–2027 2021
123
patterns of associations are not compelling. For example, the
Mexican American adolescents in the third and fourth tra-
jectory group were most often classified as ‘‘bicultural’’ by
the median split at the first assessment. However, the ado-
lescents in these two trajectory groups were declining sub-
stantially in their endorsement of the mainstream cultural
values as they got older. Those Mexican American adoles-
cents who were maintaining a relatively high endorsement of
Mexican American cultural values while increasingly
endorsing mainstream cultural values as they got older (i.e.,
trajectory group 1) were least likely to be classified as
‘‘bicultural’’ by the median split procedure at the first
assessment. Furthermore, the pattern of associations
between the trajectory groups and the median split classifi-
cations based upon the third assessment of culturally related
values were quite different from the associations observed
with the first assessment. For example, by the third assess-
ment the Mexican American adolescents in the fourth tra-
jectory groups were no longer very likely to be classified as
‘‘bicultural;’’ and the adolescents in the first trajectory group
were most likely to be classified as ‘‘bicultural.’’
Table 4 presents the cross-tabulation of the observed
median split classifications at the first and third assessment.
Although the associations are statistically significant
(v2 (9) = 56.47, p \ .001) the patterns of associations are not consistently compelling. Only 36.73 % of the Mexican
American adolescents were classified into the same dual
cultural adaptation group by the median split classification
at the first and third assessment. Furthermore, some of the
inconsistent classifications are difficult to explain. For
example, 22.47 % of those adolescents classified as
‘‘marginalized’’ at the first assessment were classified as
‘‘bicultural’’ at the third assessment.
Discussion
A number of authors have suggested that exposure to the
ethnic and mainstream cultures will lead to changes in a
variety of features of cultural orientation over time (e.g.,
Berry 2006; Cuellar et al. 1995; Felix-Ortiz et al. 1994;
Gonzales et al. 2002; LaFramboise et al. 1993; Tsai et al.
2002) that may create risk factors for, or protective factors
against, negative health and other life outcomes (e.g.,
Atzaba-Poria and Pike 2007; Schwartz et al. 2006;
Szapocznik and Kurtines 1993; Vega and Gil 1999). Hence
Table 3 The percentage of each trajectory group by a single-point-in-time median split categorization at the first and the third assessment
Group 1 (n = 241) (%) Group 2 (n = 253) (%) Group 3 (n = 107) (%) Group 4 (n = 148) (%) v2 (df)
Time 1 v2 (9) = 459.52***
Bicultural 14.10 7.91 91.59 60.81
Assimilated 18.68 23.72 1.87 33.78
Separated 38.59 15.02 5.61 3.38
Marginalized 28.36 53.36 0.93 2.03
(n = 206) (%) (n = 218) (%) (n = 82) (%) (n = 131) (%) v2 (9) = 345.61***
Time 3
Bicultural 52.43 7.80 56.10 10.69
Assimilated 7.28 38.99 0.00 35.11
Separated 35.44 6.42 43.90 15.27
Marginalized 4.85 46.79 0.00 38.93
*** p \ .001
Table 4 The percentage of Mexican American adolescents categorized by a median split at the first assessment compared to the median split categorization at the third assessment
Bicultural
(%)
Assimilated
(%)
Time 1 separated
(%)
Marginalized
(%)
v2 (df)
Time 3 v2 (9) = 56.47***
Bicultural 37.25 30.47 23.62 22.47
Assimilated 20.10 33.59 15.75 23.60
Separated 24.02 11.72 38.58 16.85
Marginalized 18.63 24.22 22.05 37.08
*** p \ .001
2022 J Youth Adolescence (2014) 43:2012–2027
123
the present study was designed to identify groups of
Mexican American youth demonstrating distinct trajecto-
ries of change in Mexican American and mainstream cul-
tural values across a 7 year developmental time frame. To
help gauge the validity of the resulting trajectory groups,
this study was also designed to examine the relationship
between these trajectory groups and several other indica-
tors of cultural orientation. Finally, to evaluate the utility of
the longitudinal approach to identify trajectories of change
in cultural values over time we examined the relationship
of these trajectory groups to the groups identified through
the median split of a single-point-in-time assessment.
The parallel process latent class longitudinal growth
curve modeling revealed four groups of individuals expe-
riencing different longitudinal trajectories of change in
Mexican American and mainstream cultural values.
Importantly, all four trajectory groups were initially quite
high in their endorsement of Mexican American cultural
values, indicating these value dimensions are broadly rel-
evant and prevalent in the families and communities of our
diverse sample of Mexican American youth. However,
these groups were distinguished by the extent to which
their levels of endorsement of Mexican American values
declined over time from late childhood to mid-adolescence,
and also based on changes over time in adolescents’
endorsement of mainstream cultural values. The adoles-
cents in the first two trajectory groups (Groups 1 and 2)
showed a similar pattern of moderate but increasing
endorsement of mainstream cultural values that distin-
guished them from two other groups that decreased in their
endorsement of mainstream cultural values over time
(Groups 3 and 4). However, whereas the first trajectory
group maintained their high endorsement of Mexican
American values, the second group showed significant
declines in Mexican American values over time. From a
person centered perspective, the first group also showed a
pattern of scores on culturally related variables that
reflected their strong connection to the ethnic culture with
increasing connection to the mainstream culture. The
adolescents in this group were high in ethnic pride, ethnic
identity, biculturalism, and English use; they used Spanish
moderately often and perceived relatively little ethnic
discrimination. In contrast, adolescents in the second tra-
jectory group were lower in ethnic pride, ethnic identity,
and biculturalism relative to the first trajectory group.
Adolescents in the third and fourth trajectory groups
showed a similar pattern of moderate but decreasing
endorsement of mainstream values over time that con-
trasted with the first two groups. However, contrasted
with each other, adolescents in the third group main-
tained higher levels of endorsement of Mexican Ameri-
can values over time compared to the fourth group. These
adolescents were relatively high in ethnic pride, ethnic
identity, biculturalism, and both Spanish and English use.
But, they also perceived a bit more discrimination than
the adolescents in the first two groups, potentially
explaining their relative decline in endorsement of
mainstream values over time. Adolescents in the fourth
trajectory group declined significantly over time from
initially high levels of Mexican American values and also
reported lower levels of ethnic pride, ethnic identity, and
Spanish language use relative to adolescents in the third
group. Furthermore the adolescents in these two trajec-
tory groups with declining mainstream values were
somewhat higher in ethnic pride and perceived discrim-
ination and used Spanish somewhat more often than the
first two groups with increasing endorsement of main-
stream values over time.
Although, these trajectory groups were associated with a
variety of culturally related variables in ways that support
the validity of these groups, differences between the tra-
jectory groups were very modest in size. These differences
were modest, in part, because the Mexican American and
mainstream cultural values and the culturally related vari-
ables used for validity comparisons were substantially
skewed. Hence, although the Mexican American cultural
values, the ethnic pride, ethnic identity, biculturalism, and
language use measures relied on a four- or five-point
response scale almost no participants utilized the lowest
response option and relatively few utilized the second
lowest response option. Interestingly, the perceived dis-
crimination scores were skewed positively with relatively
few participants reporting very high perceived ethnic
discrimination.
While the nature of the relationships between the
observed trajectory groups and the culturally related vari-
ables supports the validity of the trajectory findings, the
nature of these trajectory groups is not completely con-
sistent with the groups identified in the studies that have
assessed the dual cultural adaptations of individuals in
single-point-in-time assessments using median or mid-
scale splits (e.g., Berry 2007; Szapocznik et al. 1980). The
adolescents in each of the trajectory groups endorse Mex-
ican American cultural values at relatively high levels,
even though there are differences among the groups in the
maintenance of these values. Hence, it is difficult to char-
acterize any of these participants as ‘‘assimilated’’ or
‘‘marginalized.’’ The strong endorsement of Mexican
American cultural values might well be expected given the
relatively extensive Mexican American cultural experi-
ences associated with growing up in Mexican American
families. Although one might think that the adolescents
who are maintaining their Mexican American cultural
values and declining in their endorsement of the main-
stream cultural values (i.e., those in trajectory group 3)
could be characterized as ‘‘separated,’’ it is important to
J Youth Adolescence (2014) 43:2012–2027 2023
123
note that, although their endorsement of these values is
declining with age, at sixteen they are still at comparable
levels to those in the groups who are increasing in these
values. Similarly, although the adolescents who are stable
and high in Mexican American cultural values and
increasing in mainstream cultural values with age (i.e.,
those in trajectory group 1) may be conceptually close to
the definition of ‘‘bicultural’’, at 16 years of age these
adolescents are endorsing mainstream cultural values at
about the same level as the adolescents in trajectory group
3 who are similar in the maintenance of Mexican American
cultural values but declining in mainstream cultural values.
The distinctions between these four trajectory group
types of dual cultural adaptations are based primarily on
the adolescents’ longitudinal changes in their endorsement
of Mexican American and mainstream cultural values
rather than their absolute levels of endorsement of these
values. This is consistent with the perspective that ado-
lescence is a developmental period during which such
culturally related values are internalized at least in part
because this internalization may be dependent upon ele-
ments of ethnic identity development (e.g., Armenta et al.
2011). That is, before one adopts the values of an ethnic
group, there likely must be some affirmation of belonging
to that group and some exploration of the nature of that
group. Although this ethnic identity affirmation and ethnic
identity exploration may have begun at a younger age,
adolescence is the developmental period during which
these features of ethnic identity become particularly salient
(e.g., Phinney 1990; Umaña-Taylor et al. 2009).
That these four types of dual cultural adaptation are
primarily based on longitudinal changes also has important
methodological and analytical implication. A single-point-
in-time assessment could not detect these types of dual
cultural adaptations. Nor could a static analytical approach
(i.e., a median split, mid-scale split, or even a latent class
analysis) detect these types of adaptations. Clearly, longi-
tudinal assessments and person-centered analytical proce-
dures that examine individuals’ changes over time are
essential for examining the dual cultural adaptations of
psychological and behavioral variables that are undergoing
substantial developmental change or environmental impact
that produces change. There may be some variables for
which a single assessment and more static analytical
approach might reveal dual cultural adaptation types sim-
ilar to those identified in a more longitudinal approach. For
example, Knight et al. (2009) found two trajectories of
ethnic affiliation (i.e., one group that affiliated mostly with
other Mexican Americans and one group that affiliated
more equally with Mexican Americans and Anglo Ameri-
cans) that were relatively stable from 14 to 20 years of age.
Although a latent class analysis at any one assessment
would likely have detected these same two groups, a
median split or mid-scale split probably would not reveal
the same types of dual cultural adaptations. Furthermore, a
latent class analysis and single-point-in-time assessment
likely would not reveal the same two underlying types of
dual cultural adaptation if one’s sample included a sub-
stantial number of recent immigrants whose affiliation
patterns may be changing based upon their changing life
experiences. Given the development of analytical proce-
dures designed to identify longitudinal changes at the
individual level, and the identification of groups of indi-
viduals experiencing similar types of individual change, it
is time for the study of acculturation and enculturation as
processes of adaptation to a dual cultural existence to move
beyond the single point in time assessment and median or
mid-scale split approach.
Although the present developmental trajectory group
classifications are significantly related to the median split
classifications at both the first (i.e., fifth grade, approxi-
mately 11 years of age) and third assessment (i.e., tenth
grade, approximately 16 years of age), the nature of these
classification agreements suggests that these two different
types of classification procedures are identifying essen-
tially different clusters of Mexican American adolescents.
For example, the adolescents who are maintaining their
high endorsement of Mexican American cultural values as
they are increasingly endorsing mainstream cultural values
(i.e., those in trajectory group 1) are least likely to be
classified as bicultural at the first assessment using the
median split procedure. In contrast, these same adolescents
are most likely to be classified as bicultural at the third
assessment using the median split procedure. Similarly, the
adolescents whose endorsement of both the Mexican
American and mainstream cultural values are declining
substantially over time (i.e., those in trajectory group 4) are
most likely to be classified as bicultural by the median split
procedure at the first assessment, but most likely to be
classified as marginalized or assimilated by the median
split procedure at the third assessment. Hence, these two
procedures for identifying the type of dual cultural adap-
tation these Mexican American adolescents are experi-
encing are not consistent. Furthermore, although there is
significant agreement between the medial split classifica-
tions at the first and third assessments, this association is
weak, and over 63 % of the Mexican American adolescents
were in different median split categories at the two
assessments. Approximately one of every five adolescents
classified as bicultural at one of the two assessments were
classified as marginalized at the other assessment even
though these are the theoretically most distinct types of
dual cultural adaptation.
The implications are clear. In addition to the problems
associated with the use of the median split procedure with
arbitrary cut-points (e.g., DeCosta et al. 2009; MacCallum
2024 J Youth Adolescence (2014) 43:2012–2027
123
et al. 2002), this procedure cannot be used to evaluate the
nature of the dual cultural adaptation of ethnic minority
individuals on indices of cultural adaptation that are
changing over time due to some normative developmental
processes or changing experiential processes. One may
suspect that the present limitations of the median-split
findings are a sample/variable specific problem associated
with the skewedness of the scores on the culturally related
variables. However, such a skewed distribution of scores
has been shown to be relatively common among measure
of culturally related constructs in measures of accultura-
tion, ethnic identity, ethnic pride, ethnically associated
stress, and perceived discrimination (e.g., Coatsworth et al.
2005; Knight et al. 2009). Hence, the median split proce-
dure is not optimal for examining the acculturation and
enculturation of ethnic/mainstream identities, the internal-
ization of culturally related values, or the formation of
relatively complex attitudes and behavior patterns among
ethnic minority youths at certain ages because these fea-
tures of cultural orientation undergo developmental chan-
ges across a relatively broad age-span. This is not to say
that the more traditional median-split procedure is always
inappropriate. The median split procedure may be useful
for characterizing the acculturation/enculturation outcomes
of relatively simple behaviors that are generally stable after
a relatively young age and that are relatively normally
distributed. For example, there is relatively little change in
language use (i.e., Spanish and English use) patterns
among Mexican American youth after early to mid-ado-
lescence (Knight et al. 2009). Hence, the median-type split
procedure could be useful for characterizing the dual cul-
tural adaptation, indexed by language use, of Mexican
American adults, assuming one uses a more nuanced and
valid cut-point (e.g., Coatsworth et al. 2005) rather than the
arbitrary and sample specific median. Even here, though,
this type of classification could be compromised if the
sample were to include a substantial number of recent
immigrant adults who are in the process of improving their
English capabilities and increasingly being exposed to
situations in which they use the English language.
In summary, the present study identified four groups of
Mexican American adolescents that are demonstrating
distinct patterns of developmental changes in their
endorsement of Mexican American and mainstream cul-
tural values that reflect the adaptations associated with
growing up in a dual cultural context. All of these ado-
lescents were relatively high in their endorsement of
Mexican American cultural values, and even those ado-
lescents in the two trajectory groups that declined sub-
stantially in Mexican American cultural values remained
relatively high in their endorsement of these values. All of
these adolescents were more moderate in their endorsement
of mainstream cultural values and some demonstrated
modest increases with age while others demonstrated
substantial decreases with age. Hence, these trajectory
groups do not conform to those characterized by the
common four-fold typology of dual cultural adaptations
based upon median splits of single-point-in-time assess-
ments (e.g., Berry 2007; Szapcoznik et al. 1980). Further,
the adolescents in these trajectory groups differ in a num-
ber of culturally related variables (i.e., generation of
immigration, English and Spanish language use, ethnic
pride, ethnic identity, perceived discrimination, and bicul-
turalism) in ways that make conceptual sense. Hence, the
present study demonstrates the utility of modern latent
class longitudinal growth modeling as a means of assessing
the outcomes associated with the processes of acculturation
and enculturation for psychological constructs that may be
undergoing developmental changes (e.g., the internaliza-
tion of values) or changes associated with life experience
experiences (e.g., immigration). Although this method of
characterizing the process of change in cultural orientation
during a critical stage of development may be important for
advancing the field, future research is needed to evaluate
how these changes relate to other aspects of development
(i.e., academic success, mental and physical health, and
well-being, etc.) that are important for advancing theory
and culturally relevant services and policy.
Acknowledgments This research was supported, in part, by NIMH grant MH68920 (Culture, Context, and Mexican American Mental
Health). The authors are thankful for the support of Mark W. Roosa,
Jenn-Yun Tein, Marisela Torres, Jaimee Virgo, our Community
Advisory Board and interviewers, and the families who participated in
the study.
Author contributions G.P.K. participated in the conceptualization of the study and the data collection process, guided the analyses and
interpretation of the data, and took primary responsibility for the
preparation of the manuscript reporting the study; C.D.B. participated
in the conceptualization of the study and contributed in the preparation
and revision of the manuscript; H.C. conducted part of the statistical
analysis, drafted corresponding parts of the results section of the
manuscript, and made comments and contributions throughout drafts of
the manuscript; N.A.G. participated in the conceptualization of the
study and the data collection process, and contributed in the preparation
and revision of the manuscript; Y.L. conducted part of the statistical
analysis, drafted corresponding parts of the manuscript, and made
comments and contributions throughout drafts of the manuscript;
A.U.T. participated in the conceptualization of the study and interpre-
tation of the data, and also reviewed and provided feedback on the
manuscript. All authors read and approved the final manuscript.
References
Altschul, I., Oyserman, D., & Bybee, D. (2006). Racial-ethnic identity
in mid-adolescence: Content and change as predictors of
academic achievement. Child Development, 77, 1155–1169.
Arends-Tóth, J., & van de Vijver, F. J. R. (2006). Assessment of
psychological acculturation. In S. David & J. W. Berry (Eds.),
J Youth Adolescence (2014) 43:2012–2027 2025
123
The Cambridge handbook of acculturation psychology (pp.
142–160). New York, NY: Cambridge University Press.
Armenta, B. E., Knight, G. P., Carlo, G., & Jacobson, R. P. (2011).
The relation between ethnic group attachment and prosocial
tendencies: The mediating role of cultural values. European
Journal of Social Psychology, 41, 107–115.
Atzaba-Poria, N., & Pike, A. (2007). Are ethnic minority adolescents
at risk for problem behavior? Acculturation and intergenera-
tional acculturation discrepancies in early adolescence. British
Journal of Developmental Psychology, 25, 527–541.
Basilio, C., Knight, G. P., O’Donnell, M., Roosa, M., Gonzales, N. A.,
Umaña-Taylor, A. J., & Torres, M. (2013). The Mexican
American Biculturalism Scale: Bicultural comfort, ease, and
advantage for adolescents and adults. Manuscript under review.
Berry, J. W. (2006). Acculturation: A conceptual overview. In M.
H. Bornstein & L. R. Cote (Eds.), Acculturation and parent–
child relationships (pp. 13–32). Mahwah, NJ: Lawrence Erlbaum
Associates, Publishers.
Berry, J. W. (2007). Acculturation strategies and adaptation. In
J. E. Lansford, K. Deater-Deckard, & M. H. Bornstein (Eds.),
Immigrant families in contemporary society. Duke series in child
development and public policy (pp. 69–82). New York, NY:
Guilford Press.
Brown, B. B. (1990). Peer groups and peer cultures. In S. S. Feldman
& G. R. Elliott (Eds.), At the threshold: The developing
adolescent (pp. 171–196). Cambridge, MA: Harvard University
Press.
Coatsworth, J. D., Maldonado-Molina, M., Pantin, H., & Szapocznik,
J. (2005). A person-centered and ecological investigation of
acculturation strategies in Hispanic immigrant youth. Journal of
Community Psychology, 33, 157–174.
Cuellar, I., Arnold, B., & Maldonado, R. (1995). Acculturation rating
scale for Mexican Americans-II: A revision of the original
ARSMA scale. Hispanic Journal of Behavioral Sciences, 17,
275–304.
Daniel, E., Schiefer, D., Möllering, A., Benish-Weisman, M.,
Boehnke, K., & Knafo, A. (2012). Value differentiation in
adolescence: The role of age and culture. Child Development, 83,
322–336.
DeCosta, J., Iselin, A.-M. R., & Gallucci, M. (2009). A conceptual
and empirical examination of justifications for dichotomization.
Psychological Methods, 14, 349–366.
Eisenberg, N., Miller, P. A., Shell, R., McNalley, S., & Shea, C.
(1991). Prosocial development in adolescence: A longitudinal
study. Developmental Psychology, 27, 849–857.
Enders, C. K., & Tofighi, D. (2008). The impact of misspecifying
class-specific residual variances in growth mixture models.
Structural Equation Modeling, 15, 75–95.
Erikson, E. H. (1968). Identity: Youth, and crisis. New York: Norton.
Felix-Ortiz, M., Newcomb, M. D., & Myers, H. (1994). A multidi-
mensional scale of cultural identity for Latino and Latina
adolescents. Hispanic Journal of Behavioral Sciences, 16,
99–115.
French, S. E., Seidman, E., Allen, L., & Aber, J. L. (2006). The
development of ethnic identity during adolescence. Develop-
mental Psychology, 42, 1–10.
Gonzales, N. A., Knight, G. P., Morgan-Lopez, A., Saenz, D., &
Sirolli, A. A. (2002). Acculturation and the mental health of
Latino youths: An integration and critique of the literature. In J.
M. Contreras, K. A. Kerns, & A. M. Neal-Barnett (Eds.), Latino
children and families in the United States: Current research and
future directions (pp. 45–74). Westport, CT: Praeger.
Grimm, K. J., Ram, N., & Hamagami, F. (2011). Nonlinear growth
curves in developmental research. Child Development, 82,
1357–1371.
Hughes, D., & Dodge, M. A. (1997). African American women in the
workplace: Relationships between job conditions, racial bias at
work, and perceived job quality. American Journal of Commu-
nity Psychology, 25, 581–599.
Kiang, L., Witkow, M. R., Baldelomar, O. A., & Fuligni, A. J. (2010).
Changes in ethnic identity across the high school years among
adolescents with Latin American, Asian, and European back-
grounds. Journal of Youth and Adolescence, 39, 683–693. Klonoff, E. A., & Landrine, H. (1995). The schedule of sexist events:
A measure of lifetime and recent sexist discrimination in
women’s lives. Psychology of Women Quarterly, 19, 439–472.
Knight, G. P., Berkel, C., Umaña-Taylor, A. J., Gonzales, N. A.,
Ettkal, I., Jaconis, M., et al. (2011). The familial socialization of
culturally related values in Mexican American families. Journal
of Marriage and Family, 73, 913–925.
Knight, G. P., Gonzales, N. A., Saenz, D. S., Bonds, D., Germán, M.,
Deardorff, J., et al. (2010). The Mexican American Cultural
Values Scale for adolescents and adults. Journal of Early
Adolescence, 30, 444–481.
Knight, G. P., Losoya, S. H., Cho, Y. I., Chassin, L., Williams, J. L.,
& Cota-Robles, S. (2012). Ethnic identity and offending
trajectories among Mexican American juvenile offenders: Gang
membership and psychosocial maturity. Journal of Research on
Adolescence, 22, 782–796.
Knight, G. P., Vargas-Chanes, D., Losoya, S. H., Cota-Robles, S.,
Chassin, L., & Lee, J. M. (2009). Acculturation and encultura-
tion trajectories among Mexican American adolescent offenders.
Journal of Research on Adolescence, 19, 625–653.
LaFramboise, T., Coleman, H. L. K., & Gerton, J. (1993). Psycho-
logical impact of biculturalism: Evidence and theory. Psycho-
logical Bulletin, 114, 395–412.
Landrine, H., & Klonoff, E. A. (1996). The schedule of racist events:
A measure of racial discrimination and a study of its negative
physical and mental health consequences. Journal of Black
Psychology, 22, 144–168.
MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D.
(2002). On the practice of dichotomization of quantitative
variables. Psychological Methods, 7, 19–40.
Muthén, B. (2004). Latent variable analysis: Growth mixture
modeling and related techniques for longitudinal data. In D.
Kaplan (Ed.), Handbook of quantitative methodology for the
social sciences (pp. 345–368). Newbury Park, CA: Sage
Publications.
Muthén, B. O., & Muthén, L. K. (2010). Mplus (Version 6.00)
[Computer software]. Los Angeles, CA: Muthén & Muthén.
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on
the number of classes in latent class analysis and growth mixture
modeling: A Monte Carlo simulation study. Structural Equation
Modeling, 14, 535–569.
Pahl, K., & Way, N. (2006). Longitudinal trajectories of ethnic
identity among urban Black and Latino adolescents. Child
Development, 77, 1403–1415.
Phinney, J. S. (1990). Ethnic identity in adolescents and adults:
Review of research. Psychological Bulletin, 108, 499–514.
Rogler, L. H., Cortes, D. E., & Malgady, R. G. (1991). Acculturation
and mental health among Hispanics. American Psychologist, 46,
585–597.
Roosa, M. W., Liu, F., Torres, M., Gonzales, N. A., Knight, G. P., &
Saenz, D. S. (2008). Sampling and recruitment in studies of
cultural influences on adjustment: A case study with Mexican
Americans. Journal of Family Psychology, 22, 293–302.
Samaniego, R. Y., & Gonzalez, N. A. (1999). Multiple mediators of
the effects of acculturation status on delinquency for Mexican
American adolescents. American Journal of Community Psy-
chology, 27, 189–210.
2026 J Youth Adolescence (2014) 43:2012–2027
123
Schwartz, S. H. (1999). A theory of cultural values and some
implications for work. Applied Psychology, 48, 23–47.
Schwartz, S. J., Des Rosier, S., Huang, S., Zamboanga, B. L., Unger,
J. B., Knight, G. P., et al. (2013). Developmental trajectories of
acculturation in Hispanic adolescents: Associations with family
functioning and adolescent risk behavior. Child Development,
84, 1355–1372.
Schwartz, S. J., Montgomery, M. J., & Briones, E. (2006). The role of
identity in acculturation among immigrant people: Theoretical
propositions, empirical questions, and applied recommendations.
Human Development, 49, 1–30.
Syed, M., & Azmitia, M. (2009). Longitudinal trajectories of ethnic
identity during the college years. Journal of Research on
Adolescence, 19, 601–624.
Szapocznik, J., & Kurtines, W. M. (1993). Family psychology and
cultural diversity: Opportunities for theory, research and appli-
cation. American Psychologist, 48, 400–407.
Szapocznik, J., Kurtines, W., & Fernandez, T. (1980). Biculturalism
and adjustment among Hispanic youths. International Journal of
Intercultural Relations, 4, 353–375.
Tadmor, C. T., & Tetlock, P. E. (2006). Biculturalism: A model of the
effects of second-culture exposure on acculturation and integra-
tive complexity. Journal of Cross-Cultural Psychology, 37,
173–190.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of
intergroup behavior. In S. Worchel & W. G. Austin (Eds.),
Psychology of intergroup relations (pp. 7–24). Chicago: Nelson-
Hall.
Thayer, S. M., Valiente, C., Hageman, D., Delgado, M., & Updegraff,
K. A. (2002). Conflict resolution in friendships: A comparison of
Mexican American and Anglo American adolescents. Paper
presented at the poster presented at National Council of Family
Relations, Houston, TX.
Tsai, J. L., Chentsova-Dutton, Y., & Wong, Y. (2002). Why and how
researchers should study ethnic identity, acculturation, and
cultural orientation. In G. C. Ngayama Hall & S. Okazaki
(Eds.), Asian American psychology: The science of lives in
context (pp. 41–65). Washington, DC: American Psychological
Association.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell,
M. S. (1987). Rediscovering the social group: A self-categori-
zation theory. Oxford, UK: Blackwell.
Umaña-Taylor, A. J., Gonzales-Backen, M. A., & Guimond, A. B.
(2009). Latino adolescents’ ethnic identity: Is there a develop-
mental progression and does growth in ethnic identity predict
growth in self-esteem? Child Development, 80, 391–405.
Umaña-Taylor, A. J., Yazedjian, A., & Bámaca-Gómez, M. (2004).
Developing the ethnic identity scale using Eriksonian and social
identity perspectives. Identity: An International Journal of
Theory and Research, 4, 9–38.
United States Census Bureau. (2012). The foreign-born population in
the United States: 2010. Retrieved May 24, 2012, from http://
www.census.gov/prod/2012pubs/acs-19.pdf.
Vega, W. A., & Gil, A. G. (1999). A model for explaining drug use
behavior among Hispanic adolescents. Drugs and Society, 14,
55–73.
Author Biographies
George P. Knight is a Professor at Arizona State University in the Department of Psychology. He is currents serving as an Associate
Editor for Developmental Psychology. He received his Ph.D. in
Psychology from the University of California—Riverside. His
research interests focus on the acculturation and enculturation in
Mexican American families, ethnic identity, methods for studying
ethnic minority families, and prosocial development.
Camille D. Basilio is a doctoral student in social psychology in the Department of Psychology at Arizona State University. Her major
research interests include biculturalism and ethnic identity and their
psychological and physical health implications.
Heining Cham is a doctoral student in quantitative psychology in the Department of Psychology at Arizona State University. His major
research interests include latent variable interactions modeling and
longitudinal analysis.
Nancy A. Gonzales is Foundation Professor of Psychology and Director of the Prevention Research Center at Arizona State
University. She received her Ph.D. in Psychology from the University
of Washington. She studies social, environmental, and cultural
influences on health disparities and the translation of these research
findings into evidence-based interventions for low-income communi-
ties and families.
Yu Liu is a doctoral student in quantitative psychology in the Department of Psychology at Arizona State University. Her major
research interests include longitudinal data analysis, missing data
analysis, measurement invariance across groups and across time,
Bayesian methods, and the application of these methods in empirical
studies.
Adriana J. Umaña-Taylor is a Professor at Arizona State University in the T. Denny Sanford School of Social and Family Dynamics. She
received her Ph.D. in Human Development and Family Studies from
the University of Missouri-Columbia. Dr. Umaña-Taylor’s research
interests focus broadly on Latino youth and families and, more
specifically, on ethnic identity formation, familial socialization
processes, culturally informed risk and protective factors, and
psychosocial functioning among Latino youth.
J Youth Adolescence (2014) 43:2012–2027 2027
123
Copyright of Journal of Youth & Adolescence is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
- Trajectories of Mexican American and Mainstream Cultural Values Among Mexican American Adolescents
- Abstract
- Introduction
- The Current Study
- Method
- Participants
- Procedure
- Measures
- Generation
- Mexican American and Mainstream Cultural Values
- Language Use
- Ethnic Pride
- Ethnic Identity
- Perceived Ethnic Discrimination
- Biculturalism
- Results
- Data Analysis Plan
- Parallel Latent Class Growth Models for Mexican American and Mainstream Values
- Relation of Trajectory Groups to Cultural Variables
- Relation among the Trajectory Groups and the Median Split Classifications
- Discussion
- Acknowledgments
- References
