. .2
531%"? a

0. a
.u!

. ‘ . ., an... 2.
«kn ‘ ,9. V .d

Ears . 9.5..)
we” , ..

l...

‘ax
..
f.
1w

«vi»;

a

‘ c _
3" - ‘1 ,—
“ﬂ

.4..an

 

 

 

.uzw HAHN; 1%... Ukauu AWL 4....

I. muzf .
J... ..

 

 

LIBRARY

Michigan State
University '

 

 

 

This is to certify that the

dissertation entitled

MICHiGAN STATE RAB E

‘willMillil1llllnlliwlll ”

3 1293 01766 8694

AUTHORITATIVE AND AUTHORITARIAN PARENTING OF
MOTHERS WITH PRESCHOOL CHILDREN IN TAIWAN

presented by

Fu-Me 1 Chen

has been accepted towards fulﬁllment

of the requirements for

Ph.D . degree in Wild Ecology

 

 

Date ////7//?CV

MS U is an Afﬁrmative Action/Equal Opportunity Institution

,K/fJ/Mu/j/ ,’

(Mayo? professor 7

,/

  

0-12771

l

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

 

MTE DUE l DATE DUE

DATE DUE

 

$111201 5 290"

 

'OOc on

 

 

mg 13 fl) 9093

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 chiHC/DdaDmpSB—p.“

AUTHORITATIVE AND AUTHORITARIAN PARENTING OF MOTHERS WITH
PRESCHOOL CHILDREN IN TAIWAN

By

Fu—mei Chen

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

DOCTOR OF PHILOSOPHY

Department of Family and Child Ecology

1998

it!”

ABSTRACT
AUTHORITATIVE AND AUTHORITARIAN PARENTING OF MOTHERS

WITH PRESCHOOL CHILDREN IN TAIWAN

By
Fu-mei Chen
Doctor of Philosophy
Michigan State University, 1998

Professor Tom Luster, Chair

The purpose of this study was to examine authoritative and authoritarian parenting
in Chinese mothers with preschoolers, and factors related to the mothers' parenting
practices. This study not only focused on the global parenting typologies (e.g.
authoritative and authoritarian parenting), but also looked into the individual component
elements of authoritative and authoritarian parenting. Four subscales in the Authoritative
scale were assessed: Warmth/Involvement, Reasoning/Induction, Democratic
participation, and Good natured! Easy going. For the Authoritarian scale, the subscales
consisted of Directiveness, Corporal punishment, Non-reasoning/Punitive strategies, and
Verbal hostility. An ecological model was used to select potential predictors of
parenting practices: 1) characteristics of the mother, 2) characteristics of the child, and 3)
contextual factors.

The sample consisted of 463 mothers with their preschool children in Taiwan.

Mothers were administered a questionnaire which assessed their parenting practices,

SI.

Chinese parenting beliefs, depression level, degree of parenting daily hassles, child's age,
perception of the child's temperament, and demographic information. Based on the data,
it was found that the majority of the mothers (over 65%) in this sample could be
classiﬁed as authoritative mothers. None of the mothers in this sample were classiﬁed as
authoritarian. The rest of the mothers practiced both authoritative and authoritarian
parenting to some degree. Twenty percent of the mothers practiced authoritarian
parenting at a medium level; however, they also scored medium to high on the
authoritative measure. Fifteen percent of the mothers were medium on the authoritative
and low on the authoritarian measure.

In the analyses examining factors related to parenting practices, Belsky's model of
the determinants of parenting was supported. Maternal characteristics, child
characteristics, and contextual factors all contributed to predicting mothers' parenting
practices. The results suggested that maternal depression, child temperament, and degree
of parenting daily hassles may have cross-culturally universal inﬂuences on parenting
practices. However, the signiﬁcant inﬂuence of culturally speciﬁc parenting beliefs on
parenting was also found. Traditional Chinese parenting beliefs were found to be
positively related to both authoritative and authoritarian parenting.

Most previous cross-cultural studies have depicted Chinese parenting as
"authoritarian", or "controlling", which has not been viewed as "optimal" parenting in
western culture. This study found that although Chinese parents have been found to
score higher on the authoritarian parenting measure than western parents, authoritative
parenting was still commonly practiced among Chinese mothers. For those who practice

authoritarian parenting, it was very unlikely to be in the extensive or punitive form, and

the use of authoritarian parenting was often accompanied by the use of a medium to high
degree of authoritative parenting. Chinese parents' use of authoritative and authoritarian
parenting were both in some degree embedded in the traditional Chinese parenting

beliefs-- the concept of training.

To

my families

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my major professor Dr. Tom
Luster, whose guidance, encouragement, and extraordinary patience throughout my
graduate study has contributed to my professional growth. I would also like to thank my
committee members, Dr. Marjorie Kostelnik, Dr. Alice Whiren, and Dr. Patric Dickson
for their guidance and suggestions on various aspects of my dissertation.

Thanks are extended to the preschools that participated in this study--Shin-Yea
Kindergarten, Jiann-Gwo Kindergarten, Dong-ku Kindergarten, Nan-ku
Kindergarten, Jin-Long Kindergarten, Shin-Shang Kindergarten, Shin-ku Kindergarten,
Academic Sinica Kindergarten, Kang-Dah Kindergarten, Jia-Yuh Kindergarten, and
Sheng-yuh Kindergarten. Thanks are also extended to the mothers who took time to ﬁll
out the questionnaires.

I owe my eternal thanks and gratitude to my parents for their unconditional love
and trust which have been the endless source of my strength. Further, I deeply thank my
sister, J en-mei, for her assistance on data collection. Thank my daughter, Rena, for the
joy that she brings to my life. Finally, I would like to give my special thanks and love to
my husband, Liang-yo, who gave me one hundred percent support through both the good

and the bad. Without his thoughtfulness and full support, I would have not been able to

vi

‘r

complete my dissertation.

vii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................ xi
CHAPTER 1
INTRODUCTION .......................................................................................................... 1
Rationale ...................................................................................................................... 1
Studies of Chinese parenting .................................................................................. l
Determinants of parenting practices ...................................................................... 5
Signiﬁcance of the study .............................................................................................. 6
Theoretical background .............................................................................................. 8
Variables examined in the study ................................................................................ 10
Research questions and hypotheses .................................................................... 13
Organization of the study ............................................................................................ 15
CHAPTER 2
REVIEW OF THE LITERATURE ................................................................................ 16
Backgrounds of Chinese parenting concepts and Euro-American parenting concepts...16
Cross-cultural studies related to Chinese parenting ............................................ l9
Arguments on the validity of Baumrind's typologies on Chinese parenting ........ 21
Assessment of Parenting ............................................................................................ 27
Determinants of Parenting ............................................................................................ 32
Studies supporting Belsky‘s model of the determinants of parenting ........ 32
Determinants of authoritative and authoritarian parenting ................................ 35
CHAPTER 3
METHOD .................................................................................................................... 40
Sample .................................................................................................................... 40
Procedure .................................................................................................................... 42
Deﬁnition of variables ............................................................................................ 42
Instrumentation ........................................................................................................ 45
Translation method ........................................................................................................ 51
Limitations .................................................................................................................... 51
Analysis .................................................................................................................... 52

viii

CHAPTER 4

RESULTS .................................................................................................................... 54
Descriptive statistics for the variables in the analyses ............................................ 54
Chinese parenting patterns ............................................................................................ 57

Correlations among the component elements of Authoritative and Authoritarian
scales .................................................................................................................... 57
Categorization of Chinese parenting patterns ............................................ 59
Factors related to Chinese parenting .................................................................... 62
Relations among the predictor variables ........................................................ 62
Zero-order correlations between the predictor variables and the dependent
variables ........................................................................................................ 62
Multiple regression analyses: Predictors of parenting factors .................... 64
Differences among the four parenting patterns on the predictor variables ........ 73
Summary of the results ............................................................................................ 75

CHAPTER 5

DISCUSSION ........................................................................................................ 80
Chinese parenting patterns ............................................................................................ 8O

Interrelations among the subscales of the Authoritative and the Authoritarian
measures ........................................................................................................ 80
Categorization of Chinese parenting patterns ............................................ 81
Factors related to Chinese mothers' parenting ........................................................ 86
Regression analyses results ................................................................................ 86
Cross cultural implicatons ................................................................................ 92
MANOVA and one-way ANOVA results ........................................................ 93
Examination of Belsky's model by differentiating different aspects of parenting
behavior .................................................................................................................... 9S
Determinants of the Authoritative v.s. Authoritarian total scales .................... 95
Determinants of the Authoritative subscales ........................................................ 96
Determinants of the Authoritarian subscales ........................................................ 97
Conclusion .................................................................................................................... 99
Directions of future research .............................................................................. 100

LIST OF REFERENCES .......................................................................................... 102

APPENDICES ...................................................................................................... 1 10
Appendix A: Instruments used in this study ...................................................... 110

Parenting Practices Questionnaire (PPQ) ...................................................... 110
Parenting Daily Hassles (PDH) ............................. 113
The "Training" Questionnaire .................................................................. 114
EAS Temperament Survey for Children: Parental Ratings .............................. l 15
CES-D Scale ...................................................................................................... 116
Demographic information ........................................................................ l 17

ix

Consent letter ...................................................................................................... l 19
Appendix B: Comparison of mean scores of the Authoritative scales and Authoritarian
scales on Chinese sample in this study and American sample in Robinson

et al.'s study .......................................................................................... 120

Appendix C: Copy of the UCRIS approval letter ...................................................... 121

LIST OF TABLES

Table l - Demographic Characteristics of Sample ........................................................ 41
Table 2 - Means, standard deviations of the variables in the analysis .................... 56

Table 3 - Intercorrelations among the components of authoritative and authoritarian

patterns ........................................................................................................ 58
Table 4 - Four parenting patterns for Chinese mothers ............................................ 60
Table 5 - Intercorrelations among predictor variables ............................................ 65
Table 6 - Zero order correlations between predictors and parenting factors .................... 66

Table 7 - Multiple regression analyses: Predictors of Authoritative and Authoritarian total

scales ........................................................................................................ 67
Table 8 - Multiple regression analyses: Predictors of Authoritative subfactors .................... 68
Table 9 - Multiple regression analyses: Predictors of Authoritarian subfactors .................... 69

xi

Chapter 1

INTRODUCTION

Katmai;
Studies of Chinese parenting

The importance of studying parenting behavior has long been recognized. For
young children, parents are the most important socialization agents. Parenting practices
have a signiﬁcant inﬂuence on child development. With respect to parenting practices,
Baumrind's three major types or patterns of child rearing are broadly studied. The three
major parenting typologies are authoritative, authoritarian, and permissive. Authoritative
parents are warm, loving, involved and responsive. They respect their children's
independence, personality characteristics, point of view, interests, and motives; they
communicate well with their children, encourage give and take in discussions, and are
clear about the reasons for directives. At the same time, they are controlling and expect
mature, independent behavior appropriate for the child’s age, and requiring their children
to contribute to family functioning by helping with household tasks. In contrast,
authoritarian parents are highly controlling, rely heavily on punitive discipline, and
provide relatively little warmth. They stress the importance of obedience to authority and
discourage verbal give and take between themselves and their children. Finally,

permissive parents are nurturant, but lax in disciplining and rewarding their children.

2

They are not controlling and make few maturity demands, allowing their children to
regulate their own activities as much as possible (Baumrind, 1967).

Quite a few cross-cultural studies have been done to compare parenting practices
among different cultural groups. In some studies, Chinese parents have been found to
score higher on "restrictive", "controlling", or "authoritarian" parenting measures than
western parents. Chiu (1987) studied the child-rearing attitudes of Chinese, Chinese-
American, and Anglo-American mothers. The Chinese mothers were most restrictive, the
Anglo-American mothers were least restrictive, and the Chinese-American mothers were
intermediate on the continuum of authoritarian-control. Lin and Fu( 1990) compared
child-rearing practices among Chinese, immigrant Chinese, and Caucasian-American
parents. It was found that Chinese and immigrant Chinese parents tended to rate higher
on parental control than Caucasian-American parents. Kelley and Tseng (1992)
compared the child-rearing practices of immigrant Chinese and Caucasian American
mothers. Immigrant Chinese mothers reported more physical punishment and yelling at
the child, and scored lower on nurturance, responsiveness to child input,
nonrestrictiveness, consistency, and rule setting.

Based on the ﬁndings reported above, Chinese parenting was often depicted as
"restrictive", "controlling", or "authoritarian". The child development literature
conducted in western countries has often claimed that such parenting is not optimal for
children's development. Authoritarian parenting was associated with rejected children
and children with lower grades in school. In contrast, children from authoritative families

were found to be instrumentally competent and had better school performance (Baumrind,

1991a; Dekovic' & Janssens, 1992; Dombusch, Ritter, Leiderman, Roberts, & Fraleigh,

3

1987; Hein & Lewko, 1994). However, studies of the relations between parenting
practices and child outcomes in a sample of Chinese parents are not consistent with that
found in the western literature. Parental control or authoritarian parenting was found to
be positively related to the child's socially competent behavior, autonomy, self-esteem,
and self-control (Chen, 1997; Chung, 1994; Xu, Wan, Mussen, Shen, Li, & Cao, 1991).
Moreover, though Chinese parents were found to score higher on "controlling" or
"authoritarian" measures than western parents, Chinese children were found to have
superior school performance (Dombusch et al., 1987).

Researchers have tried to solve the paradox regarding Chinese restrictive
parenting practices and Chinese children's outcomes. For example, it has been proposed
that the parental inﬂuences are not appropriate predictors of school performance for
Asian children (Dombusch et al., 1987), or the parenting concepts "authoritarian" and
"restrictive" are not very relevant to Asians, and the labels may not be generalizable
across cultures (Chao, 1994; Chen, 1997; Chung, 1994). It was argued that these
concepts may have different meanings for Americans and Chinese. While for Americans,
"strictness" is sometimes equated with manifestations of parental hostility, aggression, .
mistrust, and dominance, for Asians, parental obedience and strictness may be equated
with parental concern, caring, or involvement (Chao, 1994; Kim & Chun, 1994; Rohner
& Pettengill, 1985).

Most studies in the area have compared parents from two or more cultural groups
and focused on differences between the groups. In cross-cultural studies, the group with
which Chinese were compared was often Euro-Americans. Intracultural variation has not

been directly studied in previous cross-cultural research. It should be noted that though

4

Chinese parents "overall" scored higher on authoritarian parenting than American parents,
it did not mean that Chinese would use authoritarian parenting extensively. Chinese
parents did not necessarily score lower on authoritative parenting, either. Depicting
Chinese parents as "controlling" or "authoritarian" may be an oversimpliﬁcation.
Moreover, in the studies examining the relations between parenting practices and child
outcomes, single parental variables have been investigated independently. However, in
naturally occurring parenting practices, different aspects of parenting occur
simultaneously, not independently. Examining the inﬂuence of single parental variables
would fail to detect possible interaction effects. Combinations or patterns of child rearing
techniques are better predictors of children's outcomes than individual practices. Some
studies have used several parent behavior variables in nonorthogonal combinations (e. g.
authoritarian and authoritative) to assess parenting. However, since authoritative and
authoritarian parenting types have been identiﬁed by Baumrind as two of the three main
parenting typologies, authoritative and authoritarian parenting tended to be dichotomized
into two opposing types. It is often assumed that if a parent is authoritarian, he or she
can not be authoritative at the same time. Therefore, the score on one measure is often
discussed without considering the score on the other measure. There are many parents
who could not be classiﬁed as "purely" authoritative or "purely" authoritarian. Parents
may have a higher score on one measure, for example, the authoritarian measure, but their
scores on the authoritative measure should also be considered. Third, a paradox has been
identiﬁed in explaining Chinese "restrictive" parenting styles and Chinese children's
outcomes, especially in explaining their superior school performance. A hypothesis could

be made that Chinese parents might exercise some aspects of authoritarian parenting

5

practices while some aspects of authoritative behaviors are still present. The authoritative
parenting practices offset the negative consequences of authoritarian parenting practices.
The meaning and effect of parental control and strictness would differ when embedded in
parental warmth and involvement versus parental hostility and rejection. It is suggested
that Chinese children's outcomes depend jointly on their parents' use of authoritarian

parenting and their use of authoritative parenting.

Determinants of parenting practices

While great effort has been devoted to studying the characteristics and
consequences of parenting, in order to better understand parenting behavior, it is also
important to study why parents parent the way they do. Belsky (1984) proposed that
parenting is determined by factors in three domains: maternal characteristics, child
characteristics, and contextual factors. The model of the determinants of parenting was
thus broadly studied. For example, Belsky and Isabella (1988) investigated the inﬂuence
of maternal characteristics, child characteristics, and contextual inﬂuences on the quality
of the infant-mother relationship. It was found that maternal personality, mother's
perceptions of child's temperament, marital quality, and neighborhood characteristics,
were signiﬁcantly different for families with secure and insecure infant—mother
attachment relationships. Hannan and Luster (1991) examined factors related to the
quality of the home environments that mothers provide for their infants. They found that
mother's age at ﬁrst birth, mother's intelligence, child's difﬁcult temperament, presence of
spouse or male partner in the home, level of income and number of children contributed

uniquely to the home environment that mothers provided for their infants. Menaghan and

6

Parcel (1991) examined multiple determinants of parenting and concluded that the home
environments created by mothers for their young children are a function of maternal and
child characteristics, maternal working conditions, and current family characteristics.
Mothers' self-esteem, intelligence, educational levels, age, child's health, presence of a
spouse, spousal earnings, number of children in the family, maternal ethnicity and
maternal occupations were related to home environments.

In regard to the determinants of parenting, the view that parenting is multiply
determined has been generally accepted. The studies examining Belsky's model have,
however, treated parenting as a general term without specifying different aspects of
parenting behaviors. Labeling parenting behavior as "supportive" or "high quality" is
largely a social judgement. The judgment is often based on parents meeting certain
social expectations. The speciﬁcation may be especially important for studying parenting
of non at-risk families, whose parenting may need further differentiation beyond
"supportive" vs. "less supportive."

5' 'E E l 1

In response to the limitations of the previous studies in this area, this study was
designed to examine authoritative and authoritarian parenting practices in Chinese
mothers with preschool children, and factors related to their parenting behavior. First,
this study examined the use of authoritative and authoritarian parenting practices in
contemporary Chinese mothers by going beyond simple comparisons of one group versus
another to focusing on within-group differences. The variations in parenting practices
within Chinese mothers were investigated. Chinese parenting patterns were identiﬁed

based on the degree or the weighing of their scores on both the authoritative and the

7

authoritarian parenting measures. Not only were the mother's scores on the authoritative
and authoritarian parenting taken into account simultaneously, the mother's scores on the
speciﬁc factors within each global parenting typology were also investigated. Going
beyond the global parenting typologies and looking into the different components of each
of these parenting practices should contribute to a better understanding of Chinese
parenting.

In addition to examining what Chinese parents actually do with their preschoolers,
this study also investigated the context within which these parenting behaviors occur.
Factors related to Chinese mother's authoritative and authoritarian parenting were
examined utilizing Belsky's model of the determinants of parenting. Speciﬁc
determinants affecting different aspects of parenting practices were examined. Factors
related to the component elements of authoritative parenting (wannth/involvement,
reasoning/induction, democratic participation, and good natured/easy going), and
authoritarian parenting (directiveness, corporal punishment, non-reasoning/punitive
strategies, and verbal hostility) were identiﬁed. This study investigated if authoritative
parenting and authoritarian parenting, as well as their component elements, were
predicted by different factors. Moreover, earlier studies have pointed to ethnic
differences in maternal behavior (Luster & Dubow, 1990; Menaghan & Parcel, 1991). A
sample of Chinese mothers was used to further examine Belsky's model in Eastern
culture. This study examined if factors related to parenting practices for Chinese parents
would be similar to those for parents in western societies. Third, in response to the
paradox in explaining Chinese parenting style and Chinese children's outcome, this study

proposes that Chinese parents' use of authoritarian parenting is often accompanied by the

8

use of authoritative parenting. The use of authoritative parenting plays an important role
in children by facilitating compliance and the acceptance of parents' authoritarian control
in some occasions, which may offset any negative effects of authoritarian parenting on
children.

The method the researcher used was to survey mothers with preschoolers in the
city of Taipei, Taiwan. In the questionnaire, the mother's parenting practices, maternal
characteristics, childcharacteristics, and the contextual factors were assessed. Focusing
speciﬁcally on mothers instead of fathers in this study was due to the fact that mothers are
still typically the primary care-takers of their children, and are usually more involved than
fathers in most of the day to day activities of their children. Men have been shown to be
much less involved than their wives in the daily care and supervision of the children
(Sroufe, Cooper, DeHart, & Marshall, 1992). In contemporary Taiwan, though more
women have joined the labor market, mothers are still the ones who play the major role in
taking care of household chores and children (Lin, 1995).

WWW

Bronfenbrenner's ecological theory (Bronfenbrenner, 1979) and Belsky's
determinants of parenting model (Belsky, 1984) are the theories guiding this study, in
particular regarding the examination of factors related to Chinese parenting practices.
Bronfenbrenner proposed that human development is profoundly related to factors within
the ecological environment. An understanding of human development demands going
beyond the observation of behavior on the part of one or two persons in the same place; it
requires an examination of multiperson systems of interaction not limited to a single

setting and must take into account aspects of the environment beyond the immediate

9

situation containing the subject. Studying mother's parenting behavior toward their
children is complex and appears to be inﬂuenced by a multitude of dimensions in the
ecosystem. Understanding individual differences in parenting requires an understanding
of a parent not only as an individual psychological agent, but also as an agent who
interacts with other family members and environments.

Belsky's model of the determinants of parenting (1984) identiﬁed three domains
of inﬂuences on parenting: the personal, psychological resources of parents, characteristics
of the child, and contextual sources of stress and support. The model makes the
following assumptions: a) that parental functioning is multiply determined, b) that
sources of contextual stress and support can directly affect parenting, c) that sources of
contextual stress and support can indirectly affect parenting by ﬁrst inﬂuencing individual
psychological well-being, and d) that personality inﬂuences contextual support/stress,
which feeds back to shape parenting. Belsky indicated that unsupportive parenting
behavior would be expected if the parents lacked personal resources, if the child was
difﬁcult to care for, and/or if the family context was characterized by high levels of stress
and few resources. A child's chances of experiencing low quality parenting are greatest if
all three factors combine to undermine effective parenting. In order of importance, the
personal psychological resources of the parent are most effective in buffering the parent-
child relation from stress. Contextual sources of support are more effective than
characteristics of the child. When two of the three determinants of parenting are at risk, it
is proposed that parental functioning is most protected when the personal resource
subsystem still ﬁrnctions to promote sensitive involvement and least protected when only

the subsystem of child characteristics fulﬁlls this function.

lO

Authoritative and authoritarian parenting practices were examined in the study.
Their component elements: warmth and involvement, reasoning/induction, democratic
participation, good natured/easy going, verbal hostility, corporal punishment,
nonreasoning/punitive strategies, and directiveness were also identiﬁed. Given the fact
that the primary interest of the study was on the issue of the "optimal", "authoritative"
parenting in western countries vs. the "traditional", "authoritarian" parenting in Chinese
societies, permissive parenting was not examined in the present study.

As noted above, Belsky's model of the determinants of parenting was utilized to
select the potential predictors. To contribute to a high return rate from parents who were
sampled, the variables examined in this study were carefully selected in order to avoid
redundancy and keep the length of the questionnaire reasonable.

Three domains of determinants were identiﬁed in Belsky's model: maternal
characteristics, child characteristics and contextual factors. In terms of maternal
characteristics, mother's age, education, child-rearing beliefs and psychological well-
being were examined in the study. Age was considered as a marker of maturity. It was
hypothesized that younger mothers were less psychologically mature, and more likely to
practice some aspects of authoritarian parenting than older mothers. Education is viewed
as a general socialization construct. Mothers with higher education were more likely to
be inﬂuenced by western culture. Thus, it was hypothesized that mothers with higher
education were more likely to practice authoritative parenting, and less likely to practice
authoritarian parenting. "Traditional" Chinese child-rearing beliefs were also examined

in the study. The traditional Chinese child-rearing beliefs assessed in this study included

11

two important features, that of a concept of training, as well as a highly involved concern
and care for children. (These ideas will be discussed in more detail in Chapters Two and
Three). Therefore, it was hypothesized that mothers holding traditional Chinese child-
rearing beliefs were more likely to practice some aspects of authoritarian parenting.
However, it was expected that some aspects of authoritative parenting (e. g. warmth and
involvement) would also be practiced by these mothers. No speciﬁc hypothesis was
made for the relation between mother's Chinese child-rearing beliefs and education. Will
greater departures from the traditional Chinese child-rearing beliefs be shown among the
higher educated Chinese mothers? Or, are traditional child-rearing beliefs fairly
consistent across educational groups among contemporary Chinese mothers? This study
will try to answer these questions.

Maternal psychological well-being has received a lot of attention in the studies of
the determinants of parenting. In this study, maternal psychological well-being was
measured by the CES-D depression scale. It was hypothesized that mothers who were
more depressed were more likely to practices authoritarian parenting, and less likely to be
authoritative with their child. It was recognized that the inﬂuence of maternal
psychological well-being on parenting may be traced back to the experiences parents had
while growing up. According to Belsky's model, maternal developmental history
inﬂuenced parenting indirectly, by ﬁrst inﬂuencing the broader context in which parent-
child relations exist. This indirect linkage was not examined in the present study.
Therefore, mother's developmental history was not included.

Children's temperament and age were examined as child characteristics.

Regarding the child's contribution to parental functioning, temperament has received the

12

most attention, especially those behavioral styles that make parenting more or less
difﬁcult (Bates, 1980; Belsky, 1984). Each mother's perception of her child's
temperament (emotionality, activity, sociability) was examined. According to Buss and
Plomin (1975), in terms of the relations between parental practices and child's
temperament, the love dimension is clearly more important for sociability, and the control
dimension is more important for emotionality. It was hypothesized that the mothers who
perceived their children as relatively emotional were more likely to practice authoritarian
parenting. Mothers who perceived their children as more social were more likely to
practice authoritative parenting. No hypothesis was made for the relation between child's
activity level and the mother's parenting practices. In addition, according to the literature
which will be reviewed in Chapter Two, studies showed that Chinese parents tend to be
highly lenient and warm toward very young children until they reach "the age of
understanding." For children beyond the "age of understanding", strict discipline is
reported to be used. Therefore, "the age of understanding" in Chinese society was also
tested. This study examined if "the age of understanding" would occur during the
preschool years.

SES, family size, and mother's report of parenting daily hassles were the
contextual factors examined in the study. According to the literature reviewed in Chapter
Two, it was hypothesized that families with higher SES status were more likely to
practice authoritative parenting, and less likely to practice authoritarian parenting.
Families with more children were more likely to practice authoritarian parenting. In
addition, mothers were asked to rate the intensity of their parenting daily hassles. The

mother's cognitive appraisal of the intensity of these hassles was considered as the

l3

reﬂection of her contextual stress/ support. Mothers who perceived their parenting events
as big daily hassles were very likely to be the ones who had less social support and more
stress than other mothers. It should be noted, however, that although parenting daily
hassles were categorized as a contextual factor, this variable may also reﬂect, at least in
part, the mother's psychological well-being. It was hypothesized that mothers who
perceived their parenting events as big hassles were more likely to practice authoritarian
parenting, and less likely to practice authoritative parenting. While contextual
stress/support factors were examined, the "sources" of the contextual stress/support (e. g.
marriage, work) were not selected for examination in the present study.
WW

On the basis of the limitations of the prior studies in this area, this study will ask
the following research questions:

Research question 1: How do the component elements of the Authoritative
(Warmth/Involvement, Reasoning/Induction, Democratic participation, and Good
natured/Easy going) and the Authoritarian ( Directiveness, Corporal punishment,
Nonreasoning/Punitive strategies, and Verbal hostility) scales relate to each other? What
parenting patterns will be identiﬁed in Chinese mothers based on their scores on the
Authoritative and the Authoritarian parenting measures?

For the ﬁrst research question, 2 hypotheses were tested:

H l . 1. There is no relationship between the mother's use of authoritarian parenting
practices and the mother's warmth/involvement toward the child.

H1.2. There is no relationship between the mother's use of directiveness, corporal

punishment parenting practices and the mother's use of non-reasoning/punitive strategies,

l4

verbal hostility parenting practices.

Research question 2: What factors are related to the Chinese mother's
authoritarian and authoritative parenting practices? How will the mother's age, education,
child-rearing beliefs, psychological well-being, her child's age, her perception of the
child's temperament and daily parenting stress, as well as the family's SES and size, relate
to her authoritative and authoritarian parenting practices?

For the second research question, 10 hypotheses were tested:

H2. 1. There is a negative relationship between mother's age and her use of
authoritarian parenting.

H2.2. There is a positive relationship between mother's education and her use of
authoritative parenting, and a negative relationship between mother's education and her
use of authoritarian parenting.

H2.3. There is a positive relationship between Chinese parenting beliefs and the
mother's use of authoritarian parenting, and a positive relationship between Chinese
parenting beliefs and the mother's use of warmth/involvement aspect of authoritative
parenting.

H2.4. There is a negative relationship between the mother's depression and her use
of authoritative parenting, and a positive relationship between the mother's depression
and her use of authoritarian parenting.

H2.5. There is a positive relationship between the child's sociability and the
mother's use of authoritative parenting.

H2.6. There is a positive relationship between the child's emotionality and the

mother's use of authoritarian parenting.

15

H2.7. There is a positive relationship between the child's age and the mother's use
of authoritarian parenting, and a negative relationship between the child's age and the
mother's warmth/involvement, as well as her good natured/easy going interaction with the
child.

H2.8. There is a positive relationship between the family's SES status and the
mother's use of authoritative parenting, and a negative relationship between the family's
SES status and the mother's use of authoritarian parenting.

H2.9. There is a positive relationship between the number of children at home and
the mother's use of authoritarian parenting.

H2. 1 0. There is a negative relationship between the mother's perception of the
degree of her parenting daily hassles and her use of authoritative parenting, and a positive
relationship between the mother's perception of the degree of her parenting daily hassles
and her use of authoritarian parenting.

; . . I l l

The dissertation is organized in ﬁve chapters. Chapter 1 is the introduction to the
study. Chapter 2 consists of a review of related literature. Chapter 3 describes the
research methodology. Chapter 4 summarizes the results of the analyses. Chapter 5

consists of a discussion of the ﬁndings and implications for future studies.

Chapter 2

LITERATURE REVIEW

3-. r'J_0“,i! 0 LI‘ '00. ' ti!‘ J_l_‘|,-.-i! _"5_!I"JI 2.! Oaz'ti' A0,! '0.

In order to best understand Chinese childrearing practices, there is a need to
explore it in the context of Confucianism (Coll, Meyer, & Brillon, 1995; Kelley & Tseng,
1992; Lin & Fu, 1990). Parental control, obedience, strict discipline, the emphasis on
education, ﬁlial piety, respect for elders, family duties and obligations, reverence for
tradition, and minimization of conﬂict are attributed to the inﬂuence of Confucianism
(Chao, 1983; Ho, 1981; Lin & Fu, 1990; C011, Meyer, & Brillon, 1995). For example,
"ﬁlial piety" is highly valued in Chinese families. Children should try to satisfy their
parents and respect and show reverence for their elders in all circumstances (Hsu, 1981).
Parental authority over the children is traditionally stressed and children's unquestioning
obedience from them is also expected (Ryan, 1985). In Chinese society, the term chiao-
yang is the most frequently used term in child rearing, with chaio (education) being
emphasized more than yang (rearing) (Ho, 1981). In terms of the nature versus nurture
controversy, Chinese have tended to emphasize the social environment rather than the
hereditary factors (Ho & Kang, 1984). Chinese parents view the parenting role mainly as
one of teacher. They are very involved in child training (Kelly & Tseng, 1992). "Cha

chiao" in Chinese means "family education." If a child misbehaves or shows bad

16

l7

manners, it is viewed as a reﬂection of poor "cha chiao." Parents are often the ones to be
criticized for not being able to "teach" their child right.

Chao (1992, 1994) stressed the importance of understanding the "training"
concept in Chinese parenting. The training concept--in Mandarin, is "chiaoshun" or
"guan." The term "chiao" means education in terms of the proper development of
character, and "shun" means very disciplined teaching. The training concept for the
Chinese implies a disciplined teaching for the building of moral character, speciﬁcally
self-discipline. "Guan" has two intercorrelated meanings in Chinese culture. Guan
means to govern, which would be how the American culture would interpret the term. In
Chinese society, parents, relatives, teachers, and any other adults who have a particular
relationship with the child, are responsible for governing the child. However, in Chinese
culture, guan also has a very positive connotation-~it means to care for, or even to love, as
well as to govern (Tobin, Wu, & Davidson, 1989). Tobin and colleagues provided the
example of a mother saying to her disobedient child in explaining the cultural notion of
guan. "Wo bu guan ni!" ( I am not going to interfere with your life-«meaning I do not
love you, I am not going to care for you anymore). Parents often tell their children that
the governing is out of love. If parents don't care about their children, parents won't
bother to govern their children.

Chao (1992) indicated that, while the Chinese emphasize training their children,
the Americans speak of "nurturing innate ability." Stevenson and Lee (1990) pointed out
that Americans have a more nativist perspective versus the Chinese emphasis on effort.
The nativist view sees child development as "not all children are capable of the same

levels of achievement, no matter how hard they work." Children are viewed as born with

18

different ability levels. The parent's role is to provide the most supportive and rich
environment possible for the child's own innate maturation or unfolding of his or her
cognitive growth. The innate unfolding process of cognitive development should come
from within the child, and parents can not direct the process. The "humanistic" approach
has a similar viewpoint. Humanists focus on children's conceptions of themselves. The
development of each child's unique "self" is a key concern (Thomas, 1992). The original
nature of the child is believed to be good or neutral. Therefore, the goal of guiding child
development is to foster the expression of the child's inner nature.

Western literature has related child-rearing practices to socioeconomic factors.
More favored parenting characteristics were found among high SES Euro-American
parents. Research has indicated that working-class parents are more likely than middle-
class parents to use disciplinary techniques of control (Kohn, 1969). More speciﬁcally,
parents of higher SES backgrounds place greater value on self-direction than do parents
of lower socioeconomic backgrounds, who value conformity to external authority. Based
on the ﬁndings regarding the different parenting styles between working-class parents and
middle-class parents, and the more positive child outcomes found among children from
higher SES families, the least possible amount of restriction on the children was expected
from parents to support children's self exploration and development. Therefore, parenting
styles of other cultures are often regarded as being too restrictive or controlling just

because they look different from those of high SES Euro-American parents (Chao, 1992).

19
Q _ l l 1' l I :1 . .

Quite a few studies of Chinese parenting have been done by comparing Chinese
mothers and, primarily, American mothers to examine cultural differences in child
rearing. Chinese parenting has been depicted as "restrictive" and "controlling", as well as
"authoritarian." Chiu (1987) studied the child-rearing attitudes of Chinese living in
Taiwan, Chinese-Americans who immigrated from Taiwan, and Anglo-Americans.
Signiﬁcant differences were found on all 23 attitudinal scales and the three attitudinal
factors extracted ﬁom them: authoritarian-control, hostility-rejection, and democratic
attitudes. The results showed that the Chinese mothers were most restrictive, the Anglo-
American mothers were least restrictive, and the Chinese-American mothers were
intermediate on the continuum of authoritarian-control. The Chinese-American mothers
were more likely to approve the expression of hostility or rejection toward the child than
the Chinese or Anglo-American mothers. The Chinese-American mothers were more
democratic than the Chinese mothers, and the Chinese mothers were more democratic
than the Anglo-American mothers. The author concluded that the ﬁndings regarding the
authoritarian-control factor were consistent with theoretical expectations as well as with
the previous empirical studies. The ﬁndings clearly indicated that Chinese mothers were
more restrictive and controlling than the Chinese-American mothers who in turn were
more restrictive and controlling than the Anglo-American mothers.

Lin and Fu (1990) compared child-rearing practices among Chinese, immigrant
Chinese, and Caucasian-American parents in a sample of parents with children enrolled
in kindergarten, ﬁrst grade, and second grade in Taiwan and the United States. The child-

rearing variables under study were: parental control, encouragement of independence,

20

expression of affection, and emphasis on achievement. It was found that Chinese and
immigrant Chinese parents tended to rate higher on parental control, encouragement of
independence, and emphasis on achievement than Caucasian-American parents. The
authors claimed that the higher ratings on parental control and emphasis on achievement
among parents of Chinese origin reﬂected the important inﬂuence of traditional cultural
values on Chinese child-rearing practices, in spite of rapid social and political, changes or
the relocation to another country. In explaining the higher ratings on encouragement of
independence among the parents of Chinese origin than their Caucasian counterparts, it
was suggested that independence may be viewed as a prerequisite to achievement. The
result showed that parents of Chinese origin, especially fathers, tend to encourage their
children to be independent and to achieve.

Kelley and Tseng (1992) compared the child-rearing practices of immigrant
Chinese and Caucasian American mothers with 3- to 8- year-old children. Almost all
Chinese mothers were from Taiwan. Three dimensions assessed parental support
(nurturance, responsiveness to child input, and nonrestrictive attitude), and three assessed
disciplinary practices and control (consistency, amount of control [i.e., rule setting], and
type of control [e.g., reasoning and physical punishment]). Results showed that
immigrant Chinese mothers reported more physical punishment and yelling at the child,
and scored lower on nurturance, responsiveness to child input, nonrestrictiveness,
consistency, and rule setting. No group difference was found on the use of reasoning.
Both immigrant Chinese and Caucasian American mothers reported considerable use of
reasoning in response to common child-rearing occurrences.

Though Chinese parents have been found to have higher scores on authoritarian

21

related measures, it should be noted that Chinese parents were no less warm or
authoritative than American parents. In Lin and Fu's study ( 1990), no difference was
found on open expression of affection between Chinese and Caucasian-American parents.
In Chao's study (1994), Chinese mothers' and European-American mothers' scores on the
authoritative scale did not differ. In the use of reasoning strategy, there was no difference
found between Chinese and Caucasian American mothers (Kelley & Tseng, 1992). In an
earlier study (Chiu, 1987), Chinese mothers were found to be more democratic than the
Anglo-American mothers. Chiu also indicated that, while the Chinese mothers were
more restrictive and controlling than the Anglo-American mothers, they were not more
likely to approve of the expression of hostility or rejection toward the child. These
ﬁndings all suggested that, while the Chinese parents may have higher scores on
"restrictive", "controlling" or "authoritarian" measures than their American counterparts,
they do not necessarily have lower scores on the "authoritative" measure, component
elements of the authoritative measure (e.g., reasoning, democratic attitudes), or have
higher scores on a particular component element of the authoritarian measure (e. g.
hostility or rejection). Therefore, "authoritarian" is an incomplete, or even misleading,
characterization of Chinese parenting. To describe Chinese parenting as restrictive or
authoritarian would be missing other aspects that are actually at least as descriptive of the
Chinese (Chao, 1992).
a, any. n .- ,0; o :._ ”"1"“ um o 94- o ..- -..__ a. ,.

1n the literature comparing parenting in Chinese and western societies, a question
has been raised concerning the ethnocentricity of the concepts of authoritative and

authoritarian parenting developed in western culture. It is also questioned if American

22

and Chinese share the same meanings for the concepts of controlling and strictness.
While for Americans, "strictness" is sometimes equated with manifestations of parental
hostility, aggression, mistrust, and dominance (Kim & Chun, 1994; Rohner & Pettengill,
1985), for Asians, parental obedience and strictness may be equated with parental
concern, caring, or involvement (Chao, 1994). For Asians, parental control may not
always involve "domination" of children per se, but rather a more organizational type of
control for the purpose or goal of keeping the family running more smoothly and
fostering family harmony (Lau & Cheung, 1987). For Chinese, the strategies for
discipline or behavior control are love-oriented and non—physical (Suzuki, 1980). The
strictness and restriction of the Chinese mothers are meant to protect more than inhibit
(Chiu, 1987). In studies of parenting practices in Asian families, Rohner and Pettengill
(1985) found that Korean youths' perceptions of parental control were correlated
positively with perceived parental warmth and low neglect, which contrasted with the
ﬁndings of studies on North American youth (e.g. Saavedra, 1977, 1980).

In parenting studies of contemporary Taiwan's society, Chung (1994) pointed out
that, although the traditional view of parental roles does not seem to be as prevalent as in
the past, the overall view of the parent as an authority ﬁgure and that parents have a duty
to control their children is still present. However, the view does not translate into
behaviors that are punitive or insensitive to children's needs and abilities. In Lin's study
(1995), she found that fathers with preschoolers in Taiwan could possess characteristics
of being nurturant, respectful and power assertive at the same time. She claimed that this
parenting style is very similar to the "authoritative parenting style" named by Baumrind

(1967). The only difference between these two styles is that Taiwanese authoritative

23

parents punish children more than American authoritative parents do. Lin claimed that
the difference results ﬁ'om the differences embedded in two cultures. Early childhood
education in American culture has been criticizing the involvement of punishment in
child rearing and education. However, in Taiwan, punishment is still considered a
strength and a tradition in rearing children. It involves a version of care, love, and
expectation.

Regarding the inﬂuences of parenting practices on children, a paradox has also
been raised in explaining the "Chinese" parenting styles (e. g. controlling, authoritarian)
and Chinese children's outcomes, especially their superior school performance. For
example, the "Chinese" parenting styles (e. g. controlling, authoritarian) have been found
to be predictive of poor school achievement among European-Americans, and yet the
Chinese are performing quite well in school (Chao, 1994). As a resolution to the
paradox, for example, it has been proposed that the parental inﬂuences are not
appropriate predictors of school performance for Asian children (Dombusch et al., 1987),
or the parenting concepts "authoritarian" and "restrictive" are not very relevant to Asians
(Chao, 1994). However, the studies still have not yielded a consistent picture. In
Dombush et al.'s study, they examined the relations between parenting styles and
adolescent school performance in a sample of four different ethnic groups, Afro-
American, Mexican-American, Asian-American, and Euro-American. They found that
Asian-American students, along with students from the other minority groups, rated their
own parents as more "authoritarian" than Euro-American adolescents. As expected
regarding the relation between the poor school performance and authoritarian parenting,

Mexican-American and Afro-American students were found to do less well in school than

24

Euro-American students. However, Asian students had better grades than Euro-American
adolescents. In response to the ﬁnding, the researchers concluded that "Asian children in
our public schools cannot be adequately explained in terms of the parenting styles we
have studied" (p.1256).

In Chao's study (1994), the cross-cultural validity of the labels "authoritarian",
"controlling", and "restrictive" was questioned. Immigrant Chinese mothers and
European-American mothers of preschool-aged children were administered standard
measures of parental control and authoritative-authoritarian parenting style as well as
Chinese child-rearing items involving the concept of training. After controlling for
mothers' education, and their scores on the measures of parental control and authoritative-
authoritarian parenting style, the Chinese mothers were found to score signiﬁcantly
higher on the "training" ideologies. The author thus proposed that the "training" concept
has important features, beyond the authoritarian concept, that may explain Chinese school
success.

Besides children's school performance, research ﬁndings regarding the relation
between authoritarian parenting and other child outcomes in Chinese samples have also
raised questions about the cross-cultural validity of Baumrind's typologies. The western
literature has suggested that authoritarian parenting is not conductive to children's
competence. Baumrind (1991 a) summarized the inﬂuences of parental styles on children
prior to adolescence. Children from authoritative families have consistently been found
to be more instrumentally competent -- agentic, communal, and cognitively competent --
than children from authoritarian or permissive families. Dekovic' and J anssens (1992)

studied parents' child-rearing styles and the sociometric status of 6-11 year-old children.

25

The results revealed that parents of popular children are more likely to adopt an
authoritative/democratic style when interacting with their children; parents of rejected
children tend to endorse an authoritarian/restrictive style.

In a study examining the relations among parenting attitudes, parent-child
interaction, and preschool children's competent behavior in Taiwan, it was found that
children of authoritarian parents scored signiﬁcantly higher on behavioral competence
measures than did children of authoritative parents (Chung, 1994). In her study,
authoritarian parents had children who scored signiﬁcantly higher on autonomy than
authoritative parents. The authoritarian parents were more likely than authoritative
parents to score more favorably on parent-child interaction as assessed by parental
scaffolding behaviors. Chung then raised serious questions about the link between child
rearing attitudes (i.e., authoritative versus authoritarian) and children's competent
behavior. Because the ﬁnding is not consistent with what was found in the western
literature, the author claimed that the Baumrind model, as culturally deﬁned, may need to
be reinterpreted. Labeling parents as authoritative or authoritarian may not generalize
across cultures.

Chen (1997) studied 171 pairs of parents and their children in Taiwan on the
relation between parental goals, parenting practices and preschoolers' socially competent
behaviors. It was found that parental warmth, consistent control and management have
signiﬁcant effects on children's socially competent development. However, parental
disciplinary control (authoritarian control) was also positively associated with children's
social competence and negatively related to children's aggressive behaviors. It should

also be noted that the frequency of parental warmth, parental consistent control, and

26

parental management tended to increase as the frequency of parental disciplinary control
increased. The author claimed that parental actions that ﬁt within the authoritarian or
restrictive pattern within European American families do not result in an "authoritarian
syndrome" in Chinese children.

Wu (1997) examined relationship between children's self-esteern and authoritarian
parenting style in a sample of mothers with their fourth, ﬁfth, and sixth graders in
Taiwan. It was found that the self-esteem scores among children of mothers with high
levels of authoritarian parenting did not differ from those with parents with lower levels
of authoritarianism. The author suggested that the positive perceptions of authoritarian
parenting in Chinese cultures may contribute to the lack of association between
authoritarianism and children's low self-esteem.

In a study conducted in China, Xu and colleagues (1991) examined family
socialization and kindergarten children's behavior and personality development. In this
study, nine dimensions of parental practices were examined: parental control, use of
reasoning, concern, intellectual stimulation, encouragement of independence, respect for
individuality, consistency, setting good examples, and fairness in conﬂict resolution.
Children's behavior and personality development were indicated by seven dimensions:
curiosity, positive attitudes toward others, self-conﬁdence, independence, self-control,
frustration tolerance, and attitudes toward work. It was found that parental control was
positively associated with positive attitudes toward others and work, self-conﬁdence, and
self—control. Among the nine dimensions of parental practices examined in this study, the
authors concluded that parental control was the socialization variable that best predicted

the development of positive characteristics and behavior of the child.

27

In the studies examining the relations between parenting practices and child
outcomes, single parental variables have been investigated independently. Examining the
inﬂuence of single parental variables failed to detect possible interaction effects. A more
complete understanding of parenting inﬂuences must examine the joint inﬂuences of
different dimensions of parenting practices. For studies using different parenting styles
(e. g. authoritarian and authoritative) to assess their inﬂuences on children, it is proposed
that the scores on both the authoritative and the authoritarian measures should be taken
into account simultaneously. It is a danger to assume that a parent would ﬁt into and only
ﬁt into one particular parenting style. The methods of assessing parenting practices is
reviewed and discussed next.
maximum

Researchers have tried to differentiate underlying dimensions of parenting styles
with two-dimensional frameworks. For example, acceptance/rejection and
dominance/submission; emotional warmth/hostility and detachment/involvement;
love/hostility and autonomy/control; warmth and permissiveness/strictness;
warmth/hostility and restrictiveness/permissiveness (see Darling & Steinberg's review,
1993). Maccoby and Martin (1983) and Rollins and Thomas (1979) invoke two general
constructs to account for the relationships among parent and child behavior reported in
the parenting literature: parental support and control. Thomas, Gecas, Weigert, and
Rooney (1974) deﬁne parental support as behavior manifest by a parent toward a child
that makes the child feel comfortable in the presence of the parent and conﬁrms in the
child's mind that he is basically accepted and approved as a person by the parent.

Parental control involves behavior of the parent toward the child with the intent of

28

directing the behavior of the child in a manner desirable to the parent (Rollins & Thomas,
1979). Studies typically suggest that these two dimensions of parenting are independent
of each other. They found that, on the average, knowing how controlling a parent is seen
to be by the child gives no basis by itself for predicting how warm the parent will be
perceived as being (Schwarz, Barton-Henry, & Pruzinsky, 1985). Because these two
dimensions are viewed as orthogonal, studies often examine them independently.

Heilbrun and Water (1968) suggested that the relationship between parental
control attempts and child behaviors (academic achievement) might be contingent on the
level of parental support. The use of support as a contingent variable to account for
discrepant results in the relationship between parental control attempts and academic
achievement in children was examined in their study. It was found that when parental
support was low, a negative relationship existed between control attempts and
achievement, but when support was high, the relationship was reversed and was positive.
The results support the notion that the two dimensions of parenting have to be studied
simultaneously to observe any possible interaction effects of these variables.

Baumrind (1966, 1967) has used several parent behavior variables in
nonorthogonal combinations to create the three typologies of authoritarian, authoritative,
and permissive. In Baumrind's study in 1991 (Baumrind, 1991b), the fourth
classiﬁcation, rejecting-neglecting, is further derived. As Smetana (1994) pointed out,
however, these global typologies may give little information about ways speciﬁc
parenting practices are related to children's behavior. For example, within the
authoritative typology, it would be reasonable to assume that inductive practices would be

related more to children's adaptive social cognition (Hart, DeWolf, & Burts, 1993) and

29

that parental warmth and involvement would have stronger linkages with children's
prosocial behavior. The global typology serves as a shorthand way to refer to an entire
constellation of attributes. The assumption of a typology is that the types are more than
and different from the sum of their parts (Baumrind, 1991a, b). While parents may be
roughly classiﬁed as, for example, authoritarian in their parenting style, there are
variations in degree and in the weighting of the component elements (Maccoby & Martin,
1983)

Hinde and Dennis (1986) indicated that by considering naturally occurring
patterns of interaction among parenting variables, typological analyses are often more
meaningful than linear analyses. For example, high assertive control with low supportive
control is hypothesized to have a different effect on a child than would moderate levels of
both variables. In Baumrind's longitudinal study (Baumrind, 1991a), when the children
were adolescents, parents in the project were classiﬁed into seven types: Authoritative,
Democratic, Nondirective, Authoritarian-Directive, Nonauthoritarian-Directive,
Unengaged, and Good Enough. In this study, families were classiﬁed as "high", "low" or
"medium" on the parent behavior scales for this further differentiation.

Other examples supporting the need to examine the degree or weighing of the
parenting typologies include Dorubusch et al.'s (1987) and Hein and Lewko's (1994)
studies. In both of their studies, parenting style was assessed by the high school student's
report concerning the frequency of certain family behaviors in his/her family. An
unambiguous or "pure" parenting style was considered to be reﬂected by students with
scores in the top one third of a single one of the parenting indexes. Thus, all students

who had scores that fell in the top one third of two or three indexes were considered to

30

come from families with mixed parenting styles. In both studies, about half of the
samples were classiﬁed as coming from families with a single pure parenting style
(authoritative, authoritarian, or permissive). For the rest of the subsarnples, they were
classiﬁed as "All indexes high", "Authoritarian and permissive", "Authoritarian and
authoritative", "Permissive and authoritative" or "No index high." A signiﬁcant portion
of families (25% of the total sample in Dombusch et al.'s study and 38% of the total
sample in Hein and Lewko's study) were classiﬁed as "no index high"---students with no
scores in the top one third of any of the parenting indexes. According to the studies
above, simply classifying families as authoritative, authoritarian, or permissive appeared
to be insufﬁcient for detecting the full range of variations of parenting patterns.
Therefore, this study proposes the need of assessing parenting practices by examining
parents' scores on both the authoritative and the authoritarian measures, and looking into
the component elements of the measures.

In this study, the use of authoritative and authoritarian parenting practices in
Chinese mothers with preschool children was assessed by the Parenting Behavior
Questionnaire (PBQ) (Robinson, Hart, Mandleco, & Olsen, 1996). This measure was
designed to overcome limitations of other widely used measures for parents of young
children (see Robinson, Mandleco, Olsen, & Hart's review, 1995). First, Baumrind has
used a multimethod approach to assess parenting styles that included parents' rating
scales, psychologists' Q-sorts, and behavioral observations. However, Baumrind's
approach has several disadvantages. For example, when observation and interviewing are
extensive, fewer subjects can be included in studies. In addition, the cost for this

approach is high in terms of special training required for data collection personnel and the

31

time required for collecting data. Quite a few studies assessing Baumrind's three main
typologies have relied on adolescents' reports of parents' parenting styles (e. g., Buri,
1991; Dombusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987; Greenberger, 1988;
Steinberg, Elmen, & Mounts, 1989; Steinberg, Lambom, Dombusch, & Darling, 1992;
Steinberg, Mounts, Lambom, & Dornbushch, 1991). The parenting measures completed
by adolescents eliminate several disadvantages of Baumrind's approach. However, these
methods may be inappropriate for use with young children, because they are designed
particularly for adolescents reporting on how they were parented and academic-related
content is often used in the questionnaire items.

Block's (1965) Child-rearing Practices Report is a widely used parenting practices
instruments developed for parents of young children. Robinson et al. (1995) indicated
that the 91-item Q-sort measure has several disadvantages. First, it contains a large
number of determined factors (28 to 33) with moderate to low reliabilities. Second, it
does not adequately tap Baumrind's typology. Third, it is comprised of many items which
may be outdated or inconsistent with the current literature. Kochanska, Kuczynski, and
Radke-Yarrow (1989) reduced the number of factors in the Block report and made them
more consistent with Baumrind's conceptualizations. However, the limitation with this
measure was that the authors used conceptual guidelines rather than empirical derivations
(for examples see Robinson et al.'s review, 1995), and the reliabilities of the scales were
not reported.

Due to the limitations of other parenting measures, the Parenting Behavior
Questionnaire (PBQ) was used in this study. It is an empirical means of assessing global

typologies consistent with Baumrind's conceptualizations of parenting. It is a measure

32

with a reasonable number of items and reliabilities. The version used in this study was
especially designed for parents of preschool children. In particular, not only the global
parenting typologies (e. g. authoritative and authoritarian), but also the speciﬁc parenting
practices that occur within the context of the typologies would be identiﬁed.
12 . [E .
Studies supporting Belsky's model of the determinants of parenting

As indicated in the introduction section, the theory guiding this study of the
determinants of Chinese parenting is Belsky's (1984) model of the determinants of
parenting. Though factors related to parenting practices have been broadly studied, much
of the research relevant to this area of concern remains unintegrated and underutilized.
Belsky (1984) pointed out the need for conceptual models capable of integrating the
disparate ﬁndings in the literature into a coherent whole that is greater than the sum of its
parts. Because the effects of the various inﬂuences on parenting may be additive or
cumulative, it is important to examine several potential inﬂuences in the same study.

Quite a few studies have investigated the combined effect of several factors on
parenting, and provide some support for Belsky's model. In Crockenberg and
McCluskey's study (1986), the authors evaluated the effect of maternal attitudes, infant
initability, and social support on maternal behavior over the infant's ﬁrst year. Mothers
with more irritable babies showed greater insensitivity toward their infants if they had
unresponsive attitudes about parenting, whereas mothers showed greater sensitivity if
they had adequate social support.

Belsky and Isabella (1988) investigated the inﬂuence of maternal characteristics

(developmental history, personality), child characteristics (temperament), and contextual

33

inﬂuences (marital quality, social networks, neighborhood characteristics) on the quality
of the infant-mother relationship. Although some of the variables did not differ for
families with secure and anxious infants, other factors including maternal personality,
mother's perceptions of child's temperament, marital quality, and neighborhood
characteristics were signiﬁcantly different for these two groups. Moreover, they found
that the probability of insecure infant-mother attachment relationships increased
dramatically (ﬁ'om 17% to 62 %) if more than one of the three inﬂuences studied
(maternal personality, marital quality, and infant temperament) fell below the sample
median.

Hannan and Luster (1991) examined factors related to the quality of the home
environments that mothers provide for their infants using NLSY data. They found that
parent, child and contextual characteristics all contributed uniquely to the quality of care
which the infants received. The mother's age at ﬁrst birth and intelligence contributed
uniquely to the home environments. Contextual variables, including the presence of
spouse or male partner in the home, level of income and number of children, were found
to be signiﬁcantly correlated with the home environments. They also found that infants
with difﬁcult temperaments received less supportive care than easier children. Their
ﬁndings were consistent with Belsky's model, which viewed parenting as multiply
determined by three domains of determinants. Moreover, in this study, a family risk
index was developed and the authors found a positive relation between scores on the risk
index and the probability of providing a relatively unsupportive environment. Eighty-
eight percent of the families with all six risk factors were in the low home environment

group compared to 1 1% of the families with a score of O on the family risk index.

34

Menaghan and Parcel (1991) examined multiple determinants of parenting and
concluded that the home environments created by mothers for their young children are a
function of maternal and child characteristics, maternal working conditions, and current
family characteristics. Mothers with higher self-esteem, higher intelligence, higher
educational levels, and who were older in age, provided more stimulating home
environments for their children. Presence of a spouse, spousal earnings, number of
children in the family, maternal ethnicity and maternal occupations were also related to
home environments. However, in examining the relationship between child
characteristics (gender and health) and home environments, this study reported a positive
correlation between the children's health problems and the children's home environments.
That is, children with health problems came from families that provided more supportive
care. The relation between the child's gender and home environment was not signiﬁcant.

According to the literature reviewed, there is general agreement that parenting
practices are multiply determined. Our understanding of parenting behavior should be
enhanced if we consider the combined inﬂuence of several factors rather than thinking
about each factor singly (Luster & Okagaki, 1993). However, it should be noted that
parenting is not deﬁned very speciﬁcally in Belsky's model. Parenting may involve such
differing behaviors as being punitive, being indifferent, being sensitive, or providing a
cognitively stimulating environment. It is important to distinguish between relationship
and control dimensions of parenting when building models of the determinants of
parenting, as the factors affecting the degree of parental warmth and involvement may
differ from those inﬂuencing disciplinary practices (Simons, Whitbeck, Conger, &

Melby, 1990). This study proposes the need to identify the unique determinants of

35

authoritative and authoritarian parenting. Identiﬁcation of speciﬁc factors related to each
of the component elements of the authoritative and the authoritarian measures would also

contribute to a better understanding of parenting practices.

Determinants of authoritative and authoritarian parenting

Research in the area of parenting style (e. g. authoritative and authoritarian) has
been focused on its relation with child outcomes, especially for adolescents (e. g.
Baumrind, D. 1991a, 1991b; Steinberg, Lambom, Darling, Mounts, & Dombush, 1994;
Weiss & Schwarz, 1996). Relatively few studies have been done studying the
determinants of authoritative and authoritarian parenting practices.

Parents have different levels of knowledge about child development and diverse
views about appropriate child-rearing practices. These differences in parental child-
rearing beliefs contribute to differences in parenting practices (Luster, Rhoades, & Haas,
1989). Chao (1994) indicated that Chinese may score high on "authoritarian" and
"restrictive" measures because they are related to the concept of "training" or "guan" in
Chinese child-rearing. Training or guan encompass a set standard of conduct enforced by
both the larger society and the parents. However, the motivations or intentions are to
assure the familial and societal goals of harmonious relations with others and the integrity
of the family unit, rather than to dominate the child. The concepts of training and
"authoritarian" have very different cultural roots, and thus very divergent implications.
Unlike the "authoritarian" concept which is associated with hostile, rejecting, and
somewhat uninvolved parental behaviors toward the child, the Chinese training concepts

include a highly involved concern and care for children (Chao, 1994). In terms of

36

maternal characteristics other than child-rearing beliefs, young maternal age has been
found to be related to strict, angry, or punitive attitudes toward the child during the ﬁrst
few years of life (Jones, Green, & Krauss, 1980; Ragozin, Basham, Cmic, Greenberg, &
Robinson, 1982). Fox, Platz, and Bentley (1995) also found a link between mother's
education, age, and her nurturing and discipline. Younger, less educated mothers were
more likely to use corporal punishment and yelling as discipline strategies, and were less
likely to be nurturant toward their children than older mothers.

In regard to the inﬂuence of child characteristics on parenting patterns, a few
studies examined parental behaviors as a function of child's age. Chinese parents tend to
be highly lenient, warm and affectionate toward infants and very young children until
they reach "the age of understanding", in contrast to the strict discipline they impose on
the older child (Bond, 1986; Ho, 1981; Ho & Kang, 1984; Sollenberger, 1968; Suzuki,
1980; Wolf, 1970). This indulgence is based on the belief that young children are
incapable of understanding, and therefore wrongdoing should be tolerated (Lang, 1946;
Wolf, 1970). Though Chinese emphasize the training concept, Kojima (1986) indicated
that most Asian parents stress the importance of early training, but not "earliest" training.
The various researchers found that parents believed the "age of understanding" was
around 6 years (Sollenberger, 1968; Wolf, 1970). According to Ho and Kang's study, it
may be as early as 3 or 4 years. In a more recent study by Kelly and Tseng (1992) with a
small sample of immigrant Chinese mothers and their 3- to 8-year—old children, parents'
reported parenting behavior of preschoolers and elementary school children were
examined. No abrupt changes were found in parenting behaviors of Chinese parents with

the beginning of primary school. The author suggested that it may be that the age at

37

which parents believe discipline needs to be introduced appear to be younger than the
ages examined in the study, or it is possible that changes in parenting behavior are
introduced gradually.

Child temperament has also been found to be related to parental behavior. In
general, more difficult children are likely to elicit more negative, controlling behavior
from their parents (Belsky, 1990). The mother's perception of her child as difficult was
found to be related to her destructive parenting, for example, hostility and rejection
(Simons, Whitbeck, Conger, & Melby, 1990). In a more recent study ( Fox, Platz, &
Bentley, 1995), mothers, who perceived their children as demonstrating more difﬁcult
behavior problem, were more likely to use corporal punishment and yelling as a
discipline strategic, and were less likely to be nurturant.

Contextual factors can inﬂuence what parents bring to the parenting situation.
Kohn (1963) indicated that the lower a parent's social class is, the more likely he or she is
to value conformity to external authority. Parents who value conformity emphasize such
things as obedience and good manners. The higher the parent's social class is, the more
likely he or she is to value self-control and responsibility. Higher SES parents have been
found to be less punitive than lower SES parents (Gecas, 1979; Kamii & Radin, 1967).
In Luster, Rhoades, and Haas's (1989) study, it was found that mothers' education,
mothers' occupational prestige, and family income were negatively related to mothers'
conformity values. In Dombusch et al.'s (1987) study of high school students, families
with higher parental education tended to be lower in authoritarian and permissive
parenting and higher in authoritative parenting. Fox et al. (1995) found that the tendency

of parents from the lower socioeconomic status level to use more frequent discipline (e.g.

38

corporal punishment and yelling) was moderated by higher maternal education. That is,
the mothers from the lower income levels who had more education reported less
discipline than did the mothers from the same income levels who had less education.

Bossard and Boll ( 1956) suggested that as the size of the family increases, better
internal organization and a higher degree of discipline are required for the sake of
efﬁciency and order. Nye et al. (1970) found that parents of large families were more
likely to employ corporal punishment in controlling children, whereas smaller families
were more likely to employ discussion and reasoning strategies. Elder and Bowerman
(1963) and Scheck and Emerick (1976), in studies of adolescents, found that as family
size increases, parents are reported as less communicative and more controlling, more
likely to use physical punishment, less likely to explain rules of conduct to the child and
less likely to give praise and support. These relationships generally held within social
class categories (middle class and lower class). In Fox et al.'s (1995) study of mothers
with very young children (1-4 years old), mothers who had more than one child at home
were more likely to use corporal punishment and yelling as a discipline strategy, and less
likely to be nurturant toward their children.

In sum, the previous cross-cultural studies did not seem to capture the real picture
of Chinese parenting. Intracultural variations in Chinese parenting should be examined.
In terms of studying the determinants of parenting, there is a need to identify the unique
determinants of authoritative and authoritarian parenting. Looking into the component
elements of authoritative and authoritarian parenting would also contribute to a better
understanding of parenting practices. In response to the limitations of the previous

studies, this study will examine authoritative and authoritarian parenting in Chinese

39

mothers and factors related to their parenting practices. The next chapter will describe

the methodology of this study.

Chapter 3

METHOD

This chapter describes the methodology of this study. Six sections are included:
(1) Sample; (2) Procedure; (3) Deﬁnition of the variables; (4) Instrumentation; (5)
Transformation method; and (6) Limitations.
Sample

About 1,100 mothers with children who were enrolled in preschools in the city of
Taipei were the sample of this study. A list of preschools from four regions of Taipei
were randomly sampled. Twelve preschools were selected. Eleven of them were
successﬁrlly contacted and agreed to participate in the study. Next, all families from the
selected preschools were invited to participate. The ﬁnal sample for data analyses
includes 463 mothers with 3- to 7- year-old children. The majority of the mothers had a
high school or college education, with an average income level of 4.3 on the 7 level
income scale. In American dollars, the average monthly family income for this sample
was about 3,200 dollars, or $ 38,400 annually. On average, they have 2.02 children, with
a range from 1 to 4. The demographic characteristics of the sample are presented in

Table 1.

4O

Table 1.
Demographic Characteristics of Sample

(N: 463)

1. Age

Mean

Range

II. Marital Status
Married

Single

Divorced

III. Parents' Educational Level

1. llliterate

2. Primary School

3. High School

4. College Graduate

5. Graduate School and Above

TV. Family Income‘

30,000 NT and below

30,000 NT to 50,000 NT

Above 50,000 NT to 70,000 NT
Above 70,000 NT to 90,000 NT
Above 90,000 NT to 1 1,000 NT
Above 11,000 NT to 13,000 NT
Above 13,000 NT

"‘ The family's total monthly income. NT= New Taiwan Dollars

41

mother's
34.8 (SD= 3.9)

22-48

98 %
.4%

1.3%

Mother’s
0 %
.7%

4| %

47 %

5%

2%

12%
2l%
19%
21%
9%

13%

child's
5.4 (SD= 1.0)
3-7

Father's
O %
.5%

26 %

55 %

7 %

42

PM

The translated version of the questionnaire along with a letter describing the study
were distributed to students in the randomly selected Taiwan preschools, who took the
questionnaires home for their mothers to ﬁll out. The mothers were asked to return the
questionnaires to the preschool. A reminder letter, which was orginally planned to
increase the return rate, was not sent out. Most preschool directors and teachers indicated
that the reminder letter would not help much, but would burden the teachers with extra
work. Instead, as suggested by the preschool directors and teachers, a gift certiﬁcate
drawing was used to encourage participation. Over 47 % of the mothers completed and
returned the questionnaire. Five hundred and ten questionnaires were retured. A total of
463 questionnaires were usable for analysis. Most of the unused questionnaires (47
questionnaires) were ones that only had the odd-numbered pages ﬁlled out on the double-
sided form.

L Li . . E . l 1

Chinese mothers' use of authoritative and authoritarian parenting and factors
related to these parenting practices were examined in this study. Predictors of
authoritative and authoritarian parenting practices were selected from three domains:
maternal characteristics, child characteristics, and contextual factors. The deﬁnitions of

key variables are presented next.

Authoritative parenting
Conceptual deﬁnition: The authoritative parent attempts to direct the child's

activities in a rational issue-oriented manner. He or she encourages verbal give and take,

43

shares with the child the reasoning behind parental policy, and solicits the child's
objections when the child refuses to conform. The parent enforces the adult perspective,
but recognizes the child's individual interests and special ways. The parent afﬁrms the
child's present qualities, also sets standards for future conduct, using reason as well as
power and shaping by regimen and reinforcement to achieve parental objectives
(Baumrind, 197 8).

Operational deﬁnition: The mother's scores on the Authoritative scale and four
subscales (Warmth and involvement, Reasoning/induction, Democratic participation,
Good natured/easy going) of the Parenting Behavior Questionnaire (PBQ) (Robinson,
Mandleco, Olsen, & Hart, 1996) were the indicators of authoritative parenting used in this

study.

Authoritarian parenting

Conceptual deﬁnition: The authoritarian parent values obedience as a virtue and
favors punitive, forceful measures to curb self-will at points where the child's actions or
beliefs conﬂict with what the parent thinks is right. The parent believes in keeping the
child in a subordinate role and in restricting his autonomy, and does not encourage verbal
give and take, believing that the child should accept a parent's word for what is right
(Baumrind, 197 8).

Operational deﬁnition: The mother's score on the Authoritarian scale and four
subscales (Verbal hostility, Corporal punishment, Nonreasoning/punitive strategies,
Directiveness) of the Parenting Behavior Questionnaire (PBQ) (Robinson, Madleco,

Olsen, & Hart, 1996) were used to measure authoritarian parenting.

44

Maternal Characteristics

Conceptual deﬁnition: Maternal characteristics are individual traits or attributes of
the mother, for example, age, education, child-rearing beliefs, and psychological well-
beings.

Operational: The mother's responses to the questions regarding her age and
education, her score on the "training" questionnaire items (Chao, 1994), and her score on

CES-D Depression scale (Radloff, 1977) were used to measure these characteristics.

Child Characteristics

Conceptual: Child characteristics are individual traits or attributes of the child, for
example, age and temperament.

Operational: Mother's report of the child's age and her perception of the child's
temperament on the EAS Temperament Survey (Buss & Plomin, 1984) were used. EAS

is an acronym for emotionality, activity level, and sociability.

Contextual Characteristics

Conceptual: Contextual characteristics are situational elements that can affect the
parenting behavior of mothers, for example, the family's social economic status (SES),
family size and parenting stress.

Operational: The mother's report on her husband/partner's educational level,
family monthly income, family size, and her score on the Parenting Daily Hassles (PDH)

(Cmic & Greenberg, 1990) were used to measure these factors.

45

Inﬂammation
Parenting Behavior Questionnaire (PBQ)

The pilot study of the PBQ (Robinson et al., 1995) used data from 1251 parents of
preschool and school-age children in the United States. The measure not only assesses
global typologies consistent with Baumrind's main conceptualizations for parents of
preadolescent children but also identiﬁes speciﬁc parenting practices that occur within the
context of the typologies. One hundred and thirty three items in this measure were
reduced to 62 items using principal axes factor analyses followed by varimax rotation.
Three global parenting dimensions emerged consistent with Baumrind's authoritative,
authoritarian, and permissive typologies. The parent is asked to rate how often she
exhibits the behavior with her child on a 5-point scale from (1) never to (5) always. The
authoritative scales consist of 27 items with a Cronbach alpha of .9l(with four factors:
Warmth and Involvement, Reasoning/Induction, Democratic participation, and Good
natured/ Easy going). The Authoritarian scales consist of 20 items with a Cronbach alpha
of .75 (with four factors: Verbal hostility, Corporal punishment, Nonreasoning/Punitive
strategies, and Directiveness). The items were constructed based on conceptualizations of
authoritative and authoritarian typologies drawn from the current literature that appeared
to have face validity. The use of separate items for each parenting style would avoid
forced correlations between the styles (Dombusch, et al., 1987).

A cross-cultural study of the measure was done by using data from families with
preschool-age children only (Robinson et al., 1996). The 62-item parenting questionnaire

was completed by parents from the United States, Australia, China, and Russia. Principal

46

axes factor analyses followed by varimax rotation were used to identify global parenting
styles and speciﬁc parenting practices for each culture. Though subtle parenting practices
within parenting styles differed between the four cultures, the overall parenting styles are
somewhat similar among parents in the four cultures. A comparison of the Chinese
version and American version for authoritative style showed that the China and the US.
versions share a "Reasoning/Induction" factor. China did not have Good natured/Easy
going and Warmth/Involvement factors. For China, the Warmth factor separated from
Involvement and China had a unique Respect/Conﬁdence factor. For authoritarian style,
China only had two factors: Verbal/Physical punishment and Verbal directiveness. For
the US. version, Verbal hostility and Corporal punishment were separate factors. The
US. version also had a Non-reasoning/Punitive strategies factor. For this sample, the
alphas of the China version were no better than the alphas of the US. version. Therefore,
due to this researcher's intention to examine how Chinese parents score on the US.
parenting measure as well as to relate these ﬁndings with previous cross-cultural studies,
the US. version of the measure was used in this study.

The US. preschool-age children version of the PBQ, which was used in this
study, was developed using data from 456 parents. The authoritative items consist of 22
questions with a reported Cronbach alpha of .88 (Warmth and Involvement with 7 items,
alpha=. 81; Reasoning/Induction with 7 items, alpha=.8 1; Democratic participation with 4
items, alpha=. 61; Good natured/Easy going with 4 items, alpha= .66). The Authoritarian
items consist of 18 questions with a Cronbach alpha of .85. (Directiveness with 4 items,
alpha=.63; Corporal punishment with 4 items, alpha= .73; Non-reasoning/Punitive

strategies with 6 items, alpha= .70; Verbal hostility with 4 items, alpha=.67). The item

47

scores were summed to produce the total score. There was an Authoritative total score
with four subscores and an Authoritarian total score with four subscores. The permissive
parenting pattern was not examined in this study.

There were slight modiﬁcations of the Authoritarian subscales for this study.
Based on a reliability analysis, three items which may not be suitable for a Chinese
sample were excluded from the analyses. The item "I tell my child what to do" was
deleted from the "Directiveness" subscale and thus increased the alpha from .57 to .64.
The items "I appear to be more concerned with own feelings than with my child's
feelings" and "When two children are ﬁghting, I discipline my child ﬁrst and ask
questions later" were deleted ﬁ'om the "Non-reasoning/Punitive strategies" subscale and
the alpha increased from .48 to .61. The alpha for the "Corporal punishment" subscale
for this sample was .77, and the alpha for the "Verbal hostility" subscale was .64. Fifteen
items remained in Authoritarian scale and the alpha for the total scale for this sample was
.84. Alphas of the Authoritative scale and its subscales for this sample were similar to
the reported alphas. For my sample, the alpha for the total Authoritative scale was .90;
the alpha for the Warmth/Involvement subscale was .77; the alpha for the
Reasoning/Induction subscale was .80; the alpha for the Democratic participation
subscale was .65; the alpha for the Good natured/Easy going subscale was .68. A copy of
the measure used in this study can be found in Appendix A.

There are several reasons for using the self-report measure. First, observations
and interviewing are expensive, so fewer subjects can be included in studies. Secondly, it
is doubtful that certain important features of Chinese mother's parenting behavior can be

observed in a limited period of time with a stranger present. Compared to American

48

parents, Chinese parents may be less accustomed to such intrusive techniques and have
relatively more concerns about their "performance". Moreover, they may have been less

likely to participate if more intrusive methods had been used.

The "training" questionnaire items

Seven "training" questionnaire items scored on a 5-point scale (ranging from 1 =
strongly disagree to 5 = strongly agree) were used to assess the mother's child-rearing
beliefs (Chao, 1994). The training questionnaire covered two areas, "ideologies on child
development and learning" (involving 4 items) and "ideologies on the mother-child
relationship" (involving 3 items), that were derived from the concepts of chiao shun (or
training or guan). The items were derived from the literature on Chinese child rearing
and involved ideas such as: the earliest possible introduction of training, the promotion of
training through exposing the young child as much as possible to the adult world and
through explicit example or comparison to other children, helping the child to succeed in
school as a primary way of expressing parent's love, the child being in the constant care of
the mother or family, and sleeping with the mother. These child-rearing beliefs may be
related to the concept of "authoritarian." However, the beliefs also imply for the Chinese
a very involved care and concern for the child. Thirteen items were originally developed
and summed to assess the Chinese "training" concept. Seven items distinguished between
the Chinese mothers and the European-American, with the Chinese scoring higher (see
Chao, 1994). The alpha for the questionnaire items was not reported by Chao. The
Chinese version of the questionnaire items are used in this study.

For this sample, item 7 "Child should be allowed to sleep in mother's bed "was

49

only signiﬁcantly related to item 6 "Child should be in the constant care of their mothers
or family", but not to the rest of the items. Deleting item 7 from the scale increased the
alpha from .61 to .64. Therefore, item 7 was deleted for this study. Six items remained
for the "training" questionnaire items" and the alpha was .64. The scores of the 6 items
were summed to produce the Chinese parenting belief score. A copy of the measure used

in this study can be found in Appendix A.

CES-D depression scale

Maternal depression was assessed with the CES-D, a widely used 20-item
depression inventory, with a reported Cronbach's alpha of .82 (Radloff, 1977). Each item
is scored on a 4-point scale in terms of frequency of occurrence during the previous week.
A score of 0 means that the symptom occurred rarely or none of the time, and a score of 3
means that the symptom occurred most of the time. The items are summed to produce a
total score and may range from 0 to 60. Higher scores indicate higher levels of I
depression. The usual cutoff to indicate "clinical depression" is 16. The CES-D
depression scale has been linked to parenting behavior in earlier studies (Radloff, 1977;
Reis, Barbera-Stein & Bennett, 1986). The measure was translated into Chinese and used
in this study. The alpha for this sample was .76. A copy of this measure can be found in

Appendix A.

EAS Temperament survey for children: parental ratings
This is a 20-item questionnaire (Buss & Plomin, 1984), assessing activity,

emotionality, sociability/shyness (the sociability subscale is viewed as a mixture of

50

sociability and shyness). Each item was rated on a scale of 1 (not characteristic or typical
of your child) to 5 (very characteristic or typical of your child). All but the Sociability
items have been taken from the Colorado Childhood Temperament Inventory (Rowe &
Plomin, 1977), which was developed in a study that included the Children's EASI-II items
(Buss & Plomin, 1975) and items based on the nine temperament dimensions of the New
York Longitudinal Study (Thomas, Chess, & Birch, 1968; Thomas, Chess, Birch,
Hertzig, & Korn, 1963). The CCTI "Sociability" scale is a measure of shyness, so the
Sociability scale is based on the adult EAS temperament survey items. The internal
consistencies of the three scales averaged .83. Test-retest reliabilities are available for 31
children with an average age of 3.6 years, the interval between ratings being one week.
The test-retest correlations were .72 for emotionality, .80 for activity, and .58 for
sociability/shyness. For this sample, the alpha for activity was .71; for emotionality it
was .77; for sociability/shyness it was .58. The measure was translated into Chinese and

used in this study. A copy of this measure can be found in Appendix A.

Parenting Daily Hassles (PDH)

This is a lS-item questionnaire assessing the degree or intensity of hassles rated
on a 5-point scale from no hassle = 1 to big hassle = 5 (Cmic & Greenberg, 1990). Two
subscales were included: (1) Parenting task: 8 items involve typical duties or tasks
normally associated with parenting (the alpha for this sample =. 78). (2) Challenging
behavior: items describing challenging child behaviors (reported alpha = .86; alpha for
this sample = .79). The alpha of the total PDH scale for this sample was .86.

Hassles are conceptualized as the imitating, ﬁ'ustrating, annoying, and distressing

51

demands that to some degree characterize everyday transactions with the environment
(Cmic & Greenberg, 1990). It was hypothesized that an individual's cognitive appraisal
of the signiﬁcance of these events for one's own well-being is the primary factor
predicting the impact of the stressor. The individual's appraisal of daily hassles are
considered better predictors of psychological well-being than are life events and operate
independently in the prediction of depression and anxiety across various adult populations
(DeLongis, Coyne, DaKof, Folkrnan, & Lazarus, 1982; Lazarus, Delongis, Folkrnan, &
Gruen, 1985; Cmic & Greenberg, 1990). The items were translated into Chinese and
used in this study. A copy of this measure can be found in Appendix A.
Imslatianmeihad

The instrument was translated into Chinese, and then translated back into English
by another person. A third person was consulted to resolve any disagreement between the
back translation of the items and the original. The back translation method was employed
for the purpose of validating the authenticity of the original.

The instruments used in this study, such as the Parenting Behavior Questionnaire
(PBQ), are self-report instruments. While the advantages of using self-report instruments
have been discussed above, there are problems related to the use of them also. As
Maccoby and Martin (1983) noted, parents may not be aware of certain aspects of their
own behavior and subjects may vary in their subjective interpretation of questions.
Susceptibility to response sets, particularly to social desirability, is also a problem. The
limitations of using self-report instruments must be considered when interpreting the data.

Using a single source of data is another limitation of the study. However, getting

52

information from the teacher would be extremely difﬁcult for an individual student's
project, and the preschool children are too young to report either their own or their
parents' behavior. Therefore, only the mother's self report is used in the study.

Finally, in regard to the sample in this study, this study has no intention of
generalizing the result to parents living in rural areas of Taiwan, or Chinese parents living
outside of Taiwan. However, it might apply to Chinese parents living in different
geographical locations, to some degree, due to the fact that they are all inﬂuenced by
Confucian ideology. Generalizing the ﬁndings to Chinese parenting with children in
different age groups (beyond preschool) also must be done cautiously.

Analxsis

In order to examine the ﬁrst research question concerning the identiﬁcation of
Chinese parenting patterns, the intercorrelations among the different factors from the
authoritative and authoritarian scales were computed. Second, a procedure was followed
to classify mother's scores on the Authoritative and Authoritarian scales as "high", "low",
or "medium". Four parenting patterns were then identiﬁed.

To examine the second research question regarding the determinants of parenting,
the zero-order correlations among the predictors were computed ﬁrst. Second, zero order
correlations between predictors and outcome variables (the Authoritative total scale, the
Authoritarian total scale, and each of their subscales) were computed. Multiple
regression was also used to examine the individual and joint contribution of the
predictors. Finally, a MANOVA followed by one-way ANOVAs were computed to
compare the differences of each predictors among the parenting patterns identiﬁed above.

For the signiﬁcance level, the most commonly used alpha levels, .001 and .05 were

53

chosen and presented in the analyses.

In this chapter, the methods used to address the research questions have been
described. To summarize, about 1,100 mothers with children who were enrolled in
preschools in the city of Taipei were sampled. F our hundred and sixty three mothers
returned completed questionnaires. Key variables assessed by the questionnaire included
mother's authoritative and authoritarian parenting practices, mother's age, education,
depression level, Chinese parenting beliefs, child's age, child's temperament, family's
SES, parenting daily hassles, and family size. In the next chapter, the results of the

analyses are presented.

Chapter 4

RESULTS

This chapter presents the results of the data analyses. Three sections are included:

_..1.-— - .-

 

“..a-Fn—o—g '-

(1) Descriptive statistics for the variables in the analyses; (2) Chinese parenting patterns;
and (3) Factors related to Chinese parenting.
I2 . . . . E l . l l . l 1

Table 2 presents the means and standard deviations of the predictor variables in
the analysis. Some predictor variables (e. g. mother's age, education, father's education,
and family income level) were described in the demographic characteristics of the sample
section of Chapter 3 (see Table 1) and are not repeated here.

Mothers had a mean score of 20.6 on the Chinese parenting beliefs measure,
which may range from 6 to 30, with higher score indicating stronger beliefs in the
Chinese "training" concepts. On average, mothers agreed with the Chinese parenting
beliefs assessed in this study to some degree. Maternal depression, assessed by CES-D
depression scale, had a mean score of 17.8, which was slightly higher than the usual
cutoff for "clinical depression" (cutoff score = 16). This may be a reﬂection of stressful
life in the big city, Taipei. The cutoff score was also based on a US. sample. The
average child age was 5.4 years-old, and the child's temperament was assessed on the 5-

point scale with an average score of approximately 3 for all three temperament indicators

54

55

(activity, emotionality, and sociability) assessed in this study. Mothers had a mean score
of 32.7 on the Parenting Daily Hassles scale, which may range from 15-75, with higher
scores indicating more parenting daily hassles perceived by the mothers.

On the parenting practices measures, mothers had mean scores ranging from 3.6
to 4.1 on the 5-point scale for the four authoritative subscales, and mean scores ranging
ﬁ'om 1.7 to 2.2 for the four authoritarian subscales, indicating a generally high level of
authoritative parenting and low level of authoritarian parenting across all the mothers.
The instrument used in this study for assessing parenting practices was an existing
instrument developed in Robinson et al.'s study (1996). For the comparison of mean
scores on the Authoritative scales and Authoritarian scales between this sample and the
the American sample in Robinson et al.'s study, please see Appendix B. This is for
reference only. Due to the fact that there were modiﬁcations of the original scale for this
study, as well as the sampling differences between the two studies, any comparison

should be made cautiously.

56

Table 2 Means, standard deviations of the variables in the analysis

Variable Mean Standard Deviation

Mother's characteristics

Parenting beliefs 20.6 3.9
CES-D scores 17.8 6.1
Child's characteristics

Age 5.4 1.0
Temperament

Activity 3.6 .7
Emotionality 2.4 .8
Sociability 3.4 .5

Contextual factors

PDH scores 32.7 8.3

Parenting Scales

Authoritative scale
Warmth/Involvement 4. 1 .6
Reasoning Induction 3.9 .6
Democratic participation 3.6 .7
Good natured/Easy going 3.9 .6
Authoritarian scale
Directiveness 2.2 .6
Corporal punishment 2.0 .6
Non-reasoning/Punitive strategies 1.7 .5
Verbal hostility 2.1 .5

57
:1 . .
Chinese parenting patterns were categorized based on the mother's scores on the
Authoritative scale, Authoritarian scale, and their subscales. Correlations among the

mother's scores on the Authoritative and Authoritarian subscales were examined ﬁrst.

Correlations among the component elements of A uthoritative and Authoritarian scales

The correlation analysis found that all four subscales of the Authoritative scale
(Warmth/Involvement, Reasoning/Induction, Democratic participation, and Good
natured/Easy going) were positively related to each other. Subscales of the Authoritarian
scale (Directiveness, Corporal punishment, Non-reasoning/Punitive strategies, and Verbal
hostility) were also signiﬁcantly related to each other in the positive direction. Regarding
the relation between the Authoritative and the Authoritarian scales, the Authoritative total
scale and the Authoritarian total scale were negatively related to each other (r=-.38,
p<.001). All subscales of the Authoritative scale were also negatively related to subscales
of Authoritarian scale at p<.00 1 level (see Table 3).

The hypotheses regarding the relation between the mother's use of authoritarian
parenting practices and her warmth/involvement toward the child, as well as the relation
between the mother's use of directiveness, corporal punishment and her use of non-
reasoning/punitive, verbal hostility parenting practices were not supported. The mother's
scores on the Authoritarian measures were negatively related to her score on the
Warmth/Involvement subscale. The mother's scores on the Directiveness and Corporal
punishment subscales were positively related to her scores on the Non-reasoning/Punitive

and Verbal hostility subscales.

58

Table 3 Intercorrelations among the components of authoritative and authoritarian

patterns.
(N=463)
A: Authoritative Scale B: Authoritarian Scale
A1 A2 A3 A4 Bl BZ B3 B4
A1
A2 .63**
A3 .56" .64**
A4 .66“ .68“ .58M
BI -.27** -.10* -.l6** -.33**
82 -.29** -.17** -.18** -.36** .45"
B3 -.34** -.26"‘* -.26** -.41"‘* .48" .41"
B4 -.18** -.l7** -.17** -.40** .53“ .49" .42M
A: Authoritative subfactors B: Authoritarian subfactors
Al: Warmth/ Involvement Bl : Directiveness
A2: Reasoning/Induction B2: Corporal punishment
A3: Democratic participation B3: Non-reasoning/Punitive strategies
A4: Good natured/Easy going B4: Verbal hostility
** P<.001

*P< .05

59

Categorization of Chinese parenting pattern

Cluster analyses were done based on the mother's scores on 8 parenting subscales.
The Quick Cluster method was used with 3- and 4- cluster solutions. The results
indicated that the mother’s scores on the subfactors of the Authoritative and Authoritarian
scales did not further differentiate Chinese parenting patterns than the scores on the
global parenting measures (i.e., Authoritative total scale and Authoritarian total scale)
did. For example, in a 4-cluster solution, mothers who were classiﬁed into cluster 1 had
the highest scores across the four Authoritative subfactors; mothers in cluster 2 had the
second highest scores across the four Authoritative subfactors; mothers in cluster 3 had
the third highest scores across the four Authoritative subfactors; mothers in cluster 4 had
the lowest scores across the four Authoritative subfactors. The same pattern also applied
to the scores across the Authoritarian subfactors. The ﬁnding suggested that, for
categorizing Chinese parenting patterns based on their scores on the parenting measures,
looking into the component elements of the scales was a redundancy. Therefore, a
decision was made to perform the categorization analysis based on the Authoritative and
Authoritarian total scores.

First, mothers were classiﬁed as "low", "medium", and "high", based on their
mean scores on the Authoritative and Authoritarian total scales. On the 5-point scale,
mothers with mean scores between 1 and 2.3 were classiﬁed as "low", above 2.3 through
3.6 as "medium", above 3.6 through 5 as "high". After the classiﬁcation, no mother in
this sample fell into the "high Authoritarian" category. Only 2 mothers were in the "low
authoritative " category; thus, they were excluded from the analysis. Therefore, four

parenting patterns were identiﬁed for this sample-- "high authoritative, low authoritarian",

60
"high authoritative, medium authoritarian", "medium authoritative, low authoritarian",
and "medium authoritative, medium authoritarian". Over 65% of the mothers fell into the
"high authoritative high, low authoritarian" group; 15% of the mothers were in the
"medium authoritative, low authoritarian"group; 12% were in the "high authoritative,

medium authoritarian" group", and 7% were in the "medium authoritative, medium

authoritarian" group (see Table 4).

Table 4. Four parenting patterns for Chinese mothers

Authoritative
Medium High
Authoritarian
Low N=69 N=295
( 1 5%) (65%)
Medium N=3 1 N=55

(7%) (12%)

61

Mothers' scores on the subscales of the Authoritative and the Authoritarian scales
were also classiﬁed as "low", "medium", and "high", based on the criteria described
above. For the Authoritative subfactors, 80% of the mothers were high in
Warmth/Involvement, and 20% were medium. No one in this study fell in the low group
for this factor. For the Reasoning/Induction subfactor, 73% were high, 26% were
medium, and only 1% were low. For Democratic participation, 55% of mothers were in
the high group, and 45% of mothers were in the medium group. Seventy-one percent of
the mothers were high on the Good natured/Easy going factor, 28% were medium, and
1% were low. For the Authoritarian subfactors, 57% of mothers were in the low
Directiveness group, and 42% of mothers were in the medium Directiveness group.
Seventy-eight percent of mothers were low in corporal punishment, 21% were medium,
and 1% were high. Most mothers (92%) in this sample were low in Non-
reasoning/Punitive strategies. Seventy-nine percent of mothers were low in Verbal
hostility, 20% were medium, and less than 1% were high.

It was noticed that the majority of the mothers, over 70%, were high in most of
the authoritative factors and low in most of the authoritarian factors. The exceptions
were Democratic participation and Directiveness. Mothers varied more widely in their
scores on these two subfactors. Approximately 56% of the mothers were high in
Democratic participation, and/or low in Directiveness, while approximately 46% of the
mothers practiced Democratic participation and Directiveness parenting at a medium

level.

62

E l 1 El . .
Relations among the predictor variables

To answer the second research question regarding factors related to Chinese
authoritative and authoritarian parenting, ﬁrst, correlations were computed to examine the
pattern of associations among the predictor variables (i.e., maternal characteristics, child
characteristics, and contextual factors). The results are presented in Table 5. Older
mothers had higher levels of family income and more children. Depressed mothers were
more likely to rate their child as emotional, and had higher scores on the Parenting Daily
Hassles scale. Among the three indicators of child temperament, sociability was
positively related to activity. Mother's education, father's education, and family income

were positively related to each other.

Zero-order correlations between the predictor variables and the dependent variables
Zero-order correlations between the predictor variables and the parenting
measures were examined to show the relations among them (see Table 6). It should be
noted that although many of the correlations were statistically signiﬁcant with this large
sample, they were modest in magnitude. All maternal characteristics were signiﬁcantly
related to the Authoritarian total scale and subscales, except for the relation between
mothers' education and verbal hostility. Mothers who were more likely to practice
Authoritarian parenting were younger, less educated, more psychological distressed, and
held more traditional Chinese parenting beliefs. All maternal characteristics were
signiﬁcantly related to the Authoritative total scale. When looking into the subscales,

older, highly educated mothers were more likely to be warm, involved, and had easy

63

going interactions with their children. Mothers' parenting beliefs were positively related
to Authoritative parenting, except for the Warmth/Involvement subscale. Mothers' CES-D
scores were negatively related to their scores on the Authoritative scales, except for
democratic participation.

Child's age was related to the Authoritative total scale. Mothers with older
children were less warm and involved with them. Mothers with more active children
were more likely to be warm, involved, had easy going interactions with their children,
and reasoned with their children. Child's age and child's activity level were not related to
the Authoritarian subscales. Mothers who perceived their children as relatively emotional
were more likely to practice Authoritarian parenting, and were less warm and easy going
with their children. Mothers with more social children were more likely to practice
Authoritative parenting. Child's sociability was also negatively related to the
Authoritarian total scale. Mothers who perceived their children as less social were more
likely to use Non-reasoning/Punitive strategies.

Among the contextual factors, Parenting Daily Hassles scores were signiﬁcantly
related to all parenting factors, with negative relations with the Authoritative factors and
positive relations with the Authoritarian factors. F athers' education was related to the
Authoritative total scale. Mothers with well-educated husbands were more likely to be
warm and involved with their children. Mothers with higher family income levels were
more likely to practice Authoritative parenting. Family income levels were also related to
the Authoritarian total scale. Mothers with higher family income levels were less likely
to use non-reasoning, punitive strategies. Mothers with more children in the family were

less likely to be warm and involved with their children.

64
Multiple regression analyses: Predictors of parenting factors

Multiple regression analyses were conducted to determine which of the predictor
variables were related to the parenting factors when other predictor variables were
controlled, and to assess how much of the variance in each outcome of interest was
explained by sets of predictor variables. The analyses were conducted for the total
Authoritative scale, the total Authoritarian scale, and separately for each of the subfactors
of the parenting scales.

In the ﬁrst step of the analyses, the predictor variables were entered in sets based
on the domain to which they had been assigned (e.g., maternal characteristics). Four
maternal characteristics were entered as predictor variables: mother's age, education,
parenting beliefs, and CES-D scores. Four child characteristics, child's age, child's
activity level, emotionality, and sociability were entered in the second analysis. The
contextual factors, Parenting Daily Hassles scores, father's education, family income and
size, were entered in the third analysis. In the ﬁnal analysis, all of the predictor variables
that were related to the outcomes in the preliminary analyses were entered
simultaneously. The results of the regression analyses for the total Authoritative and

Authoritarian scales and their subscales are presented in Tables 7, 8, and 9.

65

NOV *Nﬁr
econ.
N~ ~_ o_

nor
Nor
vo.

no.
we.
nor
wor

vor wor
nor _or
vor wor
camv. go.
*mor samv.
mo.

5 0

3mm.
nor

INN:
5.-

.59.
v0...
vor

00..

nor

met
1.3.

No..
a... _ N.

oo.

mo.

1 _.-
No..
we...

*3.

mo.
*2.
mo.
5.

*m_.

*v_r
snow.
eaﬁo.

No.

No.

cor

cor
1.3.-

mo...
*m_r

a... ~ N.

econ.

an m
cor

no.
_..:.-
...Z..
*2.

wor
5.
**0~.

8. v9. :5. va 3.
same no 598:2 .2

0885 .2
83836 935mm .3
.888 Sam .a

mug—Dab —QBNO~EGU
$328.0. .m
$8085 s
363. .0

own @230 .m
mutmtoauauaac 9230
8.88 ammo .v
as? Sauna .m
53833 $6502 .N

own area: ._

mozmtcuoaaaﬁ 9.3532

A816

85%? 33605 macaw mecca—0:088": .m 2an

66

Table 6. Zero order correlations between predictors and parenting factors

Authoritative Scales Authoritarian Scale
Total Subscales Total Subscales

Mother's 1 2 3 4 1 2 3 4
characteristics
Mother's age .11* .13* 0.04 0.1 .15* -.18* -.15* -.11* -.19** -.10*
Mother's education .17** .28** .08 .08 .10* -.15* -.14* -.15* -.16* -.03
Parenting beliefs .19** .08 .28** .14* .1 1* .31 ** .31 ** .27** .23** .14*
(353-1) scores -.17** -.i6"‘ -.14* -.09 -.21** .26" .26“ .12* .22** .20**
Child's
characteristics
Child's age -.10* -.18** -.04 -.03 -05 .05 .09 -.01 .07 .00
Activity . 12* .1 1* . 13* .05 .10* .05 .01 .08 -.02 .07
Emotionality -. 12* -. 10* -.08 -.05 -.21** .34** .24** .25** .23** .33**
Sociability .23** .16* .25** .16* .22** -. 10* -.04 -.07 -.1 1* -.08
Contextual factors
PDH scores -.22** -.15* -.17** -.15* -.30** .39** .34** .29** .20** .39**
Father's education .10* .20** .03 .04 .07 -.05 -.07 -.08 -.04 .03
Income .20** .28** .12* .14* .12* -.1 1* -.08 -.08 -.12* -.06
Number of children -.1 1* -.17** -.08 -.03 -05 -.01 .02 .00 -.03 -.02

Authoritative subfactors Authoritarian subfactors

1 .Warmth/Involvement l.Directiveness

2.Reasoning/Induction 2.Corporal punishment

3.Democratic participation 3.Non-reasoning/Punitive strategies

4.Good natured/Easy going 4.Verbal hostility

** p<.001 *p<.05

67

Table 7. Multiple regression analyses: Predictors of Authoritative and Authoritarian total

scales

(Standardized betas are presented in the table)

Authoritative total scale
Step 1 Step 2
Mother's characteristics
Mother's age .07
Mother's education .18" .12“
Parenting beliefs .26" .25"
CES-D scores -.18** -.13"
R2 .13
Child's characteristics
Child's age -.12*"' -.07
Activity .02
Emotionality -.1 1 "' -.02
Sociability .21 " .19“
R2 .07
Contextual factors
PDH scores -.23"”" -.21"”"
Father's education .04
Family income .18" .12*
Number of children -.1 1* -.09
R2 .11
Final model
R2 .23

** p< .001 * p< .05

Authoritarian total scale
Step 1 Step 2
-.14"' -.14*
-.08

.27" .24"
.19“ .14“
.16

.08

.08

.32" .13“
-. 10"

.13

.40" .26“
-.05
-.07
-.OO

.17

.29

68

Table 8. Multiple regression analyses: Predictors of Authoritative subfactor
(Standardized betas are presented in the table)

Warmth Reasoning Democratic Good natured

llnvolvement llnduction participation lEasy going
Mother's Step Step Step Step
characteristics 1 2 l 2 1 2 l 2
Mother's age .09 .02 .04 .l 1* .06
Mother's education .29** .15* .11* .07 .10* .08 .09
Parenting beliefs .16" .14* .33" .3 l ** .19" .19” .16** .18**
CES-D scores -. 14* —. 10* -. l7** -. 12* -.09 -.20""I -. 15*
R2 .14 .13 .05 .08
Child's
characteristics
Child's age -.20** -. 12* -.06 -.05 -.08
Activity .04 .02 -.03 .01
Emotionality -. 10* -.04 -.O6 -.04 -.19** -.06
Sociability .14* .15* .23** .23** .16* .14* .l9** .14*
R2 .07 .06 .03 .09
Contextual factors
PDH scores -.17** -.13* -.18** -.18** -.l7** -.18** -.30** -.27**
Father's education .12* .03 -.01 -.00 .04
Family income .23* .17" .12* .07 .14* .10* .10* .07
Number of children -. 13* -. 10* -. 10* -.09* -.04 -.06
R2 .14 .06 .05 .11
Final model
1?.2 .21 .21 .10 20

**p<.001

* p< .05

69

Table 9. Multiple regression analyses: Predictors of Authoritarian subfactors
(Standardized betas are presented in the table)

Directiveness Corporal Non-reasoning Verbal hostility
punishment [Punitive strategies

Mother's Step Step Step Step
characteristics 1 2 1 2 l 2 1 2
Mother's age -. 14* -. 13* -.O8 -.15* -. 14* -.07
Mother's education -.07 -.09 -.11* -.12* .02
Parenting beliefs .28" .26" .23** .21“ .20** .19" .12* .06
CES-D scores .19** .16" .07 .l6** .14* .17** .10*
R2 .16 .09 .12 .05
Child's
characteristics
Child's age .10* .l 1* .02 .09 .03
Activity .02 .l 1* .04 .02 . 10* .06
Emotionality .24** .04 .24** .13* .22** .11* .31** .16"
Sociability -.03 -.09 -.09 -.09
R2 .07 .08 .07 .12
Contextual factors
PDH scores .34** .24" .29** .20** .21** .11* .40** .28**
Father's education -.07 -.08 -.03 .03
Income -.05 -.02 -.09 -.05
Number of children .01 .00 -.02 -.00
R2 .13 .09 .06 .16
Final model
It2 .25 .15 .17 .20

** p< .001 *p< .05

7O

Predictors of the Authoritative scale: Mother's education, parenting beliefs,
depression, child's sociability, degree of parenting daily hassles, and family income were
signiﬁcant predictors of the mother's Authoritative total score. Parenting beliefs, child's
sociability, and degree of parenting daily hassles were signiﬁcant predictors for all four
authoritative subfactors. Mothers who held stronger Chinese parenting beliefs, perceived
their children as more social, and perceived their daily parenting tasks as less problematic
were more likely to practice authoritative parenting. The CES-D score was also a
signiﬁcant predictor of the Authoritative subfactors, except for the Democratic
Participation subfactor. Relatively depressed mothers were less likely to be warm, and
involved, use reasoning and induction discipline strategies, and have easy going
interactions with their children. Mother's education and child's age were predictive of the
Warmth/Involvement subfactor, but not the other subfactors. Mothers with higher levels
of education, and/or with younger children, were more likely to be warm and involved
with their children. Child's sociability was predictive of the Authoritative total scale and
all Authoritative subscales. Mothers who perceived their children as more social were
more likely to practice authoritative parenting in all aspects. Among the contextual
factors, family income was a signiﬁcant predictor of the Wannth/Involvement and the
Democratic participation subfactors, but was not a signiﬁcant predictor of the
Reasoning/Induction and Good natured/Easy going subfactors. Mothers with higher
family income levels were more likely to be warm and involved with their children, and
to use democratic participation strategies. Mothers with more children were less likely to
be warm and involved, and to reason with their children. The amount of variance in the

Authoritative subfactors explained by each model is presented in Table 8.

71

Predictors of the Authoritarian scale: Mother's age, parenting beliefs, depression,
the child's emotionality, and degree of parenting daily hassles were signiﬁcant predictors
of the mother's Authoritarian total score. Younger mothers were more likely to scold and
criticize their children, and use non-reasoning, punitive strategies. Parenting beliefs were
predictive of three of the Authoritarian subfactors, but were not signiﬁcantly related to the
Verbal hostility subfactor when other factors were controlled. Mothers with more
traditional Chinese parenting beliefs were more likely to be demanding, to scold or
criticize their children, as well as to use corporal punishment, non-reasoning, and punitive
discipline strategies. The CES-D score was a signiﬁcant predictor of three of the
Authoritarian subscales, but was not related to the Corporal punishment subscale.
Depressed mothers were more likely to practice authoritarian parenting. Moreover,
mothers with lower levels of education were more likely to use non-reasoning/punitive
strategies. Of the child's characteristics, child's emotionality was predictive of the total
Authoritarian scale, and three of the Authoritarian subscales, but was not a signiﬁcant
predictor of the Directiveness subscale when other factors were controlled. Mothers who
perceived their children as relatively emotional were more likely to practice authoritarian
parenting. Child's age was a signiﬁcant predictor of the Directiveness subscale. Mothers
with older children were more likely to be demanding, and to use scolding or criticizing
as discipline strategies. The Parenting Daily Hassles measure was a signiﬁcant factor for
all four subfactors. It was also the only contextual factor that was predictive of the
Authoritarian factors when other factors were controlled. Mothers who experienced
higher degrees of parenting daily hassles were more likely to practice authoritarian

parenting. The R2 for each model is presented in Table 9.

72

The most commonly identiﬁed predictor across the eight subfactors was degree of
parenting daily hassles, which was positively related to all of the Authoritarian factors,
and negatively related to all of the Authoritative factors. It indicated that mothers who
perceived their parenting daily tasks as big hassles were more likely to practice
authoritarian parenting, and less likely to practice authoritative parenting. Second in
frequency of occurrence as a predictive variable was the parenting beliefs measure, which
was positively related to all of the parenting factors, except for the Verbal hostility
subfactor, when other factors were controlled. This indicated that mothers with more
traditional Chinese parenting beliefs were more likely to practice various aspects of
authoritative and authoritarian parenting practices.

Of the child's temperament variables, child's sociability was predictive of all four
authoritative subfactors, and child's emotionality was predictive of all authoritarian
subfactors, except for the Directiveness subfactor. Child's activity level was not
predictive of any of the parenting measures. Child's age was related to the
Warmth/Involvement and Directiveness subfactors. Mothers with older children were
less likely to be warm and involved, more likely to be demanding, and to use scolding or
criticizing as discipline techniques. To test the "age of understanding" hypothesis as
discussed in the literature review section, a posthoc analysis was done by splitting
children into two groups, and comparing their scores on the Warmth/Involvement and
Directiveness subscales. Because the age range of the children was from 3 to 7, age 4, 5,
and 6 were tested individually as the splitting point. First, children were divided into two
groups--one group with children age 4 and under, the other group with children above age

4. A T-test analysis was done to compare the two groups on the Warmth/Involvement

73

and Directiveness scores. No differences were found between the two groups. Second,
children were divided into two groups--one group with children age 5 and under, the
other group with children above age 5; their scores on the Warmth/Involvement and
Directiveness scores subscales were compared. Finally, children were divided into two
groups with age 6 as the splitting point. No group differences were found regardless of

the age used to split the two groups.

Diﬂerences among the four parenting patterns on the predictor variables

In addition to examining the parenting predictors linearly, a MANOVA analysis,
followed by one-way ANOVAs were done to examine predictors of the four parenting
patterns identiﬁed earlier: "high authoritative, low authoritarian"group, "high
authoritative, medium authoritarian" group, "medium authoritative, low authoritarian"
group, and "medium authoritative, medium authoritarian" group. The MANOVA
analysis revealed that the four parenting groups differed signiﬁcantly on several predictor
variables: mother's age, mother's education, parenting beliefs, maternal depression, child's
emotionality, degree of parenting daily hassles, and family income. One-way ANOVAs
using Turkey-B tests were conducted to further examine the differences among the four
parenting patterns. For mother's age, mothers in the "medium authoritative, medium
authoritarian" group were the youngest, mothers in the "high authoritative, low
authoritarian" group were the oldest, and the other two groups fell in between. However,
no two groups were signiﬁcantly different at the .05 level. Mothers in the "high
authoritative, low authoritarian" group had the highest mean educational level among the

four groups, while mothers in the "medium authoritative, medium authoritarian" group

74

were the lowest in mother's education level. These two groups were signiﬁcantly
different in term of mother's education. For the parenting beliefs measure, the "high
authoritative, medium authoritarian" group had the highest scores on the Chinese
parenting beliefs, and was signiﬁcantly different from the other three groups. The
"medium authoritative, medium authoritarian" was the second highest, followed by the
"high authoritative, low authoritarian" group, and the "medium authoritative, low
authoritarian" group. The lowest parenting belief group, "medium authoritative, low
authoritarian", was also signiﬁcantly lower than the other three groups. For mother's
depression, "high authoritative, low authoritarian" mothers had the lowest CES-D scores.
The "medium authoritative, low authoritarian" group was the second lowest, followed by
the "high authoritative, medium authoritarian" group, and the "medium authoritative,
medium authoritarian" mothers had the highest CES-D scores among the four groups.
The lowest group, "high authoritative, low authoritarian" group, was signiﬁcantly lower
than the two highest groups.

Emotionality was the only child characteristics that differed among the four
parenting groups. Among the four groups, mothers in the "high authoritative, medium
authoritarian" group perceived their children as the most emotional, while mothers in the
"high authoritative, low authoritarian" group perceived their children as the least
emotional. These two groups' scores on child's emotionality were signiﬁcantly different.

The four parenting groups differed on two of the contextual factors--parenting
daily hassle scores and family income. The "medium authoritative, medium
authoritarian" mothers had the highest scores on the parenting daily hassles measure, the

"high authoritative, medium authoritarian" mothers were the second, followed by the

75

"medium authoritative, low authoritarian" group, and the "high authoritative, low
authoritarian." The group with the lowest scores was signiﬁcantly different from the two
highest groups. The "high authoritative, low authoritarian" group was highest on family
income, while the "medium authoritative, medium authoritarian" group was the lowest.
These two groups were signiﬁcantly different.
W115

To summarize the results, the resulst for each research question/hypothesis are
chart listed below:

Research question/hypothesis Result

 

Q1 .1 How do the component elements of All four subscales of the Authoritative

the Authoritative (Warmth/ Involvement, scale were positively related to each other.

Reasoning/ Induction, Democratic All four subscales of the Authoritarian
participation, and Good natured/Easy scale were positively related to each other.
going) and the Authoritarian All subscales of the Authoritative scale
(Directiveness, Corporal punishment, were negatively related to subscales of the
Nonreasoning/Punitive strategies, and Authoritarian scale.

Verbal hostility, ) scales relate to each

 

 

other?

 

 

76

 

Hl.l. There is no relationship between the
mother's use of authoritarian parenting
practices and the mother's

warmth/involvement toward the child.

Not supported

The mother's scores on the Authoritarian
measures were negatively related to her
score on the Warmth/Involvement

subscale

 

H l .2. There is no relationship between the
mother's use of directiveness, corporal
punishment parenting practices and the
mother's use of non-reasoning/punitive
strategies, verbal hostility parenting

practices.

Not supported

The mother's scores on the Directiveness
and Corporal punishment subscales were
positively related to her scores on the
Non-reasoning/Punitive and Verbal

hostility subscales.

 

 

Q12. What parenting patterns will be
identiﬁed in Chinese mothers based on
their scores on the Authoritative and the

Authoritarian parenting measures?

Four parenting patterns were identiﬁed
"high authoritative, low authoritarian",

"high authoritative, medium

N If
9

authoritarian medium authoritative, low
authoritarian", and "medium authoritative,

medium authoritarian".

 

 

Q2. What factors are related to the
Chinese mother's authoritative and

authoritarian parenting practices?

 

Maternal characteristics, child
characteristics, and contextual factors
were all predictive of Chinese mother's
authoritative and authoritarian parenting

practices.

 

 

77

 

 

H2. 1. There is a negative relationship Supported
between mother's age and her use of

authoritarian parenting.

H2.2. There is a positive relationship Partially supported

between mother's education and her use of
authoritative parenting, and a negative
relationship between mother's education

and her use of authoritarian parenting.

Mother's education was positively related
to the mother's scores on the Authoritative
total scale and the Warmth/Involvement
subscale.

Mother's education was negatively related
to the Non-reasoning/Punitive strategies
subscale, but not to the Authoritarian total
scale and the other three Authoritarian

subscales.

 

 

H2.3. There is a positive relationship
between Chinese parenting beliefs and the
mother's use of authoritarian parenting,
and a positive relationship between
Chinese parenting beliefs and the mother's
use of warmth/involvement aspect of

authoritative parenting.

 

Supported

 

 

78

 

H2.4. There is a negative relationship
between the mother's depression and her
use of authoritative parenting, and a
positive relationship between the mother's
depression and her use of authoritarian

parenting

Supported

 

H2.5. There is a positive relationship
between the child's sociability and the

mother's use of authoritative parenting.

Supported

 

H2.6. There is a positive relationship
between the child's emotionality and the

mother's use of authoritarian parenting.

Supported

 

 

H2.7. There is a positive relationship
between the child's age and the mother's
use of authoritarian parenting, and a
negative relationship between the child's
age and the mother's warmth/involvement,
as well as her good natured/easy going

interaction with the child.

Supported

 

 

 

79

 

H2.8. There is a positive relationship Partially supported

between the family's SES status and the There is a positive relationship between
mother's use of authoritative parenting, the family's SES status and the mother's
and a negative relationship between the use of authoritative parenting.

family's SES status and the mother's use There is no relationship between the

 

of authoritarian parenting family's SES status and the mother's use
of authoritarian parenting.
H2.9. There is a positive relationship Not supported

between the number of children at home There is no relationship between the
and the mother's use of authoritarian number of children and the mother's use

parenting. of authoritarian parenting.

 

H2. 10. There is a negative relationship Supported
between the mother's perception of the
degree of her parenting daily hassles and
her use of authoritative parenting, and a
positive relationship between the mother's
perception of the degree of her parenting

daily hassles and her use of authoritarian

 

 

parenting.

 

 

In the next Chapter, interpretations of the results are presented and the

implications of this study are discussed.

Chapter 5

DISCUSSION

In this discussion section, results are summarized and interpreted under the
following headings: (1) Chinese parenting patterns; (2) Factors related to Chinese
parenting practices; (3) Examination of Belsky's model by differentiating different aspects
of parenting behavior; (4) Conclusions; and (5) Directions for future research,
Chinesenatcntingnanenn
Interrelations among the subscales of the Authoritative and the Authoritarian measures

In response to the concern regarding the appropriateness of applying the
authoritative and authoritarian parenting typologies to assess Chinese parenting, the
interrelations among the subscales of the Authoritative and the Authoritarian scales were
computed to examine if the relations among them for Chinese would be different from
what was expected based on the western parenting literature. According to the literature
reviewed in Chapter Two, Chinese mothers' "restrictive", "authoritarian" parenting is
unlikely to be punitive, and has the component of parental concern and involvement.
Therefore, one would expect, for example, that Chinese mothers' use of authoritarian
parenting may not be related to their warmth and involvement toward children. Chinese
mothers' use of some aspects of authoritarian parenting practices (e. g. directiveness and

corporal punishment) may not be related to punitiveness and hostility. However, the

80

81

hypothesis was not supported in this study. The relations found within and between the
subscales of the Authoritative and Authoritarian scales were consistent with the literature
on western parents. Subscales within the Authoritative scales and the Authoritarian
scales were all positively and signiﬁcantly related to each others, and correlations
between the Authoritative subscales and the Authoritarian subscales were all negative and
signiﬁcant. Mothers who practiced one aspect of authoritative parenting, were more
likely to practice other aspects of authoritative parenting, and were less likely to practice
all aspects of authoritarian parenting. Mothers who practiced authoritarian parenting in
on aspect, were more likely to practice all other aspects of authoritarian parenting, and
less likely to practice authoritative parenting in all aspects. Based on this analysis, the
patterns of Chinese parenting did not seem to depart from the parenting patterns found in

western cultures.

Categorization of Chinese parenting patterns

An analysis was then done to categorize Chinese parenting patterns based on their
scores on the total Authoritative and Authoritarian scales. Mothers were divided into
"low", "medium", and "high" groups, based on their mean scores on the Authoritative and
Authoritarian total scales. Contrary to the impression from many previous cross-cultural
studies which seem to present Chinese as relatively "controlling" and "authoritarian", the
results of this study showed that the authoritative parenting pattern was predominant
among participants. Over 65% of Chinese mothers were highly authoritative and low in
authoritarian parenting. None of the mother in this sample met the criteria for high

authoritarian parenting practices. There were approximately 20% of the mothers who

82

scored in the medium range for authoritarian parenting. However, it should be noted that
the use of authoritarian parenting was accompanied by medium to high scores on the
authoritative parenting measure. Moreover, there were also about 15% of the mothers
who were medium on the authoritative and low on the authoritarian parenting measures.
The correlational analysis in this study indicated that mothers who were more likely to
practice authoritarian parenting were less likely to practice authoritative parenting. The
present analysis provides a clearer picture of Chinese parenting by going beyond
examining the scores linearly. It showed that the use of authoritative and authoritarian
parenting, though negatively correlated, were not necessarily independent from each
other. Some Chinese mothers practice authoritarian parenting at a medium level, and also
practice authoritative parenting at a medium to high level.

If a "purely" authoritative or authoritarian parenting style was deﬁned by having
scores that were high on one measure and low on the other, 65% of the Chinese mothers
in this sample would be identiﬁed as "purely" authoritative. No one was in the "purely"
authoritarian group. Thirty-ﬁve percent of the mothers in this sample could not be simply
described as "authoritative" or "authoritarian." The results support the argument that
depicting Chinese as "authoritarian" based on their higher "authoritarian" scores relative
to western samples, is not only misleading, but also an over simpliﬁcation.

In the Literature Review section, the importance of looking into the component
elements of global parenting typologies was stressed. Contrary to the researcher's
expectations, the degree and weighing of the component elements of authoritative and
authoritarian parenting scales did not differentiate further Chinese parenting patterns than

the Authoritative and Authoritarian total scales did. However, going beyond the global

83

parenting typologies and examining the subfactors did provide some insights into Chinese
parenting. As noted in the result section, over 70 % of the mothers were classiﬁed as
"high" in most of the authoritative subfactors, and "low" in most of the authoritarian
subfactors. The exceptions were the Democratic participation and the Directiveness
subfactors; scores on these subscales varied more widely among the mothers. Fifty-ﬁve
percent of the mothers were high on Democratic participation, and 45% were medium.
Fifty-seven percent of the mothers were low on the Directiveness subfactors, and 42%
were medium. In addition, most mothers in the sample (92%) hardly used non-reasoning,
punitive discipline strategies. It is suggested that, although Chinese parents have higher
scores on the authoritarian measures than Euro-American parents, it may be due to their
use of directive discipline strategies (demanding, scolding, and criticizing), and not
because they use non-reasoning, punitive strategies (e.g., punishing children with no
explanation or justiﬁcation). Instead of punishing children with no explanation, Chinese
parents are more likely to use directive discipline strategies to let children know what was
expected from them. Moreover, the results indicated that a certain percentage of Chinese
mothers may not highly encourage democratic participation in the family, but again, they
were very unlikely to be punitive toward their children. Baumrind (1968) has identiﬁed
several subpattems of parental socialization. There are two authoritarian subpattems: one
involves parents who are rejecting or indifferent, while the other involves parents who do
not encourage independence and individuality but are not rejecting. This study suggested
that Chinese's use of authoritarian parenting is very likely to be similar to the latter
subpattem identiﬁed by Baumrind. However, the independence that Chinese parents do

not particularly encourage is family independence; individual independence is not

84

necessarily discouraged (Chen & Uttal, 1988; Lin & Fu, 1990).

Quite a few studies provide support for the distinction between family
interdependence and individual interdependence. Lin and Fu (1990) found that both
Chinese in Taiwan and immigrant Chinese-American parents tended to have higher
ratings on encouragement of independence than did Caucasian-American parents. In
explaining the unexpected ﬁnding, they suggested that although the Chinese emphasize
family interdependence, children are also encouraged to be independent outside the
family to fulﬁll personal goals and to adjust to the changing demand of the society.
Kagitcibasi (1989) studied socialization values in Taiwan. The author found that being
obedient to parents and being independent and self-reliant both were viewed as desirable
traits by parents in Taiwan. Xu and the colleague (1991) invited a panel of Chinese
psychologists to select family socialization dimensions and children's characteristics
which were most critical and desirable in contemporary Chinese culture. It was found
that for family socialization dimensions, parameters such as democracy in making family
decisions which are usually investigated in American studies, were not included.
However, independence was included as one of the most desirable child characteristics.
In the present study, the results showed that approximately 70% to 80% of the mothers
would practice authoritative parenting (warmth/involvement, reasoninyinduction, and
having good natured/easy going interaction with their children) extensively, but only 55%
of the mothers would practice extensive democratic participation with their children.
Though mothers with more traditional Chinese parenting beliefs were more likely to
practice authoritative parenting, including democratic participation, a lower percentage of

mothers would practice democratic participation at a high degree compared to other

85

aspects of authoritative parenting. It seems that when making family rules is involved,
independent thinking and individuality, which may challenge parental authority, are not
particularly encouraged.

The results of this study revealed that the majority of the mothers in this sample
were highly authoritative and rarely practiced authoritarian parenting. For those who
practiced authoritarian parenting, the use of the authoritarian discipline was at "medium"
level, and was often accompanied by the use of authoritative parenting at a medium to
high level. As Chung (1994) pointed out, parental control, strict discipline, and children's
obedience may not represent the same inﬂexibility and rigidness in Taiwan as they do in
American studies. It may be that Chinese parents would typically use authoritative
parenting, and would use authoritarian discipline techniques when the authoritative
discipline does not work and parental authority is challenged. Stevenson, Chen, and Lee
(1992) also reported that in response to young children's wrong doings, Chinese parents
often try to discuss and explain misbehavior with their children. However, if they
continue the negative behaviors, direct forms of discipline, such as threats of punishment
or actual use of physical punishment, would be used. To these parents, the use of
authoritarian parenting was well-intentioned, rather than punitive. To the children, the
parents' use of authoritative parenting makes them emotionally secure, and more receptive
to their parents' occasional use of authoritarian discipline. However, it should be noted
that there was not extreme use of authoritarian parenting found in this sample. Rohner
and Pettengil's study (1985) on Korean parenting found that even though mothers high in
reported control were perceived as being high in warmth and low in neglect, strict

maternal control was also seen by Korean youths as being slightly hostile and rejecting.

86

While it is suggested that, for Chinese, the use of authoritarian parenting is very likely to
be out of concern and love, and is often accompanied by the use of authoritative
parenting, it may not apply to the case in which extensive authoritarian parenting is used.
WWW

Regression analyses results

The second research objective was to examine factors related to Chinese mothers'
parenting. Belsky's model of the determinants of parenting was supported. Maternal
characteristics, child characteristics, and contextual factors all were predictive of Chinese
mothers' parenting practices.

Maternal characteristics: Regarding maternal age, consistent with the hypothesis,
younger mothers were more likely to be punitive, or to use scolding and criticizing as
discipline techniques. The hypothesis about the relation between maternal education and
parenting practices was partially supported. Mothers with higher levels of education were
more likely to be involved with and warm toward their child, and less likely to be
punitive. Mother's education was negatively related to the Authoritarian factors in the
bivariate analyses; however, it was not predictive of most of the authoritarian parenting
subfactors when other maternal characteristics were controlled. The fact that maternal
education was signiﬁcantly related to both mother's age and parenting beliefs may explain
why education was not a signiﬁcant predictor of parenting practices in the regression
analysis. It may be Chinese parents' use of authoritarian parenting is less likely to be
inﬂuenced by their education than their parenting beliefs. Another explanation could be
that the average education level that people in Taiwan receive is relatively high,

especially in urban areas. Thus, the inﬂuence of education on parenting practices might

87

be reduced if there is a restricted range on the education variable.

Maternal depression was predictive of both authoritative and authoritarian
parenting practices. As it was hypothesized, depressed mothers were less likely to
practice authoritative parenting and more likely to practice authoritarian parenting.
However, mother's depression was not related to the use of democratic participation and
corporal punishment. Mother's use of democratic participation and corporal punishment
were more strongly related to parenting beliefs, than to mother's age, education level, or
psychological distress. Chinese training concepts do support the practice of democratic
participation toward children. However, as discussed above, democratic participation
may not be particularly encouraged due to the emphasis on family interdependence and
parental authority among the Chinese. Regarding the use of corporal punishment, Chen,
Lu, Hung, and Chen ( 1980) surveyed Chinese teachers, students, and parents regarding
corporal punishment in Taiwan. They found that most of the teachers, students, and
parents (over 92%) think that as long as no injury is inﬂicted, appropriate physical
punishment (e.g., palm swatting or standing and facing the wall for a long period of time)
is permissible or even necessary to correct students' misbehavior. It showed that
traditional Chinese parenting does support the use of corporal punishment in child
training to a certain degree.

Chinese parenting beliefs appeared to be very important predictors of both
authoritative and authoritarian parenting. As expected, mothers with more traditional
Chinese parenting beliefs were more likely to practice authoritarian parenting (though it
lost its signiﬁcance in predicting Verbal hostility when other factors were controlled).

The results also support the hypothesis that mothers with more traditional Chinese

88

parenting beliefs were more likely to be warm and involved with their children.

However, interestingly, Chinese parenting beliefs were also predictive of the other three
authoritative subfactors: Reasoning/Induction, Democratic participation, and Good
natured/Easy going. The seemingly contradictory ﬁnding-Chinese parenting beliefs are
positively related to both authoritative and authoritarian parenting-is of particular
interest. Among the Chinese, the use of authoritative and authoritarian parenting
practices are both imbedded in the concept of training. "Training" children is viewed as a
major parental task, and is primarily motivated by the parents' intense concern for their
children. This may explain the reason why the use of the two parenting patterns are not
viewed as inconsistent for the parents who hold traditional Chinese parenting beliefs.
These parents would use both inductive and power-assertive disciplinary techniques to
"train", or "guan" their children. Xu et al.'s ﬁndings (1991) also supported this notion. In
their study of family socialization and kindergarten children's personality characteristics
in China, parental control was signiﬁcantly correlated with the use of reasoning, inductive
strategies. It seems that Chinese parents set rules and restrictions on the children, but
they do "reason" with their children, instead of saying "do it because I say so." Mild
verbal or physical punishment may be used when the child continues to disobey parents.
The fact that authoritarian parenting is inﬂuenced by the concept of training may also
explain the relation between Chinese parenting and child outcomes. Baumrind (1993)
claimed the importance of parents' belief in their own effectiveness on child-rearing-
causal attributions that assign primary responsibility for child outcomes to genetic factors,
the effect of which parents believe they cannot change, undermine parents' beliefs in their

own effectiveness, whereas parents' attribution of responsibility for their children's

89

outcomes to parents' own actions is associated with more effective caregiving, which in
turn is associated with more positive child outcomes. The Chinese training beliefs may
contribute to parents' more effective use of both authoritative and authoritarian parenting,
and in turn result in more positive outcomes in Chinese children.

Regarding the question concerning the relation between mothers' Chinese child-
rearing beliefs and education, although in general, mothers with more education were less
likely to hold traditional Chinese child-rearing beliefs, even the mothers with the highest
education level (graduate and above) still agree with most of the parenting beliefs items
examined in this study. The results suggested that, although a departure from the
traditional Chinese child-rearing beliefs was shown among the higher educated Chinese
mothers, traditional Chinese child-rearing beliefs were still held among most
contemporary Chinese mothers.

Child characteristics: Regarding child characteristics, child '5 age was negatively
related to Warmth/Involvement, and positively related to Directiveness. The results are
consistent with the hypothesis that Chinese parents tend to be highly lenient, warm and
affectionate toward infants and young children, but impose more strict discipline on
older children. However, post hoc analyses suggested that the change seems to be
gradual, instead of being in response to an agreed upon "age of understanding."

In terms of child temperament, child's sociability was predictive of all four
authoritative subfactors, and child's emotionality was related to all four authoritarian
subfactors (though losing its signiﬁcance in predicting Directiveness when other
predictors were controlled). Mothers who perceived their children as more social were

more likely to practice authoritative parenting. Mothers with emotional children were

90

more likely to practice authoritarian parenting. The results support the hypothesis based
on Buss and Plomin's study (1975). According to Buss and Plomin, it is usually the
troublesome child who forces the parent to take special pains, and the trouble typically
concerns the child's excessive impulsivity or emotionality. The highly social child is
more responsive and offers more feedback from the mother's use of authoritative
practices, while a highly emotional child is less likely to develop self-control, which may
contribute to the parents' use of power assertive discipline strategies.

No speciﬁc hypothesis regarding the relation between child activity and parenting
practices was made. The result showed that the child's activity level was not predictive of
the mother's parenting practices. Child's activity was positively related to the mother's
authoritative parenting in the zero-order correlation analysis. However, because the
child's activity level was related to the child's sociability (r=.45, p<.001), after controlling
for other child characteristics in the regression analysis, child's activity was not related to
the mother's authoritative parenting. Child's activity was not related to the Authoritarian
factors either. It should be noted that the activity level assessed by the EAS
Temperament survey did not appear to assess hyperactivity. It may be that children who
have a high score on activity level are relatively active, but the level is not so high as to
create problems for parents. In general, compared to the other child temperament
indicators-- sociability and emotionality--child activity does not seem to be an important
predictor of the parenting practices assessed in this study.

Contextual factors: Among the contextual factors, the degree of parenting daily
hassles was predictive of all eight parenting outcomes. As expected, mothers

experiencing intensive parenting daily hassles were less likely to practice authoritative

91

parenting, and were more likely to practice authoritarian parenting. Parenting daily
hassles appear to be an important indicator of contextual stress. They not only contribute
to the mother's psychological distress, but also are meaningful independent predictors of
maternal parenting practices. As Cmic and Greenberg (1990) mentioned, it is important
to emphasize that the parenting daily hassles examined in this study represent typical or
normal events. In this study, when mothers' perceptions of children's temperament were
controlled, the parenting hassles measure continued to be a signiﬁcant predictor of
parenting practices. This suggests that hassles involving challenging behavior cannot be
considered simply another measure of child's difﬁcult behavior. Mother's cognitive
appraisal of the signiﬁcance of the events as a stressor contributes independently to the
prediction of parenting behavior (Cmic & Greenberg, 1990).

/ 1; Of the family’s SES, family income level was predictive of the
Warmth/Involvement and Democratic Participation subscales. However, another
indicator of the family's SES, father's education, was not predictive of the mother's
authoritative parenting. Contrary to expectations, neither indicator of the family's SES,
family's income or father's education, was related to the mother's authoritarian parenting.
Though parents' occupations were not assessed in this study, the use of authoritarian
parenting was not predicted by family income, father's education, or mother's education
(with the exception of the Non-reasoning/Punitive strategies subfactor). For Chinese
mothers, the family's SES did not appear to be a signiﬁcant predictor of their use of the
authoritarian parenting. However, it is recognized that the variability in SES in this
sample is not large.

Number of children was predictive of the Warmth/Involvement and

92

Reasoning/Induction subfactors. However, contrary to expectations, it was not related to
the mother's use of authoritarian parenting. This may be due to the fact that the average
number of children in this sample is small, which is 2.02. The range is from 1 to 4, and
only 2 percent of the mothers had 4 children. The stress caused by having many children
therefore may not be detected in this sample. However, having more children at home
does seem to reduce the time and energy for being involved, and reasoning with each

child.

Cross cultural implications

One reason for examining the determinants of Chinese parenting was to determine
if factors related to parenting practices for Chinese parents would be similar to those for
parents in western societies. This study suggests that there are certain factors that predict
parenting behavior across cultures--speciﬁcally, mother's psychological distress, child
temperament, and parenting daily hassles. However, this study also suggests that
"culturally speciﬁc" parenting beliefs plays an important role in inﬂuencing parenting
behavior. Studying Chinese parenting practices without taking their cultural roots into
consideration could lead to a misinterpretation of certain parenting behaviors, and thus
fail to understand their inﬂuences on child outcomes. Besides the inﬂuence of Chinese
parenting beliefs on Chinese mothers' parenting, the relation between SES and Chinese
mothers' use of authoritarian parenting was not consistent with what would be expected

based on western literature. The family's SES, assessed by the father's education level

and family income, was not predictive of the mother‘s authoritarian parenting. Among

the Chinese, the use of authoritarian parenting was more likely to be related to parenting

93

beliefs, along with other factors, such as parenting daily hassles, the child's emotionality

and the mother's psychological distress.

MANOVA and One-way ANOVA results

MANOVA analyses followed by one-way ANOVAs were done to examine
predictors of the four parenting patterns identiﬁed in this study: "high authoritative, low
authoritarian"group, "high authoritative, medium authoritarian" group, "medium
authoritative, low authoritarian" group, and "medium authoritative, medium
authoritarian" group. Mother's age, mother's education, parenting beliefs, maternal
depression, child emotionality, degree of parenting daily hassles, and family income were
signiﬁcant predictors of group status when the mother's scores on the Authoritative and
Authoritarian scales were considered simultaneously. In general, mothers in the "high
authoritative, low authoritarian" group were the most educated, least likely to be
depressed or perceived their parenting tasks as big hassles, and had the highest average
family income level among the four groups, while mothers in the "medium authoritative,
medium authoritarian" group were least educated, most likely to be depressed and
perceived their parenting tasks as big hassles, and had the lowest average family income
level. The "high authoritative, medium authoritarian" group and the "medium
authoritative, low authoritarian" group fell in between these two groups.

It should to be noted that, among those mothers who practiced authoritarian
parenting at a medium level, approximately 37% of them had meditun scores on the
authoritative measure ("medium authoritative, medium authoritarian" group ), and the

majority of them scored high on the authoritative measure ("high authoritative, medium

94

authoritarian" group). Comparing these mothers who practiced authoritarian parenting to
some degree, with those who practiced "purely" authoritative parenting ("high
authoritative, low authoritarian" group), showed that "purely" authoritative mothers and
mothers who practiced both authoritarian and authoritative parenting at medium levels
differed on several factors, as indicated above. Mothers who practice authoritarian
parenting, but also practice authoritative parenting at a high level, did not differ
signiﬁcantly from the "purely" authoritative mothers on maternal education and family
income level. They were, however, more likely to be depressed and perceived the
parenting daily tasks as bigger hassles than the "purely" authoritative mothers. Moreover,
Chinese parenting beliefs also differentiated between the two groups. In fact, mothers
who practiced authoritarian parenting at medium level, while also practicing authoritative
parenting at high level, held the strongest Chinese parenting beliefs among the four
groups. This ﬁnding again supports the results from the regression analysis. Chinese
parenting beliefs contribute to the use of both authoritative and authoritarian parenting.
It is suggested that special attention should be paid to the "high authoritative, medium
authoritarian" mothers. These mothers did not differ from the "purely" authoritative
mothers on mother's age, education, child's age and temperament, as well as family's SES
and size. Future studies should be done to compare children's outcomes for these two
groups. It is proposed that the use of authoritative parenting at a high level would offset
any negative effects of authoritarian parenting on child outcomes.

Regarding child characteristics, when authoritative and authoritarian measures
were considered together, only child emotionality differed among the four parenting

groups. Mothers in the "high authoritative, medium authoritarian" group rated their

95

children signiﬁcantly higher on the "emotionality" measure than mothers in the "high
authoritative, low authoritarian" group. It appeared that child's emotionality, which may
make parenting more difﬁcult, inﬂuenced parents' use of authoritarian parenting.
However, it may not necessarily inﬂuence parents' use of authoritative parenting.

.-__111-.0103‘ . 1nd 1 95am '1' its: an a- u". I .1. ‘1 -1° 0'11 'r

In general, Belsky's model of the determinants of parenting was supported.

Maternal characteristics, child characteristics, and contextual factors all contributed to
predicting individual differences in Chinese mothers' parenting. This study supported
Belsky's assertion that individual parenting differences are multiply determined, and no
one variable may be said to be the one "driving force" behind parenting practices. Unlike
most previous studies which examined Belsky's model by treating "parenting" as a
general term, this study not only examined determinants of authoritative and authoritarian
parenting individually, but also examined factors related to each of their subfactors. It
was found that different factors predicted authoritative versus authoritarian parenting.

Different sets of factors were also found to be predictive of different components of

authoritative and authoritarian parenting.

Determinants of the Authoritative v.s. Authoritarian total scales

The Authoritative and Authoritarian parenting total scales shared several
determinants--parenting beliefs, mother's depression, and mother's perception of
parenting daily hassles. Mothers who were depressed and perceived their parenting tasks
as big hassles were more likely to practice authoritarian parenting, and less likely to

practice authoritative parenting. Mothers who held more traditional Chinese parenting

96

beliefs were more likely to practice both authoritative and authoritarian parenting. In
terms of maternal characteristics other than parenting beliefs and depression, mother's age
was predictive of the Authoritarian total scale, but not the Authoritative total scale;
mother's education was predictive of the Authoritative total scale, but not the
Authoritarian scale. Younger mothers were more likely to practice authoritarian
parenting, but were not necessarily less likely to practice authoritative parenting than
older mothers. Less educated mothers were less likely to practice authoritative parenting,
but were not more likely to practice authoritarian parenting than higher educated mothers.
In terms of the relation between child characteristics and parenting measures,
authoritative and authoritarian parenting were predicted by different aspects of child
temperament. Child's sociability was predictive of the mother's authoritative parenting,
while child's emotionality was predictive of the mother's authoritarian parenting. Mothers
who perceived their children as more social were more likely to practice authoritative
parenting, while mothers who perceived their children as relatively emotional were more
likely to practice authoritarian parenting. Moreover, family income was predictive of the
Authoritative total scale, but not the Authoritarian total scale. Mothers with higher

family incomes were more likely to practice authoritative parenting.

Determinants of the Authoritative subscales

When looking into the component elements of the Authoritative scale, parenting
beliefs, child's sociability, and the degree of parenting daily hassles were predictive of all
four subscales of the Authoritative scale. As discussed above, mother's education was

predictive of the Authoritative total scale. When looking into the subscales, mother's

97

education was predictive of the Warmth/Involvement scale, but not the other
Authoritative subscales. It appeared that the inﬂuence of maternal education on
authoritative parenting was mainly on the warmth/involvement aspect. In general,
depressed mothers were less likely to practice authoritative parenting. However, it may
not apply to their use of democratic participation. According to the regression results, the
use of democratic participation was related to Chinese parenting beliefs, the child's
sociability, and the degree of perceived parenting daily hassles.

Child's age was predictive of the Warmth/Involvement subscale, but not the other
three Authoritative subscales. When children get older, mothers may be less warm and
involved, but other aspects of authoritative parenting may not be affected. In terms of the
contextual factors, mothers with higher family incomes were more likely to be warm and
involved with their children, and more likely to use democratic participation strategies.
Mothers with more children were less likely to be warm, involved with, and reason with

their children.

Determinants of the Authoritarian subscales

Regarding the determinants of the Authoritarian subscales, younger mothers are
more likely to use demanding, scolding, criticizing, or non-reasoning/punitive strategies;
however, they do not use more corporal punishment or show more hostility verbally
toward their children. Though mother's education, in general, did not affect her use of
authoritarian parenting, mothers with lower levels of education were more likely to use
non-reasoning/punitive discipline strategies toward their children.

The mother's depression level was positively related to her scores on most of the

98

authoritarian subscales, but not to her use of corporal punishment. Depressed mothers
were more likely to scold, criticize, yell at their children, or use non-reasoning/punitive
strategies. However, punishing children physically may be mainly a matter of parental
beliefs, and triggered by other factors, such as stress from parenting daily hassles or
children being temperamentally difﬁcult. Parenting beliefs were predictive of the use of
directiveness, corporal punishment, non-reasoning, punitive strategies, but not verbal
hostility, when other factors were controlled. One possible explanation is that Chinese
mothers may use scolding or criticizing in child training, but getting into arguments or
yelling and shouting at children was not supported by the Chinese training concepts.

Child's emotionality was predictive of the mother's use of corporal punishment,
non-reasoning, punitive strategies, verbal hostility, but not to her use of directiveness
when other factors were controlled. Among the child's characteristics, the mother's use of
directiveness was more likely to be predicted by the child's age. Mothers were more
likely to demand, scold, or criticize their children when their children were older. This
may due to the higher expectations that mothers place on their older children.

In sum, the analyses support the need of differentiating the determinants of
authoritative and authoritarian parenting. Several factors are shared as predictors of both
authoritative and authoritarian parenting. However, there are some factors which appear
to be speciﬁc for predicting certain parenting patterns. It is also strongly suggested that
the determinants of parenting would be better understood by looking into the component

elements of the global parenting typologies.

99
Conclusion

This study found that authoritative parenting is a predominant parenting style
among contemporary Chinese mothers in Taiwan. It is suggested that Chinese parents
who practice authoritarian parenting when necessary could be very warm and caring. In
this study, mothers who used verbal or physical punishment at moderate levels were
likely to demonstrate characteristics of "authoritative" parenting at the same time. The
use of authoritative and authoritarian parenting are both embedded in the Chinese training
concepts. Parents who hold traditional Chinese parenting beliefs would use both
authoritative and authoritarian parenting in different situations and at different times.
Both inductive and power-assertive disciplinary techniques would be used as a mean of
"training" their children. The use of authoritative parenting may play an important role in
children by facilitating compliance and the acceptance of parental control.

Belsky's model of the determinants of parenting was supported in this sample of
Chinese mothers in Taiwan. Maternal characteristics, child characteristics, and
contextual factors all explained some of the variance in Chinese parenting. In general,
factors related to Chinese parenting were similar to those found for western parents.
"Purely" authoritative mothers were the most educated, least likely to be depressed or to
perceive their parenting tasks as big hassles, and had the highest family income levels
among the four parenting groups identiﬁed in this study. It was found that different
factors predicted authoritative and authoritarian parenting. Different sets of factors were
also found to be predictive of different components of authoritative and authoritarian
parenting. Moreover, this study suggests that there are certain factors that predict

parenting behavior across cultures. However, "culturally speciﬁc" parenting beliefs also

100

play an important role in inﬂuencing parenting behavior.

In general, instead of raising questions about using Baumrind's typologies (i.e.
authoritative and authoritarian) to assess Chinese parenting, this study suggests that
Chinese parents' scores on the authoritative and the authoritarian measures should be
considered simultaneously. The degree of their use of certain parenting practices should
be taken into account, too. Having relatively higher score on the authoritarian measure
than Euro-American parents does not mean that Chinese parents use authoritarian
parenting extensively, or that they do not practice authoritative parenting at all. The
effect of authoritarian parenting on children's development is likely to be different if
parents also use authoritative methods to guide their children.
Dimsh’onsfcnﬁinuetesearsh

This study does not intend to generalize the results to parents in rural areas of
Taiwan, or Chinese parents outside of Taiwan. Future investigations should address
some of this study's limitations by including both urban and rural Chinese mothers with a
wider SES range, as these limitations of the sample could well have affected the ﬁndings
of the current investigation. This study could also contribute to ﬁrture studies which try
to explain the paradox involving Chinese parenting and the academic success of Chinese
children. Though Chinese parents have been found to score higher on "authoritarian"
measures than western parents, ﬁrture studies should asked the following questions. First,
how high is the score? Do most Chinese parents' practice authoritarian parenting
extensively, and are their scores in the "medium" range? Second, to what degree is the
Chinese parents' use of authoritarian parenting accompanied by medium to high levels of

authoritative parenting? There is a high percentage of parents who do not simply practice

101

"purely" authoritative or "purely" authoritarian" parenting. Thus, "authoritarian"
parenting by itself is not likely to explain the relation between parenting and chid
outcomes. Moreover, in comparisons of Chinese and American parenting, the component
elements of the authoritative and authoritarian parenting measures should be examined to
detect the speciﬁc differences in parenting between the two cultures. Are Chinese
parents' higher scores mainly due to the use of directive discipline techniques (e. g.
demanding, scolding, and criticizing), or other "punitive" discipline strategies? This study
suggested that Chinese parents' higher scores on the authoritarian measure may be largely
due to the use of "directiveness" techniques (e. g. demanding, scolding, and criticizing),
and less likely due to the use of punitive discipline strategies.

The present study should be viewed as a ﬁrst step toward greater understanding of
the parenting practices within the Chinese population. It should be noted that in this
study, there is no intention of making judgement about what is a "good" parenting style
for Chinese parents. Further studies should be done examining the relations between
different parenting styles and child outcomes. For future studies related to explaining
Chinese children's outcome and parenting, it is suggested that the joint inﬂuences of

Chinese parents' use of authoritative and authoritarian parenting should be considered.

LIST OF REFERENCES

LIST OF REFERENCES

Bates, J. (1980). The concept of difﬁcult temperament. W, 2_6,
299-319.

Baumrind, D. (1966). Effects of authoritative control on child behavior. Chili
Wis-ﬂ, 887'9O7*

Baumrind, D. (1967). Child care practices anteceding three patterns of preschool
behavior. MW, 75. 43-88.

Baumrind, D. (1978). Parental disciplinary patterns. Wight, 2, 223-276.

Baumrind, D. (1991 a). The inﬂuence of parenting style on adolescent competence and

substance use. Wee. 1.1, 56-95.

Baumrind, D. (1991b). Parenting styles and adolescent development. In R. Learner, A.

C. Petersen & J - Brooks-Gum (Eds.), Whip-
746-758). New York: Garland.

Baumrind, D. (1993). The average expectable environment is not good enough: A

response to Scarr. CW, ﬂ, 1299-1317.

Belsky J. (1984). The determinants of parenting: A process model. Wm,
55, 83-96.

Belsky J. (1990). Parental and nonparental child care and children's socioemotional

development: A decade in review. W, 52, 885-
903.

Belsky, J ., & Isabella, R. (1988). Maternal, infant, and social-contextual determinants of

attachment security. In J. Belsky & T. Nezworski (Eds), gljnjgg] implications Qf
ﬁlinghmem (pp. 41-94). Hillsdale, NJ: Lawrence Erlbaum Associates.

102

103

Block, J. H. (1965). . - 1° . v-.
W Berkeley, CA: Univer. Of Calrfomra, Institute

of Human Development.

Bond, M H. (Ed.). (1986). W. Hong Kong: Oxford

University Press.

 

Bossard, J. H., & Boll, E. (1956). Iholargofamilysystom. Philadelphia: University of

Pennsylvania Press.

Bronfenbrenner, U. (1979). W- Cambridge, MA:

Harvard University Press.

Buri, J. R. (1991). Parental authority questionnaire. loomaloﬂqualﬂAssossmom,

 

51, 110-119.
Buss, A., Plomin, R. (1975). a an. 1 ‘-1 1-11. 611
York: Wiley.
Buss, A4 Plomin, R. (1984). W. New

Jersey: Lawrence Erlbaum.

Chao, P. (1983). Chioosrﬂgjnsln'o. London: Kegan Paul International.

 

aohlcyomcm, Dissertation. University of California, Los Angeles.

Chao, R. K. (1994). Beyond parental control and authoritarian parenting style:
Understanding Chinese parenting through the cultural notion of training. Chili

mum,as,1111-1119.

Chen J L (1997) WWW
sociallymmnetcmhehaximinlaim Dissertation University of

Wisonsin-Madison.

Chen. C., & Uttal, D. H. (1988). Cultural values, parents' beliefs, and children's
achievement in the United States and China. Hommﬂoycloomom, 51, 351-358.

Chen, Y, Lu, C. ,Hung, Y., & Chen, L. (1980). A survey on the opinions about corporal
punishment by teachers, students and parents. Bulloﬁnoﬂﬁoooaﬂonol
Psychology, 15, 57- 74. (In Chinese)

Chiu, L. H. (1987). Child-rearing attitudes of Chinese Chinese-American, and Anglo-
American mothers Internationalloumalafﬂsxchologx 2.2. 409-419

104

Chung, F. C. (1994). 1- ' - 1. a ' .
WW Dissertation Then University

of Wisconsin-Madison.

 

Coll, C. T. G, Meyer, E. C, & Brillon, L. (1995). Ethnic and minority parenting. In
Bomstein (Eds. ), 1 1.11 1 1 1, .
(Vol. 2). Mahwah, NJ: Lawrence Erlbaum.

 

Crinc, K. A, & Greenberg, M. T. (1990). Minor parenting stresses with young children.
chilﬂlexelonment {11, 1628- 1637

Crockenberg, S., & McCluskey, K. (1986). Change in maternal behavior during the
baby’s ﬁrst year of life. Wm]; 51. 746-753.

Darling N. ,& Steinberg, L. (1993). Parenting style as context. An integrative model.
Wm. 11:, 487-496.

Dekovic', M., & J anssens, Jan M. A. M. (1992). Parents' child-rearing styles and child's
sociometric status. MonmootaLPsyohology, 23, 925-932.

DeLongis, A., Coyne, J. C., Dakof, G., Folkman, S., & Lazarus, R. S. (1982).
Relationship of daily hassles, uplifts, and major life events to health status. Hoalth
Psychology1l1 119-136-

Dombusch, S. M., Ritter, P. L., Leiderman, H. P., Roberts, D. F., & Fraleigh, M. J.
(1987). The relation of parenting style to adolescent school performance. Chili

W1 ﬂ. 1244-1257.

Elder, G. H, Jr., & Bowennan, C. E. (1963). Family structure and childrearing patterns.

The effect of family size and sex composition. W 18.,
891 -.905

Fox, R., Platz, D., Bentley, K. (1995). Maternal factors related to parenting practices,
developmental expectations, and perceptions of child behavior problems.

WM, 431-441-

Gecas, V. (1979). The 1nﬂuence of soc1al class on socialization. InW. R. Burr, R. Hill,

F 1 Nye1&ReiSS1I L (EdS) WWQD
365—404). New York: Free Press.

Greenbergen E (1988) WWW
socjohzahonoﬂchjldron. Unpublished manuscript. Irvine, CA: Univer. of
California, Program 1n Social Ecology.

105

Harman, K., & Luster, T. (1991). Inﬂuence of parent, child, and contextual factors on the
quality of the home environment. lnfahLMomalﬂoalmioumal, 12, 17-30.

Hart, C. H. ,,DeWolf M, & Burts, D. C. (1993). Parental disciplinary strategies and
preschoolers' play behaviors 1n playground settings. In C. H. Hart (Ed.,) £1an
1. 1. '11 (pp. 83- -.98) Albany, NY

 

State University of New York Press.

Heilbrun, A. B., & Waters, D. B. (1968). Underachievement as related to perceived
maternal child-rearing and academic conditions of reinforcement. Child

W1 321913-9211

Hein, C., & Lewko, J. H., (1994). Gender difference in factors related to parenting

style: A study of high performing science students. MaloﬁAoolosconL
W121 262-281 1

Hinde, R. A., & Dennis, A. (1986). Categorizing individuals: An alternative to linear
analysis mtemaﬁonalemaloﬂMaxismLﬂelelonment 2 105- 119

Ho, D. Y. F. (1981). Traditional patterns of socialization' 1n Chinese society. Aota
WW 2;, 81 -95

Ho, D. Y. F ., & Kang, T. K. (1984). Intergenerational comparisons of child-rearing

attitudes and practices in Hong Kong. DcyeloomomalPsyohology. 20, 1004-
1016.

Hsu1F1 L K- (1981). mmmmmmmmﬂmnm (3rd at) Honolulur

University Press of Hawaii.

Jones, F. A., Green, V., & Krauss, D. R. (1980). Maternal responsiveness of primiparous
mothers during the posme period: Age differences. mm o5. 56-63.

Kagitcibasi, C. (1989). Family and socialization' 1n cross-cultural perspective: A model

of change InJ J Herman (Ed) W
W. Lincoln. University of Nebraska Press.

Kamii, C. K. & Radin, N. L. (1967). Class differences' 1n the socialization practices of
Negro mothers W131 22 302- 310

Kelley, M. L., & Tseng, H. (1992). Cultural differences in child rearing: A comparison of
immigrant Chinese and Caucasian American mothers. WW

Rmhologz 23, 444-4551

Kim, U., & Chun, M. (1994). Educational "success" of Asian Americans: An indigenous
pmpectivemaoimmﬂeﬁmmmmmggx Li 329- 339

106

Kochanska, B., Kuczynski, L., & Radke-Yarrow, M. (1989). Correspondence between
mothers' self-reported and observed child-rearing practices. Chﬂiﬂcyelopmcot,
60, 56-63.

Kohn, M. L. (1963). Social class and parent-child relationships: An interpretation.
AmericmloumaloLSociolm 68, 471-480

Kohn, M. L. (1969). WWW. Homewood, IL: Dorsey.

Kojima, H. (1986). Japanese concepts of child development from the mid-17th to mid-
19th century. ‘ 1 1 - ,,2 315- 329.

 

Lang, 0. (1946). W. New Haven: Yale University Press.

Lau, S., & Cheung, P. C. (1987). Relations between Chinese adolescents' perception of
parental control and organization and their perception of parental warmth.

W1 211 726-7291

Lazarus, R. S., DeLongis, A. ,,Folkman S. & Gruen, R. (1985). Stress and adaptational

outcomes: The problem of confounded measures. Amoﬁoanﬂychologist, ﬂ),
770-779.

Lin, C. C., & Fu, V. R. (1990). A comparison of child-rearing practices among Chinese,
immigrant Chinese, and Caucasian-American parents. ChildDeyoloomonL o1,
429-433.

Lin, H. J. (1995). __ .1 ..
W Dissertation Arizona State

University.

 

Luster, T., & Dubow, E. (1990). Predictors of the quality of the home environment that
adolescent mothers provide for their school-aged children. W1

Adolescent. 19. 475-4941

Luster, T., & Okagaki, L. (1993). Multiple inﬂuences on parenting: Ecological and life-
course perspectives. In T. Luster, & L. Okagaki (Ed.,) Parontmgom

omlogioaLoeLspgcﬂyo. New Jersey: Lawrence Erlbaum.

Luster T, Rhoades K., & Haas B. (1989). The relation between parental values and

parenting behavior: A test of the Kohn hypothesis. MW
Earm'ly, 5_1_, 139-147.

Maccoby, E. E., & Martin, J. A. (1983). Socialization 1n the context of the family:

Parent-child interaction. In P. H. Mussen (Ed. ), Hondhoohoﬂchiloosychology
(4th ed., Vol. 4, pp. 1-101). New York: John Wiley& Sons.

107

Menaghan, E. G., & Parcel, T. L. (1991). Determining children's home environments:
The impact of maternal characteristics and current occupational and family

conditions. WWW LL 1-241

Nye, F. I., Carlson, J., &Garrett, G. (1970). Family size, interaction, affect, and stress.
W}; 216-26

Radloff, L. S. (1977). The CES- D scale. A self-report depression scale for research 1n the
general population W11. 385-401

Ragozin, A. S., Basham, R. B., Cmic, K. A., Greenberg, M. T., & Robinson, N. M.

(1982). Effects of maternal age on parenting role. QeyoloomontalPsyohology,
18, 627-634.

Reis, J., Barbera-Stein, L. & Bennett, S. (1986). Ecological determinants of parentmg
Eamﬂxlielations 3.2 547- 5541

Robinson, C. C., Mandleco, B. L., Olsen, S. F., &Hart,C. H. (1995). Authoritative,
authoritarian, and permissive parenting practices: Development of a new measure.

EmhologieaLRenons 7.2 819- 830

Robinson, C. C., Hart, C. H., Mandleco, B. L., &Olsen, S. F. (1996). Psychometric
support for a new measure of authoritative, authoritarian, and permissive

parenting practices: Cross-cultural connections. W

 

oon'foronoe. Quebec City, Canada

Rohner, R. P., & Pettengill, S. M (1985). Perceived parental acceptance-rejection and
parental control among Korean adolescents. thliDcyeloomom, 5o, 524-528.

Rollins, B. C., & Thomas, D. L. (1979). "Parental support, power, and control techniques
in the socialization of chlldren " InW. R. Burre, R. Hill, F. I. Nye, &I. L. Reiss

 

317- 364). New York: The Free Press. '

Rowe, D., & Plomin, R. (1977). Temperament in early childhood. LoumaloﬁPersonolng
Assessment 4.11 150-1561

Ryan, A. S. (1985). Cultural factors in casework with Chinese-Americans. Social work:

W166, 333-3401

Saavedra,J. M. (1977).

111 1110111 01.111.1'1111111‘1w11 "-111

Dissertation. Catholic University
of America.

 

108

Saavedra, J. M. (1980). Effects of perceived parental warmth and control on the self-

evaluation of Puerto Rican adolescent males. WM, 15, 41-
54.

Scheck, D. C., & Emerick, R. (1976). The young male adolescent's perception of early
childrearing behavior: The differential effects of socioeconomic status and family

size. 3212121119111. 22, 39-52.

Schwarz, J. C., Barton-Henry, M. L., & Pruzinsky, T. (1985). Assessing child-rearing
behaviors: A comparison of ratings made by mother, father, child, and sibling on

the CRPBI. W126, 462-4791

Simons, R., Whitbeck, L. ,,Conger R., & Melby, J. (1990). Husband and wife differences
in determinants of parenting: A social learning and exchange model of parental

behavior W31 52, 375- 392

Smetana, J. G. (1994). Parenting styles and beliefs about parental authority. In J. G.
Smetana (Ed.), - _ . . .. 1
(pp. 21 -.36) San Francisco, CA: Jossey-Bass.

 

Sollenberger, R. T. (1968). Chinese-American child-rearing practices and juvenile

dellnquenoy1_loumal_of_S_anLRmholo2x1141l3-231

Sroufe, L. A., Cooper, R. G., DeHart, G. B., Marshall, M. E. (1992). W;
W- New York, NY: McGraw-Hill.

Steinberg, L., Elmen, D. J ., & Mounts, N. S. (1989). Authoritative parenting,
psychosocial maturity, and academic success among adolescents. Chili

W1 691 1424-14361

Steinberg, L., Lambom, S. D., Darling, N, Mounts, N. S., & Dombusch, S. M. (1994).
Over-time changes in adjustment and competence among adolescents from
authoritative, authoritarian, indulgent, and neglectful families. Chili

W1 Q21 754-7701

Steinberg, L., Lambom, S. D., Dombusch, S. M., & Darling, N. (1992). Impact of
parenting practices on adolescent achievement: authoritative parenting, school
involvement, and encouragement to succeed. WW1 o3, 1266-1281.

Steinberg, L., Mounts, N. S., Lambom, S. D., & Dombusch, S. M. (1991). Authoritative
parenting and adolescent adjustment across varied ecological niches. W

W111 19-361

109

Stevenson, H. W. ,,Chen C. & Lee, S. (1992). Chinese families. In Roopnarine, J. L., &

Carter1D B (Eds11) Barentzehﬂisoeializationindimmmnes. Norwood, NJ:
Ablex Publishing Company.

Stevenson, H. W., & Lee, S. Y. (1990). Contexts of achlevement Monographsoﬂho
WW1 25(1-21 Senal No 221)

Suzuki, B. H. (1980). The Asian-American family. In: M. D. Fantini andR. Cardenas
(Eds11) W013 74-102) New York! Longtnan

Thomas, A., Chess, 8., & Birch, H. G. (1968). WWII.
Children New York: University Press.

Thomas, A., Chess, 8., Birch, H. G., Hertzig, M. E., &Kom, S. (1963). BehayioroL
W New York: University Press

Thomas, D. L., Gecas, V., Weigert, A., & Rooney, E. (1974). Emily sggjah'zan'gn and
111mm. Lexington, MA: Health.

Thomas, R- M (1992) ﬂomnanngthmriemtehilideyelonmem (3nd ed) Belmont,
CA: Wadsworth Publishing Company.

Tobin, J. J ., Wu, D. Y. H., & Davidson, D. H. (1989). ProschooLlnjhxocﬂnmm New

HavenzYale University Press.

Weiss, L. H., & Schwarz, J. C. (1996). The relationship between parenting types and
older adolescents' personality, academic achievement adjustment, and substance

use- WW1QZ12101-21141

Wolf, M. (1970). Child training and the Chinese family. In M. Freeman (Ed.,) Pamjly
andkinsln'ojnﬂm'nosojocjom (pp. 37-62). Stanford, CA: Stanford University

Press.

Wu, Y. J. (1997). - - - -
mlajm. Dissertation. The University of Tennessee.

 

Xu, Z. Wan, C. W., Mussen, P. Shen, J, Li, C. & Cao, Z. (1991). Family socialization
and children behavior and personality development' 1n China. _Iho1oumal_of_

W1 12, 239- 2531

APPENDICES

Appendix A
Measures Used In The Study

E'E':"EEEEI

Directions:
The following pages contain a list of behaviors that parents may exhibit when
interacting with their children. The questions are designed to measure how often you

exhibit certain behaviors towards your child.

I l .1 . l . l l . .
l = Never

2 = Once in a while

3 = About half of the time
4 = Very often

5 = Always

1. I encourage my child to talk about the child's troubles.
. I guide my child by punishment more than by reason.
. I know the names of my child's friends.
. I ﬁnd it difﬁcult to discipline my child.

2
3
4
5. I give praise when my child is good.
6. I spank when my child is disobedient.

7. I show sympathy when my child is hurt or frustrated.
8. I punish by taking privileges away from our child with little if any explanations.
9. I give comfort and understanding when my child is upset.
_ 10. I yell or shout when my child misbehaves.

_ 1 l. I am easy going and relaxed with my child.

_ 12. I tell child my expectations regarding behavior before the child engages in an

110

1 l 1
activity.
_ 13. I scold and criticize to make my child improve.
_ 14. I show patience with child.
_ 15. I grab my child when being disobedient.
16. I am responsive to my child's feelings or needs.

17. I allow my child to give input into family rules.

18. I argue with my child.

19. I appear conﬁdent about parenting abilities.

20. I give my child reasons why rules should be obeyed.

21. I appear to be more concerned with own feelings than with my child's feelings.

22. I tell my child that we appreciate what the child tries or accomplishes.

23. I punish by putting my child off somewhere alone with little if any explanations.

_ 24. I help my child to understand the impact of behavior by encouraging our child to
talk about the consequences of own actions.

25. I am afraid that disciplining my child for misbehavior will cause the child to not
like me.

26. I take my child's desires into account before asking the child to do something.

27. I explode in anger towards my child.

28. I am aware of problems or concerns about my child in school.

29. I threaten my child with punishment more oﬁen than actually giving it.

30. I express affection by hugging, kissing, and holding my child.

31. I use physical punishment as a way of disciplining my child.

32. I tell my child what to do.

33. I talk it over and reason with my child when the child misbehaves.

34. I slap my child when the child misbehaves.

35. I disagree with my child.

36. When two children are ﬁghting, I discipline my child ﬁrst and ask questions later.

37. I encourage my child to freely express (himself)(herself) even when disagreeing
with parents.

38. I scold or criticize when my child's behavior doesn't meet my expectations.

112

_ 39. I show respect for my child's opinions by encouraging my child to express them.

40. I explain to our child how we feel about the child's good and bad behavior.

41. I use threats as punishment with little or no justiﬁcation.

42. I take into account my child's preferences in making plans for the family.

43. When my child asks why (he)(she) has to conform, I state: because I said so, or I
am your parent and I want you to.

44. I explain the consequences of the child's misbehavior.

45. I demand that my child does things.

46. I channel my child's misbehavior into a more acceptable activity.

__ 47. I emphasize the reasons for rules.

Authoritative pattern: subfactor l (warmth/involvement): items 1, 5, 7, 9, 16, 22, 30.
subfactor 2 (reasoning/induction): items 12, 20, 24, 33, 40, 44, 47.
subfactor 3 (democratic participation): items 17, 26, 37, 42.
subfactor 4 (good natured/easy going): items ll, 14, 39, 46.

Authoritarian pattern: subfactor l (directiveness): items 13, 32, 38, 45.

subfactor 2 (corporal punishment): items 6, 15, 31, 34

subfactor 3 (non-reasoning/punitive strategies): items 8, 21, 23, 36,

41,43.

subfactor 4 (verbal hostility): items 10, 18, 27, 35.

113

E . 12.1“ 1 IEEIII

How hassled you felt by the following events? Please rate each of the items on a scale of
1 (no hassle) to 5 (big hassle).
.nohmle 11131135111;
1 2 3 4 5

_ 1. Continually cleaning up kid's messes

_ 2. Kid's schedules interfere with parent/household needs
_ 3. Sibling arguments require referee.

_ 4. Kids are constantly under foot or in the way

_ 5. Having to change plans to meet kids' needs

_ 6. Kids get dirty and need to have clothes changed

_ 7. Difﬁculties getting kids ready for outings on time
_ 8. Having to run extra errands just for kids

_ 9. Being nagged, whined at, or complained to

10. Kids don't listen, won't do what they are asked without being nagged
l 1. Kids demand to be entertained or played with

12. Kids resist or struggle over bedtimes

13. Need to keep a constant eye on what kids are doing
14. Kids interrupt adult conversation or interaction

15. Kids are difﬁcult to mange in public places

114

N ' ' N i

Parents have diverse views about appropriate child-rearing practices. Please rate each of

the items on the scale of 1 (strongly disagree) to 5 (strongly agree).

strongly strongly
disagree agree
1 2 3 4 5

_ 1. Parents must begin training child as soon as ready.

_ 2. Mothers must train child to work very hard and be disciplined.

__ 3. Mothers teach child by pointing out good behavior in others.

_ 4. When child continues to disobey you, he/she deserves a spanking.

_ 5. Mothers primarily express love by helping child succeed, esp, in school.
_ 6. Child should be in the constant care of their mothers or family.

_ 7. Child should be allowed to sleep in mother's bed.

115

5551 S E :1.“ 'E 113'

Rate each of the items for your child on a scale of 1 (not characteristic or typical of your

child) to 5 (very characteristic or typical of your child).

1 .. l ..
'11: 1'11 '11: 1'11
1 2' 3 4 5

1. Child tends to be shy. (Shyness)

2. Child cries easily. (Emotionality)

3. Child likes to be with people. (Sociability)

4. Child is always on the go. (Activity)

5. Child prefers playing with others rather than alone. (Sociability)

6. Child tends to be somewhat emotional. (Emotionality)

7. When child moves about, he usually moves slowly. (reversed, Activity)

8. Child makes friends easily. (reversed, Shyness)

9. Child is off and running as soon as he wakes up in the morning. (Activity)

10. Child ﬁnds people more stimulating than anything else. (Sociability)

_ l 1. Child often fusses and cries. (Emotionality)

_ 12. Child is very sociable. (reversed, Shyness)

~

——

13. Child is very energetic. (Activity)

14. Child takes a long time to warm up to strangers. (Shyness)

15. Child gets upset easily. (Emotionality)

16. Child is something of a loner. (reversed, Sociability)

17. Child prefers quiet, inactive games to more active ones. (reversed, Activity)
18. When alone, child feels isolated. (Sociability)

19. Child reacts intensely when upset. (Emotionality)

__ 20. Child is very friendly with strangers. (reversed, Shyness)

116

CW:

Instructions for questions: Below is a list of the ways you might have felt or behaved.

Please rate how often you have felt this way during the past week.

During
__ l.
_2.
_3.
_4
_5
_6.
_7
_8
9
10
ll

12.
13.
14.
15.
l6.
l7.
18.
19.

20

0 = Rarely or none of the time (Less than 1 day)

l = Some or a little of the time (1-2 days)

2 = Occasionally or a moderate amount of time (3-4 days)

3 = Most or all of the time (5-7 days)

the past week:

I was bothered by things that usually don't bother me.

I did not feel like eating; my appetite was poor.

I felt that I could not shake off the blues even with help from my family or
ﬁ-iends.

. I felt that I was just as good as other people. (reversed)

. I had trouble keeping my mind on what I was doing.

I felt depressed.

. I felt that everything I did was an effort.
. I felt hopeful about the future. (reversed)
. I thought my life had been a failure.

. I felt fearful.

. My sleep was restless.

I was happy. (reversed)

I talked less than usual.

I felt lonely.

People were unfriendly.

I enjoyed life. (reversed)

I had crying spells.

I felt sad.

I felt that people dislike me.

. I could not get "going."

117

E 1"ﬁ .

Your age:

Marital Status:

_Married Single Divorced

T he family's total monthly income:

_ 30,000 NT and below __ above 13, 000 NT
_ 30,000 NT to 50,000 NT

_ 50,000 NT to 70,000 NT

__ 70,000 NT to 90,000 NT

_ 90,000 NT to 1 1,000 NT

_11,000 NT to 13,000 NT

* NT: New Taiwan Dollars

* They are coded as, from 1=30,000NT and below, to 7=above 13,000NT for analyses

 

 

 

Education:

You Your husband/partner
illiterate
primary school
high school
college graduate

 

graduate school and above

 

* Number of years is not used as a measure of education level in Taiwan
* They are coded as followed for analyses:
illiterate: 1; primary school: 2; high school: 3; college graduate: 4; graduate school and

abov:5

118

Occupation:
You

Your husband/partner

 

119

Qonsenﬂetter

Dear Parent:

I graduated from the Department of Sociology at National Taiwan University, and
now I am a doctoral student at Michigan State University. I am working on my
dissertation designed to study mothers' parenting behaviors in Taiwan. The purpose of
this research is understand the different approaches that mothers use while parenting
preschool children. Attached is a questionnaire which you are being asked to complete, if
you choose to participate in this research project. The questionnaire asks about your
parenting practices, about your beliefs about how children should be raised, and about
your child's characteristics. There are also questions about you and your family.

Anonymity of the information is guaranteed. There is no personal risk involved
with this research. It is hoped that you will take time to complete these materials. It
should take no more than half an hour of your time. You indicate your voluntary
agreement to participate by completing and returning this questionnaire. In the event that
you choose not to participate, it would be appreciated if you would still return the
materials. Thank you in advance for your time and cooperation. Without the help of
people like yourself, research on parenting would not be possible.

If you have any questions that may arise in connection with your participation in

this research study, please feel free to contact me.
Sincerely,
Fu-mei Chen

Ph.D. student at Michigan State University
Phone # in Taipei: 633-4337

120

Appendix B
Comparison of mean scores of the Authoritative scales and Authoritarian scales on

Chinese sample in this study and American sample in Robinson et al.'s study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chinese sample American sample

(N=463) (N=245)

Mean* SD Mean SD
Authoritative total score 4.0 .5 4.0 .4
Warmth/Involvement 4. 1 .6 4.3 .5
Reasoning/Induction 3.9 .6 3.9 .6
Democratic participation 3.6 .7 3.4 .6
Good natured/Easy going 3.9 .6 4.0 .4
Authoritarian total score 2.0 .4 2.1 .4
Directiveness 2.2 .6 2.3 .7
Corporal punishment 2.0 .6 2.0 .6
Non-reasoning/Punitive strategies 1.7 .5 1.8 .6
Verbal hostility 2.1 .5 2.3 .5

 

 

 

 

* The mean scores are the average item scores

121

Appendix C: Copy of the UCRIHS Approval Letter

 

RESEARCH
MID
GRADUATE
STUDIES

mes-mecca
mm

Mensa '
(me)

mums
zen-mam

NMW
area-1015

5171355121»
FAX' 517%"?!

mans-seem
“BMW
was“

MICHIGANSTATE
UNIVERSITY

DOC”! O. 1’97

70: Ton Luster
101 Morrill Ball

ms: :14)

um: mourn-m am normals-um sum-rm IN
cum ms 11m mscnoon cantata

"WV”. “”3“” 1’6

APPROVAL om: 12/08/97

The nniversit Comittee on lasearch Involving ltunan Sub ects' (W)
review or chi: reject is complete. I am pleased to advise that the
rights and wel lretg {the human subjects appear to be adequately
rotocted and aethods to or-ed consent are appropriate.
&retore. the scams approved this project and any revisions listed

WOULD IMWMASMICS‘IAWR mm
gmmnagntsgrnmmmm romnmnmsunm

QUESTIONNAIRE.
am scams roval val for one cayearlendar beginning
' the i “till!” he mange. Investigators “.2le with
contnueaproje e arms usethegreenrcnewal
torn (enclosed 11111111911133 roval letter or when a
are zect is renewed) to seal; goat c rtitication. There is a
. mum or tour such t eneuam ssih ihle. Investigators
“hills to continue a projectbeyond tine need to suhnit it
or couplete review.
amsroas: Uterus lust review any Ugandan! Ving human
subj acts. prior to initiation 0: change. If this is done at
the3 tine o m, please use use renewal form. To

revise an approved protocol at anti; time during the year

send your written request to the ms Chair. requesting revised

approval and referencing the project 5 no tend title. Include

in your request a description o! the change and any revised

instruments. consent terms or advertisements that are applicable.
raosnus/ ,
m: should either or the tollow' arise during the course ofthe
work. investigators Inlet 11°33}; UCRIBS promptly: _(11 ans
unexpected side effects coup aints. etc.) invol Ving guan
subgectgig (2) 3:119“ long: gangsta ‘greiv‘iiromenﬁb or new
mf n. mdi ins gre man s cots than
existed when the protocol was previously reviewed and approved.

If we can be or any tuture hel . lease do not hesitat
at (517)355-2100 or tax (517)4g2-gl71. . to mu“ “’

8 mt“? e

 
  

no 11.111119111112113.

Dashed

cc: NR1 Chen

"11111111111111