Personality and Individual Differences 47 (2009) 42–46

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

The relationship between attachment style and self-concept clarity: The

mediation effect of self-esteem
Chia-huei Wu
Institute of Work Psychology, University of Sheffield, Mushroom Lane, Sheffield, S10 2TN, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: The mediation effect of self-esteem on the relationship between attachment style and self-concept clarity
Received 3 September 2008 was investigated. In Study 1, the relationship between attachment style and self-concept clarity was
Received in revised form 24 December 2008 examined by correlation and regression analysis. Results from 189 and 85 students at National Taiwan
Accepted 28 January 2009
University (NTU) showed that anxiety and avoidance attachment tendencies were negatively related with
Available online 5 March 2009
self-certainty and self-concept clarity. In Study 2, self-esteem was included to determine whether the
negative relation between attachment tendency and self-concept clarity was mediated by self-esteem.
Keywords:
Participants were 123 NTU students. The mediation effect of self-esteem on the relationship between
Adult attachment
Self-esteem
attachment tendency and self-concept clarity was supported in a path model, indicating people with
Self-concept clarity secure attachment have higher self-esteem, which results in greater clarity of self-concept.
Path analysis Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction
Attachment theory, proposed by Bowlby (1969), has become a
framework for understanding an individual’s model of self. People
with different attachment styles have different self-views. A well-
documented aspect is self-esteem. Previous studies have indicated
that an insecure attachment style is associated with lower self-es-
teem (e.g., Bartholomew & Horowitz, 1991; Brennan & Bosson,
1998; Brennan & Morris, 1997; Collins & Read, 1990; Griffin & Bar-
tholomew, 1994; Park, Crocker, & Mickelson, 2004; Roberts, Gotlib,
& Kassel, 1996), supporting the argument that an insecure person
tends to feel worthless or of little value.
In addition to self-esteem, attachment style also plays an
important role in the knowledge component of self-concept. Sev-
eral studies have indicated that people with different attachment
styles also differ in self-perception accuracy (e.g., Berger, 2001;
Dozier & Lee, 1995; Kobak & Sceery, 1988). These studies investi-
gated the relationship between attachment and self-perception
accuracy by examining discrepancies between self- and other-re-
port of participants’ characteristics for different attachment
groups. Kobak and Sceery (1988) found that self- and peer-reports
of social competence and distress lacked congruence among avoid-
ance (dismissing) persons, which suggests that avoidance attach-
ment may be related to distorted self-perception characterized
by less acknowledgment of or even denial of distress (Berger,
2001). In addition, Dozier and Lee (1995) found that persons with
either avoidant or ambivalent attachment distort their self-con-
E-mail address: pcp08cw@sheffield.ac.uk
cept. However, the type of distortion is not the same. Avoidant per-
sons tend to deny their feelings of distress, while ambivalent
persons tend to amplify their feelings of distress. Similarly, Berger
(2001) indicated that a person with insecure attachment is more
likely to have a greater discrepancy between self- and other-re-
ports of psychosocial symptoms than a securely attached person
suggesting that insecure attachment corresponds to a distortion
in self-perception. Additional evidence comes from a study which
investigated the relationships between attachment style and self-
concept in cyberspace. Wu and Lin (2005) found that persons with
secure attachment had a higher score on certainty of cyber self-
concept than both avoidant and ambivalent attachment persons,
implying that persons with different attachment styles tend to ex-
hibit characteristics of self-concept clarity in cyberspace similar to
in real life. Although the studies demonstrated differences in self-
concept clarity among attachment styles, they did not investigate
why different types of attachment are related to different levels
of self-concept clarity. The main purpose of this study is to provide
a theoretical explanation of this phenomenon by examining the
role of self-esteem in attachment theory.
According to attachment theory, secure individuals have higher
self-esteem than insecure individuals because of their earlier social
interaction experiences. For instance, in secure attachment, care-
givers always give feedback in an effective way at the proper time,
which allows children to develop feelings of trust and dependence
on their caregivers. In addition, they develop a positive self-con-
cept through the stable and predictable feedback from their care-
givers, and consider themselves to be lovable, resulting in higher
self-esteem. In contrast, children with insecure attachment receive
little or at best unstable feedback from their caregivers. With this

0191-8869/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.paid.2009.01.043
 C.-h. Wu / Personality and Individual Differences 47 (2009) 42–46
unsupportive or unstable situation, these children develop a model
of other people as untrustworthy and unpredictable and think of
themselves as unlovable, resulting in lower self-esteem. In other
words, attachment theory posits a process which is consistent with
the sociometer hypothesis (Leary, Tambor, Terdal, & Downs, 1995)
that being liked by others led to more positive self-evaluations. The
sociometer hypothesis has been supported in various studies (e.g.,
Leary, Haupt, Strausser, & Chokel, 1998; Leary et al., 1995; Nezlek,
Kowalski, Leary, Blevins, & Holgate, 1997; Srivastava & Beer, 2005)
and Srivastava and Beer (2005) especially indicated that anxiety
attachment potentiates the sociometer process by showing that
the effect of others’ liking on self-evaluations was stronger for
those individuals who were more anxiously attached.
The difference in self-esteem results in a different basis for
forming self-concepts. The existing literature indicates that the le-
vel of self-esteem has a positive relation with self-concept clarity.
Baumgardner (1990) and Campbell (1990) reported that people
with low self-esteem seem to know less about themselves than
people with high self-esteem. Campbell and Lavallee (1993) indi-
cated that low-esteem people exhibit less extremity in self-
descriptions, lower confidence on self-descriptive tasks, less stabil-
ity of self-schema, and less internal consistency of self-belief than
high self-esteem people. In terms of behavior, since low self-es-
teem people tend to be less sure of themselves, they are more sus-
ceptible to external self-relevant cues, and more behaviorally
plastic than high self-esteem people (Brockner, Wiesenfeld, & Ras-
kas, 1993; Campbell & Lavallee, 1993). Although the causal direc-
tion between self-esteem and self-concept clarity is unclear (e.g.,
Campbell, 1990; Campbell & Lavallee, 1993; Setterlund & Nieden-
thal, 1993), a longitudinal study by Nezlek and Plesko (2001)
showed that daily events lead to changes in self-esteem, which
in turn lead to changes in self-concept clarity, suggesting that
the level of self-concept clarity is influenced by the level of self-es-
teem. In addition, past studies have usually regarded self-esteem
as a personality trait and the level of self-concept clarity as a char-
acteristic of high versus low self-esteem people. For example,
Brockner (1984) reviewed the reactions-to-feedback literature
and advanced the concept of plasticity for low self-esteem people.
He indicated that low self-esteem people are more susceptible to
the effects of self-relevant social cues than are high self-esteem
people. Specifically, high self-esteem people tend to accept and
be affected by external cues that are consistent with their self-con-
cept, but low self-esteem people tend to be susceptible to a broad
range of self-relevant stimuli, with the result that high and low
self-esteem people differ in self-concept clarity (Campbell, 1990).
However, if we take a perspective from attachment theory, the
role of self-esteem can be regarded as a characteristic of security
that results from attachment experiences. As stated previously, a se-
cure attachment relationship leads to a positive self-view because of
the stable and reliable external social environment. Accordingly, the
difference between high and low self-esteem in self-concept clarity
is not due to the trait of high versus low self-esteem, but due to the
underlying psychological sense of security inherent in an individ-
ual’s attachment style. For example, secure attachment persons
usually get feedback in an effective way and at the proper time from
their caregivers. This stable and reliable environment leads them to
consider themselves lovable and allows them to form a positive self-
view. With this positive self-regard, they have more confidence in
interacting with others, creating opportunities to gain others’ feed-
back, and can build a clear self-concept out of the feedback received
in their positive stable environment. In contrast, insecure attach-
ment people did not have those features in their social environment
and thus lack the resulting self-characteristics. From this perspec-
tive, the relationship between self-esteem and self-concept clarity
can be viewed as a function of the sense of security rather than as
individual differences in self-esteem trait and self-esteem can be
43
viewed as a basis of self-concept clarity. Therefore, according to
attachment theory and studies on self-esteem and self-concept clar-
ity, it is expected that secure attachment persons would have higher
self-concept clarity due to their higher self-esteem than insecure
attachment persons.
Thus, this study aims to examine whether the mediation effect
of self-esteem on the relationship between attachment style and
self-concept clarity can be supported. Two studies were conducted.
Study 1 examined the relationship between attachment style and
self-concept clarity. In Study 2, self-esteem was included to exam-
ine whether the effect of attachment style on self-concept clarity
would be mediated by self-esteem.
2. Study 1
This study examined the relationship between attachment style
and self-concept clarity. Two measurements were used to opera-
tionally define self-concept clarity (in accordance with Baumgard-
ner (1990) and Campbell’s (1990) definition). The first measure
was a self-certainty rating scale (Baumgardner, 1990; Campbell,
1990). Confidence in personality trait ratings indicates the level
of self-certainty. The second measure was the self-concept clarity
scale developed by Campbell et al. (1996). This scale measures per-
ceived internal consistency and temporal stability of self-beliefs,
along with more generic self-certainty items. The self-concept clar-
ity scale is broader than the self-certainty rating scale in capturing
the meaning of self-concept clarity. A two-dimension (anxiety and
avoidance) attachment scale was used and correlation and regres-
sion analysis was conducted to see if there are significant negative
relations between the two attachment styles and self-certainty/
self-concept clarity.
2.1. Method
2.1.1. Participants
(1a) Participants in the first sample were 82 male and 107 fe-
male undergraduate students at National Taiwan University
(NTU). Their ages ranged from 18 to 28 years (M = 20.76,
SD = 1.34). They spent about 30 min filling out the adult attach-
ment scale and self-certainty rating scale in exchange for extra
credit in their introductory psychology course.
(1b) Participants in the second sample were 36 male and 49 fe-
male (one participant did not report his/her gender) undergraduate
students at NTU. Age data were not included in this sample. They
spent an average of 10 min completing the adult attachment scale
and the self-concept clarity scale in exchange for extra credit in
their social psychology course. Before participating in the study,
participants did not take a course on adult attachment.
2.1.2. Instruments
2.1.2.1. Adult attachment scale Taiwan version (AAS-TW). The adult
attachment scale Taiwan version (AAS-TW) was revised from the
adult attachment scale (AAS, Collins & Read, 1990) by Wu (2005)
to measure participants’ attachment style. The AAS-TW has 10
items. Participants scored each item using a five-point Likert scale
ranging from not at all to very. There are two factors underlying
the AAS-TW. The first factor is anxiety (four items), which measures
the extent to which an individual is anxious or fearful about aban-
donment or being unloved. The second factor is avoidance (six
items), which measures the extent to which an individual is uncom-
fortable with closeness and dependence on others emotionally
close to them. In four different samples, the internal consistency
(coefficient a) was ranged from 0.71 to 0.80 for the anxiety subscale
and from 0.72 to 0.81 for the avoidance subscale (Wu, 2005). In
Wu’s (2005) study, results of exploratory and confirmatory factor
44 C.-h. Wu / Personality and Individual Differences 47 (2009) 42–46
analysis on different samples revealed that the two-factor structure
was reliable and acceptable. In the current study, the internal con-
sistency of the anxiety and avoidance subscales was 0.72 and 0.68
for the first sample, and 0.72 and 0.80 for the second sample.
2.1.2.2. Self-certainty rating scale. This scale was constructed by ref-
erencing Baumgardner (1990) and Campbell’s (1990) method for
measuring self-certainty. In this scale, 16 adjective traits are pre-
sented and participants indicate how well the traits describe them
on a seven-point Likert scale anchored by strongly disagree and
strongly agree for each trait. Then, participants rate how confident
they feel about each of their ratings on seven-point Likert scale an-
chored by not at all confident and very confident. Self-certainty is
calculated by averaging the confidence ratings of the 16 traits
(Cronbach’s alpha was 0.91 for the 16 confidence ratings for the
first sample). Higher average scores on confidence ratings repre-
sent higher self-certainty.
2.1.2.3. Self-concept clarity scale (SCC). This scale was developed by
Campbell et al. (1996). There are 12 items in this scale, each an-
chored by not at all and very much on a seven-point Likert scale.
After recoding reversed items, the mean score was calculated by
averaging the ratings of the 12 items. A higher score represents
greater self-clarity. This scale has satisfactory psychometric prop-
erties (see, Campbell et al., 1996) with high average internal con-
sistency reliability among three samples (0.86), and test–retest
reliability (0.79 for 4 months and 0.70 for 5 months). The SCC is
unidimensional and is positively correlated with self-esteem, tem-
poral stability, and internal consistency of individuals’ self-descrip-
tions, and negatively correlated with negative affect, neuroticism,
anxiety, depression, and the trait of repression–sensitization. In
the current study, Cronbach’s alpha of the SCC scale was 0.88 for
the second sample.
2.2. Results
2.2.1. Demographic analysis
Regarding the background variable for the first sample, gender
and age had no significant main effect and interaction effect on
anxiety, avoidance attachment, and self-certainty. For the second
sample, gender also did not have a significant effect on anxiety,
avoidance attachment, and self-concept clarity.
2.2.2. Correlation and regression analysis
For the first sample, anxiety attachment had a positive correla-
tion with avoidance attachment (r = 0.32, p < 0.01). In addition,
both anxiety and avoidance attachment had a negative correlation
with self-certainty (r = 0.19, p < 0.01; r = 0.16, p < 0.01), sup-
porting the hypothesis that insecure attachment tendency would
have a negative relation with self-certainty. However, when self-
certainty was regressed on anxiety and avoidance attachment, only
anxiety attachment had a significant effect (b = 0.15, b = 0.16,
t(186) = 2.10, p < 0.05).
For the second sample, anxiety attachment had a positive corre-
lation with avoidance attachment (r = 0.24, p < 0.05), in addition,
both anxiety and avoidance attachment had a negative correlation
with self-concept clarity (r = 0.38, p < 0.01; r = 0.21, p < 0.05).
When self-concept clarity was regressed on anxiety and avoidance
attachment, only anxiety attachment had a significant effect
(b = 0.46, b = 0.35, t(83) = 3.42, p < 0.01).
3. Study 2
In this study, the mediation role of self-esteem between attach-
ment style and self-concept clarity was examined. Participants
were asked to complete questionnaires for anxiety and avoidance
attachment style, self-esteem, and the self-concept clarity scale.
Only the self-concept clarity scale was used here because it is
shorter than the self-certainty rating scale and Study 1 showed
that these two measurements yield similar results, and the effect
was stronger for the self-concept clarity scale. Correlation and
regression analysis was conducted first. Next, a path model was
examined to check whether the influence of attachment on self-
concept clarity was mediated by self-esteem.
3.1. Method
3.1.1. Participants
Participants in this study were 34 male and 89 female under-
graduate students at NTU. They were on average 20 years old
(M = 20.15, SD = 1.70), with a range from 17 to 27 years. They par-
ticipated for an average of 30 min in exchange for extra credit in
their introductory psychology course.
3.1.2. Instruments
3.1.2.1. Adult attachment scale Taiwan version (AAS-TW). In the cur-
rent sample, the internal consistency of the anxiety and avoidance
subscales was 0.78 for each scale.
3.1.2.2. Self-concept clarity scale (SCC). In the current sample, the
internal consistency of this scale was 0.88.
3.1.2.3. Self-esteem scale. The Rosenberg’s (1965) 10-item self-es-
teem scale was used to measure participants’ generalized, global
feelings of self-worth. Participants respond to statements on posi-
tive and negative feelings about the self, such as ‘‘I am able to do
things as well as most other people” on a seven-point Likert scale
anchored by strongly disagree and strongly agree. This self-esteem
scale has shown good reliability and validity. Internal reliability
ranged from .77 to .88, and the test–retest reliability coefficient
ranged from .82 (two-week interval) to .85 (one-week interval).
This self-esteem scale has also demonstrated adequate construct
validity, convergent validity, and discriminant validity (see, Robin-
son, Shaver, & Wrightsman, 1991). The internal reliability in the
current sample was 0.93.
3.2. Results
3.2.1. Demographic analysis
Gender had no significant effect on all research variables. Age
had a significant correlation with anxiety attachment (r = 0.30,
p < 0.01) and self-concept clarity (r = 0.28, p < 0.05), but was not
correlated with avoidance attachment and self-esteem. There
was no interaction effect of age and gender on the four main re-
search variables. In the following analysis, the age effect was in-
cluded as a control variable.
3.2.2. Correlation and regression analysis
Because age had a significant relationship with anxiety attach-
ment and self-concept clarity, partial correlations controlling for
the age effect were reported here. First, anxiety attachment was
positively correlated with avoidance attachment (r = 0.25, p < .01)
and negatively correlated with self-esteem and self-concept clarity
(r = 0.43, p < 0.01; r = 0.34, p < 0.01). Avoidance attachment was
negatively correlated with self-esteem and self-concept clarity
(r = 0.36, p < 0.01; r = 0.22, p < 0.01). Finally, self-esteem and
self-concept clarity are positively correlated (r = 0.59, p < 0.01).
When self-concept clarity was regressed on anxiety and avoidance
attachment, only anxiety attachment had a significant effect
(b = 0.43, b = 0.37, t(120) = 4.32, p < 0.01). When self-esteem
was predicted, both anxiety (b = 0.53, b = 0.39, t(120) = 4.84,
 C.-h. Wu / Personality and Individual Differences 47 (2009) 42–46
AGE
-0.63* -0.22*
ANX
-0.43**
0.12
-0.19*
AVO
Fig. 1. Path model for mediation effect of self-esteem on the relationship between attachment styles and self-concept clarity. Note: Uniqueness parameters were ignored for
simplicity. ANX; anxiety attachment, AVO; avoidance attachment, EST; self-esteem, SCC; self-concept clarity, *p < 0.05, and **p < 0.01.
p < 0.01) and avoidance attachment had significant effects
(b = 0.41, b = 0.27, t(120) = 3.31, p < 0.01).
3.2.3. Path analysis
The mediation path model is displayed in Fig. 1, in which anxiety
and avoidance attachment had a direct effect on self-esteem, and
then, self-esteem had a direct effect on self-concept clarity.
Although avoidance attachment did not have a significant unique ef-
fect on self-concept clarity in the regression analysis, it was still in-
cluded in the path model, because the independent variable is
unnecessary to have a direct effect on the dependent variable in
testing for mediation effects (James, Mulaik, & Brett, 2006; MacKin-
non, Lockwood, Hoffman, West, & Sheets, 2002). Since age had a sig-
nificant relation with anxiety attachment and self-concept clarity, it
was also included as an exogenous variable in the model to predict
self-concept clarity directly. The model was analyzed by maximum-
likelihood method with robust estimator with Mplus (Muthén &
Muthén, 2007). Fig. 1 presents the standardized estimates of the
model. The Satorra–Bentler chi-square test retained the model
(SB-v2(3) = 5.00, p > .05), and fit indices also suggested this model
was acceptable (TLI = 0.96; CFI = 0.98; RMSEA = 0.074; SRMR =
0.03).
In order to test the mediation effect, indirect effects of self-es-
teem on the relationship between attachment and self-concept
clarity were tested in Mplus (Muthén & Muthén, 2007) as well.
In the path model, the mediation effect can be tested by the indi-
rect effect computed by the product of involving path coefficients
(James et al., 2006; MacKinnon et al., 2002). Specifically, the medi-
ation effect of self-esteem on the relationship between anxiety
attachment and self-concept clarity is the product of the path load-
ing between anxiety attachment and self-esteem and the path
loading between self-esteem and self-concept clarity. Both the
indirect effects of self-esteem on the relationship between anxiety
(and avoidance) attachment and self-concept clarity are significant
(standardized indirect effect = 0.25, p < 0.01; standardized indi-
rect effect = 0.11, p < 0.05). Moreover, when two direct effects
from anxiety (and avoidance) attachment to self-concept clarity
are added in the model, these two direct effects are not significant.
Thus, it can be concluded that self-esteem completely mediates the
relationship between attachment style and self-concept clarity.
In order to certify the direction of mediation in relationships
among attachment style, self-esteem, and self-concept clarity, an
alternative model in which self-concept clarity was assumed to
mediate the relationship between attachment style and self-es-
teem was examined. In that model, the path loading from avoid-
ance attachment and self-concept clarity was not significant. The
indirect effect of self-concept clarity on the relationship between
anxiety attachment and self-esteem was significant (standardized
indirect effect = 0.22, p < 0.01). However, the Satorra–Bentler
chi-square test rejected the alternative model (SB-v2(3) = 26.14,
p < .01), and various fit indices also suggested that the model was
poor (TLI = 0.62; CFI = 0.84; RMSEA = 0.25; SRMR = 0.06). There-
45
0.17**
EST SCC
0.57**
fore, this model was rejected. Accordingly, results from the path
model support the hypothesis that self-esteem mediates the effect
of attachment style on self-concept clarity.
4. Discussion
This research investigated the mediation role of self-esteem in
the relationship between attachment style and self-concept clarity.
First, in all samples, anxiety and avoidance attachment tendency
had negative relations with self-certainty or self-concept clarity,
suggesting that the insecure attachment tendency is really related
to an unclear self-concept. However, regression analysis showed
that only anxiety attachment had a unique effect in predicting
self-certainty or self-concept clarity when both anxiety and avoid-
ance attachment were included as predictors. Study 2 further indi-
cated that the negative relation of insecure attachment tendency to
self-concept clarity is mediated by self-esteem. Although the medi-
ation effects of self-esteem for anxiety and avoidance attachment
were both significant, it was also found that the mediation effect
of self-esteem for avoidance attachment was not large. Therefore,
it can be concluded that the dimension of anxiety attachment plays
a key determinant and self-esteem mediates its effect on self-con-
cept clarity.
However, this study only tests the mediation model using cross-
sectional data. It is desirable to conduct a longitudinal study to
examine the direction between self-esteem and self-concept clar-
ity with daily records. A longitudinal study can facilitate us to
investigate the self-mechanism within attachment theory by
including the stability of self-esteem in the model. Self-esteem sta-
bility refers to the magnitude of fluctuations in momentary, con-
textually based self-esteem (Kernis, Cornell, Sun, Berry, &
Harlow, 1993). It is usually measured by the standard deviation
of people’s self-esteem scores across multiple occasions within
several days and is independent from the level of self-esteem. Re-
cently, Foster, Kernis, and Goldman (2007) link adult attachment to
self-esteem stability by showing that anxiety attachment was re-
lated to unstable self-esteem but avoidance attachment, con-
versely, was unrelated to self-esteem stability. Their findings are
consistent with the results obtained by Srivastava and Beer
(2005) that anxiety attachment potentiates the sociometer process
and avoidance attachment is less impacted by the sociometer pro-
cess, revealing the ambivalence and avoidance nature of the anxi-
ety and avoidance attachment in their social worlds. If we take the
stability of self-esteem into account in the perspective proposed in
this study (i.e., self-esteem is an indicator of attachment security
which influences the confidence and quality in social interactions),
then, it would be expected that the inherent self-esteem (in)stabil-
ity of different attachment styles may also have an impact on the
development of self-concept clarity. Therefore, in future studies,
it is desirable to conduct a longitudinal study to examine the
mechanism behind attachment style, self-esteem (level and stabil-
ity) and self-concept clarity with daily records.
46 C.-h. Wu / Personality and Individual Differences 47 (2009) 42–46
Generally, this study brings a new perspective by treating self-
esteem as a characteristic of security results from attachment
experiences in discussing the relationship between self-esteem
and self-concept clarity and may facilitate integration of different
self-concept systems under attachment theory. For example, stud-
ies on the relationship between attachment styles and models of
self have been conducted from different self-aspects such as self-
esteem (level and stability) and self-perception, the self-concept
clarity proposed in this research, and attributes of self-representa-
tion (self-structure, self-complexity, and self-discrepancy; Mikul-
incer, 1995). All these findings are consistent with attachment
theory; secure attachment persons have higher self-esteem, stable
self-esteem, better self-perception accuracy, greater self-clarity,
and a better organized self-structure than insecure persons. How-
ever, these empirical studies did not provide a systematic perspec-
tive describing the relationships between attachment style and
these self-aspects. In the current study, self-esteem and self-con-
cept clarity are linked to explain individual differences in attach-
ment styles with respect to self-concept clarity, and the sense of
security is introduced as a theoretical basis for incorporating these
two self-aspects into the attachment system. Similarly, it is worth
linking other self-aspects to discover the relation among these self-
related constructs for systematically understanding a complete
model of self in the context of attachment theory. For example,
based on previous findings, links may be made among the knowl-
edge components of self-concept, such as the self-representation
structure, self-concept clarity, and the consequence of self-percep-
tion. Additional links with the evaluative component of self-con-
cept, self-esteem, can be added to explain the affective effect on
the self-system from the perspective of security. Linking these
self-components will provide a clear picture of the relevant self-
information processing mechanism of each attachment style and
strengthen the power of attachment theory to explain the norma-
tive process of a model of self.
References
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults:
A test of a four-category model. Journal of Personality and Social Psychology, 61,
226–244.
Baumgardner, A. H. (1990). To know oneself is to like oneself self-certainty and self-
affect. Journal of Personality and Social Psychology, 58, 1062–1072.
Berger, L. (2001). The relationship between accuracy of self-perception and attachment
organization in adolescence. Unpublished undergraduate thesis, University of
Virginia, Charlottesville.
Bowlby, J. (1969). Attachment and loss. Attachment (Vol. 1). New York: Basic Books.
Brennan, K. A., & Bosson, J. K. (1998). Attachment-style differences in attitudes
toward and reactions to feedback from romantic partners: An exploration of the
relational bases of self-esteem. Personality and Social Psychology Bulletin, 24,
699–714.
Brennan, K. A., & Morris, K. A. (1997). Attachment styles, self-esteem, and patterns
of seeking feedback from romantic partners. Personality and Social Psychology
Bulletin, 23, 23–31.
Brockner, J. (1984). Low self-esteem and behavioral plasticity: Some implications
for personality and social psychology. In L. Wheeler (Ed.)., Review of personality
and social psychology (Vol. 4, pp. 237–271). Beverly Hills, CA: Sage.
Brockner, J., Wiesenfeld, B. M., & Raskas, D. F. (1993). Self esteem and expectancy-
value discrepancy: The effects of believing that you can (or can’t) get what you
want. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard
(pp. 219–240). New York: Plenum Press.
Campbell, J. D. (1990). Self-esteem and certainty of the self-concept. Journal of
Personality and Social Psychology, 56, 538–549.
Campbell, J. D., & Lavallee, L. F. (1993). Who am I? The role of self-concept confusion
in understanding the behavior of people with low self-esteem. In R. F.
Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 3–20). New
York: Plenum.
Campbell, J. D., Trapnell, P., Heine, S. J., Katz, I. M., Lavallee, L. F., & Lehman, D. R.
(1996). Self-concept clarity: Measurement, personality correlates, and cultural
boundaries. Journal of Personality and Social Psychology, 70, 141–156.
Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and
relationship quality in dating couples. Journal of Personality and Social
Psychology, 58, 644–663.
Dozier, M. L., & Lee, S. W. (1995). Discrepancies between self- and other-report of
psychiatric symptomatology: Effects of dismissing attachment strategies.
Development and Psychopathology, 7, 217–226.
Foster, J. D., Kernis, M. H., & Goldman, B. M. (2007). Linking adult attachment to self-
esteem stability. Self and Identity, 6, 64–73.
Griffin, D., & Bartholomew, K. (1994). Models of the self and other: Fundamental
dimensions underlying measures of adult attachment. Journal of Personality and
Social Psychology, 67, 430–445.
James, L. R., Mulaik, S. A., & Brett, J. M. (2006). A tale of two methods. Organizational
Research Methods, 9, 233–244.
Kernis, M. H., Cornell, D. P., Sun, C., Berry, A., & Harlow, T. (1993). There’s more to
self-esteem than whether it is high or low: The importance of stability of self-
esteem. Journal of Personality and Social Psychology, 65, 1190–1204.
Kobak, R. R., & Sceery, A. (1988). Attachment in late adolescence: Working models,
affect regulation, and representations of self and others. Child Development, 59,
135–146.
Leary, M. R., Haupt, A. L., Strausser, K. S., & Chokel, J. T. (1998). Calibrating the
sociometer: The relationship between interpersonal appraisals and the state
self-esteem. Journal of Personality and Social Psychology, 74, 1290–1299.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an
interpersonal monitor: The sociometer hypothesis. Journal of Personality and
Social Psychology, 68, 518–530.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A
comparison of methods to test mediation and other intervening variable effects.
Psychological Methods, 7, 83–104.
Mikulincer, M. (1995). Attachment style and the mental representation of the self.
Journal of Personality and Social Psychology, 69, 1203–1215.
Muthén, L. K., & Muthén, B. O. (2007). Mplus user’s guide (4th ed.). Los Angeles, CA:
Muthén & Muthén.
Nezlek, J. B., Kowalski, R. M., Leary, M. R., Blevins, T., & Holgate, S. (1997).
Personality moderators of reactions to interpersonal rejection: Depression and
trait self-esteem. Personality and Social Psychology Bulletin, 23, 1235–1244.
Nezlek, J. B., & Plesko, R. M. (2001). Day-to-day relationships among self-concept
clarity, self-esteem, daily events, and mood. Personality and Social Psychology
Bulletin, 27, 201–211.
Park, L. E., Crocker, J., & Mickelson, K. D. (2004). Attachment styles and
contingencies of self-worth. Personality and Social Psychology Bulletin, 30,
1243–1254.
Roberts, J. E., Gotlib, I. H., & Kassel, J. D. (1996). Adult attachment security and
symptoms of depression: The mediating roles of dysfunctional attitudes and
low self-esteem. Journal of Personality and Social Psychology, 70, 310–320.
Robinson, J. P., Shaver, P. R., & Wrightsman, L. S. (1991). Measures of personality and
social psychological attitudes. San Diego, CA: Academic Press.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.
Setterlund, M. B., & Niedenthal, P. M. (1993). ‘‘Who am I? Why am I here?”: Self-
esteem, self-clarity, and prototype-matching. Journal of Personality and Social
Psychology, 4, 769–780.
Srivastava, S., & Beer, J. S. (2005). How self-evaluations relate to being liked by
others: Integrating sociometer and attachment perspectives. Journal of
Personality and Social Psychology, 89, 966–977.
Wu, C. H. (2005). Factor analysis of the adult attachment scale – Taiwan version.
Poster presented at the 44th annual conference of the Taiwanese psychological
association. Tau-Yuan, Taiwan.
Wu, C. H., & Lin, Y. C. (2005). The influence of adult attachment styles on the sense
of trust to cyber-others and the cyber self-certainty (in Chinese). Journal of
Cyber Culture and Information Society, 9, 325–342.