>~1wu...uz..x.

23...: , .4 .
1 1: z .t. 1.: A: I 3.: .1
lxrraf. 5...! £1. . , frfrulxax

1.1!)..5 afltv 7:5)...
5.1.311... LflxlD‘itx ‘11.).

. .a 1..
y VOL!!!
£441.15 .

1.3.1.1: .1...
7...? Kilt,

. a .
Ln...r4.....
ii

XafL Oil 1
.lltkhui: 5:1...

I.
1.35... x.
29.5., S .u
”Luna”
y -........as..€.£.
\ it.

iv...)
5.! .xincn 5... 5.9..

.xi ‘IL‘IIvvllxtisilxhu .I
l . . .! ,1 It . .

 

 

 

 

i. a 1 ‘ .
.vl..)|..n

 

333$. “wagfiafimfifimfigfigfi, _ . .. ., . . _ .13». .

 

SlTY L BRAR

l in Hill

itil‘iiilllfiilljls

3129

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

AUTONOMY AND CONNECTEDNESS
IN ASIAN-AMERICAN UNIVERSITY STUDENTS

presented by

Linda P. Juang

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Psychologx

 

‘ 7/) "7 f4?
Date ] {14/ 12;"; / //7
U

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

_ _,.'.- H -———“- “A 4—

   
   
    

  

LIBRARY
Mlc’llgan State
University

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE
JAN-H4000 1W3
OfiAQZ 31 a) 2003

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
czblmmpma-M

AUTONOMY AND CONNECTEDNESS 1N
ASIAN-AIVIERICAN UNIVERSITY STUDENTS

By

Linda P. Juang

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1997

ABSTRACT

AUTONOMY AND CONNECTEDNESS
IN ASIAN-AMERICAN UNIVERSITY STUDENTS

By

Linda P. Juang

The purpose of this study was to examine issues of autonomy and connectedness in Asian-
American university students. The “goodness of fit” model and the concept of
acculturation were used to guide the study. All 99 participants were of Asian descent,
with parents originating from either China, Japan, the Phillippines, Thailand, Korea, or
Vietnam. The majority (87.9%) were second generation. The mean age was 19.8 years
and the range was 17 - 24 years. The results indicate that late adolescents who strongly
endorsed Asian attitudes and values reported less behavioral and emotional autonomy
compared to those less Asian. Late adolescents who experienced either low or high
autonomy, and, concurrently remained connected to their parents, fared better than those
not connected to their parents. In addition, late adolescents who experienced a good fit
with parental expectations of autonomy (i.e, the desires of the late adolescent were
congruent with the expectations of the parents regarding the timetable of autonomy
behaviors) reported higher levels of self-esteem and emotional closeness to their parents,
and lower levels of depression, behavior misconduct, and insecurity with parents,
compared to their poor fitting counterparts. Furthermore, late adolescents who better fit
into the cultural context (of this Midwestem university town) reported lower levels of
depression and higher levels of self-esteem than those who poorly fit. It is proposed that

the “goodness of fit” model and the concept of acculturation are useful in the investigation

of autonomy development and subsequent adjustment in Asian-American late adolescents.

ACKNOWLEDGMENTS

I would like to gratefiilly acknowledge the people that have helped me throughout
the past five years of graduate school. To Jacqueline Lerner, an excellent role model of a
dedicated mother and researcher. Thank you for being such a warm, supporting advisor.
To John McKinney, for his patience, critiques, and guidance in the writing of this
dissertation. You were the professor who initially helped me turn my personal interests
into a research possibility. To Alexander von Eye, whom I deeply respect for his
seemingly infinite wisdom in methodology, statistics, and good food. To Lillian Phenice
and Esther Onaga for their helpfiil suggestions and comments on my dissertation.

To Maggie Chen and the Minority Student Aides, my connections to the Asian
student population. Your willingness to help me out is greatly appreciated. To Jody
Reimer and Michael Tuma, valuable research assistants who helped in the coding and
entering of data.

To Domini Castellino, Nancy Hill, and Alex Loukas, for being there to lean on as
we helped each other through classes, comps, and dissertations. I will always remember
the times we spent studying, talking, and laughing together.

To Jennifer Juras, Ferit Kivanc, David Waldschmidt, Brad White, and Erik
Tryggestad, my good fiiends for a lifetime. Thank you for making graduate school fun.
To Harald Harte], for being a such a patient listener and for understanding me so well.

iv

You have always encouraged me when I needed it most. To Huong Nguyen, my dear
fi'iend, roommate, and other sister. I have learned so much about the important things in
life fiom you.

Finally, I am especially grateful to my mom and dad and Margaret, for calling, e-
mailing, and visiting me. You have brought me encouragement and support fiom the very

beginning of this journey all the way to the very end. Thank you.

TABLE OF CONTENTS

LIST OF TABLES ................................................... ix

LIST OF FIGURES .................................................... x
Chapter 1

INTRODUCTION .................................................... 1
Chapter 2

LITERATURE REVIEW ............................................... 3

Definition of Autonomy ........................................... 3

The Changing Perspectives on Autonomy ............................. 5

Autonomy and Connectedness ...................................... 6

American and Asian Cultures Contrasted .............................. 9

Collectivism versus individualism .............................. 9

Confucianism and filial piety ................................. 10

Degree and Timing of Autonomy ................................... 11

Adolescent-Parent Conflict Regarding Autonomy ....................... 12

Goodness of Fit ................................................ 12

Acculturation .................................................. 14

Definition of acculturation .................................. 15

Four acculturation styles .................................... 15

Acculturation styles and the concept of goodness of fit ............. 16

Acculturation and the timetable of behavioral autonomy ............ 18

Outcomes related to acculturation styles ........................ 18

Summary of Literature Review ..................................... 19
Chapter 3

THE PRESENT STUDY .............................................. 21

Definition of Major Variables ...................................... 22

Research Questions and Specific Hypotheses .......................... 24

Chapter 4

METHOD .......................................................... 26
Participants ................................................... 26
Procedure .................................................... 27
Consent ...................... -. ............................... 28
Measures ..................................................... 28

Autonomy .............................................. 28
Goodness of fit with parents ................................. 29
Acculturation ............................................ 30
Adolescent adjustment ..................................... 31
Pilot Study .................................................... 33
Missing Data .................................................. 33

Chapter 5

RESULTS .......................................................... 34
A General Overview ............................................. 34

The firll model ........................................... 34

A comparison of the full model with submodels 1 and 2 ............ 36

An alternate model ........................................ 42
Acculturation and Autonomy ...................................... 42
Emotional Autonomy, Connectedness, and Fit in Parental Expectations ...... 45
Autonomy, Connectedness, and Adjustment ........................... 46
Fit in Parental Expectations and Adjustment ........................... 54
Fit in Parental Expectations and Timetable of Autonomy ................. 60
Acculturation and Adjustment ..................................... 64
Additional Findings ............................................. 67

Chapter 6

DISCUSSION ....................................................... 69
Acculturation and the Timetable of Autonomy ......................... 69
Autonomy, Connectedness and Adjustment ........................... 69
Fit in Autonomy Expectations and Adjustment ......................... 73
Acculturation and Adjustment ..................................... 75
Overall Model ................................................. 78
Additional Findings ............................................. 79
Limitations .................................................... 81
Implications for Future Research ................................... 82

Chapter 7

CONCLUSION ...................................................... 86

vii

APPENDIX A CONSENT FOR PARTICIPATION FORM ................... 88

APPENDIX B: COVARIANCE MATRIX OF FULL LISREL MODEL .......... 89
APPENDIX C: CORRELATED ERROR TERMS FOR FULL LISREL MODEL . . . 91
APPENDIX D: PARAMETER ESTIMATES FOR FULL LISREL MODEL ....... 93
APPENDIX E: COVARIANCE MATRIX OF LISREL SUBMODEL 1 .......... 94
APPENDIX F: COVARIANCE MATRIX FOR LISREL SUBMODEL 2 ......... 95
APPENDIX G: PARAMETER ESTIMATES FOR LISREL SUBMODEL 1 ....... 96
APPENDIX H: PARAMETER ESTIMATES FOR LISREL SUBMODEL 2 ....... 97

LIST OF REFERENCES .............................................. 98

viii

Table 1.
Table 2.
Table 3.
Table 4.

Table 5.
Table 6.

Table 7.
Table 8.
Table 9.
Table 10.
Table 11.
Table 12.
Table 13.

Table 14.
Table 15.

Table 16.

LIST OF TABLES

Description of Study Variables .................................. 35
LISREL Goodness of Fit Indices ................................. 37
Correlations for Indicators of Fit with Emotional Autonomy and
Connectedness .............................................. 45
MANOVA Results for Autonomy/Connectedness Groups with

Adjustment ................................................ 48
AN OVA Results for Autonomy/Connected Groups and Adjustment ...... 49
Means and Standard Deviations for Autonomy/Connected Groups and
Adjustment ................................................ 49
Hierarchical Regression of Emotional Autonomy, Connectedness, and

Their Interaction onto Insecurity with Parents. ...................... 51
Hierarchical Regression of Emotional Autonomy, Connectedness

and Local Interaction Vectors onto Insecurity with Parents ............. 53
Multiple Regression of the Interaction Vectors of Emotional Autonomy

and Connectedness onto Insecurity with Parents ..................... 55
MANOVA Results for Autonomy Timetable and Fit in Autonomy

Timetable with Adjustment ..................................... 57
Means and Standard Deviations for Significant Adjustment Variables and
Timetable Fit Groups ......................................... 59
MANOVA Results for How Decisions are Made in the Family with

Adjustment ................................................. 61
MANOVA Results for How Often Decisions are Made in the Family and
Adjustment ................................................. 62
Correlations between Allowed Autonomy and Indicators of Fit .......... 63
Multiple Regression of Emotional Autonomy and Connectedness

Predicting Level of Fit ........................................ 65
Correlations between Acculturation and Adjustment .................. 66

ix

LIST OF FIGURES

Figure 1. Berry’s (1980) Four Acculturation Styles ........................... 17
Figure 2. t-values for Parameters of the Full Autonomy Model .................. 38
Figure 3. t-values for Parameters of Submodel 1 ............................. 40
Figure 4. t-values for Parameters of Submodel 2 ................ . ............ 41
Figure 5. An Alternate Model .......................................... 43
Figure 6. Contrasts for Testing Local Interactions for Insecurity with

Parents Variable .............................................. 52
Figure 7. Graph of the Emotional Autonomy and Connectedness Interaction

with Insecurity with parents ..................................... 56

Chapter 1

INTRODUCTION

Throughout the lifespan, individuals deal with the issue of autonomy (Baltes &
Silverberg, 1994; Erikson, 1963; Steinberg, 1990). Autonomy is especially salient during
adolescence as expectations and opportunities for independence broaden at this time. The
successfirl development of a healthy sense of autonomy, while simultaneously remaining
connected to one's- parents, is a major task to negotiate.

The development of autonomy is one potential area of disagreement for
adolescents whose parents are immigrants. Sue and Sue (1990) state that adolescents of
immigrant parents are likely to experience conflict due to the stressfulness of attempting to
“balance/reconcile traditional values with nontraditional values which, at certain times, are
at odds with one another". Parents immigrating to a new culture must deal with the reality
that while some of their values will be passed on to their children, their children will also
acquire values in the new culture that may differ from their traditional values. This

discrepancy in cultural values may create a context where values clash.l

 

1For instance, Sue and Sue (1990) describe in one of their case studies a 20-year
old Asian-American male who was experiencing severe headaches and bodily complaints.
He sought counseling when there seemed to be no medical, organic
reasons for his symptoms. It became apparent he was upset that his parents expected him
to graduate quickly and find a good job, and thereafter assume financial responsibility for

1

2

Nevertheless, not all adolescents of immigrant parents undergo such struggles.
This study seeks to understand the meaning and implications to the variations in Asian
immigrant families in the United States regarding the development of late adolescent

autonomy.

 

his younger brothers and sisters. He viewed this as a burden that was becoming
overwhelming. It was a conflict of independence between the young adult and his parents
where each held differing values, or beliefs on the appropriate level of autonomy regarding
decisions on career and family responsibilities. In other words, as a result of living in the
United States this young adult acquired values regarding independence that were
incongruent to his parents' values.

Chapter 2

LITERATURE REVIEW

The review of literature begins by defining autonomy and describing how the
conceptualization of autonomy has changed over time. Then, literature on the two
concepts that are used to guide the study - goodness of fit and acculturation - will be
presented.

11 fl . . E I

In order to clarify the process of the development of autonomy, one must first
understand the meaning of this construct. In reviewing the literature, it is clear that this is
not a simple task. Not only is there an assortment of labels describing this construct, there
is also an assortment of definitions. Leaper (1989) has compiled a list entitled "The Yin
and Yang of Psychosocial Development" that documents the variety of labels formulated
over seven decades regarding autonomy (the individual domain) and connectedness (the
interpersonal domain). To give a few examples, Loevinger (1976) conceptualized these as
individuality and mutuality, Grotevant and Cooper (1986) as individuality and
connectedness, Hill and Holmbeck (1986) as autonomy and attachment, and Selrnan
(1981) as autonomy and intimacy.

In addition to the many different labels for autonomy, there exist many definitions.

4
There is not a consensus among researchers regarding an exact definition. Different
investigators have focused on different aspects of autonomy depending on their area of
interest. However, one point of agreement is that autonomy is a multidimensional
construct. Thus, the various conceptualizations of autonomy are related to each other on
some dimensions, yet difl‘er on others.

Steinberg and Silverberg (1986) have defined autonomy as encompassing: 1) the
relationship with the parent, or emotional autonomy, 2) resistance to peer pressure, and 3)
a subjective sense of self-reliance. Other researchers have defined autonomy as self-
govemance and self-regulation (Ryan & Lynch, 1989), as psychological separation
(Homnan, 1984), as individuation, where individuation is defined as becoming distinct
within the relational context in which s/he is embedded (Karpel, 1976), and also as a
relationship property comprised of a balance between individuality (self-assertion and
separateness) and connectedness (permeability and mutuality) (Cooper, Grotevant &
Condon, 1983). To summarize, autonomy has been defined by researchers as self-
reliance, as self-reliance in the context of a relationship, and as self-reliance and
connectedness in the context of a relationship.

Autonomy development involves two domains: autonomy with respect to parents,
and autonomy with respect to peers. Autonomy with respect to peers can be defined as
resistance to peer pressure. Autonomy with respect to parents can be divided into three
parts: 1) emotional autonomy from parents, 2) behavioral autonomy from parents, and 3)
connectedness to parents. This review will focus on the three aspects of autonomy

development in relation to parents.

1] Cl . E . !

Steinberg (1990) has traced the evolution of various perspectives on the
development of autonomy in adolescence. Researchers' views of autonomy have evolved
fi'om focusing exclusively on autonomy from parents to one that emphasizes both
autonomy and connectedness in relation to parents. The first perspective is the
psychoanalytic view which was mainly developed through the ideas of Anna Freud. Her
ideas led researchers to describe and understand the detachment process of adolescents
from their parents. In this view, autonomy held a negative connotation. The detachment
process was characterized by the very familiar notion of " storm and stress" during
adolescence. However, there is much evidence that harmony, not dissonance, is by far the
norm of family life during this period (Douvan & Adelson, 1966; Montemayor, 1983).

The second perspective that Steinberg describes is the "neoanalytic view." For
example, Blos (1979) views adolescent autonomy as a process of individuation. Unlike
the psychoanalytic view which emphasizes detachment in a negative light, autonomy, or
individuation, carries with it a positive connotation where its development is a desired
achievement. However, it can only be achieved by breaking away from parents and
severing the now inappropriate infantile ties. In Blos' perspective there is no emphasis on
connectedness. This leads to what Steinberg holds is the third, revised perspective that
emerged in the mid-70$.

This final perspective acknowledges that major transformations in family relations
occur during the adolescent years, however, "it challenges the view that these

realignments necessarily occur against a backdrop of...emotional detachmen " (Steinberg,

6
1990). This view does not support the notion of storm and stress during adolescence, or
complete detachment fi'om parents, is normative. Instead, there are several different
emphases that Steinberg clarifies. First, that most adolescents develop responsible
autonomy without severing emotional bonds to their parents. Second, that demographic
and individual characteristics influence the realignment of the parent-adolescent
relationship. The third emphasis is that other contextual factors, such as the psychological
development of the parents, must be taken into account in order to better understand the
changes in family relations. The fourth emphasis is that the family is a system of
interrelationships, with multiple and reciprocal directions of influence. These four
emphases depict a developmental contextualist perspective on autonomy.

In summary, the view of autonomy has gradually shified over time as researchers
attempt to unravel this phenomena. Originally, autonomy from parents was perceived as a
detaching, negative, stressful process during adolescence. The concept then changed so
that autonomy was viewed as a positive process that also involved connection. Finally,
this view has evolved to recognize that autonomy from and connection to parents during
adolescence can be either a positive or negative process depending on various contextual
factors in which the adolescent develops.

Aumncmundficrmemdness

Various studies address some, but not all, dimensions of autonomy that
adolescents experience with their parents. Most have emphasized the detachment, or self-
reliance aspect, while ignoring the connectedness dimension of autonomy. The following

studies lend support to the notion that connectedness is an important component in

understanding autonomy in adolescence.

Lambom and Steinberg (1993) examined emotional autonomy in the context of a
supportive adolescent-parent relationship. They identified four categories of adolescents:
individuated, those with a high degree of emotional attachment to their parents and also a
high degree of support from their parents; detached, those with high emotional attachment
and low support; connected, those with low emotional attachment and high support; and
ambivalent, those with low emotional attachment and low parental support. The results
revealed that individuated, compared to detached, connected, or ambivalent adolescents,
experienced the best outcomes regarding psychosocial maturity (e. g., having pride in being
able to complete tasks, adopting an appropriate work attitude) and academic competence.
Nevertheless, individuated adolescents also reported experiencing more internal distress
than connected adolescents. The researchers argue that having a higher level of emotional
autonomy may be somewhat stressfirl for the adolescent. However, the successful
management of this freedom may lead to healthier adjustment later in life.

Another group of researchers (Cooper et al., 1983) also investigated the effects of
individuality and connectedness on psychosocial competence. Individuality was comprised
of two elements, self-assertion (having own opinions and being able to communicate
them) and separateness (possessing the ability to express difi‘erentness of self from others).
Likewise, connectedness consisted of two elements, permeability (i.e., being open to the
views of others) and mutuality (being sensitive or respectful of others). Cooper et al.
(1983) concluded that adolescents who express a balance between individuality (where

the adolescent is encouraged to explore and develop him or herself) and connectedness

8
(where the adolescent has a "secure base" from which to explore) in the parent-adolescent
relationship will experience the most adaptive outcomes (e. g., being capable of expressing
their own separate opinions and, at the same time, understanding others’ points of views).

Sullivan and Sullivan (1980) tested how adolescent-parent separation (leaving for
college) afl‘ected their relationship. Only males were included in this particular study. The
results showed that these late adolescents experienced an increase in aflectiom
communication, satisfaction, and at the same time, an increase in functional autonomy in
relation to their parents. These late adolescents were becoming more independent while
simultaneously strengthening their emotional ties to their parents.

Finally, Frank, Pirsch, and Wright (1990) found that adolescents who were
insecure or disengaged from their parents reported lower levels of emotional autonomy
from their parents. In contrast, those adolescents who experienced greater closeness and
less insecurity in relation to their parents reported greater levels of autonomy. In other
words, autonomy flourished in the context of a supportive environment.

To conclude, investigating both processes of becoming autonomous while
remaining connected are needed in describing the changing parent-adolescent relationship.
A more complete picture of autonomy can be constructed by adopting a multidimensional
approach where multiple aspects of autonomy are measured.

The next section will focus on literature that has involved Asian-American
autonomy development. It is argued that Asian and American cultures differ in their views
on autonomy, and, as a consequence, some Asian—American late adolescents may have

dimculty integrating these two competing viewpoints.

i . l i . C l 2 1

American and Asian2 cultures differ from each other in two major ways. For one,
there is a collectivism versus individualistic orientation difference between Americans and
Asians (Triandis, 1988; 1994). In addition, Asians are deeply influenced by Confircianism
and the notion of filial piety. These two world views influence the way individuals of
Asian culture view autonomy concerning the age-appropriateness of certain behaviors and
the appropriate level of emotional autonomy and connectedness.

'vimvr in ivi lim

In Asian societies the welfare of the community is emphasized rather than the
individual. Group needs take precedence over individual needs. Moreover, the expression
of individual needs and desires are considered selfish (Matsuoka, 1990). What is valued is
conformity, not independence. For example, the Japanese have a saying that exemplifies
their belief in collectivism: "The nail that sticks out gets pounded." Individuals who adopt
the perspective of collectivism will emphasize maintaining connectedness to the family and
to the community, rather than autonomy (Hui & Villareal, 1989).

In contrast, the notion of individualism operates in the United States. The needs
and rights of an individual are emphasized. Self-reliance, independence and personal

freedom are positively valued and emphasized (Rosenthal & Bomholt, 1988).

 

2The use of the terms “American” and “Asian” in no way suggests that these are
homogeneous groups. Furthermore, it is recognized that there is no "typical" Asian-
American late adolescent. However, because this study aimed to understand Asian—
American late adolescents in general, it was considered appropriate to use the term
“Asian” to describe Asians of various backgrounds.

The family as well as the community is primary to the individual in Asian societies.
Confucianism and the concept of filial piety is central in understanding the relationship
between a parent and child in Asian families (Chao, 1994). Sih (1960) describes filial piety
as "loyalty, respect, and devotion to parents. It represents one of the basic social and
religious concepts of Chinese people. It is considered the virtue of all virtues and the soul
of Chinese culture." Honoring and obeying a parent's wishes are of utmost importance
(Ho, 1986).

Traditionally, the Asian culture has placed greater emphasis on achieving one's
identity and sense of worth through close relationships with the family and also through
being a member of an established lineage and extended family system (Matsuoka, 1990).
Family interests are placed above personal interests (Sue & Sue, 1990). For example, an
adolescent may be persuaded to choose a particular career (sometimes one that he or she
may not particularly enjoy) that the parents select because it is seen as the most beneficial
to the family.

Another example that illustrates the primacy of the family is the way in which
members address one another. Unlike Western cultures, the last name is stated first
followed by the first name, an indication of the importance of family over the individual.
Again, the encouragement of autonomy and independence is not central in this framework
of beliefs. Asian parents who hold these traditional values may desire and encourage
autonomy expectations that are incongruent with the expectations or values that their

Asian-American adolescents are internalizing.

E l I. . E !

Researchers studying the patterns of competence and adjustment among
adolescents of various ethnic groups, found that Asian-Americans scored lower on self—
reliance scales than European-Americans (Lambom, Mounts, Steinberg, & Dombusch,
1991). This supports the notion that Asian parents place a greater emphasis on
connectedness rather than independence. In addition to being less functionally
autonomous, Asian-Americans have a later timetable regarding behavioral autonomy.

Feldman and Rosenthal (1991) compared age expectations of behavioral autonomy
in Hong Kong and American adolescents. In general, Hong Kong youths expected
behavioral autonomy (i.e,, everyday life management in certain domains) at a later age
than their American peers. This could be attributed to the culture of collectivism in Asia
where conformity, not independence, is highly valued. As such, behaviors demonstrating
independence are not required until a later age (Rosenthal & Bomholt, 1988). Indeed,
Hong Kong adolescents described their families as “placing less value on individualism,
outward success and individual competence, and more value on tradition, prosocial, and
well socialized outcomes."

Feldman and Rosenthal's study (1991) examined Asians, not Asian-Americans.
However, their findings can be extended to Asian-Americans. This assumption is based on
the fact that Asian-American adolescents are still connected to their Asian heritage
through their parents. Their immigrant parents may continue to espouse and encourage
their adolescents to hold traditional values that contrast those values of the host culture.

Hence, it is hypothesized that Asian-American adolescents who are more Asian will

12
experience a later timetable of autonomy than those who are more Westernized.
-P . 'n

Due to the difi‘erences between Western and Asian cultures, discrepant, and
sometimes opposing values, may arise for the adolescent and their parent (Nguyen, 1992).
A recent study by Copeland, Hwang and Brody (1996), compared Asian-American, Asian-
Intemational and European-American late adolescents on issues involving family
relationships and adolescent turmoil. It was found that Asian-Americans described
themselves as being in more turmoil over issues of independence (for example, "I feel I
have obligations to my mother/father that I wish I didn't have") compared to European-
Americans. Late adolescents who felt more conflicted over these independence issues
were more depressed, more lonely, and reported lower self-esteem.

Given that adolescents may be more exposed to the ideals of the majority culture
(e. g., by attending school and interacting with peers), they may acquire values of the
majority culture more quickly than their parents (Rosenthal, Bell, Demetriou & Efldides,
1989). This may create a context where adolescents “clash” with their more traditional
parents over attitudes and behaviors (Szapoznik & Kurtines, 1980). To investigate the
implications of these clashing values, the " goodness of fit" model will be used.
W

The concept of "goodness of fit" (Lemer & Lerner, 1983; Thomas & Chess,
1977) is demonstrated to be useful as a means to understand adolescent development and
adjustment in several contexts - the home, school, and peers (Lerner, Lerner & Zabski,

1985; Talwar, Nitz & Lerner, 1990). According to the model, the degree of fit between

 

13
the adolescent and their surroundings (whether it be physical surroundings, teacher, or
parental expectations, etc.) is predictive of adolescent fimctioning. Ifthere is a poor fit,
the adolescent is more likely to experience poor adjustment, such as more negative parent-
adolescent relationships or academic troubles (Talwar et al., 1990). In contrast, if there is
a good fit, the adolescent is more likely to experience positive outcomes, such as greater
academic competence (Lerner, 1983) or more positive relationships with peers (East,
Lerner, Lerner, Soni, Ohannessian, & Jacobson, 1992).

Eccles and her colleagues (1991) used a similar model of “stage-environment fit”
to hypothesize that the goodness of fit between the amount of control a parent yields and
an adolescent's desire for autonomy, will play a part in determining whether an adolescent
will experience negative or positive outcomes. These researchers found that poor fit was
associated with more conflict in families. For example, in issues such as choosing the
amount of involvement with peers or how much say they had in family decision-making,
some adolescents desired more autonomy than their parents were willing to give. This
lack of fit was linked to lower self-esteem and more misbehavior in the adolescent when
compared to adolescents and parents who experienced good fit.

Previous studies on adolescent development that test the goodness of fit model
have examined "poor fit” versus " good fit" between adolescent temperamental
characteristics and parent’s expectations (for example, Talwar et al., 1990). However, it
has been found that "poor fit" needs to be defined more precisely. Different types of poor
fit (e.g., exceeding parental expectations, or falling below parental expectations) have been

linked to difi‘erent outcomes (Juang, Castellino, & Lerner, 1995; Lerner et al., 1985).

14

In contrast, Eccles et al. (1991) found that two types of poor fit led to similar
outcomes. In this study, poor fit was divided into two categories - adolescents who
thought their parents exercised too much control, and adolescents who thought their
parents permitted too much fieedom. Adolescents who wanted more freedom as well as
adolescents who wanted more control, relied on their peers more than their parents for
advice and were more willing to engage in deviant behavior with their peers. The
researchers argue that optimal adolescent development requires changing the level of
parental control (i.e, the level of autonomy granted) to fit the changing developmental
needs of the adolescent.

No studies are known to have examined the goodness of fit in autonomy
expectations of Asian-American late adolescents and their parents. Asian—American late
adolescents who hold very different autonomy expectations from their parents experience
a lack of fit. The goodness of fit framework can be used to investigate whether Asian-
American late adolescents and their immigrant parents hold conflicting autonomy
expectations, and, whether there are consequences to an incongruity in expectations. In
addition to fit in parental demands, fit in cultural expectations will also be explored. This
leads us to consider the second research dimension, acculturation.

AQQDIIIILflIiQn

Since the 1970's, researchers in psychology have stressed the important role of
context in understanding human development. For example, Bronfenbrenner's (1989)
ecological systems theory and Lemer's (1984, 1993) developmental contextualism are

representative of researchers' current awareness of and interest in the importance of the

15

dynamic relationship between an individual and his or her environment. This emphasis on
context has extended across all stages of the lifespan, including adolescence (Silbereisen &
Todt, 1994). The concept of acculturation, with its emphasis on the impact of culture on
an individual’s development, underscores the importance of contextual influences.

Acculturation has been conceived of as a linear bipolar continuum (Andujo, 1988;
Sirnic, 1987). On one end was the traditional individual who held on to traditional values.
On the other end was an individual who denied the traditional culture and completely
adopted the values of the new culture. Presently, the concept of acculturation has evolved
from a simple continuum to a more complex, two dimensional view of acculturation in
which "both the relationship with the traditional or ethnic culture and the relationship with
the new or dominant culture are considered, and these two relationships may be
independent, " (Phinney, 1990). In this view, involvement in both the new culture and the
old culture are emphasized. In other words, acculturation is not simply the adoption of
new values. It also involves retaining, modifying, and sometimes giving up old values
(Berry, 1980; Phinney, 1990).

MW

Berry's (1980) model of acculturation has generated numerous studies that test his
concept of different acculturation styles. He states that there are four main styles -
marginal, assimilated, traditional (or separated) and integrated (or bicultural). He
proposes that these styles depend on two key issues: 1) the extent to which identifying and

maintaining cultural characteristics (e.g., attitudes and behaviors) of one's ethnic group are

l6
valued, and 2) the extent to which maintaining relationships with other groups outside of
one's ethnic group is valued (see Figure 1). Based on these two issues, four acculturation
strategies are identified.

Marginalized individuals feel they belong to neither their ethnic group nor the
majority culture. Individuals operating from an assimilated style choose to relinquish their
ethnicity, or cultural identity, and take on the majority culture's values and behaviors.
Traditional, or separated individuals exclusively focus on their own ethnic group values
and practices and have withdrawn from interaction with the majority culture. Finally,
individuals characterized by the integrated style strongly identify and are involved with
both their own ethnic group as well as the majority culture.

WWW

These acculturation styles can be conceptualized in terms of goodness of fit. For
example, Padilla, Wagatsuma, and Lindholm (1984) describe the goodness of fit for
immigrant populations:

"Role expectations in the new social environments may differ from those of the

immigrant's native society. Discrepancies in role expectations between the two

social systems may create conflict. By adhering to the role expectations of the
homeland, the immigrant may not fulfill the role expectations of the new social
environment. As a result, the immigrant may be forced to make a decision to

adhere to only one social system or to compromise," (p. 296).

In Padilla et al.'s (1984) study, individuals experiencing a lack of fit with cultural demands
reported higher levels of stress and lower self-esteem. One can likewise examine the

implications for Asian-American late adolescents who fit or do not fit into their social

environment.

17

mowcm acumen—2381‘ Son 8me Ransom A Emmi

 

 

 

umNHHmcflaLms Hmcofiwflumch

umpmaflsfiww< uwwmaamucH

 

 

 

 

«maaoau Locwo

gozs spa; waacmcoawmaoc
camuCfimE

op maam> co on

=wm>= ow umcmuflmcoo ufi mm

 

 

 

 

 

 

=02: =ww>=

«moflwmficmuomcmco
new >paucccfi Hmczpazo

cflmucams ow o3Hm> +0
on op uwaouawcoo pa mH

 

 

P wsmmH

m mammH

 

18
E l . l l . l l E] l . l

Feldman and Rosenthal (1990) analyzed the acculturation of autonomy
expectations for behavioral autonomy. Their sample included Hong Kong adolescents,
first and second generation Chinese adolescents residing in the United States and
Australia, and European-American and Anglo-Australian adolescents. Second generation
Chinese adolescents in both the US and Australia held significantly later expectations for
autonomy than their Western counterparts, and were more similar to Hong Kong
adolescents. The researchers argue that this is evidence that acculturation does not
progress rapidly, and that "the result of extensive contact between Chinese students of
immigrant families and their host culture is not one of assimilation, where the values of the
host culture are incorporated as a whole and the values of the culture of origin are
relinquished," (Feldman & Rosenthal, 1990). In other words, acculturation does not seem
to proceed at an equal pace across all realms of autonomy. For example, Chinese
adolescents' age expectations for items such as "being able to watch whatever t.v., video
or movie show they want," or "stay home alone if sick" were somewhat similar to Western
adolescents' age expectations. In contrast, expectations were most dissimilar for items
such as "go out on dates," "smoke cigarettes, " and "choose your own friends even if your
parents disapprove."

WW

Several researchers have proposed that the most psychologically adaptive style is
the integrated style (LaFromboise, Coleman & Gerton, 1993; Szapocznik & Kurtines,

1980). Nevertheless, when taking context into account, adoption of the other two styles,

19
assimilation and separation, "may also provide the basis for a good self-concept, if the
person is comfortable with these alternatives and is in an environment that supports them,"
(Phinney, Locher & Murphy, 1990).

For instance, an adolescent who is very assimilated (fits well into Western cultural
demands) may do very well in the Midwest, where Asians are relatively few in number.
There may be a lack of support for adhering to traditional values in this context.
However, for the same adolescent who lives in a diverse context such as California, it may
be more advantageous to adopt an bicultural style (fits well in Western as well as Asian
cultural demands), as there may be more support for both the traditional and the majority
cultures. In another instance, an adolescent may do very well to adopt a separated, or
traditional style (fits well in Asian cultural demands) if he or she lives in a strong ethnic
community. Ifthis separated adolescent has chosen not to be involved in Western culture,
this lack of fit may not necessarily lead to negative outcomes. Phinney et al. (1990)
conclude that adopting one style is not necessarily superior to another in terms of mental
health. One strategy may work well in one context, but not in another.

5 [I . E .

Currently, researchers agree that the development of autonomy from parents is a
multifaceted process that also includes maintaining a connectedness to parents. Context
(e. g., family and culture) contributes significantly to this process of becoming an
independently functioning adult. For example, an individual’s degree of acculturation
impacts ones’ ideas of appropriate autonomy levels. Moreover, there are consequences

for adolescents who do not fit well into their parent’s, or prevailing culture’s, autonomy

20
expectations. The development of autonomy is one very important aspect of the parent-

adolescent relationship that is the focus of this study.

Chapter 3

THE PRESENT STUDY

One purpose of this study is to contribute to the understanding of autonomy
development in a different population than previous studies have ofifered. The primary
goal was not to carry out a comparative evaluation, but rather to describe autonomy
development in a population rarely seen in the autonomy literature.

The present study addresses the multidimensionality of autonomy by examining
both emotional and behavioral autonomy in adolescents. In addition, the goodness of fit
of autonomy expectations is explored in the two contexts of family and culture. Most
importantly, the relationship between the different dimensions of autonomy and goodness
of fit to certain adolescent adjustment variables is investigated.

This study examines the implications for Asian-American late adolescents who
experience conflict within their families regarding issues of autonomy. This study also
examines how well the Asian-American late adolescent fits into cultural expectations (of
the U. S. and Asian cultures) regarding general behaviors and attitudes. In addition, and
more specifically regarding autonomy, this study investigates how adolescent acculturation
afiects the adolescent’s emotional and behavioral autonomy. Moreover, the level of

agreement between Asian-American late adolescents and their immigrant parents on

21

 

22

behavioral autonomy expectations is explored. Finally, there is an attempt to uncover
whether the degree of fit between adolescent expectations in the family and culture relates
to adolescent adjustment in terms of the quality of the parent-adolescent relationship,
adolescent internal distress, behavior problems, self-esteem, and psychosocial maturity.
The following section will define the major variables more precisely before discussing the
main research questions and hypotheses.

' i f M ' r ri l

Autonomy is defined as an aspect of a relationship rather than as a psychological
characteristic (Lambom & Steinberg, 1993). For instance, an adolescent can be described
as being "autonomous". Yet, this description does not distinguish who or what the
adolescent is autonomous from. This study defines adolescent autonomy in relation to the
adolescent’s parents.

This study focused on two aspects of autonomy (emotional and behavioral) and
one aspect of connectedness. EmmjgnaLaQOmy is defined as having a “mature,
realistic, and balanced perception of parents that accompanies the acceptance of primary
responsibility for personal decision making, values, and emotional stability" (Lambom &
Steinberg, 1993).

W is defined as being self-reliant from parents (e.g., having the
ability to make decisions regarding day to day life management on one's own) (F eldman &
Rosenthal, 1990). For example, adolescents demonstrate behavioral autonomy from their
parents by being allowed to stay home alone at night when their parents go out, or being

able to choose their own clothes, hairstyle, or friends. These behaviors indicate that the

 

23

adolescent can rely on his or her own judgement in these situations. In this study,
adolescents assessed the level of behavioral autonomy they experienced while growing up,
not of their current level of behavioral autonomy.

Connectedness is defined as the extent to which adolescents feel their parents are
supportive and understanding.

Won is defined as the level of involvement in both American and Asian
culture. For instance, an adolescent can demonstrate involvement in Asian culture
behaviorally by eating Asian food or listening to Asian music. Adolescents can also
demonstrate involvement in Asian culture attitudinally by endorsing Asian values, such as
behaving one should always be respectfirl to elders. Acculturation is also used as a
measure of goodness of fit.

W in the family is defined as the extent to which the adolescent’s
behavior matches parental expectations of autonomy development. For example, if
parents expected their adolescent to adhere to a curfew until he or she was 18 years old,
yet the adolescent believed that imposing a curfew after 16 years of age was unreasonable,
there is a poor fit between parental and adolescent expectations of autonomy. Or,
adolescents may match their behavior to parental expectations with regard to the level of
control the parent exercises. A good fit would be evidenced if the adolescent feels his or
her parents are granting the appropriate amount of freedom and control. A poor fit would
be evidenced with adolescents who feel their parents are not permitting enough autonomy,
or, alternatively, not providing enough guidance or control. The next section details the

major research questions and specific hypotheses of this study.

 

24
Reseamhflumtinnmifinecifisflmtheses

The first research question deals with whether information on fit in two contexts,
family and culture, is a better predictor of adolescent adjustment than including
information on fit in just one context. Recognizing that the parent-adolescent relationship
must be understood in its cultural setting, it is hypothesized that information on goodness
of fit with parental expectations and acculturation together, better predict adolescent
outcome than simply knowing whether the adolescent fits into familial expectations or
cultural expectations alone.

The second question concerns the relationship between acculturation styles and the
level of autonomy an adolescent experiences. The hypothesis is that assimilated
adolescents will report the earliest timetable of autonomy, that traditional adolescents will
report the latest, and that bicultural adolescents will report a timetable later than
assimilated yet earlier than traditional adolescents. There is not sufiicient literature to
support a hypothesis about the direction of the relationship between an adolescent who
adopts a marginal acculturation style and their timetable of autonomy. Therefore, an
analysis of this relationship is exploratory.

Another hypothesis pertaining to the first research question is that assimilated
adolescents will be the most emotionally autonomous compared to bicultural and
traditional adolescents, and traditional adolescents to be the least. There is not sufficient
literature to justify a supposition regarding the level of emotional autonomy that marginal
adolescents will demonstrate. Therefore, an analysis of this relationship is exploratory.

The third research question addresses how different levels of emotional autonomy

25

and connectedness relate to adolescent adjustment. It is hypothesized that adolescents
who experience high or low levels of emotional autonomy in the context of a high level of
parental support will show more positive adjustment than adolescents who experience high
or low levels of autonomy in the context of a low level of parental support.

The fourth research question explores the implications for a lack of fit between
adolescent autonomy expectations and parental expectations. It is hypothesized that
Asian-American late adolescents who experience a lack of fit with parental expectations
on issues of autonomy will report more negative adjustment than adolescents who
experience a good fit with parental expectations.

The fifth research question examines the outcomes associated with the
adolescent's acculturation style. It is hypothesized that in the context of a Midwestem
state such as Michigan, assimilated and bicultural adolescents will experience more
positive outcomes than either traditional or marginal adolescents, with marginal
adolescents experiencing the least positive outcomes. The following chapter describes the

methods used to test these hypotheses.

 

Chapter 4

METHOD

E . .

All 99 participants were undergraduate students3 living in a large, Midwestern
University town. The mean age was 19.8 years, and ranged from 17 to 24 years. There
were 55 females. The participants were of Asian descent, with parents originating from
either China, Japan, the Philippines, Thailand, Korea, or Vietnam. The majority (87 9%)
were second generation (e.g., their parents immigrated to the US, and the participant was
born in the US.) The remaining were either first, or 1.5 generation (e.g., they were not
born in the US. but immigrated with their parents). Most (80.8%) grew up in the
Midwest region of the US. The average length of time the participants lived in the US.
was 17.3 years (range = 7 to 24 years; SD = 3.43). The majority of participants (94.9%)
grew up living with both parents.

The parents of the participants were a well-educated group: 79.8% of the fathers

 

3 Some may not agree that college students belong to the period of “late
adolescence.” However, it can be argued that college students, even those in their early
208, are experiencing a prolonged adolescence, based on the fact that it is still acceptable
in our society for this population to delay taking on adult responsibilities (e.g., having a
career or starting a family). Furthermore, Erikson (1963) has argued that adolescence is a
time for exploration. College students are likely to be engaged in an exploration of their
life path choices.

26

27

and 26.3% of the mothers were high school graduates, and, of these, 48.5% of the fathers
and 41.4% of the mothers were college graduates. The average number of years mothers
lived in the US. was 20.7 years (range = 4 to 42; SD = 8.20). Fathers, on average, had
lived in the US. for 22.3 years (range = 7 to 48 years; SD = 9.46).

W

Participants were recruited in several ways. The Coordinator for Asian Pacific
Student Affairs of the Ofice of Minority Student Affairs (OMSA) arranged for the
principal investigator to meet with the leaders of the Asian student groups. At these
meetings, the purpose and nature of the study were explained. The student leaders were
asked to assist the investigator in contacting potential participants, either by announcing
the opportunity to participate in the study to members at a regular group meeting, or by
providing a list of members to contact. Also, Asian students who attended psychology
classes could sign up to participate in the study. By doing so, they received credit for their
class. Finally, participants were also recruited via e-mail. Information on the study was
sent to over 1000 Asian students. Approximately 8% responded.

In order to obtain a wider range of participants, data were collected under three
difi‘erent survey conditions. Participants either completed the survey in a classroom on
campus individually, in groups of 2 - 6, or at home, mailing back their responses.
Preliminary analyses were run to check whether survey condition was significantly related
to the study variables, thus acting as a covariate. Based on Bonferonni’s adjustment to
control for an inflation of Type I error, it was found that none of the study variables

correlated significantly with survey condition.

28

Participation in this study was voluntary, with no compensation for the majority of
the participants. However, psychology students received credit, and three randomly
chosen participants were monetarily awarded ($25, $25, & $50) afier completion of data
collection.
Consent

See Appendix A for a copy of the consent form.
Meamres
The following are self-report questionnaires:
Demographic information. Participants provided information on their age, sex, country
and state of birth, generational status, year in college, birth order, length of residence in
the US, and ethnic identity. In addition, participants provided information on their
parents’ country of birth, length of stay in the US, occupation, and education level.

AW!
Behavioral timetable (Feldman & Rosenthal, 1990). This is a 19-item scale that assessed
the timetable of expectations for autonomous behaviors by asking what adolescents
actually experienced with their parents, and what adolescents would have liked to
experience. For example, "At what age did you expect to be able to choose your own hair
style even if your parents disapproved?" and "At what age would you have liked to choose
your own hairstyle even if your parents disapproved?". Participants rated each item on a
5-point scale: 1 = before age 14, 2 = 14-15 years, 3 = 16-17 years, 4 = 18 years or older, 5
= never. Cronbach’s alpha for the actual timetable and preferred timetable were found to

be .86, and .85, respectively.

 

29

Emotional Autonomy Scale (EAS) (Steinberg & Silverberg, 1986). This measure required
adolescents to indicate how strongly they agreed or disagreed (on a 4-point Likert scale)
with 20 items concerning four aspects of emotional autonomy - perceives parents as
people, parental deidealization, nondependency on parents, and individuation. A sample
item is, "My parents and I agree on everything." Cronbach’s alpha in this study was .80.
Connectedness (Stutman & Lich, 1984). This 11-item scale measured the extent to which
the respondent felt close to and supported by his or her parents. Respondents indicated
how strongly they agreed or disagreed with statements using a 4-point scale. Sample
items are, "My parents and I feel like strangers to one another" and "When I am feeling
bad I can count on my parents to remind me of my worth." Cronbach’s alpha for this
sample was .89.

Warrants
Goodness of fit index. Goodness of fit was measured in several ways. First, the
discrepancy between adolescent expectations (preferred timetable) and the adolescent's
ratings of parental expectations (actual timetable) on the behavioral timetables were
calculated. The score on each item from the preferred timetable scale was subtracted from
the actual timetable scale. These difference scores were summed to produce a total fit
index. The absolute values of the scores were used. A higher score (higher discrepancy)
signified poorer fit.

The goodness of fit between adolescent and parental expectations were also
assessed by asking four questions (adapted from Eccles et al., 1991) to measure the actual

level of parental control and the adolescent’s desired level of control. Sample questions

 

30

are, "How often do you take part in family decisions that concern you?" and "How often
do you think you ought to take part in family decisions that concern you?". Respondents
indicated the frequency on a scale ranging from 1 = never to 4 = always. Discrepancy
scores were calculated by subtracting the score from the item that indicated the actual
level of parental control from the item that indicated the desired level of control. For
example, if an adolescent answered that he or she never (1) takes part in family decisions,
yet he or she thinks they ought to take part in family decisions sometimes (3), the
discrepancy score equaled 2. The absolute values of the scores were used. A higher score
represented a poorer fit.

When
Acculturation Scale (adapted from Nguyen, 1995). This scale measured adolescents’ level
of acculturation in two ways - by their endorsement of Asian and US. attitudes and
values, and by the fi'equency of engaging in Asian and US. behaviors. The degree to
which the adolescent endorsed values such as, "It is important for me to preserve my
Asian heritage" and "As far as behaviors and values, I am 'American'" were assessed on a
5-point scale ranging from "strongly agree" to "strongly disagree." By using these types
of statements, the adolescent’s involvement in Asian culture (IAS) and involvement in
US. culture (IUS) could be assessed separately. By combining scores on these two
subscales, four different acculturation styles were identified (based on Berry’s (1980)
model).

For example, an adolescent who scored high on the IAS and low on the IUS was

identified as adopting a traditional acculturation style. An adolescent who scored low on

31

the IAS and high on the IUSwas identified as adopting an assimilated acculturation style.
High scores on both the IAS and IUS indicated an integrated (bicultural) adolescent, while
low scores on both the IAS and IUS indicated a marginalized adolescent. Cronbach’s
alphas for IAS and IUS were found to be .86 and .82, respectively.

This acculturation scale also measured late adolescents’ involvement in Asian and
US. behaviors. For example, The frequency of certain behaviors such as "How often do
you eat Asian food?" were indicated on a 5-point scale ranging from "never" to "always".
Cronbach’s alpha for involvement in Asian behaviors and US. behaviors were .86 and .80,
respectively.

Warrant
The Psychosocial Maturity Inventory (Greenberger, Josselson, Knerr, & Knerr, 1974).
This measure was created to assess how well the adolescent functions socially. One
subscale consisting of 10 items was used - work orientation (sample item: "I believe in
working only as hard as I have to"). Respondents rated how strongly they agreed or
disagreed on a 4-point scale. A higher score represented a more mature response.
Cronbach’s alpha for this sample was .70.
The Misconduct Scale (adapted from Feldman & Rosenthal, 1991). This is an 8-item
scale that measured the frequency (0 = never to 3 = often) of misconduct behaviors that
the adolescent has engaged in. A sample item is, "Have you ever cheated on a test?"
Cronbach’s alpha was .76 in this study.
Center for Epidemiological Studies - Depression (CES-D) (Radloff, 1977). This is a 20-

item scale that measured psychological depression. Respondents were asked to indicate

 

32
how often (ranging from "rarely" to "most of the time") they felt or behaved during the
past week. Sample items are, "I was bothered by things that usually don't bother me", and
"I enjoyed life". Cronbach’s alpha in this study was .90.
Familial Insecurity Scale (FIS) (Ainsworth & Ainsworth, 1958). Adolescents were
asked to rate 12 items such as "I feel discouraged that it is so difiicult to live up to my
parents' expectations," on a 4-point scale ranging fi'om "very false" to "very true".
Cronbach’s alpha was found to be .81.
Self-Perception Profile for College Students (N eemann & Harter, 1986). This instrument
measured the students' perceptions of their level of competence in various domains. Of
the original 13 subscales, 5 were used: 1) academic competence, 2) social ability 3) close
fiiendships, 4) intellectual ability, and 5) global self-worth. The first 4 subscales each have
4 items to assess the student’s perceived competence in that particular area. Global self-
worth is assessed with 6 items. For example, to measure the student's perceived
competence in the academic arena, he or she responded to items such as, "Some students
feel confident that they are mastering their coursework, but other students do not feel so
confident." The student must first decide which statement is most like themselves, and
then decide whether the statement is sort of true or false or really true or false for them.
The student's score for each subscale was calculated as a mean of the items from that
subscale. Cronbach’s alpha for the subscales used in this study are as follows: academic
competence = .77, social ability = .82, close fiiendships = .84, intellectual ability = .76,
and global self-worth = .85.

Grade Point Average (GPA). Respondents were asked to report their grade point average

33
on a scale from 0.0 - 4.0.
mm

A pilot study was conducted with 10 Asian-American college students to verify
whether directions for filling out the questionnaires were clear and understandable, and to
determine the length of time needed to complete the questionnaires. These responses
were included in the final sample.

Managua.

There were a small amount (3%) of missing data. Missing data were estimated
using multiple regression. Several predictor variables that significantly correlated with the
variable with missing data were identified. Then, scores for each missing case were
predicted using these predictor variables. These predicted scores replaced the missing

information.

 

Chapter 5

RESULTS

Discrimination analyses revealed that there were no significant sex differences
found in the study variables. Therefore, for all analyses males’ and females’ data were
combined. The means, standard deviations, and ranges of the study variables are
presented in Table l.

@93me

In order to attain a general overview of the relationships between acculturation,
level of fit in the family, autonomy, and adjustment, LISREL (LInear Structural
RELationships; Joreskog & Sorbom, 1993) analyses were conducted. The covariance
matrix of the 20 study variables was analyzed (see Appendix B for the covariance matrix).
The generalized least squares (GLS) procedure was used to estimate model parameters.
The GLS procedure was chosen because of the study’s small sample size.

Ihzflrllmodsl

When the model was initially run, several of the goodness of fit indices indicated

that the data did not fit the model well‘. An examination of the modification indices

 

"Goodness of Fit Index = .71, Norrned Fit Index = .61,
Comparative Fit Index = .71

34

Table 1. Description of Study Variables

35

 

 

M SD Range
Adiustmem
Depression 1.85 .54 1.05 - 3.30
Emotional connectedness to parents 2.93 .53 1.64 - 3.82
Self-esteem: Academic 2.51 .66 1.00 - 4.00
Close fiiends 3.06 .80 1.00 - 4.00
Intelligence 2.74 .63 1.25 - 4.00
Social ability 2.99 .75 1.25 - 4.00
Global 2.87 .61 1.00 - 4.00
GPA 2.92 .47 1.80 - 3.90
Behavior misconduct .78 .42 0.00 - 2.13
Insecurity with parents 2.56 .54 1.17 - 3.92
Work attitude 2.81 .39 2.10 - 3.80
ludicatorsoLFit
Autonomy Timetable Fit .52 .37 0.00 - 1.79
How Decisions are Made in the Family .62 .75 0.00 - 3 .00
How Often are Decisions Made .82 .87 0.00 - 4.00
while including the adolescent
Autonomy
Autonomy Timetable:
Allowed 2.85 .57 1.53 - 4.11
Would Have Liked 2.36 .46 1.32 - 3.84
Emotional Autonomy fiom parents 2.80 .32 2.00 - 3.60
Emotional Connectedness to parents 2.93 .53 1.64 - 3.82
Acculhuatim
Endorsement of Values:
Asian 3.51 .44 2.23 - 4.41
Western 3.92 .40 2.91 - 4.71
Engaging in Behaviors:
Asian 3.09 .69 1.60 - 4.60
Western 4.08 .56 2.38 - 5.00

 

 

36

indicated that the model could be improved if the latent variable, fit in family, was allowed
to correlate with the observed variable of autonomy timetable. This made sense since this
observed variable had been used to calculate the fit in family scores. Furthermore, by
allowing a number of error terms to be correlated between the observed variables (38 of
the possible 196 (19.4%) correlations between error terms were set free) the model could
be improved fiirther (see Appendix C for a listing of the correlated error terms). The
modified model demonstrated good fit with the data. See Table 2 for a listing of the
goodness of fit indices.

The t-values for the parameters of the measurement model were all significant (see
Figure 2; see Appendix D for parameter estimates). Thus, the observed variables seemed
to adequately measure the latent constructs.

The t-values for the parameters of the structural model were also all significant
(see Figure 2 for t-values; see Appendix D for parameter estimates). Thus, the hypothesis
that autonomy, acculturation, and level of fit in the family predicted the late adolescent’s
well-being (e.g., depression, self-esteem, behavior misconduct, etc.) was confirmed.

AmmnafimoflthefiallmodelflifluuhmodflaLaDiZ

To assess whether the full model (autonomy and fit in two contexts - family and
culture) was better at predicting adjustment than a submodel with autonomy and only one
fit context predicting adjustment, two nested models were tested. One nested model
tested whether autonomy and fit in the family adequately predicted adjustment and the

other model tested whether autonomy and acculturation predicted adjustment.

37

com: 33
cows—:33 qu 8:33 octane.— 8: 8a 02 3on can A23 :otBtU :28585 8:82 ooaowcoaoufi 2: ”802

 

952E Baaecsmv Egan

 

3. S. 2. 033cm :82 “com commences—Sm
8a: 3.4: 2 a: 28 z :25
co; co; co; A58 “8?: “E 03:59:00
3. 8. am. 55 cans E Bebz
em. 8. 8. EB 52: E E0 aoéooo
€2.qu Gare :8.an
x: "as .Nma: E new 62% as ”he .33 0333.20
N 38:55 _ 3255 382 5823 =E

 

823 E do aoéooo Emmi .N 23

 

 

38

~8de
5,288 a a.
SE23 8%

See
mmimoeo

Mans
9885 an

aim:
gumma

23

$6

Eve—2 inc—0:834 zen 05 mo muouofifium com $29-“ .N 8sz

2mm.—

bwfim

3a.—

8: FREQ

.5 30m
.Re

I
LE

3.... E

8

mm

 

FEMS:
EHO
a
OS<OEm
m _ N w
m<><m
..
mm
2 _ .o m<AO>7=
woof mDAO>ZH

0|

39
The covariance matrix of the observed variables was analyzed using the generalized least
squares procedure for the two nested models (see Appendices E and F for the covariance
matrices of submodel 1 and submodel 2). i

The autonomy and fit in the family model (submodel 1) did not significantly differ
from the full model, based on the results from a x 2 difference test (x 2 = 32.45, df = 52).
The goodness of fit indices were comparable to the firll model (see Table 2), indicating
that this more parsimonious model seemed to fit the data just as well as the fiill model.
However, both the measurement and structural models were not as strong. Two observed
variables did not load significantly on their latent constructs (autonomy timetable with the
autonomy construct, and GPA with the adjustment construct). Furthermore, the latent
variables of autonomy and fit in the context of the family, did not significantly predict
adjustment (see Figure 3 for parameter t-values of submodel 1; see Appendix G for
parameter estimates).

The other nested model tested whether autonomy and acculturation predicted
adjustment. The x 2 difference test between the hill model and submodel 2 was not
significant (x 2 = 27.46, df = 25), indicating that this more parsimonious model seemed to
explain the data just as well as the full model. However, an inspection of the goodness of
fit indices (see Table 2) and t-values of the parameter estimates revealed that model 2 was
not as strong structurally. As with submodel 1, including autonomy along with only one
fit context (this time culture instead of family) did not significantly predict adjustment (see

Figure 4 for parameter t-values of submodel 2; see Appendix H for parameter estimates).

40

IIIIEIIEI

EC?

Emmi:

mmiHOOm

Hme<O<

mmémoqw

QZOUmH—z

0mm;

«SO

02+

2%

have

road

89:

wode

agar

.. .v-

m8.—

won..-

vac

H EnoEnsm .«o $08583 co.“ 83?.“ .m oSwE

Eur—rho
me .5

SEE 7: Hz Bo:
Em 3.0

mmm. . BEEP

a: $5055.,
83.
83- Homzzoo

83 0535

lull-Iii

N flouoEpsm mo mcoaofieam com 83a?» 4. beam—h
Rhea—MOB n _ m4.

E
«L.

E983,
E .
@5508 $2 Homzzoo
$33 2 . .02
Ohio—E
$2-
m<><mmm
o . o . WI
mo><mmm
$2

. 283:
o. .m- 22:-
mm: :3 3,82,:

<m0 .:.N

H
E
._l
IEI

$6.0

mm‘dmex—O omm
N — mg: .

mg;
.85.

mmmmnmma wad- @

 

 

l"
U %
51, 8

 

42

Therefore, the full autonomy model that includes autonomy along with two fit contexts
seems to better predict adjustment.

Anfltematemdel

To test whether an alternate model would adequately explain the data, LISREL
analyses were run on the model depicted in Figure 5. In this model it is hypothesized that
acculturation is linked to both fit in the family and the level of autonomy the adolescent
experiences, and that these two latent constructs are, in turn, linked to adolescent
adjustment. This model did not fit the data well, as demonstrated by the chi-square
significance test and the goodness of fit indices’. Although this model makes sense
theoretically, it may be that the sample size is too small to adequately estimate the extra
parameters.
WWW

To uncover more specifically the relationships within the full autonomy model,
parts of this model were tested. To test the hypothesis that the four acculturation styles
(e.g., bicultural, traditional, assimilated, marginalized) were related to behavioral and
emotional autonomy that the adolescents experienced with their parents, prediction
analyses (PA) were conducted. “Prediction analysis compares observed frequencies with a
hypothesized or predicted pattern of outcome” (von Eye & Brandstadter, 1988). Results
reveal that acculturation style did not significantly predict the level of autonomy the

adolescent experienced, behavioral or emotional. Therefore, fiirther analyses involved the

 

5x2 = 428.26, p = 0.0, GFI = .72, CFI = .71, NFI = .61, CN = 48.6,
Standardized RMSR = .16

43

IIIIIIIE

Hum—‘5

mm‘FAUOw

Hme<O<

9200mm:

“um—ma

4&0

 

 

   

n. .HOHD< HOMZZOO OED<OEM

$32 BEES? =< .m QEmE

m<><flmm

magma

m<AO>ZH

9.502

 

44

use of two dimensions of Asian and Western involvement rather than a combination of
these two dimensions to form four acculturation style groups. Nguyen (1995) found the
two dimensions to provide more information on Vietnamese adolescent adjustment more
so than classifying adolescents into four acculturation styles.

Prediction analysis was used to examine whether one could predict the
adolescent’s level of autonomy from an assessment of their commitment to Asian or
Western attitudes and their involvement in Asian or Western behaviors. It was
hypothesized that more Asian adolescents would be less behaviorally and emotionally
autonomous, and that more Western adolescents would be more autonomous. Results of
the PA show the two acculturation dimensions to successfully predict emotional but not
behavioral autonomy.

There was strong descriptive (Del6 = .21) and statistical support (z = 2.99, p < .01;
Pearson x2 = 12.04, p < .05) indicating that those who were more Asian in their attitudes
and values were less emotionally autonomous compared to those who were less Asian.
There was also some support indicating that those who were more involved in Western
behaviors were more emotionally autonomous than those less involved (Del = .09;
Pearson x2 = 15.47, p < .01; z = 1.27, p = .10).

Correlational analyses also partially confirm the first hypothesis. Those who

endorsed more Asian attitudes were less emotionally autonomous ([ = -.29, p < .01),

 

‘“Del” is a statistic that indicates the reduction in the percentage of the number of
errors made by utilizing the predictor variable (Szabat, 1990). For example, Del = .21
indicates that the number of errors made in predicting emotional autonomy was reduced
by 21% when the variable of Asian values was applied.

45
experienced a later timetable of autonomy, ([ = .25, p < .05), and desired a later timetable
of autonomy ([ = .40, p < .001) than those who endorsed less Asian attitudes. In
comparison, those who endorsed more Western attitudes preferred an earlier autonomy
timetable (r; = -.25, p < .05) than those who endorsed less. Engaging in more Asian or
Western behaviors, was not significantly related to level of emotional or behavioral
autonomy of the adolescent. It is interesting to note that Asian and Western attitudes
were related to adolescent autonomy, but did not significantly relate to how connected the
adolescent felt to their parents.
m n i ' r l ' n
Next, the relationship between the amount of autonomy the adolescent
experienced and how well the adolescent fit with parental expectations were examined.
Correlational analyses revealed that emotional autonomy and connectedness were
significantly related to the three measures of fit. More emotionally autonomous
individuals experienced poorer fit, and more connected individuals experienced better fit
with their families. See Table 3.

Table 3. Correlations for Indicators of Fit Measures with Emotional Autonomy and
Connectedness

 

Emotional Emotional
Autonomy Connectedness
Fit in autonomy timetable expectations .47 -.48
How decisions are made in the family .43 -.48
How often decisions are made that .37 -.59

include the adolescent

 

Note: all correlations significant at p < .0001

46

WW

Emotional autonomy and connectedness also significantly correlated with several
adjustment variables. More specifically, individuals who reported being more emotionally
autonomous from their parents also reported higher levels of behavior misconduct (r =
.23, p < .05), and insecurity with parents (1: = .43, p < .001). In contrast, those who
reported being more emotionally connected to their parents reported higher levels of self-
esteem in having close fiiends ([ = .24, p < .05) and social ability (I = .21, p < .05), and
less insecurity with their parents (1 = -.50, p < .001). That more emotional autonomy is
related to negative adjustment while more connectedness is related to positive adjustment
is not surprising, as emotional autonomy and connectedness are inversely related to one
another (I = -.68, p < .0001).

Nevertheless, previous studies have provided evidence that there is a more
complex relationship between emotional autonomy and connectedness and their
relationship to individual adjustment. It is not necessarily the case that emotional
autonomy leads to poorer functioning. In the context of a supportive relationship with
parents, some very autonomous adolescents can experience good outcomes (e.g., having
good relationships with their parents, having a high sense of competence). Therefore, a
MANOVA with Lambom and Steinberg’s (1993) classifications of autonomy and
connected individuals (e.g., individuated, detached, connected, and ambivalent; see page 7
in the literature review for review of these categories) was used to test firrther how

difi'erent combinations of autonomy and connectedness would affect adolescent well-

47
being. Preliminary analyses of normality, linearity, and homogeneity of
variance/covariance matrices assumptions were met. Table 4 displays the results of the
MANOVA. Two of the 10 adjustment variables were significantly related to the four
groups. Post-hoe analyses with the Bonferonni adjustment were then conducted to
identify specifically which groups difi‘ered on the adjustment variables. Results of two
one-way AN OVAs (see Table 5) revealed that ambivalent and detached adolescents were
more insecure in their relationships with their parents compared to connected adolescents.
Finally, connected adolescents reported more appropriate work attitudes than ambivalent
adolescents. See Table 6 for the means and standard deviations for the four groups and
their scores on the adjustment variables.

In addition to the MANOVA and AN OVA runs, hierarchial multiple regression
was performed to examine the independent and interactive efi’ects of emotional autonomy
and connectedness on the adjustment variables. Evaluations of assumptions of normality,
linearity, and homoscedasticity of residuals were satisfactory. Several outliers were found
for a few of the adjustment variables. When the surveys with outliers were checked, there
were no signs of unusual responding (e. g., simply checking all 1's). Also, scatterplots
revealed that the outliers followed the same trend as the majority of the other scores, they
were just a bit more extreme in their values. Thus, the outliers were left in the analyses.

In these analyses, emotional autonomy and connectedness were entered in the first

step, and the interaction term (the product of the two main effects) was entered next.

48

Table 4. MANOVA Results for Autonomy/Connectedness Groups with Adjustment

 

F-tests with (3, 95) df

Autonomy and
Connectedness Groups

 

F
Depression 1 .02
Self-esteem: Academic .59
Close fiiends 1.93
Intelligence .28
Social ability .66
Global .40
GPA .43
Behavior misconduct 1.03
Insecurity with parents 14.00"
Work attitude 2.70*

 

* p = .05
** p < .0001

49

Table 5. ANOVA Results for Autonomy/Connected Groups and Adjustment

 

Insecurity with Parents

 

 

 

 

 

 

Source df SS MS F
Between Groups 3 8.90 2.97 14.00”
Within Groups 95 20.14 .21
Total 98 29.04

Work Attitude
Source df SS MS F
Between Groups 3 1.14 .38 2.70*
Within Groups 95 13.37 .14
Total 98 14.51
* p < .05
** p < .0001

Table 6. Means and Standard Deviations for Autonomy/Connected Groups and

Adjustment

 

 

Insecurity Work

With Parents Attitude

Ambivalent M 2.64 2.63
SD .54 .38

Connected M 2.17 2.93
SD .48 .42

Detached M 2.87 2.78
SD .41 .36
Individuated M 2.56 2.83
SD .42 .25

 

50
After adjusting the significance level (using Bonferroni’s adjustment) to deal with the
increase in Type I error by running separate regressions on 11 dependent variables, results
reveal that only one adolescent functioning variable, namely, insecurity with parents was
significantly predicted from emotional autonomy, connectedness and their interaction.
Table 7 displays the correlations, B, (i, R, R2, and Adjusted R2 for these analyses with
insecurity with parents as the dependent variable.

To test for the specific location of the interaction, three contrast vectors were used
(Rovine & von Eye, 1996). First, emotional autonomy and connectedness were each split
into three levels - low, medium, and high. Then, three different parts of the 3 x 3 table of
these two variables were analyzed: 1) the main diagonal, where those high in both
autonomy and connectedness were contrasted to those medium in both autonomy and
connectedness and low in both autonomy and connectedness, 2) the area below the
diagonal where those who were high in autonomy and medium in connectedness were
contrasted to those who were low in connectedness and either medium or high in
autonomy, and 3) the area above the diagonal where those who were low on autonomy
and high on connectedness were contrasted to those low in autonomy and medium in
connectedness and those who were medium in autonomy and high in connectedness (see
Figure 6).

Emotional autonomy and connectedness were entered on the first step, and the
three vectors on the second. See Table 8 for a summary of the results. Since the two
main effects did not significantly contribute to predicting insecurity with parents, yet two

of the interactions did, a more parsimonious model with no main effects was also tested to

51

 

 

 

So. v a t.
S. v a *
mm. u R Baa?
ov. H mm
aimo. H M
3.2 385::
Ion; mm. 2.. 2 .- 3.- 5:083:
53:0... mod- 3.. mm: 32880550 Ecosofim
acm. T and- 3. >Eo=o§< EcouoEm
mmocvoaooqcoo >Eo=o§< 3:83 5:5
o m RectoEm Ecouofim 3.5085 mofiat~>
.853.“

5m? £582: :0 5:08an :2; new .mmoevoaooecoo 5:25:24 Reese—cm mo commmocwom 3020355 .5 03$.

52

Emotional Connectedness

 

 

 

Low Medium High
L°w 1 1 1 2 1 3
Emotional
Autonomy Medium
2 l 2 2 2 3
High
3 l 3 2 3 3

 

 

 

 

 

Vector 1: 33 contrasted with 11 and 22
Vector 2: 32 contrasted with 21 and 31
Vector 3: 13 contrasted with 12 and 23

Figure 6, Contrasts for Testing Local Interactions for Insecurity with Parents Variable

53

Table 8. Hierarchical Regression of Emotional Autonomy, Connectedness and Local
Interaction Vectors onto Insecurity with Parents.

 

Insecurity with Parents

B—_——|§_

 

 

Emotional Autonomy .37 .22
Emotional Connectedness -.19 -. 19
Interactions: Vector 1 .39 .24“
Vector 2 .20 .13
Vector 3 .28 .25*
R = .64"
R2 = .41
Adjusted R2 = .38
*p < .01

** p< .0001

54
examine whether the three interaction effects would adequately predict insecurity with
parents.

The three interaction vectors were entered on one step (see Table 9). An F-test
showed that the more parsimonious model was significantly difi‘erent from the fill model
(E (2,93) = 7.09, p < .01). Therefore, the unconstrained model was accepted. This model
suggested that those who were both high on emotional autonomy and connectedness
reported less parental insecurity than those who were not as autonomous nor as connected
to their parents. In addition, those who were low on autonomy and high on
connectedness reported less insecurity with their parents compared to those low in
autonomy and medium in connectedness and those who were medium in autonomy and
high in connectedness. In other words, it seems as if adolescents who experience low or
high levels of autonomy, and, at the same time, maintain high levels of connectedness,
report a more secure parent-adolescent relationship. See Figure 7 for a graph of the
interaction.

15' . E l E . l 5 ii

To address how 1) the adolescents’ autonomy timetable and, 2) the fit between the
adolescents’ autonomy timetable expectations with their parents, relate to adolescent
adjustment, a one-way between subjects MANOVA was performed. Eleven dependent
variables (DVs) were first tested with the actual timetable of autonomy behaviors. Next,

the DVs were tested with the fit between adolescent and parental expectations of the

55

Table 9. Multiple Regression of the Interaction Vectors of Emotional Autonomy and
Connectedness onto Insecurity with Parents

 

Insecurity with Parents

W

 

 

Interactions: Vector 1 .42 .25*
Vector 2 .42 .28*
Vector 3 .44 .40"
R = .56"
R2 = .32
Adjusted R2 = .30
*p < .01
** p < .0001

autonomy timetable. Results of normality, linearity, and homogeneity of
variance/covariance matrices were satisfactory for both runs. Outliers were found, but
again, when the surveys were checked and the scatterplots examined, there was no reason
to exclude them.

First, the relationship between the timetable of autonomy expectations to the 11
DVs were assessed. Adolescents were divided into two groups on the independent
variable (IV) - those who experienced an earlier timetable (at or below the mean), and
those who experienced a later timetable (above the mean). The combined DVs were
significantly, moderately related to the timetable of autonomy (Wilks’ criterion: E (1,96) =
4.13, p < .0001; n2 = .34). Results are summarized in Table 10. Three of the DVs made a
significant contribution to the combined set of DVs that differentiated between

adolescents who experienced an earlier or later timetable of autonomy.

56

can:

53H002

>Eocou3<
Hmcofiuosu

3:83 5:» 3:385 55, 5:22an 32860550 Ea hEocoS< .mcozoEm 05 no .380 H oSwE

wmmcuwuooccoo Hmcofiuosu

Ezfiums

304

 

 

 

 

 

SlUBJBd UlIM AlIJHOBSUI JOJ. BJOOS UEBW

57

Table 10. MANOVA Results for Autonomy Timetable and Fit in Autonomy Timetable

with Adjustment

 

F-tests with (1, 96) df

 

 

Timetable of Fit in Timetable of
Autonomy Behaviors Autonomy Behaviors
F F
Depression .42 5.60“
Emotional connectedness with parents 3.27 11.12M
Self-esteem: Academic 2.24 4.00*
Close friends 5.73 * 4.97 *
Intelligence 490* 3.09
Social ability . 13 631*
Global .79 3.37
GPA 1.39 .00
Behavior misconduct .99 8.33 **
Insecurity with parents 26.95 ** * 2562*”
Work attitude 1.00 4.59"‘
* p < .05
** p < .01

W p < .001

58

Next, to test the hypothesis that the fit in timetable expectations is a better
predictor of adjustment than simply knowing the timetable, another MAN OVA was run.
Adolescents were divided into two groups for the IV- those who experienced good fit and
those who experienced poor fit in timetable expectations with their parents.

Based on Wilks’ criterion, it was found that the combined DVs were significantly
and moderately related to fit group (E (1,96) = 3.56, p < .01; n2 = .31). Eight of the
dependent variables made a significant contribution to the combined set of DVs that best
distinguished between adolescents who experienced poor fit versus good fit in their
timetable of autonomy behavior expectations. Adolescents who fit well with parental
expectations of autonomy reported closer relationships to their parents, more appropriate
work attitudes, higher levels of self-esteem in three areas (e. g., academic, close fiiends,
and social ability), and lower levels of depression, behavior misconduct, and insecurity
with parents, than those who did not fit well (see Table 10). Table 11 summarizes the
means and standard deviations of these adjustment variables for the two fit groups.

In addition to the fit in timetable expectations, two other goodness of fit indicators
(by Eccles et al., 1991) were analyzed, 1) how decisions were made in the family that
concerned adolescent compared to how the adolescent would have liked these decisions to
be made, and 2) how often decisions were made that concerned the adolescent compared
to how often the adolescent would have {liked these decisions to be made. A higher score
indicated there was a greater discrepancy between what the adolescent desired and what
his/her parent allowed.

Adolescents were again divided into two groups for the IV - those who

59

Table 11. Means and Standard Deviations for Significant Adjustment Variables and
Timetable Fit Groups

 

 

Good Fit Poor Fit

Depression M 1.74 1.99
SD .53 .53

Emotional connectedness with parents M 3.07 2.74
SD .40 .60

Self-esteem: Academic M 2.62 2.36
SD .63 .66

Close fiiends M 3.21 2.86

SD .69 .88

Social ability M 3.15 2.78

SD .68 .79

Behavior misconduct M .67 .91
SD .39 .41

Insecurity with parents M 2.34 2.84
SD .50 .48

Work attitude M 2.88 2.71

SD .38 .37

 

60
experienced good fit and those who experienced poor fit. The MANOVA runs show
similar findings to the fit in the timetable of autonomy analyses. These two fit indices
were better predictors of adolescent adjustment than simply knowing how or how ofien
decisions were made in the family that concerned the adolescent. See Tables 12 and 13
for a summary of the findings.

Fit was also related to the timetable of autonomy the adolescent experienced.
Those who were allowed to engage in autonomous behaviors at an earlier age experienced
better fit. See Table 14.

Originally, it was intended to examine whether there was a difference between
those who wanted more autonomy than their parents allowed, those who wanted less (two
types of poor fit), and those who were satisfied (good fit) with the amount of autonomy
they were granted. Fit indices of 1) how decisions were made and 2) how often decisions
were made that included the adolescent, were split into those three groups. However, a
MANOVA run with the three fit groups predicting 11 adjustment variables, showed no
significant results. One reason for this may be the small size of some of the groups. For
example, there were very few adolescents who wanted less autonomy than their parents
allowed. Therefore, the two poor fit groups were combined and compared against those
with good fit.

E. . E l E . l I. l l E 5

Means for “age allowed to engage in certain autonomous behaviors” between poor

and good fit groups were compared using a one-way ANOVA. A significant mean age

difference was found between the two fit groups (E (1,97) = 24.51, p < .0001).

61

Table 12. MANOVA Results for How Decisions are Made in the Family with Adjustment

 

F-tests with (1,96) df

How decisions are made

Fit in how decisions are

 

 

in the family made in the family
F F
Depression 4.40“ 549*
Connectedness to parents 24.22*** 19.00***
Self-esteem: Academic 1.04 2.94
Friendship 3.10 3.59
Intelligence 2.23 4.70*
Social .33 1.5 1
Global 2.28 4.25*
GPA .58 .06
Behavior misconduct .19 1.10
Insecurity with parents 43.32*** 22.89***
Work attitude .17 .36
Wilks lambda .58 .74
F-value 5.76*** 2.79**
n2 .42 .26
*p < .05
** p < .01

***p < .0001

62

Table 13. MANOVA Results for How Ofien Decisions are Made in the Family and

Adjustment

 

F-tests with (1,96) df

How ofien decisions
are made in the family

Fit in how ofien
decisions are made

 

in the family
F F

Depression 1.09 3. 10
Connectedness to parents .00 2862*

Self-esteem: Academic .81 1.93

Friendship .91 3.24

Intelligence .34 1 .54

Social .58 .00

Global .73 1.84

GPA .14 .77

Behavior misconduct .22 .06
Insecurity with parents .33 2754“

Work attitude .63 .00

Wilks lambda .97 .64
F-value .25 448*

n2 .03 .36

 

*p < .0001

63

Table 14. Correlations between Allowed Autonomy and Indicators of Fit

 

Allowed Behavioral Autonomy
How are decisions made .43
How often are decisions made .37
Fit in Autonomy Timetable .61

 

Note: All correlations significant at p < .0001

Adolescents who were allowed to engage in autonomous behaviors at an earlier age fit
better into parental expectations compared to adolescents who were allowed behavioral
autonomy at a later age. It seems as if adolescents wanted more autonomy at an earlier
age than their parents allowed.

However, the mean differences were too small to translate this into more
meaningful terms. For example, the mean age for when certain autonomous behaviors
were allowed was 3.13 for the poor fit group and 2.62 for the good fit group. While this
was significant, it was not possible to distinguish whether there was an actual age
difierence between the two groups. A 2 on the timetable scale indicated the adolescent
was allowed to engage in this autonomous behavior when they were 14 - 15 years of age.
A 3 indicated they were allowed to engage in the behavior when they were 16 - 17 years
of age. The scale was not specific enough to determine the age allowed to engage in
certain activities that would have distinguished adolescents in the poor or good fit group.
Nevertheless, as reported earlier, there were moderate and significant effect sizes between
fit and timetable of autonomy.

Interestingly, more emotional autonomy was related to poor fit, but earlier

64
behavioral autonomy was related to good fit.

Hierarchical multiple regression analysis was used to examine how emotional
autonomy and connectedness related to fit. Evaluation of assumptions of normality,
linearity, and homoscedasticity of residuals were satisfactory, and no outliers were found.
The variables of emotional autonomy and connectedness to parents were entered
simultaneously. Table 15 displays results that show that fit was significantly predicted by
both emotional autonomy and connectedness, again confirming previous analyses that
more emotional autonomy was related to poorer fit, and conversely, more connectedness
was related to better fit.

AcmlfirraricnandAdjusrmem

To test whether acculturation level related to adolescent adjustment, MANOVA
was used. The four acculturation styles were not found to be significantly related to any
of the adjustment variables. Subsequently, analyses focusing on the two dimensions of
acculturation, namely, how strongly the adolescent endorsed Asian and Western values
and behaviors were performed.

Correlational analyses, reported in Table 16, indicate that several adjustment
variables are related to the level of involvement in Asian and Western cultures.
Interestingly, level of depression and global self-esteem are negatively related to
involvement in Asian culture, yet positively related to involvement in Western culture. In
other words, late adolescents who reported being more Asian experienced a higher level of
depression and lower self-esteem compared to those who reported being less Asian. In

contrast, the more Western adolescent experiences a lower level of depression and higher

65

Table 15. Multiple Regression of Emotional Autonomy and Connectedness Predicting
Level of Fit

 

Variables Emotional Emotional B [i
Autonomy Connectedness

 

Emotional Autonomy .31 .27*
Emotional Connectedness -.68** -.20 -.29*
Level of Fit .47** -.48**

Intercept .23

R = .52**

R2 = .27

Adjusted R2 = .25

 

*p< .05
** p < .0001

66

 

#0. V Q **
8. v a ..
Ion. oo. 2. S. 0:363 x53
3 .- no. 8.- vo. 3:88 5:5 3:835
8. no. 8. mo. 83:58:: 53205
No. 8.. mo. 3.. «do
awn. 2 .- 2. 2.- 1320
*mm. 2 .- E. 2 .- bans Eoom
_ H . 2 .- 8.. ac: oo:ow=_8£
*mN. 2.. 2. 3 .- mgotm 820
m H. S .- mo. 3 .- 01:033. ”Eooumofiom
E. Z.- No.- 3. 3:88 8 30580288 figuofim
1mm- :3. *vmr .21-N. Santana
E883 :33: E88? 574

325:3 :_ m:_wum:m

82? mo EoEoEOgm

 

20833.94. :5 5:83:63. 5253 80:20:00 .2 03:9.

67
self-esteem than those who are not as Western. Endorsement of Asian attitudes was
inversely correlated with endorsement of Western attitudes (1; = -.53, p < .001). Also,
engaging in Asian behaviors was also inversely correlated with engaging in Western
behaviors (1: = -.46, p < .0001).

Correlational analyses indicated that there is a relationship between how recently
the adolescent’s parent moved to the US. and how strongly the adolescent endorsed
Asian values and attitudes. The longer the father or mother lived in the US, the less
strongly the adolescent endorsed Asian values and attitudes (I = -.24, p <.05, [ = -.21, p <
.05, respectively). The length of time the adolescent’s parents lived in the US. did not
significantly relate to how strongly the adolescent endorsed Western values.

Generational status also affected how strongly the adolescent endorsed Asian
values and Asian behaviors. A one-way ANOVA was conducted to compare adolescents
who were born in the US. to adolescents who immigrated here with their parents. It is
somewhat surprising that adolescents born in the US. more strongly endorsed Asian
values and attitudes compared to adolescents who have immigrated to the US. (E (1,97) =
5.38, p < .05). Furthermore, U.S. born Asian adolescents also were more engaged in
Asian behaviors compared to foreign-bom adolescents (E (1,97) = 3.97, p < .05).
Generational status did not significantly affect the adolescent’s involvement in Western
attitudes and behaviors.

Emotional autonomy was significantly correlated with how long the adolescent’s

mother had lived in the US. Adolescents who reported more emotional autonomy had

68
mothers who had lived in the US. longer (1; = .22, p < .05). The correlation between
adolescents’ emotional autonomy from parents and the length of time their father lived in
the US. approached significance (r = .20, p = .059). The length of time the adolescent’s
parents lived in the US. did not significantly relate to how much autonomy was allowed
to the adolescent (the timetable).

Adolescent fit into parental timetable autonomy expectations was related to the
length of time the adolescents’ parents have lived in the US. It was found that
adolescents better fit into parental expectations regarding autonomy (the fit index used -
“how ofien does your family include you when making decisions that concern you”) the
longer their parents have lived in the US (for fathers, r = -.28, p < .01; for mothers, 1 = -

.24, p < .05).

Chapter 6

DISCUSSION

The goal of this study was to investigate autonomy and connectedness in Asian-
American late adolescents. Given the significance and impact of context to development,
autonomy was examined in relation to family and cultural demands. Additionally, this
study examined the relationship between late adolescent adjustment and the level of fit
they experienced in these two contexts.

r i n h Tim 1 A n m

Feldman and Rosenthal (1990; 1991) found that individuals from Asian cultures
(e. g., adolescents living in Hong Kong and Chinese adolescents living in the US. or
Australia) traditionally do not promote or engage in autonomous behaviors as early as
individuals from Western cultures. In the present study, late adolescents who were more
Asian in their values and attitudes experienced less emotional autonomy, a later behavioral
autonomy timetable and, furthermore, preferred a later timetable than those less Asian.
These findings support the notion that the adolescent’s cultural values and attitudes play a
role in determining the timing and degree of autonomy.
WWW

Another question explored was how emotional and behavioral autonomy related to
aspects of adolescent fimctioning. In this study, late adolescents who were more
connected to their parents reported higher levels of self esteem and lower levels of

insecurity with parents. The finding that connectedness with parents is associated with

69

70
positive adjustment is not surprising, given past research on the significance of family
support (Steinberg, 1990).

An inverse relationship was found between emotional autonomy and
connectedness to parents. That is, more emotionally autonomous adolescents experienced
less connectedness with their parents. Conceivably, the more autonomous adolescents
become, the more their Westemized values and behaviors conflict with their parents’
values which may be more Asian7. It could be likely, then, that the adolescent perceives
less parental support, and, perhaps more intergenerational conflict. Conversely, it is also
plausible that the lack of connection at home could drive adolescents into the larger
society more quickly, hence, leading them to become more autonomous.

More emotional autonomy has been linked to both positive and negative outcomes,
for example, healthier identity development (Frank et al., 1990), greater substance abuse
(Turner, Irwin & Millstein, 1991), and a combination of healthier identity development
and psychological health along with more substance abuse (Frank & Burke, 1992).
Fuhrman and Holmbeck (1995) argue that emotional autonomy may be adaptive for
adolescents who do not have a supportive family environment, yet, detrimental for
adolescents who do have a supportive family environment. Research has also documented
negative outcomes associated with too much or premature autonomy given to the
adolescent (Stessa & Steinberg, 1991). Taken together, these studies paint a complex

picture of the meaning and significance of autonomy.

 

7Of course, an assessment of parental acculturation is needed to confirm the
speculations that within immigrant families, parents are more traditional than their
children.

7 1

In this study, emotional autonomy was associated with negative outcomes while
behavioral autonomy was associated with both negative and positive outcomes. More
emotional autonomy directly related to behavior misconduct, insecurity with parents and
poorer fit with parental expectations of autonomy. Earlier behavioral autonomy, on the
other hand, directly related to less insecurity with parents and better fit in parental
expectations of autonomy. Conceivably, granting behavioral autonomy at an earlier age
creates a sense of responsibility and a sense of being trusted in the adolescent. In contrast,
high levels of emotional autonomy may indicate an unhealthy emotional distancing from
parents, as Ryan and Lynch (1989) would argue.

In this study, earlier behavioral autonomy was associated with lower levels of self-
esteem regarding intellectual ability and the ability to make close friends. It may be that
earlier behavioral autonomy is beneficial for the parent-adolescent relationship. However,
it may also leave room for adolescents to feel uncertain about their own abilities.

In sum, adolescents who are more emotionally autonomous from their parents do
not fare well. In contrast, adolescents who are more behaviorally autonomous experience
a good relationship with their parents. These findings demonstrate the importance of
examining multiple aspects of autonomy and their relation to different developmental
outcomes.

The context within which these autonomy levels are played out must be taken into
account. For instance, adolescents who experience a greater degree of emotional
autonomy from their parents may have parents who either trust and encourage them to

become competent individuals, or, who simply do not care, e.g., a characteristic of

72
Baumrind’s (1971) pemiissive parent. When the interaction between emotional autonomy
and connectedness was examined, a more complex relationship to the insecurity
adolescents experience with their parents, wasfound. The main effects of autonomy and
connectedness were reduced to nonsignificance when specific interactions were
considered. Findings indicate that either high or low autonomy in the context of a
supportive relationship relates to lower insecurity with parents. This suggests that the
family environment may act as a moderator between the level of emotional autonomy
adolescents experience and their subsequent functioning.

To test further the interaction between emotional autonomy and connectedness,
Lambom and Steinberg’s (1993) autonomy and connectedness classifications, e.g.,
individuated, connected, detached, ambivalent were used. These classifications were
somewhat successful in identifying differences in predicting two outcomes - insecurity in
the parent-adolescent relationship and work attitude, an aspect of psychosocial maturity.
Connected adolescents were less insecure in their relationships with their parents
compared to ambivalent and detached adolescents. Furthermore, connected adolescents
reported more appropriate work attitudes than ambivalent adolescents.

These results did not replicate Lambom and Steinberg’s finding (1993) that
individuated adolescents were the most psychologically and academically competent. One
explanation could be the differences in the age of the two samples. Lambom and
Steinberg’s sample consisted of high school adolescents. Perhaps at this stage it is more
important for adolescents to be somewhat emotionally autonomous from their parents so

that they can be free to develop their abilities and competencies. In contrast, the late

73
adolescents of this study attend college and have many more opportunities to be
independent than when they were in high school. Hence, maintaining roots to their family
(being connected) to balance this greater autonomy may be more adaptive at this time.
Fi ' n x i n n ' n

Another research question explored how well late adolescents fit into their parents’
autonomy expectations. Depending on the developmental stage of the adolescent, optimal
levels of autonomy from and connectedness to parents will change. Parents then have the
responsibility to respond sensitively to the changing needs of the adolescent by allowing
more freedom at certain times and offering more support at others (Eccles et al., 1991).
Thus, instead of generalizing about the optimal level of autonomy parents should grant
their adolescent, the fit between how much autonomy the adolescent desires and how
much their parents are willing to allow should be considered.

This study supported the use of the goodness of fit model as a means to
understand adolescent functioning. Three indicators of fit in the family were used. The
first indicator, fit in autonomy timetable expectations, successfully predicted depression,
several areas of self-esteem, the quality of the parent-adolescent relationship, behavior
misconduct, and work attitude. Hypotheses based on the goodness of fit model were
confirmed. Namely, poor fit was associated with negative outcomes whereas good fit was
associated with positive outcomes. The knowledge of fit in autonomy timetables provided
more information regarding adolescent adjustment compared to a simple examination of
the adolescent’s actual timetable. That is, taking into account both parental expectations

and adolescent’s desired expectations proved to be more useful in understanding

 

74
adolescent functioning.

The structural equation analyses suggest a causal relationship between good fit and
positive adjustment. However, it is also plausible that the security and closeness that
adolescents enjoy with their parents contributes to the adolescent’s good fit with parental
autonomy expectations. Or, adolescents who have an insecure and distant relationship
with their parents may demand more freedom to engage in activities outside the family.
This desire for more autonomy may not match what their parents find appropriate.

The second indicator of fit - how are decisions made in the family that concern the
adolescent - also predicted adolescent adjustment. This was a better predictor compared
to the third indicator of fit - how ofien are decisions made in the family that include the
adolescent. This is not surprising when considered in the goodness of fit framework.
Some adolescents do not care if they are not included in the family decision-making
process, even if the decision concerns them. Instead, they are content allowing their
parents to take control. Consequently, it would not matter to the adolescent how often he
or she was included in the decision-making. The findings suggest that the process of
decision making (e. g. haw decisions are made) is more important to adolescents than the
number of instances that they are included.

In sum, each of the three different indicators of fit (behavioral timetable
expectations, how are decisions concerning the adolescent made in the family, and, how
often are they included in decisions that affect them) were usefirl predictors of late
adolescent adjustment, yet some to a greater extent than others. Each adolescent differs

cognitively, socially, and biologically, and each may demand different timetables and

75
degrees of autonomy that are developmentally appropriate and optimal. Each adolescent
also has unique parents with unique expectations that result in differences in expectation
fit for each adolescent-parent dyad. Researchers who adopt the goodness of fit model
take these individual difl‘erences into account.

The results demonstrate that not all Asian-American late adolescents experience
poor fit with parental expectations. In other words, even though cultural variations exist
within immigrant families, the values and beliefs that family members hold may still be
consistent across generations.

AminrrafionjndAdjustmem

Since issues of autonomy are embedded in the attitudes, values and behaviors
characteristic of each culture, the four acculturation styles (Beny, 1980) in relation to
adolescent adjustment were investigated. Research has reported the integrated
acculturation style to be the most adaptive (LaFromboise et al., 1993; Szapocznik &
Kurtines, 1980). However, in this study, none of the four styles significantly related to
any late adolescent adjustment variables. Perhaps these classifications were too complex
to uncover the relationship between acculturation and outcomes, as they assume an
interaction between involvement in Asian attitudes and involvement in Western attitudes.
In this study, as in Nguyen’s study (1995) cited earlier, a two dimensional model of
acculturation that assesses involvement in each culture separately, was a better predictor
of adolescent adjustment than the four acculturation styles. Therefore, instead of using
the four classifications of acculturation, individuals were examined according to how

involved they were in Western culture and Asian culture, separately.

76

Several researchers contend that it is adaptive for adolescents to adopt some of the
majority culture’s values and attitudes. For example, a study conducted with a sample of
Chinese-Americans living in a Midwestern town found that more assimilated Chinese-
Americans reported less psychological distress symptomatology and lower levels of life
stress than those who were less assimilated (Yu, 1984; Yu & Har’ourg, 1980). Nguyen’s
(1995) study of middle-adolescent Vietnamese high school students reported that
adolescents who were more involved in Western culture reported lower levels of
depression and higher levels of self-esteem. In contrast, adolescents who were more
involved in Asian culture reported higher levels of depression and lower levels of self-
esteem. The results of the present investigation coincide with these findings. That is, late
adolescents who were more Western and less Asian experienced healthier outcomes.

In this particular university environment, a small minority (4%) of undergraduates
are Asian-American (this does not include international students who come from Asia to
study). Despite these small numbers, Asian-American undergraduates have access to a
variety of supportive services specifically geared toward Asians. For example, there are
Asian minority aides in each dorm, an Asian representative at the Office of Minority
Student Affairs, and an Asian Pacific American Student Organization that sponsors
cultural events, dances, and conferences for Asian undergraduates.

Nevertheless, in spite of these supportive services, it may still be somewhat
difficult for students to find opportunities that reinforce Asian values and attitudes, or
participate in activities such as watching Asian films, or speaking an Asian language, as

there are so few Asians in the University and surrounding area. It is plausible that the

77
majority of people that Asian-American adolescents interact with on a daily basis (e. g.,
fiiends, peers, professors, store owners), are likely to be individuals endorsing Western
values. Embracing similar values may be advantageous in some respects. Indeed, more
Western late adolescents reported higher levels of self-esteem in the area of social
relations. These late adolescents reported more confidence in their ability to interact with
other people and to make fiiends. Moreover, they felt accepted by others, and felt they
had close fi'iends with whom they could share their personal thoughts and feelings.
Overall, these findings support the hypothesis that adolescents who fit into the prevailing
cultural context enjoy healthy outcomes.

Being immersed in Asian culture is not necessarily detrimental, i.e., adopting a
separation, or traditional, acculturation style (Phinney et al., 1990). However, the reason
the adolescent chooses to be more involved in Asian culture is important. Are they
choosing not to fit in with the majority culture because of their own preferences or are
they are not being allowed to fit? For example, some adolescents want to engage in
Western activities, such as hanging out with Caucasian-American fiiends and going to
Caucasian-American parties and gatherings, but cannot because they are excluded, for
instance, by their peers. In this sample, a majority of the participants (84%) reported
feeling discriminated against because of their ethnicity. Asian adolescents who face
discrimination from peers or adults of the majority culture may then turn to the Asian
community for acceptance. Whether the lack of fit to the majority culture is or is not by
choice has implications for the adolescent’s well-being.

To summarize, in this sample, being highly involved in the ethnic culture may not

78
always be psychologically adaptive. In a relatively homogenous context where Western
culture predominates, being “too Asian” may be detrimental. However, before making
any definitive conclusions about the existence. of a direct inverse relationship between
involvement in one’s ethnic culture and adjustment, an investigation into the mechanisms
of this linkage is necessary. There may be some very Asian adolescents who fimction
quite well in a Western setting. It would be important to test potential moderators or
mediators (e.g., experiences of discrimination) between acculturation and outcomes.

The acculturation of adolescents did not significantly relate to how well late
adolescents fit into their parent’s autonomy expectations. Therefore, the investigator’s
prior assumption that Asian parents are more traditional than their late adolescents, may
be false. If parents were indeed more Asian, it would be expected that late adolescents
who were more Asian would experience better fit with parental autonomy expectations.
However, this was not found.

QmaHMQdel

The structural equation analyses offered support for the full autonomy model that
linked autonomy, fit in the family, and acculturation, to adolescent fimctioning. This
model predicted that the level of autonomy granted (emotional and behavioral) was
associated with late adolescent adjustment. In addition, this model predicted that fit in the
family regarding issues of autonomy, and fit in culture regarding general independence
expectations, were also related to adjustment. The full model better predicted late
adolescent adjustment than the other two submodels that examined autonomy in only one

Context In other words, examining the cultural context together with the family context is

79
important to uncovering paths to late adolescent well-being.

Although not central to the hypotheses of this study, the late adolescent’s
generational status and parents’ length of residence in the US. were examined in relation
to the adolescent’s acculturation levels, emotional autonomy, and fit with parent
autonomy expectations. Ting-Toomey’s (1981) study of first, second, third and fourth
generation Chinese American college students, found fourth generation Chinese-
Americans to have higher levels of ethnic identity than second or third generation Chinese-
Americans. Ting-Toomey argues that “the fourth generation seems to have searched back
for their roots rather than assimilated themselves toward the dominant white culture.” Lin
and Liu’s (1993) study of Chinese-American adult immigrants reported that the
immigrants were more Chinese than their non-immigrant parents. These researchers
contend that there was an “overcompensation” among the younger generation to
“reconnect with their cultural heritage.”

In this study, U.S. born Asian—American adolescents were more Asian in their
attitudes and behaviors compared to foreign born Asian-American adolescents, supporting
previous studies. Perhaps the US. born Asian-Americans felt it necessary to preserve
their cultural identity more so than foreign-bom Asian-Americans. The cultural identity of
the foreign-bom adolescents who have immigrated to the US. may be more fimrly
established. For example, they may be fluent in their native language and more familiar
with their traditional customs. Subsequently, they may not feel the need to actively pursue

Asian activities or join Asian groups to maintain their Asian identity. In contrast, U.S.

80

born Asian-Americans may seek out these activities and groups to reamrm and retain their
Asianness. Finding these generational difl‘erences highlights the diversity within this Asian
sample.

The fact that US. born Asian-Americans actively hold on to aspects of their
cultural heritage provides evidence for current models of acculturation (for example,
Berry, 1980). These models propose that individuals do not simply give up all aspects of
their traditional culture (i.e., completely assirnilating) while living in a culture different
from their own. But rather, they retain and modify certain aspects of their own culture’s
values and attitudes, and, concurrently, acquire new values and attitudes.

The length of time the late adolescent’s parents resided in the US correlated with
several factors. The longer the adolescent's father or mother lived in the US, the less
strongly the adolescent endorsed Asian values and attitudes. Conceivably, parents who
have lived in the US. for many years may be more Westemized, and thus, may not have
pushed their adolescent to strongly adhere to traditional Asian values.

Emotional autonomy was significantly correlated to the length of time the
adolescent’s mother, but not their father, lived in the US. Adolescents with mothers who
lived in the US. longer reported more emotional autonomy from their parents. Mothers
who have lived longer in the US. may have acquired more Western values, and thus, may
have promoted more emotional autonomy in their adolescent. Ifthis were true, it would
support the notion that increased exposure to an individualistic culture such as the US.
encourages individuals to adopt increasingly autonomous attitudes. However, without an

assessment of the mother’s level of acculturation, this is only speculation.

 

8 1

The longer the adolescent’s parents lived in the U.S., the better the adolescent fit
into parental autonomy expectations. Again, parents who have lived in the US. longer
may have adopted more Westemized autonomy expectations. Thus, their adolescents
might better fit their expectations compared to adolescents with parents that have been in
the US. for a shorter time.

Explanations of the associations just reported are speculations. The length of time
the parent has lived in the US. is not necessarily a marker for the level of acculturation the
parent has undergone. There will most likely be some parents who will remain very
traditional while others will become very Americanized. Thus, there is not enough
evidence to ofi‘er concrete explanations of these additional findings.

The sample consisted of various Asian groups representing 6 Asian countries.
Although these groups share commonalities (e. g., a general geography, a collectivistic
orientation, Confucian philosophy), each is distinguished by its specific history, customs,
and language. The sample size for each group was not sufficient to allow for between-
group comparisons. By combining and analyzing all the groups as a whole, unique group
differences are lost. Future studies should examine these specific Asian groups separately
to discover whether autonomy development proceeds in a similar manner for each group.
Nonetheless, this study is still useful in understanding autonomy development for Asian-
American adolescents in general.

One limitation of the study involved several of the measures. Participants were

asked to respond to questions pertaining to their relationships with their parents.

82
However, a distinction between the mother and father was not made. Some participants
expressed difficulty in answering these questions because they experienced very difi‘erent
relationships with each parent.

The importance of exploring mother-adolescent and father-adolescent relationships
separately is revealed in findings that demonstrate that these relationships predict different
adolescent outcomes. For example, Feldman and Wood (1994) reported that while
mothers and fathers held similar autonomy timetable expectations for their preadolescent
sons, only father expectations predicted their sons’ later adolescent behavior regarding
school performance, motivation, and social misconduct. The autonomy and
connectedness measures used in this study did not capture potential differences between
the predictive ability of the mother-adolescent versus the father-adolescent relationship for
adolescent outcomes.

I ' ' r r h

It would be premature to suggest that Asian-American adolescents should be
encouraged to adopt more Western and less Asian values and behaviors in order to
promote more positive adjustment. Instead, fiiture research should aim to uncover why
this relationship exists. What are the specific processes that link the endorsement of Asian
attitudes to more depression or lower self-esteem? Is it because the environment is not
accepting of those who are not as Western? Is it because the adolescent is viewed as
being too different?

This study focused on parental expectations of autonomy. It would be interesting

to examine peer expectations as well. For example, how well do adolescents fit into

83

parent versus peer expectations? Do parental and peer autonomy expectations coincide?
Ifnot, does poor fit into parental expectations prompt adolescents to find fiiends with
autonomy expectations more similar to their own? Furthermore, would Asian peers have
difi‘erent autonomy expectations compared to other peer groups?

Greenberger and Chen (1996) recommend researchers to investigate the timing of
mismatching expectations. In other words, they want to know when a mismatch between
adolescent and parent expectations begins. Based on their findings, they contend that this

mismatch may occur in late adolescence when conflicts over autonomy arise. These

 

researchers compared Asian-A..-.icau to Eur- , -A...-.icau early and late adolescents
(college students) on depressed mood. They found no difference when comparing early
adolescent Asian-Americans to European-Americans. Whereas, by late adolescence,
Asian-Americans reported having significantly more depressed mood than European-
Americans. When parent-adolescent interactions were taken into account, these ethnic
differences disappeared.

The researchers propose that the difference in depressed mood emerging in late
adolescence may signify difficulties in attaining autonomy in the home. The nature of the
conflicts involved disputes over the late adolescent’s “habits and routines, choice of
friends and the activities engaged with friends, and privacy with respect to telephone calls
and letters,” all aspects of autonomy. The researchers argue that “late adolescent efforts
at establishing autonomy are delayed in Asian American families perhaps because of their
later expected age of independence in various domains. These efforts are less supported in

Asian-American families because of cultural expectations for respect and obedience.

84
Asian Americans are more involved in a struggle for control than European-Americans.”

Eccles and her colleagues (1991), on the other hand, assert that a mismatch in
expectations may begin in early adolescence. They reason that, at this time, early
adolescents gain increasing opportunities for unsupervised interactions with their peers.
These peer relationships expose the adolescent to relationships that are equal in power and
authority. As a result, this may lead the adolescent to increasingly expect the same
equality at home. Thus, the early adolescent may push for more autonomy than their
parents are ready to allow at this developmental stage.

The current study was conducted in the Midwest region of the US. It is plausible
that different findings would emerge if this study were conducted in areas with much
stronger Asian communities (e. g., in terms of population, or political influence) such as
California or New York, or even other areas in the Midwest such as Chicago. For
instance, perhaps Asian-American adolescents in these cities would not experience lower
self-esteem or depression if they chose to immerse themselves in Asian culture because of
the greater support network available.

Another avenue for firture research would involve an investigation into the timing
of puberty and its association with the allowance of autonomous behaviors. The
adolescent’s rate of maturation may affect how people react to him or her (Tanner, 1991).
For instance, adolescents who experience the onset of puberty at an earlier age may be
allowed to engage in autonomous behaviors earlier compared to the adolescent who
remains physically and biologically more immature for a longer period of time, simply

because the early maturing adolescent looks older. Although maturational timing was not

85

found to significantly relate to autonomy timetables in adolescent immigrants from Eastern
Europe (Schmitt-Rodermund & Silbereisen, 1996), future research is needed to confirm

these findings in other populations such as Asian-Americans.

 

Chapter 7

CONCLUSION

All adolescents deal with issues of autonomy. However, the timing, difficulty, and
resulting outcomes of this task will vary for each adolescent, depending on their individual
characteristics, parental expectations, cultural expectations, and community
characteristics. For some adolescents there will be an easy transition to independence,
while for others it will be a hard-fought struggle.

It is proposed that adolescents’ autonomy from their parents, connectedness to
their parents, and resultant adjustment, can only be understood in the context of the
prevailing culture(s). The developmental goals that parents have for their adolescents
varies across cultures (Greenfield, 1994). These developmental goals will shape their
parenting beliefs, expectations, and behaviors (Goodnow & Collins, 1990). When parents
immigrate to another culture that promotes developmental goals different from their own,
their adolescents may experience incongruent, sometimes conflicting, ideals (e.g.,
regarding autonomy timetables) between those of their parents and the larger society. The
minority adolescent, then, is challenged to successfully navigate through these different
world views.

The findings of this study have underscored the importance of individual-

86

87
environment fit. Rather than generalizing about the optimal timing or degree of autonomy
adolescents should be granted, each adolescent must be assessed in relation to their
context. The goodness of fit model and the concept of acculturation highlight the
significance of the interplay between adolescents and their environmental demands.
Future research should continue to examine how adolescents fit into familial and cultural
expectations of autonomy. In doing so, we will gain a fuller understanding of their

journey towards autonomy into the larger society.

APPENDICES

 

APPENDIX A

APPENDIX A

CONSENT FOR PARTICIPATION FORM

This is a form asking you to participate in a research study on the development
of autonomy (independence) in Asian-Americans. The purpose of this study is to gain
a better understanding of autonomy and how this relates to family relationships in
Asian-Americans. You will be asked to fill out a set of questionnaires.

Your participation is completely voluntary and you will be free to refuse or stop
at any time without penalty. All the information you provide will be number-coded to
insure complete anonymity. There will be no way for anyone (including the principal
investigator) to associate your responses with your identity. The results from this study
will be treated with strict confidence and the participants will remain anonymous in any
report of research findings. With these restrictions, the results of this investigation can
be made available to participants upon request.

As a thank you for participating, you can enter a drawing after completing the
questionnaire by writing down your name and address where we can reach you. At the
end of data collection, three participants will be randomly chosen and two will receive
a $50 certificate, the other two will receive a $25 certificate.

If you have any questions (now or at a later time), please feel free to contact me:

Linda J uang, Principal Investigator
Department of Psychology
Michigan State University

129 Psychology Research Building
East Lansing, MI 48824

Phone: (517) 432—3843
E-mail: juanglin@pilot.msu.edu

Thank you for your time.

 

Please read the following. If you agree to participate, please sign below.

I understand what participation in this study involves, that any information
about me obtained from this research will not be traceable to me, and that I am free to
withdraw from participating at any time.

Signature Date
Investigator Date

 

 

88

 

 

APPENDIX B

 

APPENDIX B

COVARIAN CE MATRIX OF FULL LISREL MODEL

GPA

Insecurity with Parents

Intelligence Esteem
Work Attitude
Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable
Timetable Fit

How Decisions Fit
How Often Fit

Social Esteem
Friend Esteem
Intelligence Esteem
Work Attitude
Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable
Timetable Fit

How Decisions Fit
How Often Fit

GPA
0.222
-0.010
-0.020
-0.040
0.054
0.116
0.017
-0.042
0.069
0.039

-0.010
0.006
0.026

0.037
0.008
0.006
-0.015
-0.032

Social
Esteem
0.566
0.382
0.164
0.071
0.043
-0.050
0.095
-0.089
-0.015
0.083
0.008
-0.043

-0.042

Insecurity
with

Global

Parents Misconduct Depression Esteem

0.296
0.010
0.112
—0.125
-0.146
-0.102
—0.158
—0.120
-0.055
-0.015
0.009
-0.041
0.028
0.080
-0.156
0.157
0.103
0.191
0.234

Friend
Esteem

0.633
0.162

0.042
-0.053
0.114

-0.028
0.107
-0.109
-0.056
—0. 127
-0.153

0.172
0.067 0.292
—0.013 -0.239
-0.048 -0. 192
-0.021 -0.241
—0.006 —0.222
-0.046 -0. 192
-0.037 -0.075
0.001 —0.052
0.005 0.065
0.016 -0.099
0.022 0.106
0.031 0.022
—0.021 -0.051
-0.005 0.036
0.014 0.014
0.044 0.097
—0.009 0.067
Intelligence Work
Esteem Attitude
0.391
0.091 0.148
-0.006 0.027
—0.026 0.001
0.039 0.064
-0.065 0.001
—0.017 -0.009
0.057 0.033
-0. 101 -0.027
-0.042 -0.020
-0.082 —0.001
-0. 133 -0.009

89

0.372
0.265
0.298
0.267
0.268
0.073
0.035
-0.041
0.087
-0.082

0.057
0.047
-0.037
-0.077
-0.110

Involve
U.S.

0.161
-0.094
0.147
-0.121

0.016

-0.009
0.023
-0.014
0.003

Academic
Esteem

0.432
0.184
0.172
0.291
0.105
0.020
—0.052
0.055
-0.079
0.000
0.054
-0.076
-0.036
0.065
-0.086

Involve
Asian

0.197
—0.111
0.206

0.042
0.063
-0.011
-0.014
0.013

 

COVARIANCE MATRIX (Cont’d)

Behave Behave Emotional Autonomy Timetable
U.S. Asian Autonomy Connectedness Timetable Fit

Behave U.S. 0.316
Behave Asian -0. 177 0.477
Emotional Autonomy 0.001 -0.005 0.102
Connectedness 0.042 -0.038 -0. l 14 0.276
Autonomy Timetable -0.021 0.031 0.031 0.077 0.326
Timetable Fit 0.012 -0.020 0.055 0.091 0.127 0.134
How Decisions Fit -0.017 -0.087 0.104 -O.188 0.185 0.123
How Often Fit -0.043 0.058 0.103 -0.268 0.182 0.152
How How
Decisions Often
Fit Fit
How Decisions Fit 0.565

How Often Fit 0.350 0.757

APPENDIX C

APPENDIX C

CORRELATED ERROR TERMS FOR FULL LISREL MODEL

THETA-EPS
Insecurity
with Global Academic
GPA Parents Misconduct Depression Esteem Esteem

GPA 0.184
Insecurity with Parents - - 0.191
Misconduct - - - - 0,097
Depression - - - - - - 0036
Global Esteem - - 0.022 0.039 0.058 -0.015
Academic Esteem 0.057 - - - - - - - - 0,197
Social Esteem - - - - - - - - - - - -
Friend Esteem -0.047 - - - - - - - - - -
Intelligence Esteem - - - - - - - - - - 0,075
WorkAttitude -- -- -- -- -0.032 --

Social Friend Intelligence Work

Esteem Esteem Esteem Attitude
Social Esteem 0.282
Friend Esteem 0.133 0.342
helligence Esteem - - - - 0.168
Work Attitude - - - - - - 0.093
THETA-DELTA-EPS

Insecurity
with Global Academic

GPA Parents Misconduct Depression Esteem Esteem
Involve U.S. - - - - - - - - - - - -
Involve Asian - - - - - - - - 0,019 .. -
BehaveU.S. -- -- -- -- -- _-
BehaveAsian -- -- -- -- -- --
Emotional Autonomy - - 0.066 — - - - - - 0.011
Connectedness 0.028 -0.103 - - - - - - - -
Autonomy Timetable - - 0.099 «0.011 - - - - - -
Timetable Fit - - 0.078 - - -0.009 - - - -
How Decisions Fit - - 0.140 - - 0.025 — - - -
How Often Fit - - 0.167 -0.016 - - - - - -

91

THETA-DELTA-EPS (Cont’d)
Social Friend Intelligence Work
Esteem Esteem Esteem Attitude
Involve U.S. - - - - -0.013 0.029
Involve Asian - - - - 0.014 - -
Behave U.S. - - - - - - 0.058
Belnve Asian - - - - - - - -
Enotional Autommy - - - - - - - -
Connectedness - - - - - - - -
Autonomy Timetable 0.049 - - -0.035 - -
Timetable Fit - - - - - - - -
How Decisions Fit - - - - - - 0.023
How Often Fit 0.055 - - - - - -
THETA-DELTA
Involve Involve Behave Behave Emotional
U.S. Asian U.S. Asian
Involve U.S. 0.096
Involve Asian - - 0.004
Behave U.S. 0.068 - - 0.188
Behave Asian - - - - -0.016 0.207
Emotional Autonomy - - — - - - 0.012 0.025
Connectedness - - - - 0.014 -0.043 - -
AutonomyTimetable -- -- -- -- --
Thetable Fit - - - - - - - - - -
How Decisions Fit - - —0.046 - - -0.130 - -
HowOftenFit -- -- -- -- -0.044
How How
Autonomy Timetable Decisions Often
Timetable Fit Fit Fit
Autonomy Timetable 0.164
Timetable Fit 0.049 0.063
How Decisions Fit - - - - 0.277
How Often Fit - - - - - - 0.265

92

Autonomy Connectedness

APPENDIX D

APPENDIX D

PARAMETER ESTMATES FOR FULL LISREL MODEL

LAMBDA-Y

GPA

Insecurity with Parents
Misconduct
Depression

Global Esteem
Academic Esteem
Social Esteem

Friend Esteem
Inelligence Esteem
Work Attitude

IAMBDA-X

Acculturation Autonomy

Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable
Timetable Fit

How Decisions Fit
How Often Fit

GAMMA

Adjustment

0.095
-0.185
-0. 147
-0.450

0.570

0.393

0.394

0.336

0.404

0.163

 

0.182
-0.409
0.198
-0.417

Fit inthe

Acculturation Autonomy Family

Adjustment

 

0.416

-0.594

0.416

93

APPENDIX E

APPENDIX E

COVARIAN CE MATRIX OF LISREL SUBMODEL 1

Insecurity
with Global Academic
GPA Parents Misconduct Depression Esteem Esteem

GPA 0.222
Insecurity with Parents -0.010 0.296
Misconduct -0.020 0.010 0.172
Depression -0.040 0.112 0.067 0.292
Global Esteem 0.054 -0. 125 -0.013 -0.239 0.372
Academic Esteem 0.116 -0.146 -0.048 -0.192 0.265 0.432
Social Esteem 0.017 -0.102 -0.021 -0.241 0.298 0.184
Friend Esteem -0.042 -0. 158 -0.006 -0.222 0.267 0.172
Intelligence Esteem 0.069 -0. 120 -0.046 -0.192 0.268 0.291
Work Attitude 0.039 -0.055 -0.037 -0.075 0.073 0.105
Emotional Autonomy 0.000 0.080 0.031 0.022 -0.009 0.000
Connectedness 0.037 -0. 156 -0.021 -0.051 0.057 0.054
Autonomy Timetable 0.008 0.157 -0.005 0.036 —0.047 -0.076
Timetable Fit 0.006 0.103 0.014 0.014 -0.037 -0.036
How Decisions Fit -0.015 0.191 0.044 0.097 -0.077 -0.065
How Often Fit -0.032 0.234 —0.009 0.067 -0.110 -0.086
Social Friend Intelligence Work Emotional
Esteem Esteem Esteem Attitude Autonomy Connectedness
Social Esteem 0.566
Friend Esteem 0.382 0.633
Intelligence Esteem 0.164 0.162 0.391
Work Attitude 0.071 0.064 0.091 0.148
Emotional Autonomy -0.015 -0.028 -0.017 -0.009 0.102
Connectedness 0.083 0.107 0.057 0.033 -0.114 0.276
Autonomy Timetable 0.008 -0.109 -0.101 -0.027 0.031 -0.077
Timetable Fit -0.043 -0.056 -0.042 -0.020 0.055 -0.091
How Decisions Fit -0.060 -0.127 -0.082 -0.001 0.104 -0.188
How Often Fit 0.042 —0. 153 —0.133 -0.009 0.103 -0.268
How How
Autonomy Timetable Decisions Often
Timetable Fit Fit Fit
Autonomy Timetable 0.326
Timetable Fit 0.127 0.134
How Decisions Fit 0.185 0.123 0.565
How Often Fit 0.182 0.152 0.350 0.757

94

 

APPENDIX F

APPENDIX F

APPENDIX F

COVARIAN CE MATRIX FOR LISREL SUBMODEL 2

Friend Esteem
Intelligence Esteem
Work Attitude
Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable

Social Esteem
Friend Esteem
Intelligence Esteem
Work Attitude
Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable

Behave U.S.

Behave Asian
Emotional Autonomy
Connectedness
Autonomy Timetable

Insecurity
with
Parents

0.296
0.010
0.112
-0.125
-0. 146
-0.102
-0.158
—0. 120
-0.055
-0.015

-0.041
0.028
0.080

-0. 156
0.157

Esteem

0.633
0.162

0.042
—0.053
0.114

—0.028
0.107
-0.109

Behave
Asian

0.477
-0.005
-0.038

0.031

Misconduct Depression

0.172
0.067
-0.013
-0.048
-0.021
-0.006
-0.046
—0.037
0.001
0.005
0.016
0.022
0.031
-0.021
-0.005

0.292
-0.239
-0. 192
-O.241
-0.222
-0.192
-0.075
-0.052

0.065
—0.099

0.106

0.022
-0.051

0.036

Intelligence Work
Esteem Attitude

0.391
0.091
-0.006
-0.026
0.039
-0.065
—0.017
0.057
-0.101

Emotional

0.148
0.027
0.001
0.064
—0.001
-0.009
0.033
-0.027

Global

Academic

Esteem Esteem

0.372
0.265 0.432
0.298 0.184
0.267 0.172
0.268 0.291
0.073 0.105
0.035 0.020
-0.041 -0.052
0.087 0.055
-0.082 -0.079
-0.009 0.000
0.057 0.054
—0.047 —0.076
Involve Involve
U.S. Asian
0.161
—0.094 0.197
0.147 -0. 1 ll
-0. 121 0.206
0.016 -0.040
-0.004 0.042
-0.009 0.063
Autonomy

Autonomy Connectedness Timetable

0.102
-0.114
0.031

95

0.276
-0.077

0.326

APPENDIX G

APPENDIX G

PARAMETER ESTIMATES FOR LISREL SUBMODEL 1

LAMBDA-Y

GPA

Insecurity with Parents
Misconduct
Depression

Global Esteem
Academic Esteem
Social Esteem

Friend Esteem
Intelligence Esteem
Work Attitude

LAMBDA-X

Emotional Autonomy
Connectedness
Autonomy Timetable
Timetable Fit

How Decisions Fit
How Often Fit

GAMMA

Adjustment

Adjustment

0.085
-0.209
-0. 133
0.485

Fit in the
Autonomy Family
0.251 - -
—0.418 - -
-0.255 0.444
- - 0.223
- - 0.472
- - 0.626
Fit in the
Autonomy Family
0.026 -0 229

96

APPENDIX H

APPENDIX H

PARAMETER ESTMATES FOR LISREL SUBMODEL 2

LALIBDA-Y

GPA
Insecurity with Parents

Friend Esteem
Intelligence Esteem
Work Attitude

LAMBDA-X

Involve U.S.

Involve Asian
Behave U.S.

Behave Asian
Enotional Autonomy
Connectedness
Autonomy Timetable

GAMMA

Adjustment

Acculturation

Adjustment
0.093
-0. 138
-0. 155
-0.451
0.573
0.392
0.418
0.330
0.399
0.163

 

Acculturation
0.182 - -
~0.436 - -
0.197 - -
-0.384

Autonomy

97

LIST OF REFERENCES

LIST OF REFERENCES

Ainsworth, M. & Ainsworth, L- (1953) Maasufingaecufihaandparsanaladiustmnt,

Toronto: University of Toronto Press.

Andujo, E. (1988). Ethnic identity of transethnically adopted Hispanic adolescents.
Seeial Wer_(, 33, 531-535.

Baltes, M.M. & Silverberg, SB. (1994). The dynamics between dependency and
autonomy: Illustrations across the life span. In D.L. Featherman, Lerner, R.M., &

Perlrnutter, M. (Eds), Life-Span Development and Behevier. Hillsdale, NJ:

Lawrence Erlbaum.

Baumrind, D. (1971). Current patterns of parental authority. Develepmentel Psyehelegy
Men negraphs, 4( 1, Ban 2).

Berry, J.W. (1980). Acculturation as varieties of adaptation. In A. Padilla (Ed)

Aeenlmratien' lheery, medels and seme new findings, 9-25. Boulder, CO:

Westview.
Blos, P. (1979). Wm New York: lntemational Universities Press.

Bronfenbrenner, U. (1989). Ecological systems theory. Annals ef Child Develepment, 6,
187-249.

Chao, R.K. (1994). Beyond parental control and authoritarian parenting styles:
understanding Chinese parenting through the cultural notion of training. Child

WJHl-llm-

Cooper, C.R., Grotevant, H.D., & Condon, SM. (1983). Individuality and connectedness
in the family as a context for adolescent identity formation and role-taking skill. In

H.D. Grotevant & C.R. Cooper (Eds), Adeleseent develepment in the family.

New Directions for Child Development, no. 22. San Francisco: J ossey-Bass.

98

99

Copeland, Hwang, & Brody (1996). Asian-American adolescents: Caught between
cultures? Presented at the Society for Research on Adolescence, Boston, MA,
March.

Douvan, E. & Adelson, J. (1966). Wedge. New York: John Wiley
and Sons, Inc.

East, P.L., Lerner, R.M., Lerner, J.V., Soni, R.T., Ohannessian, C.M., & Jacobson, LP.
(1992). Early adolescent peer group fit, peer relations, and psychosocial
competence: A short-term longitudinal study. 1
11(2), 132-152.

Eccles, J. S. ,Buchanan, C..M ,Flanagan, C. ,Fuligni, A. ,Midgley, C. &Yee, D. (1991).

Control versus autonomy during early adolescence. W
53- 68.

Erikson, E. (1963). mmmdandjoatm WW. Norton & Co.; New York, NY.

Feldman, S.S., & Rosenthal, DA. (1990). The acculturation of autonomy expectations in
Chinese high schoolers residing in two Western nations. Intemmjenmm

M. 259-281.

Feldman, S.S. & Rosenthal, DA. (1991). Age expectations of behavioral autonomy in
Hong Kong, Australian and American youth: the influence of family variables and

adolescents' values. m' l l P hl 6 ,1-23.

Feldman, S.S. & Wood, D.N. (1994). Parents’ expectations for preadolescent sons’
behavioral autonomy: A longitudinal study of correlates and outcomes. Jenmalef

RasearahnnAdnlessanaaAfl). 45-70.

Frank, S. J. & Burke, L. (1992). Deidealization and autonomy in late adolescence.
Replications and extensions of earlier findings. In S. Frank (Chair), Adeleseem
? Symposium conducted at the meeting of
the Society for Research in Adolescence, Washington, D. C.

Frank, S.J., Pirsch, L.A., & Wright, V.C. (1990). Late adolescents' perceptions of their
relationships with their parents: Relationships among deidealization, autonomy,
relatedness, and insecurity and implications for adolescent adjustment and ego

identity status. Immalnfloutharrdflnlesemflfi), 571-587-

Fuhrman, T. & Holmbeck, G. (1995). A contextual-moderator analysis of emotional
autonomy and adjustment in adolescence. Child Develepment, 66(3 1, 793-811.

100

Goodnow, J. & Collins, AW. (1990). ' '
r f ’ ' . Hillsdale, NJ: Lawrence Erlbaum.

Greenberger, E. & Chen, C. (1996). Perceived family relationships and depressed mood
in early and late adolescents: A comparison of European and Asian Americans.

Dexelapmantallsychsflngyalzm, 707-716-

Greenberger, E., Josselson, R., Knerr, C., & Knerr, B. (1974). The measurement and
structure of psychosocial maturity. lenleQLXQutLandAdgleseeneeA, 127-
143.

Greenfield, P. (1994). Independence and interdependence as developmental scripts:
Implications for theory, research, and practice. In P. Greenfield & R. Cocking

(Eds), §;ress-enlmral reefs ef minen'ty enild deyelepment. Hillsdale, NJ:

Lawrence Erlbaum.

Grotevant, H. & Cooper, C. (1986). Individuation in family relationships: A perspective
on individual differences in the development of identity and role-taking skill in

adolescence. Human Develepment, 29, 82-100.

Hill, J. & Holmbeck, G. (1986). Attachment and autonomy during adolescence. In G.
Whitehurst (Ed), WW. Greenwich, CN: JAI Press.

Ho, D.Y.F. (1986). Chinese patterns ef seeializatien; A eritieal review. In M.H. Bonds
(Ed), The psychology of the Chinese people. Hong Kong: Oxford University
Press.

Hoffinan, J .A. (1984). Psychological separation of late adolescents from their parents.
Jenma] ef (lennseling Psyehelegy, 31(2), 170-178.

Hui, C.H. & Villareal, M.J. (1989). Individualism-Collectivism and psychological needs:

Their relationships in two cultures. Jeumal Qf £2ress-(2ultural Esyehelegy, 2013 ).
310-323.

Jereskog, K.G. & Sorbom, D. (1993). W. Chicago, IL:

Scientific Software lntemational.
Juang, L.P., Castellino, D.R., & Lerner, J .V. (1995). Early adolescent adjustment: Links
to the "goodness of fit" between temperament and contextual demands. Presented

at the Society for Research on Child Development, Indianapolis, IN, March.

Karpel, M. (1976). Individuation: From fusion to dialogue. Family Preeess, 15, 65-82.

101

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

Lambom, S.D., Mounts, N.S., Steinberg, L., & Dombusch, SM. (1991). Patterns of
competence and adjustment among adolescent from authoritative, authoritarian,

indulgent, and neglectful families. W, 1049-1065.

Lambom, S.D. & Steinberg, L. (1993). Emotional autonomy redux: Revisiting Ryan and
Lynch. Child Delelepmem, 64I 483-499.

Leaper, C. (1989). Discussant's comments: Self and other in theories of adolescent
psychosocial development. Society for Research on Child Development, Kansas
City, April.

Lerner, J. V. (1983). A "goodness of fit" model of the role of temperament in psychosocial
adaptation m early adolescents. MW, 149-157.

Lerner, RM. (1984). On the nature of human plasticity. mm
W. Cambridge, MA: Cambridge University Press.

Lerner, R. M (1993). A developmental contextual view of human development. In S. C.
Hayes, L J Hayes, H. W. Reese, & T. R. Sarbin (Eds), Varietiemflsciemjfie
Malian. Reno, NV. Context Press.

Lerner, R. M. & Lerner, J. V. (1983). Temperament and adaptation across life: Theoretical
and empirical issues In P. B. Baltes & O. G. Brim, Jr. (Eds),
maximum 197- 231

Lerner, R.M. & Lerner, J .V. & Zabski, S. (1985). Temperament and elementary school
children’s actual and rated academic performance: A test of the goodness of fit

model. mammammmwm 125-136.

Lin, C. & Liu, W T. (1993). Intergenerational relationships among Chinese immigrant

families from Taiwan. In H. P. McAdoo (Ed. ), W
dim, 271-286. Newbury Park, CA: Sage.

Loevinger, J. (1976). W San Francisco:

Jossey-Bass

Matsuoka, IX. (1990). Differential acculturation among Vietnamese refugees. Social
, 341-345.

102

Montemayor, R. (1983). Parents and adolescents in conflict: All families some of the time

and some families most of the time. WM, 83-103.
Neeman, J. & Harter, S. (1986). lf- i fil

smdem. University of Denver, CO.

Nguyen, RH. (1995). Bridging two worlds: Acculturation and adjustment in Vietnamese
adolescents. Unpublished thesis.

Nguyen, N A (1992). Living between two cultures. Treating first-generation Asian
Americans. In L. A. Vargas & J. D. Koss-Chioino (Eds. ) W
W

n1 r n ,
204-222. San Francisco: Jossey-Bass.

Padilla, A. M. ,Wagatsuma, Y. ,Lindholrn, K. J. (1984). Acculturation and personality as
predictors of stress in Japanese and J apanese~Amencans W11
W.) 295- 305

Phinney, J. S. (1990). Ethnic identity adolescents and adults. Review of research.
108 3 ,499- 514.

Phinney, J. S. ,Locher, B. T. & Murphy, R (1990). Ethnic identity development and
psychological adjustment in adolescence In A. Stifiinan & L. Davis (Eds ), Etlmie
mmmemmmmnn Newbury Park CA Sage-

Radlofi‘, L. S. (1977). A CES-D scale: A self-report depression scale for research in the
general population. AnphedfisychdggrealMeasyLemenLJ, 385-401.

Rosenthal, D, Bell, R., Demetrious, A. &Efldides, A. (1989). From collectivism to
individualism? The acculturation of Greek immigrants in Australia. lntematjgnal
leumal QfPsyethng, 24, 57- 71

Rosenthal, D. & Bomholt, L. (1988). Expectations about development in Greek and
Anglo-Australian families. W, 19—34.

Rovine, M. & von Eye, A. (1996). Regression interaction with continuous variables:
Difficulties and computational suggestions. Technical Report Series #95-2 for The
Methodology Center, College of Health and Human Development, The
Pennsylvania State University.

Ryan, R.M. & Lynch, J.H. (1989). Emotional autonomy versus detachment: Revisiting
the vicissitudes of adolescence and young adulthood. Child Deyelepment, 60, 346-
356.

103

Schmitt-Rodermund, E. & Silbereisen, R (1996). Acculturation of developmental
timetables among adolescent immigrants. Paper presented at the International
Congress of Psychology, Montreal, Canada, August 16 - 21, 1996.

Selrnan, R. (1981). The development of interpersonal competence: The role of
understanding in conduct. W13), 401-422.

Sih, P.K.T. (1960). Translated by M.L. Makra, W- New York: St. John's
University Press.

Silbereisen, R.K. & Todt, E. (Eds) (1994). Adolescence in context: The interplay of
family, school, peers, and work in adjustment. New York: Springer-Verlag.

Sirnic, A. (1987) Ethnicity as a career for the elderly. The Serbian-American case.
W. 113-126

Steinberg, L. (1990). Autonomy, conflict, and harmony. In S. Feldman & G. Elliot (Eds)
WW Cambridge MA: Harvard University
Press.

Steinberg, L. & Silverberg, SB. (1986). The vicissitudes of autonomy in early

adolescence. ChildDexelgpmenLfl, 341-851.

Stessa, M. & Steinberg, L. (1991). Family structure and the development of autonomy
during adolescence. W, 38-55.

Stutman, S & Lich, S (1984) WWW
Lelatignshipjnxentgm, Los Angeles, CA.

Sue, D.W. & Sue, D. (1990). Mamammmamd New York:
Wiley.

Sullivan, K. & Sullivan, A. (1980). Adolescent-parent separation. Memental
Bsyehelegy, [6(2), 93-99.

Szabat, K. A. (1990). Prediction analysis In A von Eye (Ed), Sjgfisfiealmclhgdsjn
W New York: Academic Press.

Szapocznik, J. & Kurtines, W. (1980). Acculturation , biculturalism, and adjustment
among Cuban Americans. In AM. Padilla (Ed) W
W, 139-159. Boulder, CO: Westview.

Talwar, R., Nitz, K., & Lerner, RM. (1990). Relations among early adolescent

104

temperament, parent and peer demands, and adjustment: A test of the goodness of

fit model. W343, 279-298-

Tanner, J.M. (1991). Secular trend in age of menarche. In RM. Lerner, AC. Petersen, &

 

 

J. Brooks-Gum (Eds), F ‘ " of n-‘-- ‘ (pp. 637-641). New York:
Garland.

Thomas, A & Chess, S. (1977). W. New York:
Brunner/Mazel.

Ting-Toomey, S. (1981). Ethnic identity and close fiiendship in Chinese-American
college students. WA 383-406.

T1iandis,H C (1988) Collectivism and individualism: A reconceptualization of a basic
concept in cross-cultural psychology. In G. K. Venna & C. Bagley (Eds)
MW, 60- 95 London: Macmillan

Triandis, H. C (1994). Theoretical and methodological approaches to the study of
collectivism and individualism. In U. Kim & H. S. Hakhoe (Eds ) lndmduahsm

WW5, 41- 51 Thousand Oaks CA:

Sage Publications.

Turner, A., Irwin, C., & Millstein, S. (1991). Family structure, family process, and
experimenting with substances during adolescence. ,[Qumalfleseambm
Adolescence; 93-104.

von Eye, A. & Brandstadter, J. (1988). Application of prediction analysis to cross
classifications of ordinal data. Biemetrieal JQumal, 30, 651-665.

Yu, L C. (1984). Acculturation and stress within Chinese American families. humalgf
W, 77 94

Yu, L. C. & Harburg, E. (1980). Acculturation and stress among Chinese Americans in a
university town. Intematienal lgumal Qf (heupT ensi ens, IO, 99- 119.

 

 

 

 

NICHIGRN STATE UNIV. LIBRARIES
llHlHllllllllHINllHlllllHWMlllllllllllllllllll
31293015956869