Advertisement

Journal of Youth and Adolescence

, Volume 47, Issue 4, pp 717–730 | Cite as

Longitudinal Links Between Identity Consolidation and Psychosocial Problems in Adolescence: Using Bi-Factor Latent Change and Cross-Lagged Effect Models

  • Kai Hatano
  • Kazumi Sugimura
  • Seth J. Schwartz
Empirical Research

Abstract

Most previous identity research has focused on relationships between identity synthesis, confusion, and psychosocial problems. However, these studies did not take into account Erikson’s notion of identity consolidation, that is, the dynamic interplay between identity synthesis and confusion. This study aimed to examine longitudinal relationships and the directionality of the effects between identity consolidation and psychosocial problems during adolescence, using two waves of longitudinal data from 793 Japanese adolescents (49.7% girls; ages 13–14 and 16–17 at Time 1). A bi-factor latent change model revealed that levels and changes in identity consolidation were negatively associated with levels and changes in psychosocial problems. Furthermore, a bi-factor cross-lagged effects model provided evidence that identity consolidation negatively predicted psychosocial problems, and vice versa. Our study facilitates a better understanding of the importance of identity consolidation in the relations between identity components and psychosocial problems.

Keywords

Identity consolidation Identity development Adolescence Psychosocial problems 

Introduction

Adolescence is a critical time when many changes occur involving biological (e.g., brain development, secondary sex characteristics), cognitive (e.g., acquisition of formal abstract reasoning), and social (e.g., restructuring of parent-adolescent relationships) processes (Kroger 2004). These changes often introduce instability into adolescents’ lives and may increase risks for psychosocial problems (e.g., Ames et al. 2015; Castelao and Kröner-Herwig 2014; Moffitt 1993). Internalizing symptoms (e.g., anxiety and depressive symptoms) refer to the individual’s internal psychological environment (Achenbach et al. 2002). Externalizing symptoms refer to acting out (e.g., aggression, delinquent behaviors, and health risk behaviors) that may violate other people’s rights (Achenbach et al. 2002). These psychosocial problems (i.e., internalizing and externalizing symptoms) are rooted in several developmental predictors, one of which is a confused or disorganized sense of identity.

Erikson (1968) proposed that adolescence is the time of life when identity comes into ascendance, and that identity represents a fundamental developmental task during adolescence and the transition to adulthood. Erikson (1950) conceptualized identity in complex ways based on his clinical observations. With regard to the person’s internal experience, Erikson’s ideas about identity include (at least) three types of components, that is, synthesis, confusion, and consolidation (Schwartz 2007). Synthesis represents a coherent and consistent sense of self over time and across situations, whereas confusion represents a fragmented, changeable, and haphazard sense of self. Identity consolidation represents the overall concept based on the dynamic interplay between identity synthesis and confusion, where both synthesis and confusion are necessary for health development. Thus far, researchers (e.g., Schwartz et al. 2015b; Schwartz Zamboanga et al. 2013) have demonstrated that, during adolescence and young adulthood, identity synthesis and identity consolidation are often accompanied by low levels of internalizing and externalizing symptoms (e.g., depression, anxiety, delinquent behaviors, and drug use), whereas identity confusion is associated with the reverse pattern (e.g., Schwartz et al. 2009a, b, c, 2015a). However, much of this research has been cross-sectional, and there is a lack of longitudinal studies examining the relationships between identity consolidation and psychosocial problems. Further, much of the extant literature involves studies conducted in North America and Western Europe, with little work conducted in other parts of the world. The present study was intended to expand the Eriksonian identity literature through a longitudinal study of Japanese youth.

Identity has taken on an even greater degree of importance given the extension of Erikson’s “psychosocial moratorium” into the late teens and 20s (Arnett 2000, 2007). Furthermore, with the increasing complexity of social systems, the impact of identity synthesis on psychosocial problems may have changed compared with previous periods. Late modern societies are characterized by extended periods of education and deferral of marriage and parenthood (Côté and Levine 2016; Furlong and Cartmel 2009). The period of uncertainty associated with identity development has therefore been lengthened. For some writers, these social changes have raised the question as to whether an integrated identity (or identity synthesis) serves adaptive functions for contemporary young people (e.g., Rattansi and Phoenix 2005).

Japan, the country where the current study was conducted, exemplifies these social changes that have occurred. In Japan, the modernization of education advanced after World War II. In the 1940s, only about 10% of high school students went on to college. By the 1970s, after the end of high economic growth, the number had increased to approximately 30%. In 2016, 54.7 and 16.4% of high school graduates went on to college (or junior college) and vocational school, respectively (Statistics Bureau of Japan 2016). Furthermore, in line with the movement of women into society, the timing of marriage became later. In the 1940s, the average age of women getting married was about 23 years. It increased to 25 years in the 1970s, then 29.4 years in 2014 (Ministry of Health, Labor and Welfare 2014). In addition, 70% of Japanese unmarried persons aged 20–24 live with their parents (Statistics Bureau of Japan 2017). These data suggest the presence of an extended transition to adulthood in Japan and imply that Japan is an appropriate place to examine whether identity components (i.e., identity consolidation, synthesis, and confusion) may protect against or exacerbate adolescent psychosocial problems in a context characterized by an extended transition to adulthood. Through examining the longitudinal relationships and the directionality of effect between identity components and psychosocial problems in Japan, we attempt to highlight the dynamic relations among identity components and psychosocial problems in a non-Western late modern society.

Identity Synthesis, Confusion, and Psychosocial Problems

Erikson’s theory serves as the foundation for the majority of developmentally based identity research (Kroger et al. 2011), and developmental identity research can be divided into two general approaches—neo—Eriksonian approaches that focus on identity processes and those Eriksonian approaches that focus on identity coherence (Schwartz 2007). The identity status approach (Marcia 1966; Kroger et al. 2011) focuses on the processes involved in the formulation of identity. Identity status theory is based on the assumedly independent processes of exploration and commitment, where exploration represents sorting through prospective alternatives and commitment represents adherence to one or more of the alternatives considered.

In contrast, other approaches emphasize the importance of coherence (for a review, see Syed and McLean 2016). Erikson’s notions of synthesis and confusion speak directly to the presence (synthesis) or absence (confusion) of such a sense of identity coherence. Within the identity status model, exploration sometimes leads to firm and adaptive commitments, and sometimes it does not (Luyckx et al. 2008). Moreover, it is possible for commitments to be enacted but not to reflect one’s true sense of self (Waterman et al. 2013). A focus on identity synthesis and confusion may address these issues by focusing on the outcomes of identity development—that is, the extent to which one’s identity is synthesized and the extent to which it is confused.

From a clinical viewpoint, Erikson (1950, 1968) stressed that identity synthesis and confusion play important roles in either potentiating or protecting against internalizing and externalizing symptoms. Previous cross-sectional studies have indicated that identity synthesis was negatively associated with internalizing symptoms (e.g., depressive symptoms and non-suicidal self-injury) (Azmitia et al. 2013; Claes et al. 2014). On the other hand, identity confusion was positively related to internalizing and externalizing symptoms (e.g., delinquent behavior, cigarette smoking, alcohol use, early sex initiation, and unprotected sex) (Donovan et al. 2013; Ferrer-Wreder et al. 2008; Schwartz et al. 2005, 2008). Furthermore, a few longitudinal studies, conducted in North America, have provided evidence for directional relationships between identity and internalizing and externalizing symptoms. Schwartz and colleagues found that trajectories of identity confusion predicted externalizing problems over 3 years among samples of Hispanic immigrant adolescents in the United States (Schwartz et al. 2008, 2017). These studies suggest that synthesis and confusion (especially confusion) play important roles predicting problematic adolescent outcomes.

Appling the Bi-Factor Model to Identity Consolidation

The studies reviewed here did not, however, take into account Erikson’s notion of identity consolidation. For Erikson (1950, 1968), identity is constructed as an interplay between identity synthesis and confusion, and healthy identity development involves a preponderance of synthesis over confusion. Methodologically speaking, from such a perspective identity is understood as a “hybrid,” where identity represents both one’s overall sense of self and a sense of balancing identity synthesis with identity confusion. Such a hybrid model may be the most closely consistent with Erikson’s work, where the “overall” identity dimension represents the overall “valence” of the identity that one holds.

Recent statistical advances have allowed us to capture the hybrid conceptualization of identity using a bi-factor model (Chen et al. 2012; see Fig. 1). Bi-factor modeling is useful when researchers are examining various dimensions of a single underlying construct, and where both the dimensions and the overall construct are theorized as predicting outcomes. Bi-factor models have been used in recent research to study the structure of self-esteem, personality traits, and the association between identity and other important outcomes (e.g., Chen et al. 2006; Mu et al. 2016). Specifically, Chen and colleagues showed that bi-factor models have potential advantages over second-order latent variable models when researchers are interested in predicting external criteria (Chen et al. 2006). Bi-factor models allow us to model Erikson’s view of identity, such that the hybrid/underlying identity dimension (i.e., identity consolidation) represents the overall valence of the identity that one holds, and where synthesis and confusion serve as additional dimensions of identity.
Fig. 1

Erikson Psychosocial Stage Inventory bi-factor model

Indeed, Schwartz and colleagues used the bi-factor model to examine the factorial structure of identity consolidation (Schwartz et al. 2009b). To use bi-factor modeling to represent identity consolidation, a measure with separate synthesis and confusion subscales is needed. The Erikson Psychosocial Stage Inventory (EPSI; Rosenthal et al. 1981) represents an example of such a scale, having six items representing synthesis and six items representing confusion. Previous factor-analytic studies have indicated that the fit of a bi-factor model was superior to that of one-factor or two-factor models among Japanese adolescents (Dimitrova et al. in press) and among young adults in the United States (Schwartz et al. 2009b). These findings suggest that using a bi-factor model might be the most optimal way to represent three identity components, both in the United States and in Japan.

Given identity synthesis and confusion coexist to some extent within a person (Marcia 2002), it would be of utmost of importance to focus on an overall profile of identity consolidation for understanding the link between identity and healthy psychosocial development, as well as consider the separate dimensions of identity synthesis and identity confusion (Schwartz et al. 2009b, 2015a). Indeed, previous studies—though these studies did not use the concept of identity consolidation—have suggested that, although a high degree of coherence in one’s sense of identity was adaptive, some degree of both synthesis and confusion would be required for healthy identity development in adolescence (cf. Crocetti et al. 2008; Luyckx et al. 2006), supporting the importance of identity consolidation in optimal psychosocial development. Furthermore, Schwartz et al. (2009b) found that the associations between identity consolidation (i.e., the “overall” identity construct in the bi-factor model) and variables measuring internalizing symptoms (i.e., anxiety and depression) were stronger and more consistent compared to the relationships of identity synthesis and confusion with indices of internalizing symptoms. This result suggests that identity consolidation contributes to psychosocial outcomes above and beyond the contributions of identity synthesis and confusion.

However, longitudinal studies examining the relationships between identity consolidation (operationalized in a bi-factor model) and psychosocial problems during adolescence are lacking. First, available cross-sectional research on identity consolidation have presented the relationships between identity consolidation and psychosocial problems at a particular point in time (Schwartz et al. 2009a, b, c); therefore, it remains unclear whether changes in identity consolidation are related to changes in psychosocial problems over time. To infer if this is the case, the present study employed a longitudinal design as the first attempt to unveil the associations between developmental changes in identity consolidation and psychosocial problems, using a latent change model. Second, there have been no previous attempts to investigate whether identity consolidation predicts psychosocial problems, or vice versa. Taking advantage of a longitudinal design, we examined the directionality of the effects in the relationship between identity consolidation and psychosocial problems, using a cross-lag model. In sum, to demonstrate the change in the relationships (rather than just associations) and the predictive relationships (the directionality of effects), it is essential to conduct longitudinal studies where earlier levels of the dependent variables can be controlled. These steps (i.e., using both latent change model and cross-lag model) have yet to be taken with regard to the links between identity consolidation and psychosocial problems.

Current Study

In the present study, we sought to examine the longitudinal and predictive associations between identity consolidation and psychosocial problems during adolescence. To achieve this goal, we first tested the longitudinal invariance of the bi-factor model of identity using confirmatory factor analysis (CFA). Second, we examined the links of the changes in three identity components (i.e., identity consolidation, identity synthesis, and identity confusion) with the changes in psychosocial problems. Previous cross-sectional studies using bi-factor models (Schwartz et al. 2009b) and longitudinal studies using other operationalizations of identity (e.g., Azmitia et al. 2013; Donovan et al. 2013; Schwartz et al. 2008) have found that various identity components related to psychosocial problems in different ways; identity consolidation and synthesis were negatively related to internalizing symptoms, whereas identity confusion was positively associated with both internalizing and externalizing symptoms. Further, although the extent to which identity synthesis predominates over identity confusion (i.e., identity consolidation) is theoretically assumed to be important for positive psychosocial functioning (Erikson 1950), both synthesis and confusion may still exert separate effects on outcomes beyond the effects of consolidation (Schwartz et al. 2009b). In line with the results of these prior studies, we hypothesized that changes in identity consolidation and synthesis would be negatively associated with changes in psychosocial problems over time, and the change in confusion would be positively linked to them over time. Specifically, we hypothesized that the relationships of identity consolidation with psychosocial problems would be stronger and more consistent compared to the relationships of identity synthesis and confusion with psychosocial problems. Lastly, we tested the directionality of effects among the three identity components (i.e., identity consolidation, identity synthesis, and identity confusion) and psychosocial problems. It is commonly assumed that identity variables predict psychosocial outcomes, although this assumption is rarely tested empirically. In one of the few empirical tests, Schwartz et al. (2012) found that identity commitments and self-concept clarity negatively predicted anxiety and depressive symptoms, more than vice versa, over time. We therefore hypothesized that identity synthesis, confusion, and consolidation would predict psychosocial problems (more than vice versa) over time. Specifically, we hypothesized that identity consolidation would more strongly predict psychosocial problems compared to the relationships of identity synthesis and confusion with psychosocial problems.

Method

Participants and Procedures

Data for the present study were drawn from a Japanese longitudinal study on identity (Hatano and Sugimura in press; Hatano et al. 2016, 2017). In this longitudinal study, data were collected using an online research company (MACROMILL: http://www.macromill.com/), which has a variety of clients (i.e., registrants) around the world. When a survey begins, the company sends an e-mail to the registrants that match the researchers’ inclusion criteria. Next, registrants can choose whether to participate in the survey. The researchers’ inclusion criteria for the present study were that participants were (1) of Japanese nationality, (2) recruited by sampling throughout Japan, and (3) contributed to our goal of reaching approximately 600 participants in each of the two age groups (i.e., 13 and 16 years old at Time 1). Because adolescents cannot register with the research company themselves, we targeted parents with adolescent children aged 13 and 16. Seventy-two percent of the participants lived in urban areas in the Kanto, Chubu, and Kansai districts (i.e., eastern, central, and middle-western parts of Japan, respectively). Prior to participation, parents and adolescents received an email with information about the purpose of the research, and if they were willing to participate, they signed informed consent/assent agreements. The emails were sent to 39,993 parents having 13- and 16-year-old adolescents, and 1233 adolescents participated in the initial survey (wave 1) conducted in March 2013 (the participation rate was 3.1%). After providing consent, parents and adolescents received an email containing a hyperlink to the web-based survey. Participants received reward points corresponding to 50 JPY (approximately US$0.50) for completing the questionnaire. Because participants were required to answer all items, there was no item-level missing data in this survey. In the present study, we used the data collected in November 2013 (wave 2) and in March 2015 (wave 4), which are called Time 1 and Time 2, respectively, in the present article. Seven hundred ninety-three Japanese adolescents (49.7% girls) participated in the Time 1 survey and 663 adolescents (50.5% girls) participated in the Time 2 survey. At baseline, 398 participants were early adolescents aged 13‒14 years (53.7% girls), and 395 were middle adolescents aged 16‒17 years (45.8% girls). At Time 2, 324 participants were early adolescents aged 15 years (53.4% girls), and 339 were middle adolescents aged 18 years (47.8% girls). Because the attrition rate (approximately 23% in the present study) may be related to Time 1 levels of synthesis, confusion, and psychosocial problems, we conducted a t-test to examine whether there was differential attrition. There were no significant differences between individuals who did and did not complete Time 2 in terms of synthesis (t(791) = −1.57, p = .89, d = 0.13), confusion (t(791) = 0.68, p = .47, d = 0.06), and psychosocial problems (t(791) = −0.33, p = .16, d = 0.03). All analyses were conducted using Amos 18.0 (Arbuckle 2009).

Measures

Identity

We assessed identity consolidation, synthesis, and confusion using the Erikson Psychosocial Stage Inventory (EPSI; Rosenthal et al. 1981; for the Japanese version, see Hatano et al. 2014). This measure consists of 12 items rated on a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). Six items represent identity synthesis, and 6 items represent identity confusion. Sample items include “I have got a clear idea of what I want to be” (synthesis) and “I change my opinion of myself a lot” (confusion). Regarding identity consolidation, we assessed it as an overall balance of identity synthesis and confusion by using the bi-factor model (Fig. 1). Values of Cronbach’s α for synthesis and confusion ranged from .78–.83 at Times 1 and 2, respectively.

Psychosocial problems

We assessed internalizing and externalizing symptoms using the Strengths and Difficulties Questionnaire (SDQ; Goodman 1997; for the Japanese version, see Sugawara et al. 2006). This measure consists of 5 subscales (5 items each), rated on a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). In the present study, we used three of the 5 SDQ subscales, which assess (a) emotional symptoms (i.e., mood of depression and anxiety), (b) peer problems (i.e., feeling of loneliness based on peer trouble) as internalizing symptoms, and (c) conduct problems (i.e., antisocial behaviors) as externalizing symptoms. The total score of the three subscales was used as the overall level of psychosocial problems1 Sample items include “I worry a lot” (emotional symptoms), “I often think it would be better to try to find a different best friend” (peer problems), and “I can make other people do what I want” (conduct problems). Values of Cronbach’s α for the SDQ total scale were .87 and .91 at Times 1 and 2, respectively.

Statistical Analysis

To examine the association between identity components and psychosocial problems, the analyses proceeded in three primary steps. First, we conducted the longitudinal invariance analyses to confirm that the structure of bi-factor model was equivalent across time. In the bi-factor model, each identity item is linked separately with its primary subscale (synthesis or confusion) and with the overall identity factor. The overall factor was hypothesized to extract the common variance among the specific factors; hence, the overall identity factor was not allowed to correlate with either synthesis or confusion and the two specific factors were set to be uncorrelated to each other. No residual correlations were allowed. Following Dimitrov (2010), we used the standard longitudinal invariance testing procedure. We tested three types of measurement invariance: (a) configural invariance (the same number of factors and pattern of fixed and freely estimated parameters across groups), (b) metric invariance (equivalence of factor loadings, indicating that respondents from each group attribute the same meaning to the latent construct), and (c) scalar invariance (invariance of both factor loadings and item intercepts, indicating that the meaning of the construct and the scaling of the latent factor are equal across groups). Configural invariance was tested by evaluating the fit of the unconstrained model (with both time points included) to the data. Metric and scalar invariance were evaluated by comparing constrained (i.e., factor loadings set equal across time for metric invariance, and both loadings and intercepts set equal across time for scalar invariance) vs. unconstrained models. For model comparisons, the χ 2 index should be as small as possible, the comparative fit index (CFI) should exceed .95, and the root mean square error of approximation (RMSEA) should be less than .05, with values less than .08 representing reasonable fit (Kline 2010). To test whether the fit of the bi-factor model was equivalent across time, we relied on differences in CFI, RMSEA, and Akaike Information Criteria (AIC) between constrained and unconstrained models. If the differences in model fit indices exceeded the following criteria, the null hypothesis of invariance would be rejected: ΔCFI > .010 and ΔRMSEA > .015 (Kline 2010). Furthermore, the AIC was used to evaluate model fit. A difference in AICs of 20 or greater indicates an absence of invariance (Burnham and Anderson 2002).

Second, we constructed a bi-factor latent change model to examine the change in the identity components and change in the psychosocial problems over time (e.g., Mu et al. 2016; see Fig. 2). Latent change models (McArdle and Nesselroade 1994) facilitate modeling complex patterns of development and change in multiple developmental processes simultaneously (Jackson and Allemand 2014), and the level of a latent construct and the change of this latent construct over time are estimated and all latent initial and change factors are allowed to covary (e.g., Takahashi et al. 2013). In this model, we examined the associations among identity components and psychosocial problems by dividing the intercepts (i.e., levels) and slopes (i.e., changes). Corresponding error terms (i.e., the same item at Times 1 and 2) were allowed to correlate. Lastly, we constructed a bi-factor cross-lagged effect model to examine the directionality of effects between identity components and psychosocial problems (see Fig. 3).
Fig. 2

Format of the bi-factor latent change model of identity components and psychosocial problems. Note: (1) The error terms associated with each item were set in the bi-factor model and these terms associated with each item were set to be correlated between Time 1 and Time 2. (2) Em emotional symptoms, Con conduct problem behaviors, Pe peer problems

Fig. 3

Bi-factor cross-lagged effect model. Note: (1) The error terms associated with each item were set in all models and these terms associated with each item were set to be correlated between Time 1 and Time 2. (2) Em emotional symptoms, Con conduct problem behaviors, Pe Peer problems

Results

Longitudinal Invariance Testing

To test the stability of the factor structure of the EPSI, we examined measurement invariance between Time 1 and Time 2 using confirmatory factor analyses. The configural invariance model fit the data adequately, χ 2(84) = 361.268, p < .001, CFI = .949, RMSEA = .043, AIC = 553.268. Furthermore, there was evidence of metric invariance, χ 2(105) = 410.288, p < .001, CFI = .943, RMSEA = .041, AIC = 560.288, ΔCFI = .006, ΔRMSEA = .002, ΔAIC < 20, and scholar invariance, χ 2(117) = 440.516, p < .001, CFI = .940, RMSEA = .040, AIC = 566.516, ΔCFI = .009, ΔRMSEA = .003, ΔAIC < 20. These results provide strong evidence for the EPSI and three types of measurement invariance over time.

Descriptive Statistics

Table 1 presents means and standard deviations for scale scores, as well as correlations among identity synthesis, identity confusion, and psychosocial problems at both time points. At both Time 1 and Time 2, identity synthesis was negatively associated, and identity confusion was positively associated, with psychosocial problems.
Table 1

Means, standard deviations, and correlations among variables

Variable

1

2

3

4

5

6

Mean

SD

1. Time 1 Identity synthesis

1

−.13***

−.27***

.37***

−.24***

−.23***

3.08

.60

2. Time 1 Identity confusion

 

1

.56***

−.20***

.48***

.45***

2.79

.60

3. Time 1 Psychosocial problems

  

1

−.20***

.45***

.63***

2.59

.50

4. Time 2 Identity synthesis

   

1

−.28***

−.28***

3.09

.52

5. Time 2 Identity confusion

    

1

.65***

2.77

.58

6. Time 2 Psychosocial problems

     

1

2.56

.53

***p< .001

Bi-Factor Latent Change Model

To test the change in the identity components and change in the psychosocial problems, we constructed a bi-factor latent change model (e.g., Mu et al. 2016; see Fig. 2). By using this model, we could test the relations among levels, which would correspond to average overall levels of both the overall factor (i.e., identity consolidation) and the specific factors (i.e., identity synthesis and identity confusion), as well as the relations among changes in overall and specific factors over time. Fit indices suggested that the model fit the data adequately, χ 2(350) = 1004.293, p< .001, CFI = .926, RMSEA = .046 (90% CI = .043–.049). The correlation estimates obtained through this analysis are presented in Table 2. Level in identity consolidation was negatively associated, and level in identity confusion was positively associated, with levels in psychosocial problems. Furthermore, change in identity consolidation was negatively related to change in psychosocial problems, while change in identity confusion was positively related to change in psychosocial problems. There were no significant relationships between the level and change in synthesis and the level and change in psychosocial problems.
Table 2

Correlate estimates between level and change in identity components and psychosocial problems

Variables

 

Psychosocial problems

  

Level

Change

  

Standardized coefficient

Unstandardized coefficient

SE

CR

Standardized coefficient

Unstandardized coefficient

SE

CR

Identity consolodation

Level

−.55***

−.14

.02

−.83

.11

.02

.01

1.62

 

Change

.17*

.04

.02

2.14

−.28***

−.05

.01

−3.47

Identity synthesis

Level

.01

.00

.12

.25

.00

.00

.01

.00

 

Change

.01

.00

.01

.25

−.03

−.01

.01

−.57

Identity confusion

Level

.56***

.12

.01

8.60

−.22***

−.04

.01

−3.79

 

Change

−.28***

−.06

.01

−4.02

.46***

.08

.01

6.63

SE Standard error of the estimate, CR Critical Ratio

*p< .05; **p< .01; ***p< .001

Bi-Factor Cross-Lagged Effect Model

To examine the directionality of effects between identity and psychosocial problems, we estimated a bi-factor cross-lagged model (see Fig. 3). By using this model, we could test the directionality of effects between identity consolidation and psychosocial problems. This model provided an acceptable fit to the data, χ 2(350) = 967.351, p< .001, CFI = .930, RMSEA = .045 (90% CI = .041–.048). The stability paths (i.e., auto-regressive effects), concurrent relationships, and cross-lagged effects are indicated in Table 3. As for the concurrent relationships, identity consolidation was negatively related to psychosocial problems, whereas identity confusion was positively related to psychosocial problems, both for Time 1 and Time 2. With respect to the cross-lagged effects, identity consolidation negatively, and identity confusion positively, predicted psychosocial problems. On the other hand, psychosocial problems negatively predicted both identity consolidation and synthesis, and positively predicted identity confusion, both for Time 1 and Time 22 We concluded that the effects of identity components vis-à-vis psychosocial problems were likely bidirectional, and this was particularly evident in identity consolidation as well as identity confusion.
Table 3

Auto-regressive effects, concurrent relationships, and cross-lagged effects

Variable

Auto-regressive effects

Concurent relationships

Cross-lagged effects

     

Time 1

Time 2

Identity components→psychosocial problems

Psychosocial problems→identity components

 

β

b

SE

CR

β

b

SE

CR

β

b

SE

CR

β

b

SE

CR

β

b

SE

CR

Identity consolidation

.27**

.19

.07

2.62

−.54***

−.14

.02

−.7.76

−.34***

−.04

.01

−3.99

−.16*

−.17

.07

−2.32

−.37***

.50

.09

5.71

Identity synthesis

.35***

.25

.04

6.69

−.03

−.01

.01

−.79

−.08

−.01

.01

−1.56

.00

.00

.03

−.10

−.19**

−.17

.06

−2.85

Identity confusion

.17**

.17

.06

2.81

.60***

.12

.01

8.89

.61***

.07

.01

7.30

.14*

.18

.09

2.07

.42***

.29

.06

4.59

β Standardized coefficient, b Unstandardized coefficient, SE Standard error of the estimate, CR Critical Ratio.

*p < .05; **p < .01; ***p < .001

Discussion

Identity consolidation is a key concept to understand healthy identity development during adolescence (Erikson 1968). However, there is the question of whether and how identity consolidation serves adaptive psychosocial functioning for contemporary young people. With the increasing complexity of late modern societies, the impact of identity on psychosocial problems may have changed compared with previous periods. Furthermore, there is a lack of longitudinal studies examining the relationship between identity consolidation and psychosocial problems; thus, it is not yet known whether change in identity consolidation is associated with psychosocial problems, and which comes first in the period of adolescence. In this study, we aimed to address these fundamental questions arising from theoretical and empirical studies concerning the links between identity consolidation and psychosocial problems during adolescence among a Japanese sample. To accomplish this purpose, we first tested the longitudinal invariance of a bi-factor module constructed from the Erikson Psychosocial Stage Inventory (EPSI). The results of CFAs indicated an acceptable fit for, and equivalence of, the bi-factor solutions across Time 1 and Time 2. Next, we tested the longitudinal relations between identity components and psychosocial problems using a bi-factor latent change model. The results obtained from this model indicated that changes in identity consolidation and in confusion were related to changes in psychosocial problems. Lastly, we examined the directionality of effects among identity components and psychosocial problems. Bi-factor cross-lagged effects modeling results showed that all identity components (i.e., identity consolidation, identity synthesis, and identity confusion) were significantly predicted by psychosocial problems and that psychosocial problems were predicted by two of three identity components (i.e., identity consolidation and identity confusion).

The findings of longitudinal invariance supplement the evidence for the factorial robustness of the EPSI over time; that is, within the bi-factor model the “synthesis with confusion” factor emerged in addition to the separate synthesis and confusion factors. This supports the notion that a synthesized sense of identity is not the terminal state of identity formation—rather, it still leaves the possibility of “changing course” later for individuals who appear sure of the direction they are taking (Luyckx et al. 2006; Schwartz et al. 2009b).

The key finding from the present study was the relationship between identity components and psychosocial problems in adolescence over time. More specifically, although a previous study utilizing a bi-factor approach to identity was cross-sectional (Schwartz et al. 2009b), our results demonstrate longitudinally that, in line with our hypotheses, changes in identity consolidation were negatively associated with changes in psychosocial problems and that changes in identity confusion were positively linked to these problems. Thus, we provide evidence for dynamic associations of the identity components with psychosocial problems, suggesting that those adolescents who increase in identity consolidation may be likely to decrease in anxiety, depressive symptoms, problems with friends, and conduct problems. On the other hand, those adolescents who increase in identity confusion may be more likely to manifest these problems.

It is particularly intriguing that identity synthesis as a specific factor was found to have no relationship to psychosocial problems, and identity consolidation as an overall factor to be associated with it. This finding suggests that much of the covariance between positive identity outcomes and psychosocial problems is attributable to the balance between synthesis and confusion, rather than to synthesis per se. Having a synthesized sense of identity along with a feeling that there is some “room left over” for more identity work (Luyckx et al. 2006) may be more closely related to adolescents’ adaptive psychological functioning. Based on arguments regarding ego development (e.g., Loevinger 1984), tolerance for stressful life events and experiences (i.e., ego strength) may have the potential to protect adolescents from depressive experiences, problems with friends, and acting-out behavior. Despite there being no relationship between identity synthesis and psychosocial problems, the relationship between identity confusion and psychosocial problems remained relatively large with the inclusion of the overall factor. Taken together, we have made for a more rigorous test of whether the identity components are related to adolescent psychosocial problems.

Furthermore, we examined the directionality of the effects between identity components and psychosocial problems. Contrary to our hypothesis, only two of three identity components (i.e., identity consolidation and identity confusion) predicted psychosocial problems, whereas all three identity components were predicted by psychosocial problems. Additionally, the cross-lagged effects of psychosocial problems to identity components were more pronounced than the cross-lagged effects of identity components to psychosocial problems. This pattern of findings suggests that the effects of psychosocial problems on identity components may be stronger than vice versa. This means that feeling depressed, left out, and engaging in delinquent behavior may cause the person to be restricted from some important identity opportunities (e.g., Crocetti et al. 2009). A possible explanation for this stronger effect of psychosocial problems on identity components may be explained by the differences in the United States and Japanese cultural contexts. The previous studies focused on adolescents from diverse ethnic groups and immigrant youths in the United States (e.g., Schwartz et al. 2012). In the multicultural U.S. context, young people are exposed to other cultural groups and may be prompted to explore cultural as well as personal identity issues. For instance, Hispanic youth in the United States may question and examine as to what it means to be Hispanic and how their ethnic group fits into the larger national context, how Hispanic they feel, and how much Spanish they want to speak―among other cultural concerns (Syed and Juang 2014). On the other hand, Japan is a rather culturally homogenous society, and the vast majority of youth are not confronted with cultural concerns. In addition, a critical issue relating to identity formation in contemporary Japan is the prolongation of the identity formation process. In late modern Japanese society, adolescents enjoy rich opportunities for freely exploring future plans over a long period compared to youth in the past; however, such a social situation seemingly creates difficulty in identity formation for some adolescents because young people are likely to lose their way toward finding their own sense of self in a context with too many identity options (Hatano and Sugimura in press). In this situation, Japanese adolescents may experience psychosocial problems while they are groping for a sense of self before they volitionally engage in the identity formation process, and this may cause identity conflict. Given that adolescence is a time of increased internalizing (e.g., depressive symptoms) and externalizing (e.g., rule breaking and substance use) symptoms for some youth (Ames et al. 2015; Castelao and Kröner-Herwig 2014), it is conceivable that identity components are not only risk factors pertaining to or protecting against psychosocial problems, but that psychosocial problems may represent impediments to healthy identity development during adolescence.

Overall, our findings suggest the potential importance of identity consolidation as a key index for adolescent psychosocial adjustment during adolescence in a late modern society where norms and structures are increasingly unstable and uncertain (Côté and Levine 2016). Although such a description has been used primarily to refer to North American and Western European societies (e.g., Côté and Bynner 2008), Japan has been increasingly characterized by an extended transition to adulthood, an uncertain job market, and extensive relationship experience prior to marriage (Sugimura and Mizokami 2012). Our findings suggest the importance of identity components to adaptive functions for contemporary young people in late modern society.

Implications for Intervention to Prevent Problems among Adolescents

The present results provide several valuable implications for understanding, preventing, and intervening into psychosocial problems in adolescence. First, our findings add a new perspective on the role of identity consolidation in supporting adolescents’ healthy identity development. Previous studies have primarily emphasized the importance of promoting identity synthesis and reducing identity confusion, and hence have proposed intervention programs to support or help adolescents with a confused sense of identity (e.g., Markstrom-Adams et al. 1993). On the other hand, our findings suggest the importance of supporting adolescents in reaching a balance between identity synthesis and confusion. This view can be particularly well applied to adolescents growing up in late modern societies, such as the United States, Japan, and Western European countries, because the transition to adulthood has tended to be postponed in those countries (Côté and Levine 2016). In such societies, it is hard for adolescents to resolve identity conflicts and distress that are likely to be prolonged (e.g., Berman et al. 2011). Therefore, advocates for adolescents (e.g., parents, teachers, and counselors) may be better served by considering the interplay between identity synthesis and confusion from a long-term perspective. They may need to develop intervention programs in school (as well as family) contexts, and help adolescents understand the role of identity conflict (i.e., opportunities and room to further develop a sense of identity) to prevent psychosocial problems and navigate successfully through the adolescent period.

Second, because many researchers have thus far hypothesized that identity synthesis would predict against psychosocial problems, they have focused on increasing a sense of identity synthesis or strengthening identity commitments to prevent adolescent psychosocial problems (e.g., Eischas et al. 2017). However, considering that early adolescents are just beginning the process of developing a sense of identity (Archer and Waterman 1983) and acquiring advanced cognitive skills (Kroger 2004), our findings suggest that it may also be wise to focus primarily on psychosocial problems, reductions in which may promote healthy identity development.

Limitations and Future Directions

Although our study does provide important evidence for associations between identity components and psychosocial problems during adolescence, some limitations should be acknowledged. First, because we used only two measurement points, we were not able to examine non-linear change. Future studies should include data from three or more waves so that latent growth curve modeling can be used (Duncan et al. 1999). Second, the present longitudinal study followed participants for a relative brief period (i.e., 2 years). Given that young people experience a series of life tasks and changes, such as the transition from school to work, as they move toward young adulthood (Côté and Levine 2016), further research following them into their 20s is needed. Finally, the attrition rate of participants in the present study might have been high because of minimal compensation (i.e., 50 JPY) for participation. Although this amount is typical in the context of research norms in Japan, we need to identify ways to better incentivize retention in future Japanese studies.

Conclusion

Despite these limitations, the present longitudinal study has provided important evidence regarding the associations between identity components and psychosocial problems. Although previous research using bi-factor models focused on cross-sectional links between identity consolidation and psychosocial problems, we examined longitudinal associations by using both a bi-factor latent change and cross-lagged effects models. By using a bi-factor latent change model, we were able to show that adolescents who increased in identity consolidation had a decrease in psychosocial problems, while those who increased in identity confusion were more likely to manifest psychosocial problems. Furthermore, by using a bi-factor cross-lagged effect model, we unraveled that feeling depressed, left out, and engaging in delinquent behavior (i.e., psychosocial problems) may cause the adolescent to be restricted from some important identity opportunities. Our findings highlight that identity consolidation and confusion play important roles in protecting against and potentiating adolescent psychosocial problems. Taken together, the bi-factor model used in the present study empirically supports Erikson’s (1968) conceptions of identity (synthesis with confusion) and provided the knowledge that an “overall” identity dimension (i.e., identity consolidation) may be a useful target for intervention efforts for adolescents living in late modern societies. We hope that the present study will inspire more cross-national, longitudinal, and applied work in this direction.

Footnotes

  1. 1.

    Previous research indicated that the SDQ subscales have three higher-order factors: internalizing factor (i.e., emotional symptoms and peer problems), externalizing factor (i.e., conduct problems and hyperactivity), and prosocial factor (Goodman et al. 2010). Prosociality does not assume psychosocial problems, and hyperactivity is slightly different from Achenbach’s definition of externalizing problems (Achenbach et al. 2002) because it does not involve violating other people’s rights. Therefore, we used 3 of the 5 subscales (i.e., emotional symptoms, peer problems, and conduct problems) as indicators of psychosocial problems.

  2. 2.

    To clarify the strength of the directionality of these effects, we compared two types of models. A constrained model (cross paths from identity components to psychosocial problems and cross paths from psychosocial problems to identity consolidation were constrained to be equal across measurement waves) was compared with an unconstrained model (with all these coefficients allowed to vary). If the unconstrained model fit was good, there would be the difference between the cross-lagged effects. The unconstrained model fit was favorable (constrained model: χ 2(355) = 1035.175, p < .001, CFI = .923, RMSEA = .047, 90% CI = .043–.050, AIC = 1315.175, AIC > 20). Consequently, the cross-lagged paths between the identity consolidation and psychosocial problems were not equal.

Notes

Authors' Contributions

K.H. conceived of the study, participated in its design and coordination and drafted the manuscript. K.S. and S.J.S. conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.

Data Sharing Declaration

This manuscript’s data will not be deposited. Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all participants included in the study.

References

  1. Achenbach, T. M., Dumenci, L., & Rescorla, L. (2002). Ten-year comparisons of problems and competencies for national samples of youth self, parent, and teacher reports. Journal of Emotional and Behavioral Disorders, 10, 194–203.  https://doi.org/10.1177/10634266020100040101.CrossRefGoogle Scholar
  2. Ames, M. E., Wintre, M. G., & Flora, D. B. (2015). Trajectories of BMI and internalizing symptoms: Associations across adolescence. Journal of Adolescence, 45, 80–88.  https://doi.org/10.1016/j.adolescence.2015.08.016.CrossRefPubMedGoogle Scholar
  3. Arbuckle, J. L. (2009). AMOS18 user’s guide/. Chicago, IL: SPSS Inc.Google Scholar
  4. Archer, S. L., & Waterman, A. S. (1983). Identity in early adolescence: A developmental perspective. Journal of Early Adolescence, 3, 203–214.  https://doi.org/10.1177/0272431683033003.CrossRefGoogle Scholar
  5. Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–480.  https://doi.org/10.1037/0003-066X.55.5.469.CrossRefPubMedGoogle Scholar
  6. Arnett, J. J. (2007). Suffering, selfish, slackers? Myth and reality about emerging adults. Journal of Youth and Adolescence, 36, 23–29.  https://doi.org/10.1007/s10964-006-9157-z.CrossRefGoogle Scholar
  7. Azmitia, M., Syed, M., & Radmacher, K. (2013). Finding your niche: Identity and emotional support in emerging adults’ adjustment to the transition to college. Journal of Research on Adolescence, 23, 744–761.  https://doi.org/10.1111/jora.12037.CrossRefGoogle Scholar
  8. Berman, S. L., You, Y.-F., Schwartz, S., Teo, G., & Mochizuki, K. (2011). Identity exploration, commitment, and distress: A cross national investigation in China, Taiwan, Japan, and the United States. Child & Youth Care Forum, 40, 65–75.  https://doi.org/10.1007/s10566-010-9127-1.CrossRefGoogle Scholar
  9. Burnham, K. P., & Anderson, D. R. (2002). Model selection and multimodel inference: A practical information-theoretic approach. New York, NY: Springer.Google Scholar
  10. Castelao, C. F., & Kröner-Herwig, B. (2014). Developmental trajectories and predictors of externalizing behavior: A comparison of girls and boys. Journal of Youth and Adolescence, 43, 775–789.  https://doi.org/10.1007/s10964-013-0011-9.CrossRefGoogle Scholar
  11. Chen, F. F., West, S. G., & Sousa, K. H. (2006). A comparison of bifactor and second-order models of quality of life. Multivariate Behavioral Research, 41, 189–225.  https://doi.org/10.1207/s15327906mbr4102_5.CrossRefPubMedGoogle Scholar
  12. Chen, F. F., Hayes, A., Carver, C. S., Laurenceau, J.-P., & Zhang, Z. (2012). Modeling general and specific variance in multifaceted constructs: A comparison of the bifactor model to other approaches. Journal of Personality, 80, 219–251.  https://doi.org/10.1111/j.1467-6494.2011.00739.x.
  13. Claes, L., Luyckx, K., & Bijttebier, P. (2014). Non-suicidal self-injury in adolescents: Prevalence and associations with identity formation above and beyond depression. Personality and Individual Differences, 61–62, 101–104.  https://doi.org/10.1016/j.paid.2013.12.019.CrossRefGoogle Scholar
  14. Côté, J. E., & Bynner, J. M. (2008). Changes in the transition to adulthood in the UK and Canada: The role of structure and agency in emerging adulthood. Journal of Youth Studies, 11, 251–268.  https://doi.org/10.1080/13676260801946464.CrossRefGoogle Scholar
  15. Côté, J. E., & Levine, C. G. (2016). Identity formation, youth, and development: A simplified approach.. New York: Psychology Press.Google Scholar
  16. Crocetti, E., Rubini, M., & Meeus, W. (2008). Capturing the dynamics of identity formation in various ethnic groups: Development and validation of a three-dimensional model. Journal of Adolescence, 31, 207–222.  http://dxdoi:10.1016/j.adolescence.2007.09.002.
  17. Crocetti, E., Klimstra, T., Keijsers, L., Hale, III, W. W., & Meeus, W. (2009). Anxiety trajectory classes and identity development in adolescence: A five-wave longitudinal study. Journal of Youth and Adolescence, 38, 839–849.  https://doi.org/10.1007/s10964-008-9302-y.CrossRefPubMedGoogle Scholar
  18. Dimitrov, D. M. (2010). Testing for factorial invariance in the context of construct validation. Measurement and Evaluation in Counseling and Development, 43, 121–149.CrossRefGoogle Scholar
  19. Dimitrova, R., Hatano, K., Sugimura, K., & Ferrer-Wreder, L. (in press). The Factorial Validity and Equivalence of Identity as Measured by the Erikson Psychosocial Stage Inventory in Adolescent Samples from the United States and Japan. European Journal of Psychological Assessment.Google Scholar
  20. Donovan, R. A., Huynh, Q.-L., Park, I. J. K., Kim, S. Y., Lee, & Robertson, E. (2013). Relationships among identity, perceived discrimination, and depressive symptoms in eight ethnic-generational groups. Journal of Clinical Psychology, 69, 397–414.  https://doi.org/10.1002/jclp.21936.CrossRefPubMedGoogle Scholar
  21. Duncan, T. E., Duncan, S. C., Strycker, L. A., Li, F., & Alpert, A. (1999). An introduction to latent variable growth curve modeling: Concepts, issues and applications. Mahwah, NJ: Erlbaum.Google Scholar
  22. Erikson, E. H. (1950). Childhood and society. New York: Norton.Google Scholar
  23. Erikson, E. H. (1968). Identity and the life cycle. Psychological issues, Vol. 1 (1), Monograph 1. New York: International University Press.Google Scholar
  24. Eischas, K., Montgomery, M. J., Meca, A., & Kurtines, W. M. (2017). Empowering marginalized youth: A self-transformative intervention for promoting positive youth development. Child Development.  https://doi.org/10.1111/cdev.12866.Google Scholar
  25. Ferrer-Wreder, L., Palchunk, A., Poyrazli, S., Small, M. L., & Domitrovich, C. E. (2008). Identity and adjustment. Identity, 8, 95–105.  https://doi.org/10.1080/15283480801938143.CrossRefGoogle Scholar
  26. Furlong, A., & Cartmel, F. (2009). Higher education and social justice. Maidenhead: Open University Press.Google Scholar
  27. Goodman, A., Ramping, D. L., & Ploubidis, G. B. (2010). When to use broader internalizing and externalizing subscales instead of the hypothesized five subscales on the Strength and Difficulties Questionnaire SDQ: Data from British parents, teachers and children. Journal of Abnormal Child Psychology, 38, 1179–1191.  https://doi.org/10.1007/s10802-010-9434-x.CrossRefPubMedGoogle Scholar
  28. Goodman, R. (1997). The Strengths and Difficulties Questionnaire: a research note. Journal of Child Psychology and Psychiatry, 38, 581–586.  https://doi.org/10.1111/j.1469-7610.1997.tb01545.x.CrossRefPubMedGoogle Scholar
  29. Hatano, K., Sugimura, K., & Crocetti, E. (2016). Looking at the dark and bright sides of identity formation: New insights from adolescents and emerging adults in Japan. Journal of Adolescence, 47, 156–168.  https://doi.org/10.1016/j.adolescence.2015.09.008.CrossRefPubMedGoogle Scholar
  30. Hatano, K., Sugimura, K., & Klimstra, T. (2017). Which comes first: Personality traits or identity during adolescence? Journal of Research in Personality, 67, 120–131.  https://doi.org/10.1016/j.jrp.2016.06.014.CrossRefGoogle Scholar
  31. Hatano, K., Sugimura, K., Nakama, R., Mizokami, S., & Tsuzuki, M. (2014). Erikson shinri shakai teki dankai mokuroku (dai 5 dankai) 12 koumoku ban no sakusei [Examining the reliability and validity of a Japanese version of the 12-item Erikson Psychosocial Stage Inventory (the 5th stage)]. Japanese Journal of Psychology, 85, 482–487.  https://doi.org/10.4992/jjpsy.8513319.CrossRefPubMedGoogle Scholar
  32. Hatano, K., & Sugimura, K. (in press). Is Adolescence a period of identity formation for all youth? Insights from a four-wave longitudinal study of identity dynamics in Japan. Developmental Psychology.Google Scholar
  33. Jackson, J. J., & Allemand, M. (2014). Moving personality development research forward: Applications using structural equation models. European Journal of Personality, 28, 300–310.  https://doi.org/10.1002/per.1964.CrossRefGoogle Scholar
  34. Kline, R. B. (2010). Principles and practice of structural equation modeling (3rd ed.). New York: Guilford Press.Google Scholar
  35. Kroger, J. (2004). Identity in adolescence: The balance between self and other (3rd ed.). London, New York: Routledge.CrossRefGoogle Scholar
  36. 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. 31–53). New York: Springer.CrossRefGoogle Scholar
  37. Loevinger, J. (1984). On the self and predicting behavior. In R. A. Zucker, J. Aronoff & A. I. Rabin (Eds.), Personality and the prediction of behavior (pp. 43–68). Orlando, FL: Academic.Google Scholar
  38. Luyckx, K., Goossens, L., Soenens, B., & Beyers, W. (2006). Unpacking commitment and exploration: Validation of an integrative model of adolescent identity formation. Journal of Adolescence, 29, 361–378.  https://doi.org/10.1016/j.adolescence.2005.03.008.CrossRefPubMedGoogle Scholar
  39. 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.CrossRefGoogle Scholar
  40. Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 3, 551–558.  https://doi.org/10.1037/h0023281.CrossRefPubMedGoogle Scholar
  41. Marcia, J. E. (2002). Identity and psychosocial development in adulthood. Identity, 2, 7–28.  http://dx.doi.org/10.1207/S1532706XID0201_02.
  42. Markstrom-Adams, C., Ascione, F. R., Braegger, D., & Adams, G. R. (1993). Promotion of ego-identity development: Can short-term intervention facilitate growth? Journal of Adolescence, 16, 217–224.  https://doi.org/10.1006/jado.1993.1020.CrossRefPubMedGoogle Scholar
  43. McArdle, J. J., & Nesselroade, J. R. (1994). Using multivariate data to structure developmental change. In H. W. Reese & S. H. Cohen (Eds.), Lifespan developmental psychology: Methodological contributions (pp. 223–267). Hillsdale, NJ: Lawrence Erlbaum.Google Scholar
  44. Ministry of Health, Labour and Welfare (2014). Heisei ni-jyuu-rokunen jinkou doutai toukei geppou nennkei no gaiyou [The report of vital statistics in 2014]. http://www.mhlw.go.jp/toukei/saikin/hw/jinkou/geppo/nengai14/index.html. Accessed 1 Nov 2017.
  45. Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674–701.  https://doi.org/10.1037/0033-295X.100.4.674.tabfigCrossRefPubMedGoogle Scholar
  46. Mu, W., Luo, J., Nickel., L., & Roberts, B. (2016). Generality or specificity? Examining the relation between personality traits and mental health outcomes using a bivariate bi-factor latent change model. European Journal of Personality, 30, 467–483.  https://doi.org/10.1002/per.2052.CrossRefGoogle Scholar
  47. Rattansi, A., & Phoenix, A. (2005). Rethinking youth identities: Modernist and postmodernist Frameworks. Identity, 5, 97–123.CrossRefGoogle Scholar
  48. Rosenthal, D. A., Gurney, R. M., & Moore, S. M. (1981). From trust to intimacy: A new inventory for examining Erikson’s stages of psychosocial development. Journal of Youth and Adolescence, 10, 525–537.  https://doi.org/10.1007/BF02087944.CrossRefPubMedGoogle Scholar
  49. Schwartz, S. J. (2007). The structure of identity consolidation: Multiple correlated constructs or one superordinate construct? Identity, 7, 27–49.  https://doi.org/10.1080/15283480701319583.CrossRefGoogle Scholar
  50. 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.  https://doi.org/10.1177/0044118X05275965.CrossRefGoogle Scholar
  51. Schwartz, S. J., Hardy, S. A., Zamboanga, B. L., Meca, A., Waterman, A. S., Picariello, S., Luyckx, K., Crocetti, E., Kim, S. Y., Brittian, A. S., Roberts, S. E., Whitbourne, S. K., Ritchie, R. A., Brown, E. J., & Forthun, L. F. (2015a). Identity in young adulthood: Links with mental health and risky behavior. Journal of Applied Developmental Psychology, 36, 39–52.  https://doi.org/10.1016/j.appdev.2014.10.001.CrossRefGoogle Scholar
  52. Schwartz, S. J., Klimstra, T. A., Luyckx, K., Hale, W. W., & Meeus, W. (2012). Characterizing the self-system over time in adolescence: Internal structure and associations with internalizing symptoms. Journal of Youth and Adolescence, 41, 1208–1225.  https://doi.org/10.1007/s10964-012-9751-1.CrossRefPubMedGoogle Scholar
  53. Schwartz, S. J., Luyckx, K., & Crocetti, E. (2015b). What have we learned since Schwartz (2001)? A reappraisal of the field of identity development. In K. C. McLean & M. Syed (Eds.), The oxford handbook of identity development (pp. 539–561). New York: Oxford University Press.Google Scholar
  54. Schwartz, S. J., Mason, C. A., Pantin, H., & Szapocznik, J. (2008). Effect of family functioning and identity confusion on substance use and sexual behavior in Hispanic immigrant early adolescents. Identity, 8, 107–124.  https://doi.org/10.1080/15283480801938440.CrossRefPubMedPubMedCentralGoogle Scholar
  55. Schwartz, S. J., Mason, C. A., Pantin, H., Wang, W., Brown, C. H., Campo, A. E., & Szapocznik, J. (2009a). Relationships of social context and identity to problem behavior among high-risk Hispanic adolescents. Youth & Society, 40, 541–570.  https://doi.org/10.1177/0044118X08327506.CrossRefGoogle Scholar
  56. Schwartz, S. J., Unger, J. B., Meca, A., Lorenzo-Blanco, E. I., Baezconde-Garbanati, L., Cano, M. Å., et al. (2017). Personal identity development in hispanic immigrant adolescents: Links with positive psychosocial functioning, depressive symptoms, and externalizing problems. Journal of Youth and Adolescence, 46, 898–913.  https://doi.org/10.1007/s10964-016-0615-y.CrossRefPubMedGoogle Scholar
  57. Schwartz, S. J., Zamboanga, B. L., Luyckx, K., Meca, A., & Ritchie, R. A. (2013). Identity in emerging adulthood: Reviewing the field and looking forward. Emerging Adulthood, 1, 96–113.  https://doi.org/10.1177/2167696813479781.CrossRefGoogle Scholar
  58. Schwartz, S. J., Zamboanga, B. L., Wang, W., & Olthuis, J. V. (2009b). Measuring identity from an Eriksonian perspective: Two sides of the same coin. Journal of Personality Assessment, 91, 143–154.  https://doi.org/10.1080/00223890802634266.CrossRefPubMedGoogle Scholar
  59. Schwartz, S. J., Zamboanga, B. L., Weisskirch, R. S., & Rodriguez, L. (2009c). The relationships of personal and ethnic identity exploration to indices of adaptive and maladaptive psychosocial functioning. International Journal of Behavioral Development, 33, 131–144.  https://doi.org/10.1177/0165025408098018.CrossRefGoogle Scholar
  60. Statistic Bureau of Japan (2016). Gakkou kihon chousa [School basic survey]. http://www.mext.go.jp/b_menu/toukei/chousa01/kihon/1267995.htm. Accessed 1 Nov 2017.
  61. Statistics Bureau of Japan (2017). 2015 population census, basic complete tabulation on occupations, Japan, Table 40: Population, by type of living with parent(s) (3 groups), marital status (4 groups), type of activity, age (single years) and sex. http://www.e-stat.go.jp/SG1/estat/GL08020103.do?_toGL08020103_&tclassID=000001093838&cycleCode=0&requestSender=search. Accessed 1 Nov 2017.Google Scholar
  62. Sugawara, M., Sakai, A., Sugiura, T., & Matsumoto, A. (2006). SDQ: The Strengths and Difficulties Questionnaire. http://www.sdqinfo.com/. Accessed 1 Nov 2017.Google Scholar
  63. Sugimura, K., & Mizokami, S. (2012). Personal identity in Japan. New Directions for Child and Adolescent Development, 138, 123–143.  https://doi.org/10.1002/cad.20025.CrossRefGoogle Scholar
  64. Syed, M., & Juang, L. P. (2014). Ethnic identity, identity coherence, and psychological functioning: Testing basic assumptions of the developmental model. Cultural Diversity and Ethnic Minority Psychology, 20, 176–190.CrossRefPubMedGoogle Scholar
  65. Syed, M., & McLean, K. (2016). Understanding identity integration: Theoretical, methodological, and applied issues. Journal of Adolescence, 47, 109–118.  https://doi.org/10.1016/j.adolescence.2015.09.005.CrossRefPubMedGoogle Scholar
  66. Takahashi, Y., Edmonds, G. W., Jackson, J. J., & Roberts, B. W. (2013). Longitudinal correlated changes in conscientiousness, preventative health-related behaviors, and self-perceived physical health. Journal of Personality, 81, 417–427.  https://doi.org/10.1111/jopy.12007.CrossRefPubMedPubMedCentralGoogle Scholar
  67. Waterman, A. S., Schwartz, S. J., Hardy, S. A., Kim, S. Y., Lee, R. M., Armenta, B. E., Whitbourne, S. K., Zamboanga, B. L., Brown, E. J., Williams, M. K., & Agocha, V. B. (2013). Good choices, poor choices: Relationship between the quality of identity commitments and psychosocial functioning. Emerging Adulthood, 1, 163–174.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2017

Authors and Affiliations

  1. 1.Development Center for Higher Education, Osaka Prefecture UniversityOsakaJapan
  2. 2.Department of PsychologyGraduate School of Education, Hiroshima UniversityHiroshimaJapan
  3. 3.Public Health Sciences at the University of Miami, Leonard M. Miller School of MedicineMiamiUSA

Personalised recommendations