Time Perspective and Identity Formation: Short-Term Longitudinal Dynamics in College Students Koen Luyckx1 Willy Lens2 Ilse Smits1 Luc Goossens1 Department of Psychology Catholic University Leuven, Belgium 1. Center for Developmental Psychology 2. Research Center for Motivation and Time Perspective Correspondence concerning this article can be sent to Koen Luyckx, Catholic University of Leuven, Department of Psychology, Tiensestraat 102, B-3000 Leuven, Belgium. E-mail: Koen.Luyckx@psy.kuleuven.be. The first author is postdoctoral researcher at the Fund for Scientific Research in Flanders (FWO). Short title: Time Perspective and Identity Formation 2 Abstract Planning for the future and developing a personalized identity are conceived of as important developmental tasks adolescents and emerging adults are confronted with on the pathway to adulthood. The present study set out to examine whether both tasks develop in tandem by using a short-term longitudinal data-set consisting of 371 college students assessed at two time-points 4 months apart. Identity formation was assessed using identity commitment and three identity processing styles underlying identity exploration; time perspective was assessed using the presenthedonistic, present-fatalistic, and future-oriented perspectives. Using cross-lagged structural equation modeling, three competing models were tested: a time-perspective main-effects model, an identity main-effects model, and a reciprocal model. In accordance with expectations, evidence was found for the reciprocal model with identity formation and time perspective mutually reinforcing one another across time. Implications and suggestions for future research are discussed. Key words: time perspective, future, identity style, exploration, commitment, cross-lagged analysis 3 Time Perspective and Identity Formation: Short-Term Longitudinal Dynamics in College Students Adolescence is typically conceived of as the period in life in which individuals are confronted with developing an integrated and self-endorsed identity and making plans and preparing for the future. By setting future-oriented goals, exploring different identity alternatives, and committing to certain life paths, adolescents can direct their own development in their social world and negotiate their passage into adulthood (Erikson, 1968; Nurmi, Poole, & Seginer, 1995; Seginer & Halabi-Kheir, 1998). Due to profound social-cultural changes occurring the last decades in many Western nations (e.g., the postponement of the completion of schooling, marriage, and becoming financially independent; Arnett, 2004; Fadjukoff, Kokko, & Pulkkinen, 2007), the transition to adulthood is postponed for many individuals until the late twenties, substantially prolonging the phase of exploration and preparing for adult roles (Schwartz, Côté, & Arnett, 2005). Arnett (2000) argued that, in current industrialized societies, the age period of the late teens and the twenties constitutes a distinct developmental period demographically, subjectively, and in terms of identity exploration and planning for the future – a developmental period which he labeled emerging adulthood. Inspired by Erikson (1968) and Levinson (1978) who focused on the prolonged role experimentation and institutionalized moratorium in this age period, Arnett stated that certain developmental tasks coming to the fore in adolescence – with identity formation and preparing for the future being the most important ones – further intensify in emerging adulthood, and especially in the college context. Previous research indeed testified to the importance of emerging adulthood and the college context in particular for identity exploration and future orientation (Luyckx, Goossens, & Soenens, 2006; Waterman, 1993). In sum, planning for the future and identity formation are conceptualized as interrelated developmental tasks adolescents and emerging adults are confronted with (Luyckx et al., 2008; Marcia, 1993; Nurmi et al., 1995). Research from a developmental perspective, however, focusing on how individuals’ identity commitments and specific styles of approaching the task of identity exploration (referred to as identity styles; Berzonsky, 1990) are related to the degree to which individuals are future- versus present-oriented is lacking. Consequently, the present study made use of a short-term longitudinal design set up in a sample of college students to examine the extent to which these different variables influence each other across time. Before we proceed to our hypotheses, we will outline the theories on identity formation and time perspective as used in the present study and developed, respectively, by Berzonsky (1990) and Zimbardo and Boyd (1999). Identity Formation: Styles of Exploration and Commitment 4 Inspired by Erikson’s (1968) seminal work on identity development throughout the life span, the processes of identity exploration and commitment have been viewed as crucial dimensions of personal identity formation (Grotevant, 1987; Waterman, 1999). Marcia (1966) was among the first to conceptualize exploration and commitment as two basic identity dimensions. He defined exploration as the degree to which individuals engage in a personalized search for different values, beliefs, and goals; commitment was defined as the adherence to a set of convictions, goals, and beliefs. By crossing these dimensions, Marcia arrived at four distinct identity statuses or types: achievement (characterized by making commitments after a period of exploration), foreclosure (characterized by making commitments without a period of prior exploration), moratorium (characterized by being in an exploratory state without settling into steady commitments yet), and diffusion (characterized by a general lack of identity work). Commitment is what tends to separate successful from unsuccessful identity development (Schwartz et al., 2005). Forming and adhering to stable and self-endorsed identity commitments is assumed to nurture feelings of having an integrated sense of self (Côté & Levine, 2002). Many studies have indeed found evidence for a positive association between identity commitments and psychological well-being (e.g., Bosma & Kunnen, 2001). In an attempt to capture the socialcognitive processes underlying exploration, Berzonsky (1990) differentiated between three identity processing styles, that is, the information-oriented, normative, and diffuse-avoidant styles, tapping into the strategies that individuals prefer in approaching or avoiding the tasks of constructing and reconstructing a sense of identity (Berzonsky & Adams, 1999; Soenens, Berzonsky, Vansteenkiste, Beyers, & Goossens, 2005). First, the information-oriented style is typical of individuals who actively construct a sense of identity by seeking out and evaluating personally relevant information. Individuals within Marcia’s (1966) achievement and moratorium statuses have been found to rely on this style of identity processing. These individuals are open to new information, critical towards their selfconcepts, and willing to revise or accommodate their identity when faced with discrepant information about themselves. This identity style was found to be positively associated with problem-focused coping strategies, mature interpersonal relationships, empathy, and an autonomous or self-determined mode of functioning, and negatively with ill-being, need for closure, and social prejudice (Berzonsky, 1990, 1992; Berzonsky, Nurmi, Kinney, & Tammi, 1999; Luyckx et al., 2007; Soenens, Duriez, & Goossens, 2005). Second, the normative style is typical of individuals who rely strongly on the prescriptions and expectations held by significant others (such as parents or other authority figures) when confronted with the identity task, such as those individuals situated in Marcia’s (1966) foreclosure status. These individuals have rigidly organized identity commitments and they are also “closed” 5 to information that threatens their values and beliefs. As such, they try blindly to preserve their identity commitments instead of engaging themselves in a profound exploratory process. Although these individuals can experience well-being, they have been found to score low on measures of experiential openness and empathy, and high on measures of conservatism and prejudice (Berzonsky, 1990, 1992; Berzonsky & Kuk, 2000; Soenens, Berzonsky, et al., 2005; Soenens, Duriez, et al., 2005). Finally, the diffuse-avoidant style typifies individuals who avoid dealing with personal conflicts and identity-relevant problems until situational demands dictate their behavior. These individuals accommodate their identity as a function of current social demands and consequences, without arriving at a well-stablished sense of identity in the long term. Such strategies are likely to result in the absence of strong and self-endorsed identity commitments and a fragmented and unintegrated identity structure, which is likely to result in a state of diffusion as outlined by Marcia (1966). Diffuse-avoidant individuals display low levels of active information processing, conscientiousness, persistence, and problem-solving coping abilities, and high levels of neuroticism, maladaptive coping strategies, and maladjustment (Berzonsky, 1990, 1992; Berzonsky & Kuk, 2000; Soenens, Berzonsky, et al., 2005; Soenens, Duriez, et al., 2005). Time Perspective: Being Present- Versus Future-Oriented Humans’ perceptions of time and how these perceptions are related to indicators of intra- and interindividual functioning have been studied under the heading of time perspective (Lens & Moreas, 1994). The study of time perspective draws from cognitive-motivational theorizing and refers to how the flow of human experience is parceled into different time frames – being past, present, and future – and, as such, looks at the notion of time as an individualized psychological phenomenon (Lens, 1986, 2006; Nuttin & Lens, 1985; Zimbardo & Boyd, 1999). Some individuals indeed are better able than others to foresee the future implications of their present behaviors and envisage how their present behavior can serve the attainment of future goals, whereas other people rather live in the present and do not anticipate the future consequences of their present activities (Simons, Vansteenkiste, Lens, & Lacante, 2004). So, time perspective refers to the relative temporal orientation that guides and influences individuals’ actions and goals (Henson, Carey, Carey, & Maisto, 2006). Zimbardo and colleagues (Keough, Zimbardo, & Boyd, 1999; Zimbardo & Boyd, 1999; Zimbardo, Keough, & Boyd, 1997) specifically focused on the extent to which individuals are present- versus future-oriented and the ramifications these different motivational perspectives have on psychosocial functioning and well-being. First, individuals with a dominant present time perspective are primarily oriented towards the here-and-now and are inclined to form goals and adopt behaviors that meet immediate desires. They are undistracted by past worries or anxieties but may also be unable to plan a realistic life 6 path. These individuals have been found to experience negative outcomes, such as mental health problems, substance abuse, and juvenile delinquency, especially when they are living in predominantly future-oriented societies or contexts (Keough et al., 1999; Zimbardo & Boyd, 1999). There are two types of being predominantly present-oriented, however, that are hypothesized to operate differently and to be grounded in different psychological mechanisms. Individuals with a dominant present-hedonistic time perspective are oriented towards present enjoyment and excitement and display a high interest in novelty and sensation seeking. They score low on ego or impulse control and avoid cost-benefit analyses and contingency planning. Individuals with a dominant present-fatalistic orientation, on the other hand, believe that the future is predestined and cannot be controlled or influenced by individual actions, resulting in hopelessness and amotivation. These individuals bear the present with fatalistic resignation, unable to see themselves as active agents of their own development (D’Alessio, Guarino, De Pascalis, & Zimbardo, 2003; Zimbardo & Boyd, 1999). Second, individuals with a dominant future time perspective, by contrast, set goals and plan strategies for achieving them and for meeting obligations in the long term (Lens & Tsuzuki, 2007). They are more capable than present-oriented individuals in articulating and visualizing future goals and, consequently, in restraining themselves from engaging in risky behaviors such as unsafe sex and substance abuse. As such, a future time perspective is a cognitive-motivational framework that may protect against impulsive and unhealthy behaviors in the present because it increases the ability to foresee negative consequences of certain actions and behaviors (Keough et al., 1999). As such, these individuals tend to be more succesful than present-oriented individuals, both academically and in their careers (Simons et al., 2004). In sum, the present and future temporal frames are used by individuals in forming expectations, goals, and life paths and, as such, might enable them to build their identity (D’Alessio et al., 2003; Keough et al., 1999; Nurmi et al., 1995; Seginer, Vermulst, & Shoyer, 2004). However, research that examines how time perspective and identity formation (as assessed by identity commitment and styles) are developmentally interrelated is lacking. Hypotheses of the Present Study The present study addressed two developmental research questions with respect to the link between identity formation and time perspective in a short-term longitudinal design using a sample of college students. Research Question 1: Stability of Concurrent Associations. As can be expected based upon the description of identity styles, commitment, and the three time perspective scales, differential hypotheses with respect to their interrelationships can be formulated. 7 First, we hypothesized that the future-oriented time perspective would be positively related to the information-oriented identity style and to identity commitment, and negatively to the diffuse-avoidant identity style. As such, typical features of being future-oriented, such as the consideration of future consequences, high levels of conscientiousness, efficiency, and ambition, and preference for consistency (Zimbardo & Boyd, 1999), would translate themselves into a personalized search for identity options and the ability to make strong and self-endorsed identity commitments. Similarly, Luyckx, Soenens, and Goossens (2006) found conscientiousness (being strongly related to a future time perspective; Zimbardo & Boyd, 1999) on the one hand and identity commitment and indices of adequate identity exploration on the other hand to be interrelated across time. The diffuse-avoidant style on the other hand might compromise goal setting and planning for the future, hence its hypothesized negative association with future time perspective (Kerpelman & Mosher, 2004; Pulkkinen & Rönka, 1994). Second, we hypothesized that both present time perspective scales – and especially the present-fatalistic scale – would be negatively related to the information-oriented style and commitment, and positively to the diffuse-avoidant style. Individuals who are present-oriented either live for pleasure today with little regard for tomorrow (in the case of a dominant presenthedonistic time orientation) or do not believe that anything they will do is likely to make a difference in their lives (in the case of a dominant present-fatalistic orientation). As such, engaging in a personalized information-oriented identity search or making firm and steady identity commitments would likely be perceived of as being low in priority for present-hedonistic individuals or as simply being unattainable for present-fatalistic individuals. Both these present time perspective scales would be more likely to be positively related to the diffuse-avoidant style, being indicative of people who postpone identity work and avoid dealing with identity issues, leaving them diffused about the direction their lives have to take. Further, we hypothesized that these predicted relationships between identity styles and commitment on the one hand and time perspective on the other hand would be relatively equivalent across the two waves of the present short-term longitudinal study. With respect to the link between time perspective and the normative style, our expectations were less clear. If any, we expected that the normative identity style (which is characteristic of individuals who attach high value to the norms and prescriptions upheld by significant others) would be positively related to the future time perspective, given that the college context strongly emphasizes academic and future achievement (as most parents of college-attending youth are also likely to do). Relatedly, we expected that the normative style would be negatively related to the present-hedonistic time perspective scale. Previous research (e.g., Marcia, 1993) also pointed to negative associations between the normative style and the foreclosure-status on the one hand and 8 substance abuse on the other hand, the latter being a possible instantiation of such a presenthedonistic lifestyle (Zimbardo & Boyd, 1999). Research Question 2: Longitudinal Cross-Lagged Associations. These concurrent relations and the stability therein are hypothesized to be reflective of developmental associations that would appear – at least to some extent – in cross-lagged analyses aimed at substantiating the direction of effects in the link between these two sets of constructs. Three competing cross-lagged models looking at independent associations of the three time perspective scales were tested in the present study: a time-perspective main-effects model, an identity main-effects model, and a reciprocal model. The time-perspective main-effects model assumes that one’s current time perspective influences or drives developmental changes in identity styles and commitment. Several authors (Boniwell & Zimbardo, 2004; Zimbardo & Boyd, 1999) indeed stated that time perspective can become a dispositional characteristic which influences individual choices, actions, and decisions and, as such, can be a prerequisite to start building one’s identity. A concrete instantiation of this model would be, for instance, that a future time perspective positively predicts the use of the information-oriented identity style across time. In support of this model, Lennings (1998) classified a future time perspective as part of an actualizer profile, with being oriented towards the future positively influencing self-actualisation and identity development. The identity main-effects model, on the other hand, assumes than one’s current state of identity formation influences developmental changes in time perspective. As noted earlier, time perspective is likely to be influenced by several factors, such as one’s life course and identity development throughout different contexts (Aspinwall, 2005; Boniwell & Zimbardo, 2004; Kerpelman & Mosher, 2004). A concrete instantiation of this model would be, for instance, that the use of the information-oriented identity style positively predicts a future time perspective across time. When individuals are exploring different identity alternatives, this exploration implies that they are motivated to decide who they might become in the future, hence broadening their time perspective into the future (Kerpelman & Mosher, 2004). When analyzing the premises of both the time-perspective main-effects model and the identity main-effects model, it becomes clear that these models are not mutually exclusive. Their integration gives rise to a third, more complex model, the reciprocal model, which assumes that identity formation and time perspective develop as part of a transactional system with both mechanisms influencing and reinforcing each other (Caspi & Roberts, 1999; Nuttin & Lens, 1985; Seginer & Noyman, 2005). As such, based on extant theoretical literature and on the existence of bidirectional associations found between identity and several personality constructs (e.g., Luyckx, 9 Soenens, et al., 2006), we hypothesized that the reciprocal model would be substantiated by our longitudinal data and would provide the best fit to the data. Method Participants and Procedure Two-wave longitudinal data (with the two waves being four months apart) were collected at a university in Flanders, the Dutch-speaking part of Belgium. This university mainly attracts Caucasian students from middle-class backgrounds. Participation in the study was voluntary, anonymity was guaranteed, and participants received course credit for attending the group testing sessions. The sample at Time 1 (i.e., November 2007) consisted of 371 freshman students (77.5% women) from the Faculty of Psychology and Educational Sciences. Mean age was 18.25 years (SD = 1.26). At Time 2, a subgroup of 309 of them participated again (representing an 83% participation rate). Participants with and without complete data were compared using Little’s (1988) Missing Completely At Random (MCAR) test. A nonsignificant test statistic suggests that missing values can be reliably estimated. We obtained such a nonsignificant chi-square value for this test, χ² (32) = 5.73, ns. Consequently, to minimize the bias associated with attrition and missing data (Schafer & Graham, 2002), we used the expectation maximization (EM) algorithm available in SPSS 11.5 to impute missing data enabling us to perform all primary analyses on the full sample of 371 participants. Measures Identity styles and commitment. We used the Identity Style Inventory – Version 4 (ISI-4; Smits et al., 2008) to measure the three identity styles (i.e., information-oriented, normative, and diffuse-avoidant styles) and commitment. Sample items are “When facing a life decision, I take into account different points of view before making a choice” (information-oriented; 7 items), “I think it is better to adopt a firm set of beliefs than to be open-minded” (normative; 8 items), “I’m not sure where I’m heading in my life; I guess things will work themselves out” (diffuse-avoidant; 9 items), and “I know basically what I believe and don’t believe” (commitment; 9 items). Participants responded to each item using a 5-point scale ranging from 1 (Not at all like me) to 5 (Very much like me). Smits and colleagues (2008) demonstrated that the ISI-4 scales had acceptable internal reliabilities, high one-week test-retest reliabilities, and adequate convergent validity with measures of identity status and identity content emphasis. In the present sample, Cronbach’s alphas for information-oriented, normative, diffuse-avoidant, and commitment were .78, .74, .77, and .79, respectively, at Time 1 and .79, .67, .77, and .84, respectively, at Time 2. Confirmatory Factor Analysis (CFA) was used to check the factor structure of the ISI-4 across both waves. For each latent variable, three parcels were created in a random fashion and 10 served as indicators of the latent variables (Marsh, Hau, Balla, & Grayson, 1998). Covariances between errors of the same parcels across the two waves were allowed (Wiley & Wiley, 1974). In all of the models estimated, we used standard model fit indices (Kline, 2006). The chi-square index, which tests the null hypothesis of perfect fit to the data, should be as small as possible; the Root Mean Square Error of Approximation (RMSEA) should be less than .08; and the Comparative Fit Index (CFI) should exceed .90 and preferably .95. Because preliminary analyses indicated some non-normality in the data, the Satorra-Bentler (1994) scaled chi-square statistic (SBS-χ²) was used. The model comprising the three identity styles and commitment at both measurement waves provided an adequate fit to the data (SBS-χ² (212) = 340.77, RMSEA = .04, CFI = .97). Additionally, the factor loadings of the same indicator variables were constrained as equal across time to assess measurement invariance across time (Marsh, 1994). The null hypothesis of invariance across time would be rejected if at least two of the following criteria were satisfied (Vandenberg & Lance, 2000): ΔSBS-χ² significant at p < .05; ΔCFI equals or is greater than .01; and the change in non-normed fit index (ΔNNFI) equals or is greater than .02. Although the NNFI was not used here to evaluate the fit of a single model, it is extremely sensitive to small deviations or differences in model fit and is a useful tool in invariance testing (Little, 1997). As indicated by the invariance test, the more parsimonious invariant model (SBS-χ² (224) = 379.60, RMSEA = .04, CFI = .97) fit the data equally well, ΔSBS-χ² (12) = 42.09, p < .001; ΔCFI < .01; and ΔNNFI < .01. Time perspective. We used the Zimbardo Time Perspective Inventory (ZTPI; Zimbardo & Boyd, 1999) to assess the present-hedonistic, present-fatalistic, and future time perspectives. Sample items are “I believe that getting together with one’s friends to party is one of life’s important pleasures” (present-hedonistic; 15 items), “Fate determines much in my life” (presentfatalistic; 9 items), and “I believe that a person’s day should be planned ahead each morning” (future; 13 items). Participants indicated the extent to which each item characterizes them on a 5point scale ranging from 1 (Very uncharacteristic) to 5 (Very characteristic). Cronbach’s alphas for present-hedonistic, present-fatalistic, and future were .78, .67, and .71, respectively, at Time 1 and .79, .69, and .72, respectively, at Time 2. CFA was again used to check the factor structure of the ZTPI across both waves. For each latent variable, three parcels were created in a random fashion which served as indicators and covariances between errors of the same parcels across the two waves were allowed. The model comprising the three perspectives at both measurement waves provided an adequate fit to the data (SBS-χ² (111) = 126.83, RMSEA = .02, CFI = .99). The factor loadings of the same indicator variables were again constrained as equal across time to assess measurement invariance. As 11 indicated by the invariance test, the more parsimonious invariant model (SBS-χ² (120) = 137.46, RMSEA = .02, CFI = .99) fit the data equally well, ΔSBS-χ² (9) = 11.87, p = .22; ΔCFI < .01; and ΔNNFI < .01. Results Stability and Mean-Level Changes Table 1 presents both the rank-order stability coefficients (i.e., the correlation between a variable assessed at the two adjacent time points) and mean-level changes in all study variables from Time 1 to Time 2. First, the stability coefficients for time perspective ranged from .58 to .70 (all ps < .001), whereas the stability coefficients for identity were somewhat lower and ranged from .47 to .59 (all ps < .001). Second, two sets (i.e., one for time perspective and one for identity styles and commitment) of multivariate repeated measures of ANOVA were conducted to asses mean-level changes across time. Overall multivariate effects of time perspective (Wilks’ λ = .92; F (3, 368) = 11.19; p < .001, η2 = .08) and identity (Wilks’ λ = .93; F (3, 368) = 7.01; p < .001, η2 = .07) were significant. Table 1 presents all follow-up univariate analyses for the two significant multivariate effects obtained. The present-hedonistic and present-fatalistic perspectives decreased across time, whereas the future perspective, information-oriented identity style, and commitment increased across time. Additional analyses with gender as a between-subjects factor indicated that stability and change in identity styles and commitment (Wilks’ λ = .98; F (4, 365) = 1.52; p = .19, η2 = .02) and in time perspective (Wilks’ λ = .99; F (3, 366) = 0.65; p = .59, η2 = .01) were not moderated by gender. Correlational Analyses Table 2 presents the correlations among all study variables at Times 1 and 2. As could be expected based upon the literature, the present-hedonistic and present-fatalistic perspectives (which were positively interrelated) were negatively related to the future perspective in both waves. Commitment was positively related to information-oriented style and negatively to diffuseavoidant style at both waves, with the latter being negatively related to the information-oriented style and positively to the normative style. As expected, the present-hedonistic perspective was positively related to the diffuse-avoidant style, the present fatalistic perspective was negatively related to information-oriented style and commitment, and positively to diffuse-avoidant style, and the future orientation was positively related to commitment and information-oriented style, and negatively to diffuse-avoidant style. A comparison between a model with these correlations left free to vary across time against a model with these correlations constrained equal across time – as outlined by Robins, Fraley, Roberts, and Trzesniewski (2001) – indicated that some of the associations among identity and time perspective were significantly different across time, ΔSBS-χ² (12) = 30.96, p < .01; ΔCFI = 12 .01; and ΔNNFI > .02. Ancillary analyses indicated – although differences were limited – that the correlations between the present-fatalistic perspective and identity were somewhat different across time (ΔSBS-χ² (4) = 16.93, p < .01; ΔCFI = .01; and ΔNNFI > .02). The correlations between the present-hedonistic perspective and identity (ΔSBS-χ² (4) = 3.74, p = .44; ΔCFI < .01; and ΔNNFI < .01) and between the future perspective and identity (ΔSBS-χ² (4) = 4.77, p = .31; ΔCFI < .01; and ΔNNFI = .01) did not differ substantially across time. Cross-Lagged Analyses Three subsequent models (i.e., a time-perspective main-effects model, an identity maineffects model, and a reciprocal model) using path analysis with Structural Equation Modeling (SEM) were tested. In all cross-lagged models being tested, all synchronous or within-time associations at Times 1 and 2 and all stability coefficients were controlled for (Asendorpf & van Aken, 2003). Fit indices of all models being tested are reported in Table 3. The baseline model including all synchronous relations and all stability coefficients (i.e., Model 1) was estimated and provided an adequate fit to the data (SBS-χ² (42) = 108.35, RMSEA = .07, CFI = .97). Second, we estimated the time-perspective main-effects model (i.e., Model 2) in which all paths from time perspective to identity commtiment and the three identity styles were allowed. Five of these paths were significant. The present-hedonistic perspective negatively predicted the normative style, the present-fatalistic perspective positively predicted the diffuseavoidant style, and, finally, the future-oriented perspective positively predicted the informationoriented style and identity commitment, and negatively the diffuse-avoidant style. The remaining paths were trimmed from the model, resulting in the more parsimonious Model 3. Third, the identity main-effects model was estimated (i.e., Model 4), in which all the paths from the three identity styles to the three time perspective scales were allowed. Four of these paths were significant. The information-oriented style and identity commitment positively predicted the future-oriented perspective, identity commitment negatively predicted the present-hedonistic perspective, and, finally, the diffuse-avoidant style positively predicted the present-fatalistic perspective. The remaining paths were trimmed from the model, resulting in the more parsimonious Model 5. Finally, the Reciprocal model (i.e., Model 6), in which all these significant paths are combined, provided the best fit to the data (SBS-χ² (33) = 59.10, RMSEA = .05, CFI = .99). Figure 1 represents a graphical presentation of Model 6. For reasons of clarity, all synchronous associations, stability coefficients, and non-significant cross-lagged paths are omitted from the figure. To assess whether the structural relationships in Model 6 were invariant across gender, a multigroup analysis was performed. As part of this analysis, we compared a constrained model (with all cross-lagged path coefficients set as equal across gender) against an unconstrained model 13 (with all cross-lagged path coefficients allowed to vary across gender). No significant difference emerged (ΔSBS-χ² (8) = 6.29, p = .61; ΔCFI < .01; and ΔNNFI < .01) indicating that the model fit equally well for men and for women. Discussion The present study was the first to examine longitudinal associations between the identity styles and commitment on the one hand and present and future time perspective orientations on the other hand. College students were sampled given that identity formation and preparing for the future and adult roles come to the fore and further intensify in emerging adulthood and especially in the college context. Our results illustrated that identity formation and time perspective are interdependent developmental processes mutually reinforcing one another across time. Development and Longitudinal Associations Although time perspective is hypothesized to be a fundamental psychological process acquired at an early age through cultural, psychosocial, and family influences, continuous development is hypothesized to occur due to contextual influences and the resolution of simultaneous developmental tasks (Hulbert & Lens, 1988; Lens & Gailly, 1980; Nurmi, 1991, 1994). For instance, the college context – being a primarily future-oriented context – may stimulate being future-oriented and lead to decreases in both present-oriented time perspectives. Despite the short time interval between both measurement waves in the present study, the freshman students of the present sample indeed became more future-oriented and less presentoriented (and especially less present-fatalistic oriented) across time. Similarly, previous research indicated that the college context with its numerous alternatives and possible life paths can foster and stimulate identity exploration and commitment (Luyckx, Goossens, et al., 2006; Montgomery & Côté, 2003). Individuals in the present sample indeed became more information-oriented (indicative of an increasing amount of pro-active identity exploration) and were more able to form steady and self-endorsed commitments across time. Apparently, these first-year college students seemed to be increasingly preparing themselves for their academic career and future adult roles by tackling self-defining identity issues more pro-actively across time and by extending their time perspective well into the future. Further, the concurrent associations and cross-lagged effects pointed to important developmental mechanisms which were supported with various degrees of convergence between the two sets of analyses conducted. Substantial concurrent associations and reciprocal influences. At both time points, future time perspective was positively related to identity commitment and the information-oriented style. Further, these two sets of variables were reciprocally interrelated across time with future time perspective positively predicting commitment and the information-oriented style, and vice versa. The present-fatalistic time perspective was positively related to the diffuse-avoidant style at both 14 time points and both these variables were also reciprocally related across time with the presentfatalistic time perspective positively predicting the diffuse-avoidant style, and vice versa. Apparently, the goal-directed and efficient work style of future-oriented individuals characterized by a tendency to invest in the construction of the individual future goes hand in hand with proactive identity exploration and the formation of identity commitments over time. A presentfatalistic orientation, on the other hand, is developmentally linked to the use of a diffuse-avoidant style. Individuals scoring high on the diffuse-avoidant style view themselves as being predetermined by fate and factors over which they have little control; they ground their sense of self in immediate considerations, such as reputation and impressions on others. Similarly, Bohart (1993) already argued that being future oriented with its internal locus of control – as opposed to the present-fatalistic time orientation which is characterized by an external locus of control and amotivation - is a fundamental asset for human development, allowing for opportunities, a sense of agency, and the making of choices. Substantial concurrent associations and unidirectional influences. In line with the arguments forwarded by Bohart (1993), future time perspective was negatively related to the diffuse-avoidant style at both time points and negatively predicted the diffuse-avoidant style across time. Further, at Time 2, the present-hedonistic time perspective scale was negatively related to the normative style and identity commitment. Whereas the present-hedonistic scale was found to negatively predict the normative style across time in the longitudinal analyses, commitment was found to negatively predict the present-hedonistic scale across time. Apparently, making strong identity commitments negatively predicted a dominant reliance on merely enjoying the present and living for the moment. By deciding on which path one’s life has to take, individuals seem to be protected against an overreliance on thrill or sensation seeking and hedonistic values. A present-hedonistic time perspective, in turn, negatively predicted accommodating to the norms and prescriptions upheld by others in the quest for one’s identity, probably given the fact that indulging oneself in a presenthedonistic lifestyle is at odds with the expectations upheld by the academic setting and by parents, who are most likely to emphasize future-oriented goals and aspirations. Substantial concurrent associations. Finally, stable concurrent associations were found which did not appear as cross-lagged effects. At both time points, the diffuse-avoidant style was positively related to the present-hedonistic perspective (besides its positive correlation with the present-fatalistic perspective as described earlier). Apparently, using a diffuse-avoidant style is associated with both an unbothered or pleasure-seeking approach strongly grounded in the present as well as with feelings of hopelessness with respect to the future. Luyckx, Goossens, Soenens, Beyers, and Vansteenkiste (2005) made a distinction within Marcia’s (1966) diffusion status that resembles this pattern of associations. Individuals in the carefree diffusion status quite liked their 15 uncommitted identity state, whereas individuals in the diffused diffusion status were characterized by maladjustment due to their lack of identity work. Future research should investigate whether individuals in carefree diffusion would be mainly characterized by a present-hedonistic orientation, whereas individuals in diffused diffusion would be mainly characterized by a presentfatalistic orientation. At both time points, the present-fatalistic time perspective was negatively related, as expected, to both the information-oriented style and identity commitment. In general, presentfatalistic individuals feel helpless and hopeless (Zimbardo & Boyd, 1999), which could leave them unmotivated to tackle identity-relevant issues. Further, the future time perspective scale was found to be positively related to the normative style, although not as strong as to the information-oriented style or to identity commitment. Apparently, being in a future-oriented context such as the college setting could lead normative individuals to adopt a future-oriented time perspective, although not as self-endorsed and profound as information-oriented individuals do (as illustrated by the modest concurrent associations and the lack of cross-lagged associations between the normative style and future time perspective). Limitations and Suggestions for Future Research The present study shows a number of limitations which need to be remedied in future research. First, with respect to the measurement of time perspective, the present study focused only on the present and future time perspective and no attention was paid to the past orientation. Although most research focused only on the present and future time perspective (Zimbardo & Boyd, 1999), the distinction between the past-positive and past-negative time perspective might prove to be fruitful in the study of identity formation. For instance, individuals who are characterized mainly by a past-negative time perspective (which is characterized by a worrisome, negative, and aversive view of the past) might be hindered in their pro-active search for a personalized identity (Luyckx et al., 2008). Or, one would expect normative individuals, being strongly focused on referent groups, to be past-oriented and to use well-established traditions and rules as a reference frame. Narrative identity research already emphasized the importance of the past in identity construction by stating that individuals build and internalize life stories to integrate the reconstructed past and the imagined future which provides life with purpose, meaning, and coherence. As such, narrative identity plays a key role in well-being and happinness (Bauer, McAdams, & Pals, 2008). Further, the use of all three time perspective scales (i.e., past, present, and future) would allow researchers to link identity formation to the balanced time perspective construct, which is characteristic of individuals scoring moderate to high on the past-positive, present-hedonistic, and future time perspectives, and, as such, is hypothesized to be the most 16 optimal outlook on psychological time (Drake, Duncan, Sutherland, Abernethy, & Henry, 2008; Zimbardo & Boyd, 1999). Second, the present sample represented a predominantly female student population with a particular background (i.e., students in Psychology and Educational sciences). Therefore, we can not state with certainty that our findings can be generalized to populations with another educational background. Further, we can not generalize our findings to individuals beyond the age range examined. As such, research units from other cultures working with diverse populations of adolescents are encouraged to focus on similar research questions. For instance, previous research indicated that a future time perspective was positively related to higher socio-economic status and higher education (Aspinwall, 2005; Zimbardo & Boyd, 1999). Given the developmental associations obtained in the present study, such findings emphasize the need to chart the process of identity formation in less-privileged youth around the globe and the barriers they face (Galambos & Martínez, 2007; Schwartz, 2005; Yoder, 2000). Third, all measures were self-administered questionnaires. Although questionnaires are most appropriate to gather information about internal and subjective processes such as identity development and time perspective, it should be noted that the sole reliance on a single informant may artificially inflate correlations among constructs (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Finally, the time-span under consideration was relatively short. Given that major developmental changes in both time perspective and identity formation are likely to occur only in the long run, long-term longitudinal research is needed that tracks time perspective and identity formation from early to mid-adolescence through emerging adulthood well into the adult years. Such research would do well to investigate how such changes in time perspective and identity relate to psychosocial outcomes, such as self-esteem, depression, drug use, and sexual risk taking, and to include additional variables (e.g., the Big Five of personality) that might influence the covariation between time perspective and identity formation. Despite these limitations, we believe that the present results broaden our knowledge of college students’ psychological functioning by indicating how one’s outlook on psychological time is related to how one undertakes the process of identity formation. Given that both are conceived of as core developmental tasks young people face in today’s society, we encourage researchers from different cultural backgrounds to examine similar research questions in different populations and settings. 17 References Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469-480. Arnett, J. J. (2004). Emerging adulthood: The winding road from the late teens through the twenties. Oxford, UK: Oxford University Press. 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. Aspinwall, L. G. (2005). The psychology of future-oriented thinking: From achievement to proactive coping, adaptation, and aging. Motivation and Emotion, 29, 203-235. Bauer, J. J., McAdams, D. P., & Pals, J. L. (2008). Narrative identity and eudaimonic well-being. Journal of Happiness Studies, 9, 81-104. Berzonsky, M. D. (1990). Self-construction over the life-span: A process perspective on identity formation. In G. J. Neimeyer & R. A. Neimeyer (Eds.), Advances in personal construct psychology (Vol. 1, pp. 155-186). Greenwich, CT: JAI Press. Berzonsky, M. D. (1992). Identity style and coping strategies. Journal of Personality, 60, 771-788. Berzonsky, M. D., & Adams, G. R. (1999). Reevaluating the identity status paradigm: Still useful after 35 years. Developmental Review, 19, 557-590. Berzonsky, M. D., & Kuk, L. S. (2000). Identity status, identity processing style, and the transition to university. Journal of Adolescent Research, 15, 81-98. Berzonsky, M. D., Nurmi, J., Kinney, A., & Tammi, K. (1999). Identity processing style and cognitive attributional strategies: Similarities and difference across different contexts. European Journal of Personality, 13, 105-120. Bohart, A. C. (1993). Emphasizing the future in empathy responses. Journal of Humanistic Psychology, 33, 12-29. Boniwell, I., & Zimbardo, P. G. (2004). Balancing one’s time perspective in pursuit of optimal functioning. In P. A. Linley & S. Joseph (Eds.), Positive psychology in practice (pp. 165180). Hoboken, NJ: Wiley. Bosma, H. A., & Kunnen, S. E. (2001). Determinants and mechanisms in ego identity development: A review and synthesis. Developmental Review, 21, 39-66. Caspi, A., & Roberts, B. W. (1999). Personality continuity and change across the life course. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 300-326). New York: Guilford Press. 18 Côté, J. E., & Levine, C. G. (2002). Identity formation, agency, and culture: A social psychological synthesis. Mahwah, NJ: Erlbaum. D’Alessio, M., Guarino, A., De Pascalis, V., & Zimbardo, P. G. (2003). Testing Zimbardo’s Stanford Time Perspective Inventory (STPI) – Short Form: An Italian study. Time and Society, 12, 333-347. Drake, L., Duncan, E., Sutherland, F., Abernethy, C., & Henry, C. (2008). Time perspective and correlates of well-being. Time and Society, 17, 47-61. Erikson, E. H. (1968). Identity, youth and crisis. New York: Norton. Fadjukoff, P., Kokko, K., & Pulkkinen, L. (2007). Implications of timing of entering adulthood for identity achievement. Journal of Adolescent Research, 22, 504-530. Galambos, N. L., & Martínez, M. L. (2007). Poised for emerging adulthood in Latin America: A pleasure for the privileged. Child Development Perspectives, 1, 109-114. Grotevant, H. D. (1987). Toward a process model of identity formation. Journal of Adolescent Research, 2, 203-222. Henson, J. M., Carey, M. P., Carey, K. B., & Maisto, S. A. (2006). Associations among health behaviors and time perspective in young adults: Model testing with boot-strapping replication. Journal of Behavioral Medicine, 29, 127-137. Hulbert, R., & Lens, W. (1988). Time and self-identity in later life. International Journal of Aging and Human Development, 27, 127-137. Keough, K. A., Zimbardo, P. G., & Boyd, J. N. (1999). Who’s smoking, drinking, and using drugs? Time perspective as a predictor of substance abuse. Basic and Applied Social Psychology, 21, 149-164. Kerpelman, J. L., & Mosher, L. S. (2004). Rural African American adolescents’ future orientation: The importance of self-efficacy, control and responsibility, and identity development. Identity: An International Journal of Theory and Research, 4, 187-208. Kline, R. B. (2006). Principles and practices of structural equation modeling (2nd edition). New York: Guilford Press. Lennings, C. J. (1998). Profiles of time perspective and personality: Developmental considerations. Journal of Psychology, 132, 629-642. Lens, W. (1986). Future time perspective: A cognitive-motivational concept. In D. R. Brown & J. Veroff (Eds.), Frontiers of motivational psychology (pp. 173-190). New York: SpringerVerlag. 19 Lens, W. (2006). Future time perspective: A psychological approach. In Z. Uchnast (Ed.), Psychology of time: Theoretical and empirical approaches (pp. 51-64). Lublin, Poland: Wydawnictwo KUL. Lens, W., & Gailly, A. (1980). Extension of future time perspective in motivational goals of different age groups. International Journal of Behavioral Development, 3, 1-17. Lens, W., & Moreas, A. (1994). Future time perspective: An individual and a societal approach. In Z. Zaleski (Ed.), Psychology of future orientation (pp. 23-38). Lublin, Poland: Towarzystwo Naukowe KUL. Lens, W., & Tsuzuki, M. (2007). The role of motivation and future time perspective in educational and career development. Psychologica, 46, 29-42. Levinson, D. J. (1978). The seasons of a man’s life. New York: Ballantine. 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. Little, T. D. (1997). Mean and covariance structures (MACS) analyses of cross-cultural data: Practical and theoretical issues. Multivariate Behavioral Research, 32, 53-76. 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., Goossens, L., Soenens, B., Beyers, W., & Vansteenkiste, M. (2005). Identity statuses based upon four rather than two identity dimensions: Extending and refining Marcia’s paradigm. Journal of Youth and Adolescence, 34, 605-618. 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, 5882. Luyckx, K., Soenens B., Berzonsky, M. D., Smits, I., Goossens, L., & Vansteenkiste, M. (2007). Information-oriented identity processing, identity consolidation, and well-being: The moderating role of autonomy, self-reflection, and self-rumination. Personality and Individual Differences, 43, 1099-1111. Luyckx, K., Soenens, B., & Goossens, L. (2006). The personality-identity link in emerging adult women: Convergent findings from complementary analyses. European Journal of Personality, 20, 195-215. 20 Marcia, J. E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3, 551-558. Marcia, J. E. (1993). The status of the statuses: Research review. In J. E. Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), Identity: A handbook for psychosocial research (pp. 22-41). New York: Springer-Verlag. Marsh, H. W. (1994). Confirmatory factor analysis models of factorial invariance: A multifaceted approach. Structural Equation Modeling, 1, 5-34. Marsh, H. W., Hau, K. T., Balla, J. R., & Grayson, D. (1998). Is more ever too much? The number of indicators per factor in confirmatory factor analysis. Multivariate Behavioral Research, 33, 181-220. 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. Nurmi, J.-E. (1991). How do adolescents see their future? A review of the development of future orientation and planning. Developmental Review, 11, 1-59. Nurmi, J.-E. (1994). The development of future orientation in life-span context. In Z. Zaleski (Ed.), Psychology of future orientation (pp. 63-74). Lublin, Poland: Towarzystwo Naukowe KUL. Nurmi, J.-E., Poole, M. E., & Seginer, R. (1995). Tracks and transitions: A comparison of adolescent future-oriented goals, explorations, and commitments in Australia, Israel, and Finland. International Journal of Psychology, 30, 355-375. Nuttin, J., & Lens, W. (1985). Future time perspective and motivation: Theory and research method. Leuven & Hillsdale, NJ: Leuven University Press and Erlbaum. Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. Pulkkinen, L., & Rönka, A. (1994). Personal control over development, identity formation, and future orientation as components of life orientation: A developmental approach. Developmental Psychology, 30, 260-271. Robins, R. W., Fraley, R. C., Roberts, B. W., & Trzesniewski, K. H. (2001). A longitudinal study of personality change in young adulthood. Journal of Personality, 69, 617-640. 21 Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. Von Eye & C. C. Clogg (Eds.), Latent variables analysis: Applications for developmental research (pp. 399-419). Thousand Oaks, CA: Sage. Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147-177. Schwartz, S. J. (2005). A new identity for identity research: Recommendations for expanding and refocusing the identity literature. Journal of Adolescent Research, 20, 293-308. 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. 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, 309328. Seginer, R., & Noyman, M. S. (2005). Future orientation, idfentity, and agency: Their relations in emerging adulthood. European Journal of Developmental Psychology, 2, 17-37. Seginer, R., Vermulst, A., & Shoyer, S. (2004). The indirect link between perceived parenting and adolescent future orientation: A multiple-step model. International Journal of Behavioral Development, 28, 365-378. Simons, J., Vansteenkiste, M., Lens, W., & Lacante, M. (2004). Placing motivation and future time perspective theory in a temporal perspective. Educational Psychology Review, 16, 121139. Smits, I., Berzonsky, M. D., Soenens, B., Luyckx, K., Goossens, L., Kunnen, S. E., & Bosma, H. A. (2008). The Identity Style Inventory – 4. Internal research report, Catholic University Leuven, Belgium. Soenens, B., Berzonsky, M. D., Vansteenkiste, M., Beyers, W., & Goossens, L. (2005). Identity styles and causality orientations: In search of the motivational underpinnings of the identity exploration process. European Journal of Personality, 19, 427-442. Soenens, B., Duriez, B., & Goossens, L. (2005). Social-psychological profiles of identity styles: Attitudinal and social-cognitive correlates in late adolescence. Journal of Adolescence, 28, 107-125. 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. 22 Waterman, A. S. (1993). Developmental perspectives on identity formation: From adolescence to adulthood. In J. E. Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), Identity: A handbook for psychosocial research (pp. 42-68). New York: SpringerVerlag. Waterman, A. S. (1999). Identity, the identity statuses, and identity status development: A contemporary statement. Developmental Review, 19, 591-621. Wiley, J. A., & Wiley, M. G. (1974). A note on correlated errors in repeated measurements. Sociological Methods and Research, 3, 172-188. Yoder, A. E. (2000). Barriers to ego identity status formation: A contextual classification of Marcia’s identity status paradigm. Journal of Adolescence, 23, 95-106. Zimbardo, P. G., & Boyd, J. N. (1999). Putting time in perspective: A valid, reliable individualdifference metric. Journal of Personality and Social Psychology, 77, 1271-1288. Zimbardo, P. G., Keough, K. A., & Boyd, J. N. (1997). Present time perspective as a predictor of risky driving. Personality and Individual Differences, 23, 1007-1023. 23 Table 1 Mean-Level Changes and Stability Coefficients of Time Perspective and Identity Formation Variable r (T1, T2) M (SD) M (SD) F (eta²) 1. Present-hedonistic perspective .68*** T1 3.41 (.45) T2 3.36 (.44) 5.77* (.01) 2. Present-fatalistic perspective .58*** 2.86 (.49) 2.78 (.49) 11.38*** (.03) 3. Future-oriented perspective .70*** 3.27 (.45) 3.36 (.46) 21.09*** (.05) 4. Commitment .59*** 3.46 (.55) 3.54 (.52) 8.81** (.02) 5. Information-oriented style .51*** 3.93 (.46) 4.04 (.45) 21.38*** (.06) 6. Normative style .47*** 2.62 (.53) 2.59 (.43) 1.52 (< .01) 7. Diffuse-avoidant style .57*** 2.57 (.59) 2.54 (.50) 1.55 (< .01) Note. r (T1, T2) = stability coefficient from T(ime) 1 to T(ime) 2; M = mean; SD = standard deviation. * p < .05. ** p < .01. *** p < .001. 24 Table 2 Correlations Between Time Perspective and Identity Formation at Times 1 and 2 Variable 1. Present-hedonistic perspective 1. 2. 3. 4. 5. 6. 7. -- .32*** -.37*** -.04 .01 -.07 .20** 2. Present-fatalistic perspective .43*** -- -.29*** -.32*** -.14* .09 .47*** 3. Future-oriented perspective -.39*** -.27*** -- .29*** .44*** .13* -.33*** 4. Commitment -.11* -.16** .34*** -- .20*** -.08 -.48*** 5. Information-oriented style -.07 -.19** .35*** .34*** -- .07 -.24*** 6. Normative style -.12* .05 .12* -.02 -.03 -- .24*** -.36*** -.49*** -.42*** 7. Diffuse-avoidant style .21*** .31*** Note. Values above the diagional are from Time 1, values below the diagonal are from Time 2. * p < .05. ** p < .01. *** p < .001. .23*** -- 25 Table 3 Fit Indices of the Various Models Tested Model Description Fit Indices df SBS-χ² RMSEA CFI NNFI Model 1: Baseline model 42 108.35 .07 .97 .94 Model 2: Time-perspective main-effects model 30 69.52 .06 .98 .95 Model 3: Trimmed Model 2 37 75.82 .05 .98 .96 Model 4: Identity main-effects model 30 88.52 .07 .98 .93 Model 5: Trimmed Model 4 38 91.95 .06 .98 .94 Model 6: Reciprocal model 33 59.10 .05 .99 .97 Note. df = degrees of freedom; SBS-χ² = Satorra-Bentler Scaled chi-square; RMSEA = Root Mean Square Error of Approximation; CFI = Comparative Fit Index; NNFI = Non-Normed Fit Index. The NNFi was not used to evaluate the model fit of signle models but was used to assess the differences in model fit between subsequent models. 26 Figure captions Figure 1. Final cross-lagged path model linking time perspective to identity commitment and exploration styles. Within-time correlations and stability coefficients are not presented for reasons of clarity. All path coefficents are standardized. * p < .05. ** p < .01. *** p < .001. 27 Figure 1. Time 1 Time 2 Informationoriented style Informationoriented style .12* Normative style Normative style -.12* Diffuseavoidant style Diffuseavoidant style .10* -.18*** Commitment Commitment .18*** -.08* Presenthedonistic Presenthedonistic .10* Presentfatalistic Futureoriented Presentfatalistic .10* .09* Futureoriented