ArticlePDF Available

Personal identity processes and self-esteem: Temporal sequences in high school and college students

Authors:

Abstract

Based on a dual-cycle identity model, we examined how identity processes were associated with self-esteem in high school and college students. Cross-lagged analyses in three longitudinal studies found that commitment making and identification with commitment were positively related and ruminative exploration was negatively related to self-esteem. A self-esteem main-effects model was supported in high school students (with self-esteem predicting these identity processes) and a reciprocal model was supported in college students (with identification with commitment and ruminative exploration being reciprocally related to self-esteem). Apparently, high self-esteem functions as a resource for tackling identity-related issues in high school and college students. When adolescents enter college and make the transition to adulthood, identity consolidation, in turn, increasingly plays into self-esteem as well.
1
Personal Identity Processes and Self-Esteem:
Temporal Sequences in High School and College Students
Koen Luyckx
Theo A. Klimstra
Bart Duriez
Catholic University Leuven, Belgium
Stijn Van Petegem
Wim Beyers
Ghent University, Belgium
Eveline Teppers
Luc Goossens
Catholic University Leuven, Belgium
Correspondence should be sent to Koen Luyckx, KULeuven, Department of Psychology,
Tiensestraat 102, 3000 Leuven, Belgium. E-mail: Koen.Luyckx@ppw.kuleuven.be. The fourth
author is a doctoral researcher at the Fund for Scientific Research Flanders (FWO).
2013, Journal of Research in Personality, 47, 159-170.
doi: http://dx.doi.org/10.1016/ j.jrp.2012.10.005
Final author version, accepted
2
Abstract
Based on a dual-cycle identity model, we examined how identity processes were associated with
self-esteem in high school and college students. Cross-lagged analyses in three longitudinal studies
found that commitment making and identification with commitment were positively related and
ruminative exploration was negatively related to self-esteem. A self-esteem main-effects model
was supported in high school students (with self-esteem predicting these identity processes) and a
reciprocal model was supported in college students (with identification with commitment and
ruminative exploration being reciprocally related to self-esteem). Apparently, high self-esteem
functions as a resource for tackling identity-related issues in high school and college students.
When adolescents enter college and make the transition to adulthood, identity consolidation, in
turn, increasingly plays into self-esteem as well.
Key words: identity; self-esteem; self; adolescence; transition to adulthood.
3
Personal Identity Processes and Self-Esteem:
Temporal Sequences in High School and College Students
Personal identity and self-esteem are accorded a prominent role in adolescence and the
transition to adulthood in various theories of personality development (Erikson, 1968; Harter,
1999). Put simply, one’s personal identity provides the answer to the question “Who am I and
what do I want to do in my life?”. Self-esteem typically refers to a global evaluation of one’s self-
worth (Rosenberg, 1965). Intimate links between personal identity and self-esteem have been
posited in various theories. Identity theorists such as Erikson, for instance, viewed identity and
identity confusion as the polar outcomes of the late adolescent psychosocial crisis. Individuals
resolve the identity crisis either by the achievement of a synthesized identity or by ending up in a
state of identity confusion, in which commitments to identity issues are vague or even non-
existent. The stronger their identity, the more aware individuals appear to be of their strengths and
weaknesses and the stronger their self-esteem. Conversely, the more diffused this identity
structure, the more confused individuals seem to be and the weaker their self-esteem. Self-esteem
theorists, for their part, also state that identity and self-esteem are interdependent and mutually
reinforcing mechanisms in a common self-system (Leary & Tangney, 2003). The concept of self-
esteem, therefore, has been an integral part of the study of personal identity development (Heppner
& Kernis, 2011).
However, no study to date investigated how personal identity processes and self-esteem
actually influence one another over time. Hence, the present studies examined longitudinal
associations linking identity processes and global self-esteem during the high school and college
periods using a dual-cycle model of personal identity formation (Luyckx, Goossens, & Soenens,
2006). Before we turn to an in-depth discussion of how identity and self-esteem develop in high
school and college, we discuss how we assessed personal identity formation in the present article.
Personal Identity Processes and Self-Esteem
4
The Classical Approach to Examining Personal Identity Formation
The construct of ego identity refers to an aspect of personality that articulates the place of
individuals within society and affords them a sense of uniqueness (Helson & Srivastava, 2001;
McAdams & Olson, 2010). Most previous studies on ego identity formation have been guided by
Marcia’s (1980) identity status paradigm, in which the processes of exploration and commitment
are distinguished. Exploration refers to the active questioning of various identity alternatives,
whereas commitment pertains to the adherence to a set of convictions, goals, and values. Based on
these two processes, Marcia defined four identity statuses: achievement (strong commitments after
exploration), foreclosure (strong commitments without past exploration), moratorium (exploring
alternatives without arriving at commitments), and diffusion (no current commitments or
exploration). A recent meta-analysis indicated that, generally speaking, foreclosed and achieved
individuals display the highest levels of self-esteem, whereas individuals in diffusion and
moratorium display the lowest levels of self-esteem (Kroger & Marcia, 2011). Put differently,
strong identity commitments seem to be accompanied by high levels of self-esteem, whereas the
absence of commitments (which may or may not be coupled with a current search for such
commitments) seems to be accompanied by lowered self-esteem. However, a detailed
understanding of how the underlying processes of commitment and exploration exactly relate to
self-esteem has not fully emerged from these studies.
A Dual-Cycle Model of Personal Identity Formation
Several identity theorists have moved beyond the identity status paradigm and have
developed broader process-oriented models of identity, in which they “unpack” exploration and
commitment into a larger set of specific processes. One example is the work of Luyckx, Schwartz,
Berzonsky, and colleagues (2008) who empirically distinguished among five separate but
interrelated identity processes. Four of these five identity processes are subsumed under two
consecutive cycles of identity formation (Luyckx, Goossens, & Soenens, 2006). Whereas the first
cycle, the commitment formation cycle, represents Marcia’s (1980) classical paradigm, the second
5
cycle, the commitment evaluation cycle, reflects more recent views on identity. When forming
their commitments, young people can consider different identity alternatives before they decide
upon a given commitment. This first cycle, therefore, may be described in terms of two processes,
that is, exploration in breadth or the pro-active exploration of various identity alternatives, and
commitment making, or the adherence to a set of convictions and values. Hence, both of these
processes map onto Marcia’s classical processes of exploration and commitment. However, as
soon as individuals have formed identity commitments, they can start to evaluate these
commitments. They engage in an in-depth exploration of the commitments that are already in
place (e.g., by gathering additional information or talking with others about the choice made) and,
if all goes well, increasingly identify with and grow certain and confident about these choices
(Grotevant, 1987). The second cycle, therefore, may also be described in terms of two processes,
that is, exploration in depth of current commitments and identification with commitment.
A fifth identity process, referred to as ruminative exploration, was later added to the model
(Luyckx, Schwartz, Berzonsky, et al., 2008). This particular form of exploration is conceptualized
as a process that delays or inhibits progress in identity development. Individuals scoring high on
this process experience difficulty settling on satisfying answers to identity questions. Partially
troubled by what they perceive as inadequate progress towards personally important identity goals,
they keep asking themselves the same questions, resulting in feelings of uncertainty and
incompetence (Trapnell & Campbell, 1999).
Among the four processes in the original dual-cycle model, both commitment variables
showed positive concurrent associations with the Rosenberg (1965) Self-Esteem Scale (RSES) –
which taps into global self-esteem – whereas exploration in breadth and exploration in depth
showed negative associations with that same measure. Ruminative exploration also showed a
negative correlation with self-esteem. When looking at unique variability in each exploration
process, ruminative exploration was negatively related to self-esteem, whereas the other two
exploration processes were rather unrelated to self-esteem. Similarly, when looking at unique
6
variability in each commitment variable, identification with commitment in particular was
positively related to self-esteem (Luyckx, Schwartz, Berzonsky, et al., 2008). However, the
prospective associations between these five identity processes and self-esteem in both high school
and college students remain unexplored. A detailed view on such prospective associations is
needed to inform intervention efforts targeting individuals struggling with their self-concept and
identity.
Self and Identity in High School and College
Development and Prospective Associations
High school and college students are likely to show substantial changes in self-esteem and
identity processes, because of the many maturation processes and life transitions occurring
(Erikson, 1968; Harter, 1999). Self-esteem, as operationalized in the present studies, refers to the
overall evaluation of one’s worth or value as a person (Harter, 1999; Rosenberg, 1965). From
middle childhood on, cognitive maturation allows the individual to construct such a global
knowledge of his or her self-worth as a person. Further cognitive maturation through adolescence
gradually allows individuals to arrive at balanced and relatively stable self-views of both positive
and negative attributes, with a realistic view on their limitations and strengths. Middle adolescents
still struggle with apparent contradictions in self-images and have difficulty in dealing with
differing standards and opinions of others. This awareness of opposing self-attributes renders these
individuals vulnerable for confusion, distress, and lowered self-worth (Fischer, 1980; Harter,
1999). However, by the time individuals reach late adolescence and make the transition to
adulthood, they increasingly rely on their own self-standards that govern personal choices and
create their own ideals and aspirations. A recent large-scale longitudinal study indeed confirmed
that global self-esteem increases through adolescence and continues to increase (although more
slowly) through the twenties (Erol & Orth, 2011).
These changes in self-esteem may be partially dependent upon and interrelated with one’s
personal identity formation. For instance, the formation of identity commitments could help
7
individuals to partially overcome the self-image uncertainty typical of middle adolescence.
Further, the self-confident use of personally endorsed standards by late adolescents and college
students motivates them to evaluate their identity choices and commitments made earlier in life
and, consequently, to arrive at an integrated and consolidated identity and subsequent feelings of
self-worth. In sum, reciprocal pathways linking identity processes and self-esteem can be expected
to emerge through adolescence and the transition to adulthood (Grotevant, 1987; Heppner &
Kernis, 2011).
Intertwined with these cognitive maturation processes, individuals experience different life
events and normative expectations in high school and in college. When adolescents attend high
school and start preparing for their future, personal identity formation comes into prominence as
they launch themselves in the exploration process (Meeus, van de Schoot, Keijsers, Schwartz, &
Branje, 2010). More specifically, high school students have to figure out what they want to
achieve in their lives, such as exploring which educational pathway would suit them best
(Skorikov & Vondracek, 2011). By exploring and setting such future-oriented goals, adolescents
can direct their own development and negotiate their passage into adulthood (Duriez, Luyckx,
Soenens, & Berzonsky, 2012; Erikson, 1968; Seginer & Halabi-Kheir, 1998). For instance, in
Belgium, where the present studies were conducted, college students need to choose a specific
major when starting with their first year of higher education. Hence, the educational system and
the societal context expect high school students to commit to a college major by the end of high
school (Kalakoski & Nurmi, 1998).
Next, college students have to rebalance their lives and find their way into college and adult
life. Most freshmen can no longer fully rely on their existing social network of friends and family
and have to deal with many life changes and choices, which can lead to substantial changes in
identity and self-esteem (Montgomery & Côté, 2003). Pascarella and Terenzini (2005) indeed
illustrated that college students change in an integrated way on a broad array of value, attitudinal,
psychosocial, and moral dimensions. For instance, freshmen not only have to adapt to a new
8
academic environment. Most of them also have to adapt to a new living situation and have to
invest in independent time management because, in Belgium, they live away from their parental
home during the week (Montgomery & Côté, 2003). At the same time, they become exposed to
peers stemming from different backgrounds (each with different values and attitudes), which can
lead to substantial self-reflections and re-evaluations of personal choices and commitments. Due to
all these changes, the in-depth assessment, evaluation, and consolidation of identity commitments
and choices have been shown to be a prominent identity task in college students (Klimstra, Hale,
Raaijmakers, Branje, & Meeus, 2010). Hence, in line with the dual-cycle model of identity
formation (Luyckx, Goossens, & Soenens, 2006), initial commitment formation primarily emerges
as a key functional process in high school, whereas commitment evaluation – although it already
emerges during the high school years – represents a key developmental task in college students
(Bosma & Kunnen, 2008; Klimstra, Luyckx, et al., 2010).
It remains to be investigated, however, whether such identity processes have different
consequences for, or are differentially grounded in, self-esteem for high school and college
students. As adolescents grow older and start preparing for adult roles, strong identity
commitments and a clear life path might increase the clarity of the self and could increasingly play
into feelings of self-worth (Schwartz, Côté, & Arnett, 2005). Meeus, Iedema, Maassen, and Engels
(2005) established that, with increasing maturation through adolescence, the making of steady
identity commitments becomes increasingly important for one’s emotional adjustment, providing
indirect evidence that identity processes may relate differently to self-esteem depending on the
developmental period. Further, in related identity research building explicitly on Marcia’s (1980)
work (such as ethnic and racial identity research; Phinney, 1990), additional evidence for the
importance of the developmental context can be found. For instance, Yip, Seaton, and Sellers
(2006) found a meaningful relationship between ethnic identity resolution and depressive
symptoms but only in college students and not in high school students. They explicitly framed
these results within a developmental framework and stated: “If ethnic identity resolution is also a
9
primary developmental task for college students, then we might also expect that there would be
greater psychosocial consequences for resolving such a task within this developmental period”
(Yip et al., 2006, p. 1515).
General Form of Associations
Despite the predominantly cross-sectional nature of studies focusing on the link between
personal identity processes and self-esteem, many researchers interpreted the results as evidence
for a dominant pathway going from identity to self-esteem - referred to as the identity main-effects
model. However, no previous research has justified the claim that processes of exploration or
commitment actually lead to increases or decreases in self-esteem. Hence, this central tenet of
identity research and theorizing (Erikson, 1968; Waterman, 1992) remains to be investigated.
Further, several identity theorists also drew attention to the reverse model – referred to as the
self-esteem main-effects model – in which self-esteem is conceived of as a driving force behind
identity processes (Heppner & Kernis, 2011; Vignoles, Regalia, Manzi, Golledge, & Scabini,
2006). Grotevant (1987), among others, stressed that this reversed path – from self-esteem to
identity formation – also merits empirical attention because it can be expected that self-esteem
exerts an influence on the identity processes outlined earlier. High self-esteem could be a resource
for making strong self-endorsed identity commitments. A sense of competence, which is closely
linked to self-esteem (Orth, Robins, & Roberts, 2008), indeed has been demonstrated to lead to
increased commitment making and subsequent identification with these commitments over time
(Luyckx, Vansteenkiste, Goossens, & Duriez, 2009).
When analyzing the mechanisms forwarded in these two main-effects models, it becomes
clear that these models are not mutually exclusive and could be combined into a third, more
complex model: the reciprocal model (cf. Orth et al., 2008). For instance, the fact that students
make self-endorsed identity choices could increase their self-esteem levels. Conversely, if students
experience high self-esteem, they might feel more competent in making identity-relevant choices
10
and, hence, identify themselves to a higher degree with them. As such, the reciprocal model asserts
that self-esteem could be both an outcome and an antecedent of identity formation processes.
The Present Studies
To ascertain the direction of effects in the identity – self-esteem link, three longitudinal
studies were conducted: one assessing high school students (Study 1) and two assessing college
students (Studies 2 and 3). A cross-lagged design is the preferred method to investigate temporal
sequences and allows for investigating how inter-individual differences in certain variables can
come about over time. In a cross-lagged design, two or more variables are measured at two or
more points in time, yielding estimates of synchronous relations, autoregressive or stability
coefficients, and cross-lagged effects. The former two refer to the association between the different
variables at each point in time, and the prediction of a variable by its level at previous time points,
respectively. The latter effects refer to the prediction of a variable by other variables that have
been measured before, controlling for the baseline level of the predicted variable (Asendorpf &
van Aken, 2003).
Inspired by previous cross-sectional research and the dual-cycle model of identity formation
(Luyckx, Goossens, & Soenens, 2006), we expected that especially identification with
commitment and ruminative exploration would be related to self-esteem over time in high school
and college students. With respect to pro-active exploration, our expectations were less clear. For
instance, on the one hand, exploration in breadth is thought to lead to the making of firm
commitments over time (Luyckx, Goossens, & Soenens, 2006), which could lead to increases in
self-esteem. On the other hand, a broad and continued exploration of alternatives, especially on the
verge of adulthood, might induce feelings of uncertainty and, hence, might be accompanied by
concurrent distress and lowered self-esteem (Schwartz, Zamboanga, Weisskirch, & Rodriguez,
2009). With respect to the reverse path, high self-esteem levels could provide individuals with the
psychological resources necessary to invest in a broad exploratory strategy (Grotevant, 1987;
Luyckx et al., 2009). But then again, low self-esteem individuals have more poorly articulated
11
notions of who they are (Campbell, 1990), also necessitating the need for a broad-based identity
exploration.
In line with the reciprocal model outlined earlier, bi-directional associations involving
identification with commitment and ruminative exploration were expected to emerge in both high
school and college. However, given that identity evaluation and consolidation become increasingly
normative when transitioning to adulthood, we expected that self-esteem would be more strongly
rooted in such identity processes in college students as compared to high school students (cf.
Lerner & Kaufman, 1985). More specifically, as adolescents grow older and start preparing for
adult roles, having strong identity commitments and a clear life path sketched out might increase
the clarity of the self and play into feelings of self-worth (Schwartz et al., 2005), possibly more so
than when adolescents are still in high school.
Additionally, we examined gender differences in identity and self-esteem and examined
prospective relationships between identity and self-esteem when controlling for these potential
gender differences. Previous research on identity and self-esteem found gender differences in
mean levels, with males scoring higher than females on self-esteem and, although not consistently
across studies, males scoring somewhat lower than females on identity exploration (Kling, Hyde,
Showers, & Buswell, 1999; Luyckx et al., 2009; Orth et al., 2008). Finally, we explored whether
the associations between identity processes and self-esteem would be moderated by gender. In line
with previous research focusing more broadly on identity and psychosocial functioning
(Berzonsky, 2011), we did not expect gender to moderate these relationships.
Study 1
In Study 1, we examined associations between all five identity processes in the expanded
model and self-esteem across two measurement waves (with a one-year interval) in high school
students. Our main expectation was that reciprocal associations would hold, with identification
with commitment and ruminative exploration being linked to self-esteem over time. As such, in
line with the reciprocal model detailed above, self-esteem was hypothesized to be both an
12
antecedent and an outcome of these identity processes.
Method
Participants and procedure. Data were collected in 2010 at one high school in the Dutch-
speaking part of Belgium during a regular class period. Students in 9th, 10th, and 11th grade at Time
1 were included in the present study. One year later, participants completed the same set of
questionnaires during a regular class period. Data collections were supervised by the sixth author.
Parental consent was obtained for this particular study. For all three studies, participants signed a
standard consent form at Time 1 and were informed that they could discontinue their participation
at any time. Participation in the study was voluntary and anonymity was guaranteed at both waves;
all participants were assigned a unique code number to protect their confidentiality. The sample
was comprised of 662 adolescents drawn from the academic, technical, and vocational tracks;
65.4% were girls. No information was available with respect to ethnic composition of the sample.
The mean age at Time 1 was 15.37 years (SD = 1.00; range 14 - 18 years).
Individuals were included in the present study if they participated in at least one of both
measurement waves. More than 99% of the sample participated both at Times 1 and 2. Hence, less
than 1% of the data at the scale level was missing. Participants with and without complete data
were compared using Little’s (1988) Missing Completely At Random (MCAR) test which was
non-significant, χ² (39) = 0.82, ns. Accordingly, we used the full information maximum likelihood
(FIML) procedure provided in Mplus 4.0 (Muthén & Muthén, 2002). This procedure uses all
available information (including information from participants with missing data) to estimate the
model parameters (Enders, 2010).
Questionnaires. All questionnaires in Studies 1-3 were in Dutch.
Identity processes. Participants completed the Dimensions of Identity Development Scale
(DIDS), which has been shown to be a highly reliable instrument in Belgian and US student
samples (Luyckx, Schwartz, Berzonsky, et al., 2008; Schwartz et al., 2011). Evidence for its
factorial structure and external validity has been provided in different samples (Luyckx, Schwartz,
13
Berzonsky, et al., 2008). The DIDS assesses identity processes with respect to future plans and
possible life-paths. Each identity process was measured by five items on a 5-point Likert-type
rating scale, ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Sample items read: “I
have decided on the direction I want to follow in my life” (commitment making), “I sense that the
direction I want to take in my life will really suit me” (identification with commitment), “I
regularly think over a number of different plans for the future” (exploration in breadth), “I
regularly talk with other people about the plans for the future I have made for myself” (exploration
in depth), and “It is hard for me to stop thinking about the direction I want to follow in my life”
(ruminative exploration). Cronbach’s alphas were .91, .86, .80, .77, and .82, respectively, at Time
1, and .93, .88, .84, .81, and .84, respectively, at Time 2.
Self-esteem. Self-esteem was measured using the Rosenberg Self-Esteem Scale (RSES;
Rosenberg, 1965). This scale contains 10 items scored on a 4-point Likert-type rating scale,
ranging from 1 (“does not apply to me at all”) to 4 (“applies to me very well”). This questionnaire
was translated into Dutch by Van der Linden, Dijkman, and Roeders (1983), who have provided
evidence for the validity and reliability of this Dutch translation. A sample item is “I feel that I
have a number of good qualities”. Cronbach’s alphas were .90 and .91 at Times 1-2, respectively.
Results and Discussion
Preliminary analyses. Table 1 shows all means and standard deviations. Using one-way
multivariate analysis of variance (MANOVA) at Time 1, a multivariate effect of gender was found
(Wilks’ λ = .93; F (6, 655) = 8.26; p < .001). Follow-up univariate analyses revealed that girls
scored higher than boys on exploration in depth (M = 3.22, SD = 0.70; and M = 3.10, SD = 0.72,
respectively; F (1, 660) = 4.51; p < .05, Cohen’s d = .17) and ruminative exploration (M = 2.86,
SD = 0.83; and M = 2.69, SD = 0.79, respectively; F (1, 660) = 6.76; p < .01, Cohen’s d = .21).
Further, girls scored lower than boys on self-esteem (M = 2.87, SD = 0.62; and M = 3.19, SD =
0.53, respectively; F (1, 660) = 44.08; p < .001, Cohen’s d = .55). Correlations at Times 1-2 are
reported in Table 1. As expected, both commitment variables related positively and ruminative
14
exploration related negatively to self-esteem across time.
Cross-lagged analyses. Structural Equation Modelling (SEM) was used to test the temporal
sequences linking identity processes and self-esteem. In the model being tested, all within-time
associations at Times 1-2 and all autoregressive paths were estimated. Further, gender and age
were controlled for by estimating paths to each of the constructs in the model (Bollen, 1989).
Finally, all lagged effects among the identity processes were included because omission of a subset
of paths may bias the estimates of the remaining paths (Cole & Maxwell, 2003; Reichardt, 2002).
All parameter estimates in the subsequent models being tested in the present studies were within
their bounds (i.e., between -1.0 and 1.0). To evaluate model fit, we used the chi-squared index,
which should be as small as possible; the Root Mean Square Error of Approximation (RMSEA),
which should be less than .10, and preferably .06, for adequate fit; the Standardized Root Mean
Square Residual (SRMR), which should be less than .10; and the Comparative Fit Index (CFI),
which should exceed .90, and preferably .95 (Hu & Bentler, 1999; Kline, 2005). The model
estimated had zero degrees of freedom and, hence, had a perfect fit to the data. Figure 1 displays
all significant standardized autoregressive paths and cross-lagged paths from identity processes to
self-esteem and vice versa1. Self-esteem positively predicted identification with commitment and
negatively predicted ruminative exploration over time. Overall, findings favored the self-esteem
main-effects model.
Multi-group cross-lagged analyses were conducted to assess whether the structural
coefficients obtained would differ between boys and girls. A constrained model (with all
coefficients set equal across gender) was compared with an unconstrained model (with all
coefficients allowed to vary across gender). The null hypothesis of invariant path coefficients
across gender would be rejected if at least two of the following criteria were satisfied (Cheung &
Rensvold, 2002; Vandenberg & Lance, 2000): Δχ² significant at p < .05; ΔCFI .01; and
ΔRMSEA .015. Invariance tests indicated that no significant differences emerged between both
models (Δχ² (36) = 39.86, ns; ΔCFI < .01; ΔRMSEA = .018), favoring the more parsimonious
15
constrained model. Consequently, we could conclude that structural paths applied equally well to
boys and girls.
In sum, in line with hypotheses, self-esteem predicted the identity processes of identification
with commitment and ruminative exploration over time. However, rather contrary to expectations,
these associations were unidirectional in nature in high school students and supported the self-
esteem main-effects model instead of the reciprocal model. In Studies 2 and 3, similar models
were tested on college students.
Study 2
Study 2 examined how the four identity processes in the original dual-cycle model (i.e.,
commitment making, identification with commitment, exploration in breadth, and exploration in
depth) related to self-esteem over time in college students. Four-wave longitudinal data (with
measurement intervals of 1 year) were used. We again assessed whether the across-time
associations would be of the unidirectional or the reciprocal type. Given the fact that identity
evaluation and consolidation become increasingly normative when transitioning to adulthood, we
expected that self-esteem would be more strongly rooted in such identity processes in college
students (assessed in Study 2) as compared to high school students (assessed in Study 1).
Method
Participants and procedure. Data were collected at a large university (mainly attracting
Caucasian students with a middle-class background) in the Dutch-speaking part of Belgium in the
context of the Leuven Trajectories of Identity Development Study (L-TIDES; Luyckx, Goossens,
& Soenens, 2006).2 The first wave was conducted in 2002. Identity and self-esteem were assessed
on a yearly basis at four measurement occasions. Questionnaires were distributed in lecture halls
or by mail. At Time 1, all participants were freshmen from the Faculty of Psychology and
Educational Sciences, which serves a predominantly female student population. Initially, 638
students were contacted but 73 of them refused to participate for reasons unknown (participation
rate of 89%). Hence, our sample was comprised of 565 Caucasian students; 85.3% were women.
16
The mean participant age at Time 1 was 18.66 years (SD = 0.66; range 17-22 years). In total, at the
scale level, 19.5% of the data was missing because of participant drop-out. A non-significant
MCAR test statistic, χ² (80) = 6.26, ns, suggested that missing values could be reliably estimated.
A one-way MANOVA pointed to mean differences on identity and self-esteem at Time 1 between
participants who dropped out after Time 1 and those who participated at all measurement times
(Wilks’ λ = .97; F (5, 559) = 3.17; p < .01). A detailed inspection at the univariate level revealed
that participants who dropped-out only scored somewhat lower on identification with commitment
than those who participated at Times 1-4 (F (1, 563) = 4.50; p < .05, Cohen’s d = .19).
Accordingly, we used the FIML procedure to deal with missing data3.
Questionnaires.
Identity processes. Participants completed the 32-item Ego Identity Process Questionnaire
(EIPQ; Balistreri, Busch-Rossnagel, & Geisinger, 1995), measuring exploration in breadth and
commitment making at a global level (i.e., across different content domains such as politics,
education, friendship, and values). In the Dutch version, Items 9, 10, 13, and 18 were dropped
because they did not pattern significantly on their hypothesized factor (Luyckx, Goossens, Beyers,
& Soenens, 2006). All items were answered on a 5-point Likert-type rating scale, ranging from 1
(“strongly disagree”) to 5 (“strongly agree”). Sample items are “I have definitely decided on the
occupation I want to pursue” (commitment making; 15 items), and “I have never questioned my
views concerning what kind of friend is best for me (reverse coded)” (exploration in breadth; 13
items). Both the English and the Dutch version showed adequate factorial and convergent validity
(Balistreri et al., 1995; Luyckx, Goossens, Beyers, & Soenens, 2006). Cronbach’s alphas were .72,
.80, .75, and .77 at Times 1-4, respectively, for commitment making; and .74, .77, .75, and .76 at
Times 1-4, respectively, for exploration in breadth.
In addition, participants completed the 26-item Utrecht-Groningen Identity Development
Scale (U-GIDS; Meeus, 1996) developed for use with Dutch-speaking adolescents, and measuring
exploration in depth and identification with commitment (again across content domains). All items
17
were answered on the same 5-point Likert-type rating scale. Sample items are “My education gives
me certainty in life” (identification with commitment; 16 items), and “I try to figure out regularly
what other people think about my best friend” (exploration in depth; 10 items). Meeus,
Oosterwegel, and Vollebergh (2002) provide an overview of concurrent and construct validity of
the measure. Cronbach’s alphas were .82, .83, .86, and .85 at Times 1-4, respectively, for
identification with commitment; and .64, .68, .68, and .69 at Times 1-4, respectively, for
exploration in depth.
Self-esteem. Self-esteem level was again measured using the RSES. Cronbach’s alphas were
.91, .91, .92, and .92 at Times 1-4, respectively.
Results and Discussion
Preliminary analyses. Table 2 shows means and standard deviations. Using a one-way
MANOVA at Time 1, a multivariate effect of gender was found (Wilks’ λ = .93; F (5, 559) = 8.27;
p < .001). Follow-up univariate analyses revealed that men scored lower than women on
identification with commitment (M = 3.39, SD = 0.51, and M = 3.50, SD = 0.44, respectively; F (1,
563) = 4.17; p < .05, Cohen’s d = .23) and exploration in depth (M = 3.47, SD = 0.46, and M =
3.62, SD = 0.39, respectively; F (1, 563) = 10.55; p < .001, Cohen’s d = .35), but significantly
higher than women on self-esteem (M = 3.23, SD = 0.57, and M = 2.98, SD = 0.56, respectively; F
(1, 563) = 14.27; p < .001, Cohen’s d = .44). Correlations at Times 1-4 are also reported in Table
2. Both commitment variables related positively and exploration in breadth related negatively to
self-esteem.
Cross-lagged analyses. As in Study 1, SEM was used with all synchronous or within-time
associations and all autoregressive paths between adjacent measurement times being included.
Gender and age were controlled for in all models. Finally, all lagged effects among the identity
processes were included. Path analyses then proceeded in two steps. In the first cross-lagged
model, the structural paths included in the model were freely estimated. This model provided an
adequate fit to the data (χ² (75) = 309.30, p < .001; RMSEA = .07; CFI = .95; SRMR = .04). In the
18
second cross-lagged model, these structural paths were constrained to be equal across all three
time intervals (χ² (125) = 382.48, p < .001; RMSEA = .06; CFI = .95; SRMR = .07). Invariance
tests indicated that the more parsimonious invariant model fitted the data equally well (Δχ² (50) =
73.18, p < .05; but ΔCFI < .01; ΔRMSEA < .015). Consequently, we retained this model with
longitudinal constraints. Figure 2 displays all significant standardized autoregressive paths and
cross-lagged paths from identity processes to self-esteem and vice versa4. Multi-group analyses
indicated that, as in Study 1, the structural paths applied equally well to men and women (Δχ² (25)
= 15.64, ns; ΔCFI < .01; ΔRMSEA < .015).
Results were in line with our expectations and favored the reciprocal model. Self-esteem
level consistently and positively predicted commitment making and identification with
commitment (i.e., partially in line with Study 1). In turn, commitment making and identification
with commitment consistently and positively predicted self-esteem level (i.e., in contrast to Study
1). As the latter findings generally differ from the results of Study 1, Study 2 suggests that identity
formation seems to have increasing repercussions for one’s self-esteem when adolescents enter
college and start making the transition to adulthood. Study 3 was conducted to replicate the latter
findings.
Study 3
Study 3 examined how the five identity processes in the expanded dual-cycle model related
to self-esteem over time in college students. We used three-wave longitudinal data (with
measurement intervals of 3 months) in a sample of college students. As in Study 2, we expected
that (a) especially commitment making and identification with commitment would exhibit
significant associations with self-esteem over time and that (b) the across-time associations would
be of the reciprocal rather than the unidirectional type (and particularly so for identification with
commitment). In addition, ruminative exploration, which was not assessed in Study 2, would be
negatively associated with self-esteem over time.
In Study 3, we also made a distinction between level and stability of self-esteem. Most
19
researchers who studied self-esteem focused exclusively on its level. More recently, however,
multiple components of self-esteem have been emphasized, with the distinction between secure
versus fragile self-esteem (i.e., the degree to which self-esteem is vulnerable to or affected by
positive or negative external influences) being an important aspect of this heterogeneity (Kernis,
2003). Self-esteem instability refers to the magnitude of short-term fluctuations in contextually-
based feelings of self-worth (Heppner & Kernis, 2011). Notably, self-esteem instability has been
found to predict psychosocial outcomes over and above self-esteem level. For instance, self-
esteem instability has been linked to more reactivity to positive and negative daily events, greater
anger and hostility proneness, and more depression in the face of daily hassles (Kernis, 2005). It
therefore seems appropriate, when examining links between identity processes and self-esteem, to
assess the stability of self-esteem in addition to level of self-esteem. To assess self-esteem
instability, individuals are generally asked to complete a measure of global self-esteem on a daily
basis during one or two weeks, with specific instructions to base their answers on how they feel at
the moment of questionnaire completion. In a next step, the standard deviation of individuals’ total
scores across these daily assessments is conceptualized as an indicator of self-esteem instability,
with larger standard deviations pointing to greater instability (Kernis, 2005).
Commitment making and identification with commitment, as an indication of a clear and
consolidated identity, were expected to be negatively related to self-esteem instability. As indirect
support for this hypothesis, self-esteem instability has been linked to lower self-concept clarity and
to less self-determination and experienced meaning in life (Kernis, 2005). We further expected that
ruminative exploration would primarily accompany unstable, fragile feelings of self-worth over
time. Having a poorly developed identity and ruminating over different identity elements might
render individuals more vulnerable for specific evaluative information, thereby enhancing unstable
feelings of self-esteem (Kernis & Waschull, 1995). In support of this hypothesis, previous research
documented associations between unstable self-esteem and feelings of incompetence, suboptimal
coping strategies, and depressive attribution styles (Kernis, 2005). Hence, besides ascertaining the
20
temporal sequence between identity processes and self-esteem level, by assessing self-esteem
instability in-between Times 2-3 we could ascertain whether identity at Time 2 would predict self-
esteem stability and whether self-esteem stability would predict identity at Time 3.
Method
Participants and procedure. Data were collected at the same university as in Study 2. The
first wave was conducted in 2009. Individuals participated in three measurement waves, each 3
months apart. In between Times 2-3 (i.e., in the midst of this time interval), individuals completed
daily assessments of self-esteem during two five-day weeks (i.e., from Monday to Friday). At
Time 1, all participants were freshmen from the Faculty of Psychology and Educational Sciences.
Due to the fact that our Time 1 assessment was organised as a collective testing session for which
students received course credit, none of the students refused participation at Time 1. Our sample
was comprised of 458 students, of whom 84.9% were women; 94% of participants were
Caucasian. Mean age at Time 1 was 18.25 years (SD = 0.97; range 17-24 years). Individuals were
included in the present study if they participated in at least one measurement wave and if they
completed at least 6 out of 10 daily self-esteem assessments, reducing our final sample to 413
participants. In this final sample, 4.1% of the data at the scale level was missing. A non-significant
MCAR test statistic, χ² (134) = 12.33, ns, suggested that missing values could be reliably
estimated. A one-way MANOVA was conducted to investigate mean differences in identity and
self-esteem at Time 1 between participants who dropped out after Time 1 and those who
participated at Times 1-3. No significant differences emerged (Wilks’ λ = .99; F (6, 368) = 0.55;
ns, η2 = .01). Accordingly, as in Studies 1 and 2, we used FIML to deal with missing data.
Questionnaires.
Identity processes. As in Study 1, the DIDS was used to assess commitment making,
identification with commitment, exploration in breadth, exploration in depth, and ruminative
exploration. Cronbach’s alphas were .92, .87, .84, .76, and .85, respectively, at Time 1, .90, .83,
.86, .81, and .86, respectively, at Time 2, and .92, .86, .88, .84, and .88, respectively, at Time 3.
21
Self-esteem. Self-esteem level was again measured using the RSES. Cronbach’s alphas were
.92, .92, and .93 at Times 1-3, respectively. In addition, over the course of two five-day weeks,
participants’ self-esteem stability was assessed by asking them to complete an internet assessment
of the RSES once every evening and to indicate how they felt at that moment (i.e., state self-
esteem). Questionnaires were put online every evening at 6:00 P.M. and participants were
instructed to complete the measures during the evening. Hence, 10 daily self-esteem assessments
were available. As recommended by Kernis, Granneman, and Mathis (1991), 10-point scales were
used, anchored by 1 (“strongly disagree”) and 10 (“strongly agree”). The standard deviation of
total scores across the multiple assessments served as the index of self-esteem instability, with
higher standard deviations indicating more unstable self-esteem (M = .78; SD = .44). We also
created a measure of average state self-esteem for each individual by computing the mean across
these daily assessments (M = 6.84; SD = 1.18) (Kernis et al., 1991). In line with previous research
(Kernis, 2005), average state self-esteem and self-esteem stability were negatively correlated (r = -
.39; p < .001).
Results and Discussion
Preliminary analyses. Table 3 shows means and standard deviations. Using one-way
MANOVA at Time 1, no multivariate effect of gender was found (Wilks’ λ = .98; F (6, 405) =
1.35; ns). Correlations at Times 1-3 are also reported in Table 3. Both commitment variables
related positively and ruminative exploration related negatively to self-esteem.
Cross-lagged analyses. As in Studies 1 and 2, SEM was used and all within-time
associations at Times 1-3 and all autoregressive paths were controlled for. Again, all lagged effects
among the identity processes were included. The measures of average state self-esteem and self-
esteem instability resulting from the daily assessments were incorporated in the cross-lagged
design as shown in Figure 3. The concurrent association between average state self-esteem and
self-esteem instability was included as an additional control. Further, additional paths were
included from self-esteem level at Time 2 to average state self-esteem and from average state self-
22
esteem to self-esteem level at Time 3, respectively. Gender and age were controlled for.
Path analyses proceeded in two steps. In the first model, structural path coefficients were
freely estimated, which provided an adequate fit to the data (χ² (58) = 132.52, p < .001; RMSEA =
.06; CFI = .98; SRMR = .03). Next, in the second model, we constrained these structural
parameters to be equal across all time intervals. With respect to the structural paths stemming from
and leading to the daily assessments of self-esteem, the paths from Time 2 identity to average state
self-esteem were constrained to be equal to those from Time 1 identity to Time 2 self-esteem level;
the paths from average state self-esteem to Time 3 identity were constrained to be equal to those
from Time 1 self-esteem level to Time 2 identity. This constrained model provided an adequate fit
to the data (χ² (96) = 340.23, p < .001; RMSEA = .08; CFI = .94; SRMR = .08) but invariance tests
indicated that this model had a substantially worse fit than the unconstrained model (Δχ² (38) =
207.71, p < .001; ΔCFI = .04; ΔRMSEA = .02). Hence, we rejected the longitudinal constraints
and retained the unconstrained model. Significant standardized structural coefficients are presented
in Figure 35. Further, multi-group cross-lagged analyses indicated that, as in Study 1, the
longitudinal paths applied equally well to men and women (Δχ² (74) = 108.65, p < .01; but ΔCFI <
.01; ΔRMSEA < .015).
As in Study 2, findings were in line with the reciprocal model. Again in line with Study 2,
self-esteem level and average state self-esteem consistently and positively predicted identification
with commitment. In addition, self-esteem level and average state self-esteem consistently and
negatively predicted ruminative exploration, and self-esteem level at Time 2 positively predicted
commitment making at Time 3. Further, whereas in Study 2 identification with commitment
consistently and positively predicted self-esteem level, this path was replicated from Time 1 to
Time 2 in Study 3. Ruminative exploration consistently and negatively predicted self-esteem level
and average state self-esteem. In addition, ruminative exploration at Time 2 positively predicted
self-esteem instability. Apparently, ruminative exploration did not only play into lower levels of
self-esteem over time but also seemed to render individuals more vulnerable for unstable self-
23
esteem.
General Discussion
Using three longitudinal datasets and measures of five personal identity processes (i.e.,
commitment making, identification with commitment, exploration in breadth, exploration in depth,
and ruminative exploration) and two components of self-esteem (i.e., level and stability), we
examined cross-lagged associations between identity processes and self-esteem. Our main concern
was the cross-sectional nature of most previous studies focusing on the link between personal
identity and self-esteem, raising questions about temporal order. In order to design intervention
efforts based on self and identity constructs, a detailed outlook on temporal sequences needs to be
established. Across studies, especially identification with commitment and ruminative exploration
were consistently related to self-esteem over time. However, the exact form of these associations
differed somewhat for high school students and college students. Whereas self-esteem predicted
these identity processes in high school students, reciprocal associations emerged in the college
setting. Apparently, as the evaluation and consolidation of identity commitments and choices
becomes increasingly functional in the college context (Bosma & Kunnen, 2008; Luyckx,
Goossens, & Soenens, 2006), these identity processes could increasingly have repercussions for
one’s self-esteem. Before we discuss the different temporal sequences obtained, readers should
note that the present studies do not allow for drawing definite conclusions with respect to how
these temporal sequences potentially differ between high school and college students. A long-term
longitudinal study in which cohorts of individuals are followed through high school and college is
needed to make more authoritative claims. Nonetheless, the present studies in combination already
shed some light on this intriguing question.
Temporal Sequences Linking Personal Identity and Self-Esteem in High School and College
Our cross-lagged analyses provided support for the self-esteem main-effects model in high
school students (Study 1) and the reciprocal model in college students (Studies 2 and 3). Readers
should note that the cross-lagged associations obtained were not moderated by gender. Hence,
24
despite the fact that some mean gender differences were obtained, findings applied equally well to
males and females, again underlining the robustness of the findings obtained. Consistent across
studies, the same identity processes were intertwined with self-esteem over time. Further, some of
the core mechanisms identified seemed to be invariant across the high school and college periods.
More specifically, in line with previous research documenting the pervasive influence of self-
esteem on human behavior (Campbell, 1990; Kernis, 2005), self-esteem was found to be an
important predictor of especially identification with commitment and ruminative exploration
across both developmental periods. These findings substantially extend previous cross-sectional
research indicating that self-esteem not only constitutes an important correlate of identity (as again
confirmed at the different time-points of our longitudinal studies) but that self-esteem could also
function as a resource for tackling identity-related questions. Feelings of self-worth facilitate an
internal frame of reference and, as such, enable both high school and college students to make and
identify themselves with identity commitments (Erikson, 1968; Harter, 1999). High self-esteem
also seemed to protect against identity worry and rumination, again testifying to the confidence
and competence individuals with high self-esteem display in addressing the many identity options
and alternatives they are confronted with. In sum, experiencing high levels of self-esteem could set
individuals on a pathway to achieving a mature and synthesized sense of identity.
Further, processes of identity evaluation and consolidation increasingly have ramifications
for individuals’ self-esteem when they enter college and start making the transition to adulthood
(Bosma & Kunnen, 2008; Luyckx, Goossens, & Soenens, 2006). In other words, a reciprocal
model with bi-directional influences was supported in Studies 2 and 3. Identification with
commitment and ruminative exploration influenced subsequent feelings of self-worth in college
students. Commitment making, however, did not provide a consistent foundation for subsequent
changes in self-esteem in college students. This finding indicates that, when looking at unique
associations, particularly the degree to which individuals identify themselves with their
commitments influences self-esteem (Luyckx, Schwartz, Berzonsky, et al., 2008). Hence, although
25
self-esteem seems to influence both commitment processes to some extent, especially
identification with commitment, in turn, seems to influence self-esteem in the college setting.
Apparently, increasing confidence over one’s identity choices could provide college students with
a sense of self-worth over time. The consolidation of identity commitments indeed constitutes a
crucial identity task in the transition to adulthood, providing opportunities for college students to
thrive and to experience increases in self-esteem over time. Hence, although the college setting
may be characterized by a diversity of choices and life options for individuals in Western societies
(Arnett, 2000), individuals need to come to grips with themselves and their lives in order to
negotiate the transition to adulthood successfully (Côté & Levine, 2002; Schwartz et al., 2005). As
a further illustration of this tenet, a ruminative approach to identity issues in which identity choices
are postponed rendered college students vulnerable for a lowered and fragile sense of self-worth in
Study 3.
In sum, two reciprocal loops were identified that link identity and self-esteem. The first loop
linked identification with commitment to self-esteem. As noted, self-esteem level was found to
serve as a catalyst for the making and internalization of identity choices in Studies 1-3 (cf.
Vignoles et al., 2006). Conversely, as noted, identification with commitment positively predicted
self-esteem over time in Studies 2-3. Collectively, these findings suggest that high self-esteem and
a strong identification with identity commitments reinforce one another across time, possibly
leading to a stable, strong, and consolidated self (Trzesniewski, Donnellan, & Robins, 2003). Such
a secure self-system could provide an important inner resource for tackling the challenging
transition to adulthood and the many psychosocial tasks college students are confronted with
(Côté, 2000; Côté & Levine, 2002; Montgomery & Côté, 2003).
A second reciprocal loop linked ruminative exploration and self-esteem. Self-esteem was
found to negatively predict ruminative exploration in Studies 1 and 3, and ruminative exploration,
in turn, negatively predicted self-esteem in Study 3, possibly leading to a negative vicious cycle.
Such a pathway could play into a fragile self, which implies that college students do not succeed in
26
arriving at a solid self (Campbell, 1990), but instead worry about the future and feel insecure about
their own self-worth. As a further illustration of this fragile self, ruminative exploration was found
to lead to greater daily instability in self-esteem in Study 3. Accordingly, especially in the college
setting, personal identity formation, self-esteem level, and self-esteem stability appeared to be
components of an interlocking system that have reciprocal effects on one another (Kernis, 2005).
Although Studies 2 and 3 sampled highly similar participants in terms of demographic and
educational background, there were a number of methodological and measurement differences.
First, Study 2 had longer measurement intervals compared to Study 3. Second, Study 2 measured
identity processes across different content domains whereas Study 3 measured identity processes
with respect to future life-plans and lifestyles. Third, in Studies 2 and 3, global self-esteem was
measured using the RSES and average state self-esteem (as based on daily assessments) was used
as an additional index of self-esteem level in Study 3 (Kernis, 2003). Regardless of these
differences, the general pattern of findings was quite similar (although the strength of the cross-
lagged paths differed somewhat), testifying to the validity of the temporal sequences obtained in
the college setting.
Theoretical and Practical Implications
In all three studies, self-esteem was found to be an important resource for the ways
individuals tackle identity-related questions and issues, and self-esteem influences were found on
commitment making, identification with commitment, and ruminative exploration. As such, the
present findings are relevant for the debate on whether self-esteem has any benefits for the
individual or not (Baumeister, Campbell, Krueger, & Vohs, 2003; Donnellan, Trzesniewski, &
Robins, 2011; Swann, Chang-Schneider, & McClarty, 2007). Although critics have indicated that
efforts to boost self-esteem are of little value because self-esteem appears to be inconsequential
(Baumeister et al., 2003), the present findings indicate that individuals’ self-esteem does matter
when it comes to forming a self-endorsed sense of identity (cf. Swann et al., 2007). Hence, due to
the fact that self-esteem was prospectively linked with core identity processes, facilitating or
27
strengthening a stable and secure self-esteem could be a pathway for strengthening and improving
identity-related work in high school and college students. Such interventions might be most
relevant in contemporary late-modern societies that lack the structure and guidance on which to
rely in forming a sense of identity (e.g., Côté, 2000). The importance of individual resources such
as self-esteem indeed becomes increasingly important for individuals to deal with the many
options they are confronted with on the road to adulthood. Feelings of self-worth make individuals
more confident in relying on personally endorsed identity standards, enabling them to make
identity choices and protecting them against regression to a state of chronic identity worry or
rumination.
Provided that future studies following a single cohort through adolescence and emerging
adulthood replicate the present findings, it could also be useful for counselors to focus their
interventions directly on a faulty identity formation process. If some individuals are guided
through a difficult identity process, some of the pain and misfortune associated with this process
may be alleviated and self-esteem may benefit when these individuals transition to adulthood.
Moreover, treatment programs promoting general competence and problem-solving skills can have
a salutary influence on self-related processes (Ferrer-Wreder et al., 2002; Petersen et al., 1993).
However, the (long-term) effects of intervention-induced changes in identity-relevant processes
remain to be investigated. The present findings could provide a rationale to investigate this
uncharted territory. Importantly, for such interventions to be successful in the long run, they
should focus on individuals’ identity and self-esteem simultaneously because, due to the
developmental interdependence of identity and self-esteem, changes in one construct must be
reinforced by corresponding changes in the other construct (Swann et al., 2007).
Limitations and Suggestions for Future Research
Some of the cross-lagged coefficients were rather small, but these coefficients were obtained
while simultaneously controlling for all within-time associations and autoregressive paths. Further,
such relatively modest coefficients could be expected given that identity and self-esteem are
28
multiply determined (Swann et al., 2007). The present studies also have some limitations which
provide suggestions for future research. First, we could not rule out the rival hypothesis that
important third variables were causing the relationships. One likely candidate could be self-
concept clarity, a variable closely linked to both personal identity development and self-esteem
(Campbell, 1990). For instance, in a recent cross-sectional study (Luyckx, Schwartz, Soenens,
Vansteenkiste, & Goossens, 2010), identity integration (which is operationally very close to self-
concept clarity; Campbell, 1990) was found to relate strongly to commitment making,
identification with commitment, and self-esteem. Due to their cognitive maturation, late
adolescents become increasingly capable of dealing with seemingly opposing or conflicting self-
attributes which can lead to a more integrated identity and, hence, an increased sense of self-worth
(Harter, 1999).
Second, identity processes and self-esteem were assessed through self-report questionnaires.
Although questionnaires are most appropriate to gather information about identity and self-esteem,
the reliance on a single informant might artificially inflate correlations among constructs.
However, such shared method variance has been statistically removed by controlling for all
within-time associations and autoregressive paths in the analyses (Orth et al., 2008).
Third, as noted, the present studies focused on very specific aspects of identity, that is,
commitment and exploration processes. Hence, the findings obtained cannot be generalized to
other relational, social, and collective identity aspects (Schwartz, Luyckx, & Vignoles, 2011).
Relatedly, the present studies primarily sampled Caucasian European participants. Previous
research has demonstrated empirical parallels and commonalities across American and European
Caucasian adolescents in personal identity processes and how they relate to psychosocial
functioning (Schwartz, Adamson, Ferrer-Wreder, Dillon, & Berman, 2006). More diverse samples
in terms of ethnic background, however, should be used in future research. Although Schwartz and
colleagues (2005) found substantial consistency across three US ethnic groups in identity
constructs such as commitment and exploration, it remains to be investigated how the different
29
variables assessed in the present studies interrelate in non-Western cultures or in non-Caucasians
living in other Western cultures. For instance, non-Caucasians may have unique ethnicity-related
identity concerns that may relate differently to self-esteem. Likewise, previous cross-cultural
research found that variables such as identity consolidation or consistency can have different
implications for one’s self-esteem depending on one’s ethnic or cultural background (Kiang &
Fuligni, 2009; Suh, 2002). Collectively, these findings urge future research to rely on ethnically
diverse samples, paying attention to the role of the broader socio-cultural context.
Similarly, future research should focus on college samples from different majors that are also
more balanced in terms of gender to make more definite claims with respect to the role of gender
in the identity - self-esteem link. Although the present studies consistently indicated that the over-
time associations were not moderated by gender, a more balanced gender distribution would allow
for making more authoritative claims with respect to the influence of gender. Finally, our college
student samples excluded individuals who do not seek higher education, a group often referred to
as the “forgotten half” (Halperin, 2001). Recent research conducted in Belgium suggested that
college students were more likely than their working counterparts to engage in ruminative
exploration, and less likely to have made commitments (Luyckx, Schwartz, Goossens, & Pollock,
2008). Indeed, the entrance into steady employment directs future decision-making and, hence,
leads to the establishment of steady future commitments. Working late adolescents and emerging
adults may be granted less time to spend on identity exploration, because they do not have access
to the psychosocial moratorium provided by the college setting (Montgomery & Côté, 2003).
Hence, it remains to be investigated if the associations between identity processes and self-esteem
differ between college students and their working peers.
Finally, future studies using techniques such as latent growth curve modelling and latent
class growth analysis should assess how inter-individual differences in intra-individual change in
identity processes and self-esteem emerge and potentially develop in tandem through the transition
to adulthood. Such a complementary view on development and change could further enhance our
30
knowledge on the exact mechanisms linking identity and self-related processes.
Despite these limitations and cautions, the present series of studies has established important
links between key processes of personal identity development and self-esteem level and stability.
A prominent strength of the present article is the use of three longitudinal data-sets tapping into
two different developmental periods. Collectively, these three studies provide preliminary insight
into the ways in which personal identity and self-esteem influence one another over time. In so
doing, these studies illustrate the strategic value of this type of research. Hence, we hope that the
present series of studies will instigate future researchers to disentangle further the fascinating link
between identity and self-esteem.
31
Endnotes
1. With respect to the significant cross-lagged paths among the identity processes, Study 1
indicated that commitment making at T2 was positively predicted by identification with
commitment (β = .11; p < .05) and exploration in depth (β = .08; p < .05) at T1. Identification
with commitment at T2 was positively predicted by commitment making (β = .24; p < .001)
and exploration in depth (β = .09; p < .05) at T1. Exploration in breadth at T2 was positively
predicted by identification with commitment (β = .11; p < .05) and exploration in depth (β =
.18; p < .001) at T1. Exploration in depth at T2 was positively predicted by exploration in
breadth (β = .10; p < .05) at T1. Finally, ruminative exploration at T2 was negatively predicted
by commitment making (β = -.11; p < .05) at T1.
2. Parts of the data of Study 2 have been used in previous articles based on L-TIDES. No
previous article focused on cross-lagged associations between identity and self-esteem.
Readers should note that at the time L-TIDES was conducted, no measure of ruminative
exploration was available yet.
3. We repeated our cross-lagged analysis on those participating at all four measurement times
(longitudinal N = 316). Results were virtually identical as the ones reported in the article. The
same applies to Study 3 (longitudinal N = 369).
4. Study 2 indicated that commitment making was negatively predicted by exploration in
breadth (T1 to T2: β = -.04; p < .05; T2 to T3: β = -.04; p < .05; and T3 to T4: β = -.04; p <
.05) and positively by identification with commitment (T1 to T2: β = .07; p < .01; T2 to T3: β
= .06; p < .01; and T3 to T4: β = .07; p < .01) over time. Identification with commitment was
positively predicted by commitment making (T1 to T2: β = .11; p < .001; T2 to T3: β = .10; p
< .001; and T3 to T4: β = .12; p < .001) over time. Exploration in breadth was negatively
predicted by commitment making (T1 to T2: β = -.05; p < .05; T2 to T3: β = -.05; p < .05; and
T3 to T4: β = -.05; p < .05) over time. Finally, exploration in depth was positively predicted
by commitment making (T1 to T2: β = .09; p < .01; T2 to T3: β = .07; p < .01; and T3 to T4: β
32
= .08; p < .01) over time.
5. Study 3 indicated that commitment making at T2 was negatively predicted by ruminative
exploration at T1 (β = -.16; p < .001) and commitment making at T3 was positively predicted
by exploration in breadth at T2 (β = .17; p < .01). Identification with commitment was
positively predicted by commitment making (T1 to T2: β = .20; p < .001; and T2 to T3: β =
.23; p < .001) and exploration in breadth (T1 to T2: β = .10; p < .05; and T2 to T3: β = .16; p <
.05). Exploration in breadth was positively predicted by exploration in depth (T1 to T2: β =
.18; p < .001; and T2 to T3: β = .19; p < .001), and exploration in breadth at T2 was positively
predicted by commitment making at T1 (β = .15; p < .05). Exploration in depth at T3 was
positively predicted by exploration in breadth at T2 (β = .16; p < .05). Finally, ruminative
exploration was positively predicted by exploration in depth (T1 to T2: β = .21; p < .001; and
T2 to T3: β = .15; p < .01) and negatively by identification with commitment (T1 to T2: β = -
.17; p < .01; and T2 to T3: β = -.13; p < .05).
33
References
Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the
twenties. American Psychologist, 55, 469-480.
Asendorpf, J. B., & van Aken, M. A. G. (2003). Personality-relationship transaction in
adolescence: Core versus surface personality characteristics. Journal of Personality, 71, 629-
666.
Balistreri, E., Busch-Rossnagel, N. A., & Geisinger, K. F. (1995). Development and preliminary
validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18, 179-192.
Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K. D. (2003). Does high self-esteem
cause better performance, interpersonal success, happiness, or healthier lifestyles?
Psychological Science in the Public Interest, 4, 1-44.
Berzonsky, M. D. (2011). A social-cognitive perspective on identity construction. In S. J.
Schwartz, K. Luyckx, & V. L. Vignoles (Eds.), Handbook of identity theory and research
(pp. 55-76). New York: Springer.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Bosma, H. A., & Kunnen, S. E. (2008). Identity-in-context is not yet identity development-in-
context. Journal of Adolescence, 31, 281-289.
Campbell, J. D. (1990). Self-esteem and clarity of the self-concept. Journal of Personality and
Social Psychology, 59, 538–549.
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing
measurement invariance. Structural Equation Modeling, 9, 233-255.
Cole, D. A., & Maxwell, S. E. (2003). Testing meditational models with longitudinal data:
Questions and tips in the use of structural equation modelling. Journal of Abnormal
Psychology, 112, 558-577.
Côté, J. E. (2000). Arrested adulthood: The changing nature of maturity and identity. New York:
New York University Press.
34
Côté, J. E., & Levine, C. G. (2002). Identity formation, agency, and culture: A social psychological
synthesis. Mahwah, NJ: Erlbaum.
Donnellan, M. B., Trzesniewski, K. H., & Robins, R. W. (2011). Self-esteem: Enduring issues and
controversies. In T. Chamorro-Premuzic, S. von Stumm, & A. Furnham (Eds.), The Wiley-
Blackwell handbook of individual differences (pp. 718-746). Chichester, UK: Wiley-
Blackwell.
Duriez, B., Luyckx, K., Soenens, B., & Berzonsky, M. D. (2012). A process-content approach to
adolescent identity formation: Examining longitudinal associations between identity styles
and goal pursuits. Journal of Personality, 80, 135-161.
Enders, C. K. (2010). Applied missing data analysis. New York: Guilford.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Erol, R. Y., & Orth, U. (2011). Self-esteem development from age 14-30 years: A longitudinal
study. Journal of Personality and Social Psychology, 101, 607-619.
Ferrer-Wreder, L., Lorente, C. C., Kurtines, W., Briones, E., Busseri, J., Berman, S., & Arrufat, O.
(2002). Promoting identity development in marginalized youth. Journal of Adolescent
Research, 17, 168-187.
Fischer, K. W. (1980). A theory of cognitive development: The control and construction of
hierarchies of skills. Psychological Review, 87, 477-531.
Grotevant, H. D. (1987). Toward a process model of identity formation. Journal of Adolescent
Research, 2, 203-222.
Halperin, S. (Ed.). (2001). The forgotten half revisited: American youth and young families, 1988-
2008. Washington, DC: American Youth Policy Forum.
Harter, S. (1999). The construction of the self: A developmental perspective. New York: Guilford
Press.
Helson, R., & Srivastava, S. (2001). Three paths of adult development: Conservers, seekers, and
achievers. Journal of Personality and Social Psychology, 80, 995-1010.
35
Heppner, W. L., & Kernis, M. H. (2011). High self-esteem: Multiple forms and their outcomes. In
S. J. Schwartz, K. Luyckx, & V. L. Vignoles (Eds.), Handbook of identity theory and
research (pp. 329-356). New York: Springer.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.
Kalakoski, V., & Nurmi, J. E. (1998). Identity and educational transitions: Age differences in
adolescent exploration and commitment related to education, occupation, and family.
Journal of Research on Adolescence, 8, 29-47.
Kernis, M. H. (2003). Toward a conceptualization of optimal self-esteem. Psychological Inquiry,
14, 1-26.
Kernis, M. H. (2005). Measuring self-esteem in context: Importance of stability of self-esteem in
psychological functioning. Journal of Personality, 73, 1-37.
Kernis, M. H., Grannemann, B. D., & Mathis, L. C. (1991). Stability of self-esteem as a moderator
of the relation between level of self-esteem and depression. Journal of Personality and
Social Psychology, 61, 80-84.
Kernis, M. H., & Waschull, S. B. (1995). The interactive roles of stability and level of self-esteem:
Research and theory. In M. P. Zanna (Ed.), Advances in experimental social psychology
(Vol. 27, pp. 93-141). San Diego, CA: Academic Press.
Kiang, L., & Fuligni, A. J. (2009). Ethnic identity in context: Variations in ethnic exploration and
belonging within parent, same-ethnic peer, and different-ethnic peer relationships. Journal of
Youth and Adolescence, 38, 732-743.
Klimstra, T. A., Hale, W. W., Raaijmakers, Q. A., W., Branje, S. J. T., & Meeus, W. H. J. (2010).
Identity formation in adolescence: Change or stability? Journal of Youth and Adolescence,
39, 150-162.
Klimstra, T. A., Luyckx, K., Hale III, W. W., Meeus, W. H. J., van Lier, P. A. C., & Frijns, T.
36
(2010). Short-term fluctuations in identity: Introducing a micro-level approach to identity
formation. Journal of Personality and Social Psychology, 99, 191-202.
Kline, R. B. (2005). Principles and practices of structural equation modeling (2nd edition). New
York: Guilford Press.
Kling, K. C., Hyde, J. S., Showers, C. J., & Buswell, B. N. (1999). Gender differences in self-
esteem: A meta-analysis. Psychological Bulletin, 125, 470-500.
Kroger, J., & Marcia, J. E. (2011). The identity statuses: Origins, meanings, and interpretations. In
S. J. Schwartz, K. Luyckx, & V. L. Vignoles (Eds.), Handbook of identity theory and
research (pp. 31-54). New York: Springer.
Leary, M. R., & Tangney, J. P. (2003). Handbook of self and identity. New York: Guilford Press.
Lerner, R. M., & Kauffman, M. B. (1985). The concept of development in contextualism.
Developmental Review, 5, 309-333.
Little, R. J. A. (1988). A test of missing completely at random for multivariate data with missing
values. Journal of the American Statistical Association, 83, 1198-1202.
Luyckx, K., Goossens, L., Beyers, W., & Soenens, B. (2006). The Ego Identity Process
Questionnaire: Factor structure, reliability, and convergent validity in Dutch-speaking late
adolescents. Journal of Adolescence, 29, 153-159.
Luyckx, K., Goossens, L., & Soenens, B. (2006). A developmental-contextual perspective on
identity construction in emerging adulthood: Change dynamics in commitment formation
and commitment evaluation. Developmental Psychology, 42, 366-380.
Luyckx, K., Schwartz, S. J., Berzonsky, M. D., Soenens, B., Vansteenkiste, M., Smits, I., &
Goossens, L. (2008). Capturing ruminative exploration: Extending the four-dimensional
model of identity formation in late adolescence. Journal of Research in Personality, 42, 58-
82.
Luyckx, K., Schwartz, S. J., Goossens, L., & Pollock, S. (2008). Employment, sense of coherence,
and identity formation: Contextual and psychological processes on the pathway to sense of
37
adulthood. Journal of Adolescent Research, 23, 566-591.
Luyckx, K., Schwartz, S. J., Soenens, B., Vansteenkiste, M., & Goossens, L. (2010). The path
from identity commitments to adjustment: Motivational underpinnings and mediating
mechanisms. Journal of Counseling and Development, 88, 52–60.
Luyckx, K., Vansteenkiste, M., Goossens, L., & Duriez, B. (2009). Basic need satisfaction and
identity formation: Bridging self-determination theory and process-oriented identity research.
Journal of Counseling Psychology, 56, 276-288.
Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent
psychology (pp. 159-186). New York: Wiley.
McAdams, D. P., & Olson, B. D. (2010). Personality development: Continuity and change over the
life course. Annual Review of Psychology, 61, 517-542.
Meeus, W. (1996). Studies on identity development in adolescence: An overview of research and
some new data. Journal of Youth and Adolescence, 25, 569-598.
Meeus, W., Iedema, J., Maassen, G., & Engels, R. (2005). Separation-individuation revisited: On
the interplay of parent-adolescent relations, identity and emotional adjustment in
adolescence. Journal of Adolescence, 28, 89-106.
Meeus, W., Oosterwegel, A., & Vollebergh, W. (2002). Parental and peer attachment and identity
development in adolescence. Journal of Adolescence, 25, 93-106.
Meeus, W., van de Schoot, R., Keijsers, L., Schwartz, S. J., & Branje, S. (2010). On the
progression and stability of adolescent identity formation: A five-wave longitudinal study in
early-to-middle and middle-to-late adolescence. Child Development, 81, 1565-1581.
Montgomery, M. J., & Côté, J. E. (2003). College as a transition to adulthood. In G. R. Adams &
M. D. Berzonsky (Eds.), Blackwell handbook of adolescence (pp. 149-172). Malden, MA:
Blackwell.
Muthén, L. K., & Muthén, B. O. (2002). Mplus user’s guide. Los Angeles, CA : Muthén &
Muthén.
38
Orth, U., Robins, R. W., & Roberts, B. W. (2008). Low self-esteem prospectively predicts
depression in adolescence and young adulthood. Journal of Personality and Social
Psychology, 95, 695-708.
Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students: A third decade of
research. Hoboken, NJ: Wiley.
Petersen, A. C., Compas, B. E., Brooks-Gunn, J., Stemmler, M., Ey, S., & Grant, K. E. (1993).
Depression in adolescence. American Psychologist, 48, 155-168.
Phinney, J. S. (1990). Ethnic identity in adolescence and adulthood: A review and integration.
Psychological Bulletin, 108, 499-514.
Reichardt, C. S. (2002). The priority of just-identified, recursive models. Psychological Methods,
7, 307-315.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University
Press.
Schwartz, S. J., Adamson, L., Ferrer-Wreder, L., Dillon, F. R., & Berman, S. L. (2006). Identity
status measurement across contexts: Variation in measurement structure and mean levels
among White American, Hispanic American, and Swedish emerging adults. Journal of
Personality Assessment, 86, 61-76.
Schwartz, S. J., Beyers, W., Luyckx, K., Soenens, B., Zamboanga, B. L., Forthun, L. F., Hardy, S.
A., Vazsonyi, A. T., Ham, L. S., Kim, S. Y., Whitbourne, S. K., & Waterman, A. S. (2011).
Examining the light and dark sides of emerging adults’ identity: A study of identity status
differences in positive and negative psychosocial functioning. Journal of Youth and
Adolescence, 40, 839-859.
Schwartz, S. J., Côté, J. E., & Arnett, J. J. (2005). Identity and agency in emerging adulthood: Two
developmental routes in the individualization process. Youth and Society, 37, 201-229.
Schwartz, S. J., Luyckx, K., & Vignoles, V. L. (Eds.) (2011). Handbook of identity theory and
39
research. New York: Springer.
Schwartz, S. J., Zamboanga, B. L., Weisskirch, R. S., & Rodriguez, L. (2009). The relationships of
personal and ethnic identity exploration to indices of adaptive and maladaptive psychosocial
functioning. International Journal of Behavioral Development, 33, 131–144.
Seginer, R., & Halabi-Kheir, H. (1998). Adolescent passage to adulthood: Future orientation in the
context of culture, age, and gender. International Journal of Intercultural Relations, 22, 309-
328.
Skorikov, V. B., & Vondracek, F. W. (2011). Occupational identity. In S. J. Schwartz, K. Luyckx,
& V. L. Vignoles (Eds.), Handbook of identity theory and research. New York: Springer.
Suh, E. M. (2002). Culture, identity consistency, and subjective well-being. Journal of Personality
and Social Psychology, 83, 1378-1391.
Swann Jr., W. B., Chang-Schneider, C., & McClarty, K. L. (2007). Do people’s self-views matter?
Self-concept and self-esteem in everyday life. American Psychologist, 62, 84-94.
Trapnell, P. D., & Campbell, J. D. (1999). Private self-consciousness and the Five-Factor model of
personality: Distinguishing rumination from reflection. Journal of Personality and Social
Psychology, 76, 284-304.
Trzesniewski, K. H., Donnellan, M. B., & Robins, R. W. (2003). Stability of self-esteem across the
life span. Journal of Personality and Social Psychology, 84, 205-220.
Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance
literature: Suggestions, practices, and recommendations for organizational research.
Organizational Research Methods, 3, 4-70.
Van der Linden, F. J., Dijkman, T. A., & Roeders, P. J. B. (1983). Metingen van kenmerken van
het persoonssysteem en sociale systeem [Measurements of features of the person system and
the social system]. Nijmegen, The Netherlands: Hoogveld Instituut.
40
Vignoles, V. L., Regalia, C., Manzi, C., Golledge, J., & Scabini, E. (2006). Beyond self-esteem:
Influence of multiple motives on identity construction. Journal of Personality and Social
Psychology, 90, 308-333.
Waterman, A. S. (1992). Identity as an aspect of optimal psychological functioning. In J. E.
Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), Identity: A
handbook for psychosocial research (pp. 50-72). New York: Springer-Verlag.
Yip, T., Seaton, E. K., & Sellers, R. M. (2006). African American racial identity across the life
span: Identity status, identity content, and depressive symptoms. Child Development, 77,
1504-1517.
41
Table 1
Descriptive Statistics and Pearson Correlations Between Identity Processes and Self-Esteem at Times 1 and 2 in Study 1
Variable
M (SD)
T1
M (SD)
T2
rs with self-esteem
T1 T2
1. Commitment making 3.38 (0.88) 3.48 (0.93) .10** .21***
2. Identification commitment 3.43 (0.73) 3.47 (0.75) .26*** .34***
3. Exploration in breadth 3.55 (0.68) 3.65 (0.70) .06 .10**
4. Exploration in depth 3.18 (0.71) 3.36 (0.73) .01 .11**
5. Ruminative exploration 2.80 (0.82) 2.84 (0.85) -.32*** -.33***
6. Self-esteem 2.98 (0.61) 3.06 (0.60) -- --
Note.T = Time; M = mean; SD = standard deviation.
* p < .05. ** p < .01. *** p < .001.
42
Table 2
Descriptive Statistics and Pearson Correlations Between Identity Processes and Self-Esteem at Times 1 Through 4 in Study 2
Variable
M (SD)
T1
M (SD)
T2
M (SD)
T3
M (SD)
T4
rs with self-esteem
T1 T2 T3 T4
1. Commitment making 3.17 (0.43) 3.20 (0.45) 3.28 (0.42) 3.31 (0.38) .24*** .39*** .38*** .40***
2. Identification 3.48 (0.45) 3.49 (0.42) 3.49 (0.43) 3.51 (0.38) .36*** .34*** .44*** .50***
3. Exploration in breadth 3.26 (0.50) 3.33 (0.48) 3.35 (0.47) 3.37 (0.41) -.11* -.17*** -.18*** -.18***
4. Exploration in depth 3.60 (0.40) 3.63 (0.39) 3.68 (0.36) 3.70 (0.36) .01 -.02 .07 .04
5. Self-esteem 3.02 (0.57) 3.14 (0.53) 3.18 (0.54) 3.25 (0.48) -- -- -- --
Note.T = Time; M = mean; SD = standard deviation.
* p < .05. ** p < .01. *** p < .001.
43
Table 3
Descriptive Statistics and Pearson Correlations Between Identity Processes and Self-Esteem at Times 1, 2, and 3 in Study 3
Variable
M (SD)
T1
M (SD)
T2
M (SD)
T3
rs with self-esteem
T1 T2 T3
1. Commitment making 3.72 (0.84) 3.58 (0.85) 3.60 (0.84) .21*** .27*** .32***
2. Identification commitment 3.49 (0.72) 3.53 (0.73) 3.52 (0.77) .39*** .39*** .36***
3. Exploration in breadth 3.69 (0.67) 3.49 (0.80) 3.54 (0.83) .01 -.08 -.01
4. Exploration in depth 3.50 (0.67) 3.31 (0.80) 3.37 (0.80) .07 .01 .02
5. Ruminative exploration 2.79 (0.83) 2.83 (0.89) 2.81 (0.94) -.41*** -.49*** -.44***
6. Self-esteem 3.01 (0.58) 3.04 (0.56) 3.14 (0.58) -- -- --
Note.T = Time; M = mean; SD = standard deviation.
* p < .05. ** p < .01. *** p < .001.
44
Figure 1.
Final cross-lagged model linking self-esteem to identity processes in Study 1. Only significant
structural path coefficients are displayed. Within-time correlations, paths from gender and age, and
cross-lagged paths among the identity processes are not presented for reasons of clarity. All path
coefficients are standardized.
* p < .05. ** p < .01. *** p < .001.
Figure 2.
Final cross-lagged model linking self-esteem to identity processes in Study 2. Only significant
structural path coefficients are displayed. Within-time correlations, paths from gender and age, and
cross-lagged paths among the identity processes are not presented for reasons of clarity. All path
coefficients are standardized.
* p < .05. ** p < .01. *** p < .001.
Figure 3.
Final cross-lagged model linking self-esteem to identity processes in Study 3. Only significant
structural path coefficients are displayed. Within-time correlations, paths from gender and age, and
cross-lagged paths among the identity processes are not presented for reasons of clarity. All path
coefficients are standardized.
* p < .05. ** p < .01. *** p < .001.
45
Time 1 Time 2
Commitment
making
Commitment
making
Identification
commitment
Identification
commitment
Exploration
In breadth
Exploration
In breadth
Exploration
In depth
Exploration
In depth
Ruminative
exploration
Ruminative
exploration
Self-esteem Self-esteem
.47***
.30***
.32***
.34***
.40***
.66***
-.10**
.09*
46
Time 1 Time 2 Time 3 Time 4
Commitment
making
Commitment
making
Commitment
making
Commitment
making
Identification
commitment
Identification
commitment
Identification
commitment
Identification
commitment
Exploration
In breadth
Exploration
In breadth
Exploration
In breadth
Exploration
In breadth
Exploration
In depth
Exploration
In depth
Exploration
In depth
Exploration
In depth
Self-esteem Self-esteem Self-esteem Self-esteem
.62***
.50***
.73***
.74*** .72***
.49***
.05*
.09***
.70***
.56***
.09***
.77***
.57***
.06*
.05*
.72***
.57***
.77***
.06*
.77***
.53***
.05*
.10***
.05* .05* .05* .05*
47
Time 1
Time 2
Time 3
Daily assessments
Commitment
making
Commitment
making
Commitment
making
Identification
commitment
Identification
commitment
Identification
commitment
Exploration
In breadth
Exploration
In breadth
Exploration
In breadth
Exploration
In depth
Exploration
In depth
Exploration
In depth
Ruminative
exploration
Ruminative
exploration
Ruminative
exploration
Self-esteem Self-esteem M state
self-esteem
Self-esteem
SD state
self-esteem
.40*** .38***
.11*
.31***
.20***
.43***
.35***
.69***
.30***
.31***
.25***
.39***
.57*** .45***
.46***
.17***
-.22*** -.13***
.16***
-.12*
.14*
.13**
-.13**
... Au fil de leurs recherches, Luyckx et ses collaborateurs (Luyckx, Schwartz, et al., 2008) ajoutent un troisième processus d'exploration : l'exploration ruminative. Si l'exploration de surface et l'exploration en profondeur demeurent des processus adaptatifs et sont définis comme des processus d'exploration réflexive, l'exploration ruminative est plutôt à concevoir comme un processus maladaptatif qui retarde ou inhibe le développement de l'identité (Luyckx et al., 2013). Ce processus est problématique car il conduit l'individu à être bloqué dans des explorations avec une incapacité à former des engagements stables au cours du temps, ce qui entraine des sentiments d'incertitude et d'incompétence (Luyckx, Soenens, et al., 2008). ...
... En lien avec des préoccupations actuelles socio-éducatives et des questions de santé, les recherches en Sciences Humaines et Sociales fournissent une littérature abondante au sujet de l'ajustement psychologique des jeunes. Plus précisément, l'objectif des développementalistes est de déterminer quels sont les antécédents en termes de ressources, et les issues, conséquences ou apports en termes d'estime de soi ou de satisfaction de vie rapportés par les individus, en lien avec par exemple leur développement identitaire (Bosma & Kunnen, 2008;Klimstra et al., 2010;Luyckx et al., 2013) ou leurs activités en ligne (Burke & Kraut, 2016;Steinfield et al., 2008). En effet, plusieurs recherches ont mis en évidence qu'une recherche identitaire intense et prolongée durant les périodes d'adolescence et d'adulte en émergence, caractérisées par une certaine instabilité ou une incertitude dans les principaux domaines de la vie (relationnel, du travail, etc.) peut être accompagnée de symptômes anxieux ou dépressifs et d'une baisse d'estime de soi et du degré de satisfaction de vie (Crocetti, Rubini, Luyckx, et al., 2008a;Luyckx et al., 2011). ...
... En effet, plusieurs recherches ont mis en évidence qu'une recherche identitaire intense et prolongée durant les périodes d'adolescence et d'adulte en émergence, caractérisées par une certaine instabilité ou une incertitude dans les principaux domaines de la vie (relationnel, du travail, etc.) peut être accompagnée de symptômes anxieux ou dépressifs et d'une baisse d'estime de soi et du degré de satisfaction de vie (Crocetti, Rubini, Luyckx, et al., 2008a;Luyckx et al., 2011). A l'inverse, une prise d'engagements satisfaisants ou ayant des conséquences profitables peut avoir des effets bénéfiques en termes d'estime de soi (Luyckx et al., 2013) Selon Marcia (1993b), les processus de formation de l'identité par le questionnement, l'exploration et l'engagement sont au centre du changement du niveau d'estime de soi. De fait, les changements d'estime de soi sont en relation avec la construction de l'identité personnelle. ...
Thesis
Full-text available
De nos jours, les configurations socio-technologiques modernes conduisent les communautés scientifiques, médicales et éducatives à de nombreux débats concernant les usages numériques des jeunes. Les périodes d’adolescence et d’émergence de l’âge adulte correspondent toutes deux à une phase d’exploration identitaire, qui consiste à tester des rôles dans différents contextes et à les intégrer dans une identité personnelle cohérente. Des recherches traitant du bien-être des jeunes au travers de leurs usages numériques suggèrent qu’une utilisation forte peut mener les jeunes à des issues défavorables sur le plan socio-émotionnel tandis que d’autres indiquent que ces espaces peuvent permettre de développer des compétences psychosociales, de se sentir soutenus émotionnellement. Ce travail de recherche vise à proposer un cadre conceptuel et méthodologique qui permet d’étudier les relations entre les expériences en ligne des jeunes et leur développement psychosocial au travers de la construction de leur identité et de leur niveau d’ajustement psychologique, tout en prenant en compte les caractéristiques spécifiques des jeunes (ie., période d’âge et sexe). Notre population se compose d’adolescents scolarisés en lycée et de jeunes adultes étudiants de l’enseignement supérieur. Ce travail de recherche comportait 3 temps de mesure : un premier temps en Décembre 2018 (N=1970), un second temps en Mars 2019 (N=970), et un troisième temps en Mai/Juin 2019 (N=819). Les participants ont répondu à un ensemble de questionnaires évaluant les usages numériques, les processus de la construction identitaire et les indices d’ajustement psychologique. Les données ont été traitées selon une double approche, à la fois centrée sur les personnes et centrée sur les variables. En complément, des entretiens semi-directifs ont été réalisés quatre mois après le recueil de données quantitatives. Nos résultats soulignent une grande diversité chez les jeunes et ont permis d’identifier des profils contrastés : 6 profils d’usages numériques et 7 statuts identitaires. Des analyses en tri-croisés indiquent des liens cohérents et stables entre les profils d’usages numériques et les statuts identitaires. Concernant l’ajustement psychologique, nos résultats indiquent que les profils d’usages numériques caractérisés par une présentation de soi authentique en ligne, et les statuts identitaires caractérisés par des engagements forts, présentent les meilleurs niveaux d’ajustement psychologique. A l’inverse, les profils d’usages numériques caractérisés par une présentation de soi en ligne falsifiée, et les statuts identitaires associés à de faibles engagements s’accompagnent des niveaux d’ajustement psychologique les plus faibles. L’étude des liens croisés-décalés entre le processus identitaire mal-adaptatif d’exploration ruminative et les variables d’usages numériques, en considérant les variabilités intra et inter-individuelles au fil du temps (ie., RI-CLPM), révèle de nombreuses associations spécifiques selon la période d’âge et le sexe des jeunes. Les entretiens semi-directifs ont permis de préciser le vécu psychologique d’une adolescente et d’une jeune adulte, et soulignent quelques différences d’usages liées à leurs motivations et à leurs niveaux de développement. Ces résultats permettent de compléter la littérature et de développer un nouveau regard sur les usages numériques des jeunes français. Des usages numériques semblent être problématiques pour le développement psychosocial de certains jeunes, tandis que pour d’autres, ils peuvent être plus favorables en termes de construction de l’identité et d’ajustement psychologique. Cette étude souligne l’importance de caractériser les usages numériques en prenant en compte la période d’âge et le sexe des jeunes de façon à apprécier leur incidence sur le développement psychosocial. Enfin, cette étude conduit à dégager des perspectives de recherche et des recommandations appliquées en matière d’éducation et de santé.
... Meanwhile, exploration in depth and identification with commitment are seen as a commitment evaluation cycle, in which individuals reflect on and manage their commitments. The commitment formation cycle is significant at the earlier stage of identity development, whereas the commitment evaluation cycle is significant at the later stage (Luyckx et al., 2013b). ...
... Based on this sense of belonging, they may proactively engage in searching, revising, and implementing their commitments. In line with this assumption, a prior study indicated that selfesteem, which is closely linked to a positive sense of self, positively predicted commitment processes and 4 negatively predicted ruminative exploration (Luyckx et al., 2013b). Conversely, individuals who hold identities with high levels of negative identity elements tend to be hostile toward others/society Knutson, 1981). ...
... Also consistent with our hypothesis (Hypothesis 2b) were the study results revealing an insignificant prediction of negative identity elements on ruminative exploration at both the between-and withinperson levels. In contrast to individuals with low selfesteem who manifest high levels of ruminative exploration (Luyckx et al., 2013b), emerging adults with identities opposing sociocultural expectations might be relatively relieved from identity-related worries because of their fixation on undesirable roles. Additionally, inconsistent with our hypothesis (Hypothesis 2b), negative identity elements did not hinder proactive exploration. ...
Article
Full-text available
Developing identities that are well-aligned with sociocultural expectations is a key psychosocial developmental task for adolescents and emerging adults. Most studies have examined identity development focusing on how individuals develop their identities (identity process), such as identity exploration and commitment. Meanwhile, researchers have emphasized incorporating the what of identity development (identity content) with identity processes to further the understanding of identity development in sociocultural contexts. This study focuses on the positive and negative valences of identity defined by desirable and undesirable images shared in sociocultural contexts. We investigated the bidirectional associations of identity exploration and commitment processes with positive and negative identity elements using longitudinal data over three measurement waves. Participants were 2,313 Japanese emerging adults enrolled in higher education (70.95% women; Mage = 20.43). The cross-lagged panel analysis and random-intercept cross-lagged panel analysis were used to estimate associations at both between- and within-person levels. Results indicated that commitment making negatively predicted negative identity elements, whereas identification with commitment positively predicted positive identity elements. Meanwhile, positive identity elements positively predicted identification with commitment only for participants with low levels of negative identity elements, while negative identity elements negatively predicted commitment making and identification with commitment. These associations were found only at the between-person level. The findings highlight that emerging adults develop identities through close interactions in which they engage in identity exploration and commitment processes, as well as construct identity content valences. Developmental sequences of identity, along with their sociocultural contexts and practical implications, are discussed. (PsycInfo Database Record (c) 2021 APA, all rights reserved).
... As such, identity uncertainty represents an important risk factor for psychopathology (Schwartz, 2016;Schulenberg et al., 2014). Indeed, compared to adolescence, identity uncertainty in emerging adulthood becomes increasingly related to ruminative exploration and depressive symptoms (Luyckx et al., 2013). This may be particularly problematic, as the transition to adulthood has become more individually directed and less socially prescribed . ...
... Indeed, in our review, only a handful of studies have sought to explore the directionality of the association between identity and outcomes. Utilizing the dual-cycle model with a sample of Belgian university-attending emerging adults, Luyckx et al. (2013) found a bidirectional relationship between commitment making and identification with commitment and self-esteem across 4-years of data collection. ...
Chapter
Full-text available
The establishment of a coherent sense of self and identity represents a central developmental task of adolescence and emerging adulthood that has been increasingly conceptualized as a “steering mechanism” that guides one's life pathways and decisions. However, although identity development often involves navigating a period of identity uncertainty, prolonged and/or severe identity distress can interfere with normal adaptive functioning. Unsurprisingly, an extensive body of research has indicated that the ability to establish an integrated sense of self is associated with the greatest levels of well-being and the lowest levels of internalizing symptoms. This current chapter serves as a review of the link between identity and health among adolescent and emerging adult populations. Towards that end, we begin by providing an overview of identity theory and contextualize identity development within adolescence and young adulthood. Next, we provide a narrative review of research documenting the association between identity and adolescent health. Finally, we conclude by reviewing the research on identity-focused intervention programs and highlight important future directions.
... In addition to negative psychosocial adjustment, individual differences in dual-cycle identity processes predict positive adjustment as well. For instance, those individuals with strong and stable identity commitments report higher self-esteem (Luyckx et al., 2013) and higher levels of experienced meaning in life (Negru-Subtirica, Pop, Luyckx, Dezutter, & Steger, 2016). Also, adolescents with an ethnic identity profile characterized by high exploration, strong resolution (resembling commitments), and high centrality of their ethnicracial background within their self-concept reported higher academic engagement, life satisfaction, and self-esteem than adolescents with an ethnic identity profile characterized by low exploration and resolution (regardless of centrality; Wantchekon & Umaña-Taylor, 2021). ...
Article
Full-text available
One of the key developmental tasks in adolescence is to develop a coherent identity. The current review addresses progress in the field of identity research between the years 2010 and 2020. Synthesizing research on the development of identity, we show that identity development during adolescence and early adulthood is characterized by both systematic maturation and substantial stability. This review discusses the role of life events and transitions for identity and the role of micro‐processes and narrative processes as a potential mechanisms of personal identity development change. It provides an overview of the linkages between identity development and developmental outcomes, specifically paying attention to within‐person processes. It additionally discusses how identity development takes place in the context of close relationships.
... According to Marcia's (1966Marcia's ( , 1993 original identity status theory, identity commitments formed by engaging in effortful, self-exploration and self-reflection should optimize adaptability and personal well-being. More recent dimensional views of identity formation have found that psychological wellbeing and protection from ill-being depends primarily on one's identity commitments independent from the process by which they were formed (e.g., Luyckx et al., 2013;Meeus, 2018). In a three-wave longitudinal study with Romanian adolescents, Negru-Subtirica et al. (2017) found that the informational style at T1 & T2 predicted stronger educational commitments at T2 & T3, respectively. ...
Article
Full-text available
Identity processing styles (i.e., normative, informational, and diffuse-avoidant) refer to differences in how individuals engage or circumvent the challenge of processing self-relevant information and negotiating identity-relevant conflicts. The role identity commitment, personal agency, and self-regulation played in mediating relationships between identity styles and how successfully students were adapting to college was investigated. Measures of identity style, commitment, agency, self-regulation, and three indicators of adaptation (i.e., depressive symptoms, loneliness, and college adjustment) were completed by 402 college freshmen. A normative style was associated with an adaptive pattern; whereas the pattern for the diffuse-avoidant style was maladaptive. An informational style was only directly associated with college adjustment. All these relationships were mediated by personal agency and self-regulation. Commitment only uniquely mediated relationships with depressive symptoms. The findings suggest that identity commitment has a minimal to negligible impact on college adjustment independent from students’ sense of agency and regulatory resources. Implications of the role agency and self-regulation play in how freshmen students with different identity styles form identity commitments and adapt to college are considered.
Article
The present paper includes two studies examining how identity functioning is related to (psychological characteristics of) somatic symptoms. Study 1 examined associations between identity and somatic symptoms, whereas Study 2 additionally examined associations between identity and psychological characteristics of somatic symptoms and investigated the mediating role of anxiety and depressive symptoms in the association between identity disturbance and (psychological characteristics of) somatic symptoms. The total sample consisted of adolescents and emerging adults aged 14–30, including 686 and 663 participants for Study 1 and 2, respectively. In both studies, youth in troubled diffusion reported the most (psychological characteristics of) somatic symptoms. Finally, significant indirect effects from identity disturbance to (psychological characteristics of) somatic symptoms through anxiety and depressive symptoms were found.
Article
Identity development is a prominent task during adolescence, and the way adolescents develop their identity is an important factor in psychopathology. The present study aimed to identify different identity trajectory classes and investigated how these classes are related to psychopathological symptoms (i.e., depressive symptoms; eating disorder symptoms; somatic symptoms and related thoughts, feelings, and behaviors; and substance use). A total of 599 Flemish adolescents aged 12–18 at Time 1 (41.3% female; Mage = 14.93) participated at three annual measurement points. Five identity trajectory classes emerged using latent class growth analysis (achievement, foreclosure, moratorium, carefree diffusion-increasing exploration, troubled diffusion). In addition, multigroup latent growth curve modeling demonstrated the co-development of identity trajectory classes with psychopathology. Adolescents in classes reflecting maladaptive identity functioning, such as moratorium and troubled diffusion, displayed significantly more psychopathological symptoms. These findings indicate the importance of targeting identity functioning in the prevention and intervention of psychopathology among adolescents.
Article
Full-text available
Objective: The aims of this investigation were to explore (a) the experiences of women who stutter in university settings; (b) the impact of stuttering related to the emotions attached to stuttering; (c) the impact of stuttering on relationships for women; and (d) coping and management of stuttering within the university setting. Method: Interpretive phenomenological analysis was used to understand the lived experiences of women who stutter when attending university. In-depth semi-structured interviews were collected from seven women who stutter and subjected to inductive analysis. Results: Qualitative data analysis indicated that participants reported four themes related to their university experiences: 'Role of support'; 'Client-centered therapy'; 'Role of authoritative figures'; and 'The stuttering stereotype exists. ' Discussion: Women are presented with unique challenges while attending university , as they must combat stereotypes related to stuttering and being a woman, and they must also overcome the related obstacles. Conclusions: There is much work to be done in the university setting, in order to change how people who stutter are portrayed and how stuttering is perceived. These data point toward the need for advocacy training within the university setting, and for more holistic approaches to be utilized in stuttering intervention.
Article
Full-text available
This research aims to further our understanding of the processes of metaperception formation and meta-accuracy by introducing the positivity-specificity model to metaperception, which can be used to disentangle two components of trait metaperceptions: metapositivity (attitudes) and trait-specificity (substance). In two North American samples (Sample 1, N = 547; Sample 2, N = 553), we used the positivity-specificity model to investigate five important aspects of metaperceptions, namely the extent to which (a) metaperceptions reflect metapositivity versus trait-specificity, (b) metapositivity reflects attitudes about the self, (c) the effects of metapositivity and trait-specificity vary across traits and acquaintances, (d) metapositivity helps or hurts meta-accuracy, and (e) metapositivity and trait-specificity are accurate independent of self-perceptions. Overall, participants' ideas about how they were seen included attitudes and substance, but the relative contribution of each depended on the trait being judged and on how well they knew an acquaintance. Participants' ideas about how positively they were seen were related to how positively they saw themselves to varying degrees depending on how much they knew and liked their acquaintances. Participants were also accurate about how positively they were seen and about how they were seen on a given trait, independent of positivity and, with close acquaintances, independent of self-perceptions. The current work demonstrates how the positivity-specificity model can be used to investigate how people think about and have insight into the impressions they make on others. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
Thesis
The purpose of this study was to examine how high-performing secondary school students perceive their school music ensemble participation in relationship to their social identity. Research questions included the following: (1) How do participants rate their primary large ensemble membership in relationship to their self-concept?; (1a) How do selected variables: type of ensemble (i.e. band, orchestra, choir), age, time dedicated per week, and leadership positions, collectively and individually predict the importance of participants’ primary large ensemble membership to their self-concept? (2) How do participants rate their personal judgments of how valuable their primary large ensemble membership is compared to their perception of how others view their ensemble membership? (3) How do participants’ scores on the Social and Personal Identities Scale compare with previous research findings involving individuals engaged in the arts? To address the research questions, adolescent band, orchestra, and choir musicians (N = 126, 86.3% response rate) participating in a summer performing arts camp completed a paper and pencil survey about their high school music ensemble experiences. The survey included general and music demographic questions as well as a modified version of the Collective Self-Esteem Scale (CSE) and the Social and Personal Identities Scale (SIPI) as a means of measuring social identity and the salience of their group memberships. In general, participants self-identified as active members of their high school music program, with 66.6% holding some level of leadership position, and participants reported devoting an average of eight hours per week to their primary ensemble. Most respondents reported taking private music lessons, participating in additional music ensembles, and holding memberships in other non-music groups at the same time. Results of this study include: (a) participants who reported holding a major, or significant, leadership position indicated that their primary large ensemble membership had a greater importance to their self-concept, (b) respondents’ perceptions of how others evaluate their large ensemble was strongly related to their personal judgments of how favorable their large ensemble was, and (c) participants in this study indicated a lower desire for uniqueness and independence within their social groups and, consequently, were more likely to emphasize conformity in their social groups when compared to previous research findings. Implications for music education practice include recommendations that music teachers: (a) aim to situate their ensemble in a socially favorable position, (b) strive to maximize leadership opportunities without diluting the value of these positions, and (c) consider how to create a greater sense of unity and inclusiveness within their ensembles. Additionally, pre-service music teachers would likely benefit from a deeper understanding of theories related to social identity and how they can be applied to their future environments. Suggestions for future research and a possible extension of theories related to identity and music participation are discussed.
Article
Full-text available
Two studies examined the rank-order stability of self-esteem from age 6 to 83: Study 1 was a meta-analysis of 50 published articles (N = 29,839) and Study 2 analyzed data from 4 large national studies (N = 74,381). Self-esteem showed substantial continuity over time (disattenuated correlations ranged from the .50s to .70s), comparable to the stability found for personality traits. Both studies provided evidence for a robust developmental trend: Self-esteem stability was low during childhood, increased throughout adolescence and young adulthood, and declined during midlife and old age. This trend could not be explained by age differences in the reliability of self-esteem measures, and generally replicated across gender, ethnicity, self-esteem scale, nationality (U.S. vs. non-U.S.), and year of publication.
Article
Full-text available
Recent scholars have dismissed the utility of self-esteem as well as programs designed to improve it. The authors challenge these contentions on conceptual, methodological, and empirical grounds. They begin by proposing that the scope of recent analyses has been overly narrow and should be broadened to include specific as well as global self-views. Using this conceptualization, the authors place recent critiques in historical context, recalling that similarly skeptical commentaries on global attitudes and traits inspired theorizing and empirical research that subsequently restored faith in the value of both constructs. Specifically, they point to 3 strategies for attaining more optimistic assessments of the predictive validity of selfviews: recognizing the utility of incorporating additional variables in predictive schemes, matching the specificity of predictors and criteria, and using theoretically informed standards for evaluating predictor– criterion relationships. The authors conclude that self-views do matter and that it is worthwhile and important to develop and implement theoretically informed programs to improve them.
Article
This study examined the development of individuals whose motivations and skills led them to develop in different but equally positive ways. C. D. Ryff's (1989) scales for Environmental Mastery (EM) and Personal Growth (PG) were used to identify three configurations of positive mental health in 111 women of the Mills Longitudinal Study: Achievers, high on both scales; Conservers, high on EM, low on PG; and Seekers, high on PG, low on EM. Each pattern showed a distinctive profile of strengths on four criteria of maturity - competence, generativity, ego development, and wisdom - and each was predicted by distinctive features of positive and negative emotionality, identity processes, and change in self-control across 31 years of adulthood. Identity at age 43 mediated the influence of personality at age 21 in predicting positive mental health pattern at age 60.
Article
A common concern when faced with multivariate data with missing values is whether the missing data are missing completely at random (MCAR); that is, whether missingness depends on the variables in the data set. One way of assessing this is to compare the means of recorded values of each variable between groups defined by whether other variables in the data set are missing or not. Although informative, this procedure yields potentially many correlated statistics for testing MCAR, resulting in multiple-comparison problems. This article proposes a single global test statistic for MCAR that uses all of the available data. The asymptotic null distribution is given, and the small-sample null distribution is derived for multivariate normal data with a monotone pattern of missing data. The test reduces to a standard t test when the data are bivariate with missing data confined to a single variable. A limited simulation study of empirical sizes for the test applied to normal and nonnormal data suggests that the test is conservative for small samples.
Article
Theories often place constraints on causal relationships, and such constraints are often assessed with causal models. Causal models should be recursive and just identified because cause is recursive and is more likely to be just identified than overidentified. A just-identified, recursive model (JIRM) is specified that satisfies both requirements and that can be used to assess a wide range of causal implications in either a norm-referenced or criterion-referenced manner. P. E. Meehl and N. G. Waller (2002) proposed an innovative method for theory appraisal called the delete one-add one (D1 -A1) method, which assesses a relatively narrow range of causal implications, allows nonrecursive models, and is only norm referenced. The JIRM and D1-A1 methods are compared.
Conference Paper
All individuals have multiple views of themselves. Whereas the consistency among the different aspects of identity is emphasized in Western cultures, the "multiple selves" are often viewed as coexisting realities in East Asian cultures. This research revisits the classic thesis in psychology that identity consistency is a prerequisite condition of psychological well-being. Between individuals (Study 1), people with a more consistent self-view had a more clear self-knowledge, were more assertive, and, most notably, had self-experiences that were less affected by the perspectives of others. Compared with North American participants. (Study 2), Koreans viewed themselves more flexibly across situations, and their subjective well-being was less predictable from levels of identity consistency. Also, consistent individuals received positive social evaluations from others in the United States but not in Korea.