Economic Insecurity and Separation–individuation in Portuguese Emerging Adults

Scholars have widely agreed that the transition to adulthood has become destandardized and more extended in time. Although this prolonged transition to adulthood is common in late-modern societies, it is not universal, and the pathways towards adulthood may vary according to contextual factors and personal resources. This study examined the association between perceived economic insecurity and the process of separation–individuation regarding the romantic partner, and explored the mediating role of personal resources (self-esteem, depression). Participants were 387 Portuguese emerging adults (mean age = 22.32 years; SD = 3.47) involved in a romantic relationship, from different economic and cultural backgrounds. Findings revealed that self-esteem and depressive symptoms totally mediated the relationship between economic insecurity and separation–individuation, with higher levels of self-esteem and lower levels of depression being associated with a more successful process of separation–individuation. This study contributes to shed some light on the interaction between developmental processes and social context variables in the transition to adulthood.


Introduction
Over the last decades, scholars have widely agreed that the transition to adulthood has become destandardized and is now more extended in time (e.g., Arnett, 2000;Billari, 2004;Buchmann and Kriesi, 2011;Žukauskienė, 2016).Young people are allowed an extended period of exploration and experimentation, trying-out different pathways (e.g., in education, work, living conditions and romantic partners) before engaging into long-term commitments (Arnett, 2007;Luyckx et al., 2011).The dissociation in the transitions between education and labour market, dating, marriage and parenting have led J. Arnett (2000) to label the period between the late teens and the twenties as emerging adulthood, a time where young people are not adolescents anymore, but have also not yet reached a full adult status.Although this prolonged transition to adulthood is common in late-modern societies, it is not universal, and the pathways towards adulthood may highly vary, according to contextual factors.Several authors have drawn attention to the mechanisms of social exclusion and precariousness that have lowered young people's status (Côté and Bynner, 2008;Moreno et al., 2012;Žukauskienė, 2016).Recent studies began to point out the role that personal resources (such as self-efficacy, agency and identity capital) have in the ability to develop a sense of adulthood and to deal with instability and precariousness (Côté, 2002;Luyckx et al., 2011;Oliveira et al., 2014).However, to our knowledge, no study has explored how an instable and precarious life situation (expressed in perceived economic insecurity) may impact personal psychological resources (e.g., self-esteem and absence of depression), and, in turn, emerging adult's individuation.This study approaches this mediating hypothesis, assessing the role of psychological functioning between perceived economic insecurity and the process of separation-individuation, in a sample of Portuguese emerging adults from different economic and cultural backgrounds.
The literature on financial stress and economic pressure has long documented the impact of the daily difficulties created by financial hardship on personal wellbeing and on the quality of adult close relationships (e.g., Conger et al., 1993Conger et al., , 1999;;Dakin and Wampler, 2008).Economic pressure concerns indicators such as being instable to purchase basic necessities or pay bills, having to reduce normal expenditures or experiencing daily difficulties that follow from limited income or high financial demands (Conger et al., 1999).Situations of financial pressure and economic insecurity, and the perception of high instability seem to be associated with depression, anxiety, anger and frustration and lower well-being (Dooley et al., 2000;Davis and Mantler, 2004;Kuwabara et al., 2007;Luyckx et al., 2011;Serido, 2010;Thönnissen et al., 2006;Thönnissen et al., 2008).Also, studies showed that couples and newlyweds experiencing economic disadvantage, financial difficulties and lack of social support report lower marital satisfaction, poor relationship quality and higher individual distress (Carlson et al., 2004;Dakin and Wampler, 2008;Karney and Bradbury, 2005).
Separation-individuation can be defined as the development of independence and autonomy in the context of close relationships with significant others (Mattanah et al., 2004).Studies showed that the resolution of the separation-individuation process has consequences in the adjustment to new contexts in emerging adulthood (e.g., Mattanah et al., 2004;Tanner, 2006).Emphasizing the co-constructional nature of the separation-individuation process, most research focused on the role of family structure and dynamics to the healthy development of a differentiated sense of self (e.g., Kruse and Walper, 2008;Kins et al., 2011).However, the period of emerging adulthood poses specific challenges to this developmental process, and the development of intimacy and autonomy within romantic relationships can be defined as one of the main developmental tasks that contributes to the understanding of the trajectories to adulthood and of the way young people negotiate romantic commitments and life plans (Aquilino, 2006;Collins and van Dulmen, 2006;Shulman and Connolly, 2013).Nevertheless, separation-individuation in the context of emerging adult's romantic relationships has been rarely addressed (see Brandão et al., 2012;Haws and Mallinckrodt, 1998;Thönnissen et al., 2010, for exceptions).
During emerging adulthood, the absence of institutional and political support towards autonomy can be overwhelming for those less equipped with social and personal resources (Côté, 2006;Moreno et al., 2012;Robinson, 2016;Schoeni and Ross, 2005;Schulenberg and Zarrett, 2006).This lack of adequate structural support poses a big challenge for emerging adults dealing with major life tasks that are milestones for the future adult life.Although it is a period marked by optimism and hope for the future, it is also the life phase with higher vulnerability for depressive and behavioural disorders (Kuwabara et al., 2007;Moreno et al., 2012;Schulenberg and Zarrett, 2006).Research supports that one in four emerging adults in the ages from 18 to 25 years will experience a depressive episode.This age group presents the highest incidence and cumulative prevalence of depression in the overall population (Kuwabara et al., 2007;O'Connor et al., 2011).
Especially in Southern Europe, family bonds appear as major resources in dealing with instability, developing life projects and a sense of well-being (Billari, 2004;Buchmann and Kriesi, 2011;Crocetti and Tagliabue, 2016;Mendonça and Fontaine, 2013;Oliveira et al., 2014;Saraiva and Matos, 2016;Vogel, 2002).More recently, Southern European countries have suffered the consequences of the global economic crisis.According to the Statistics Portugal (INE, 2014), between 2011 and 2013 the unemployment rate for people between 15 and 29 years old was almost the double of the unemployment rate for the general population (26.3 and 14.8 per cent, respectively) (see Saraiva and Matos, 2016, for more descriptors on the Portuguese society).In the European context, Portuguese and Italian parents contributed the most for their children's incomes (an average of €146 and €102 per month, respectively).
The associations between quality of romantic relationship and well-being, specially self-esteem and depression, are well documented (Demir, 2008;Galambos et al., 2006;Johnson et al., 2012;Joyner and Udry, 2000;Karney and Bradbury, 2005).Research suggests that relationship intimacy and individuation are relevant components in emerging adults' development having a beneficial impact on selfesteem and depression.Our research focuses on this association assuming its twofold nature.Not only a healthy romantic relationship contributes to boost one's sense of personal well-being, but also depression and self-esteem levels function as internal resources that influence romantic relationship quality (Chow and Ruhl, 2014;Vujeva and Furman, 2011;Shulman et al., 2014).In line with identity capital theory (Côté, 2002(Côté, , 2006)), these two variables may represent emerging adults' availability and capacity to deal with challenges associated with individuation.
In this study, we considered two crucial aspects of well-being, self-esteem and depressive mood, as psychological resources that can be seen as measures of general adaptation and psychological functioning and that come into play in the development of autonomy (Galambos et al., 2006).Self-esteem can be understood as a global appreciation of one's self-worth, and all the thoughts and feelings towards the self as a whole (Rosenberg, 1965).Depressive mood refers to the recurrent experience of feelings of loss, guilt and worthlessness, helplessness and hopelessness (Radloff, 1977).

The Present Study
This study was developed as part of the research group YAGISSP (Young Adults in Germany, Italy, Sweden, Spain and Portugal), that is concerned with the life trajectories of emerging adults across Europe.We understand 'emerging adulthood' as a generative concept that takes into account the changes in the experience of becoming an adult.Here, we intend to explore the difficulties in the process of separation-individuation towards the romantic partner in the Portuguese context, looking at contextual and personal variables.Previous studies of the same research group (Thönnissen et al., 2006(Thönnissen et al., , 2008) ) have shown that emerging adults who perceive higher economic insecurity showed lower levels of well-being and greater difficulties in separation-individuation towards parents.We expected that the perceived economic insecurity of Portuguese emerging adults was associated with difficulties in the process of separation-individuation, this time towards the romantic partner.
We examined whether the perceived economic insecurity predicted the difficulties in separation-individuation towards the romantic partner, exploring the mediating role of psychological well-being (i.e., self-esteem and depression).We expected that, the more negative the economic situation was perceived, the more difficulties in the process of separation-individuation would be reported.Since changes in one's economic insecurity seem to have an impact on well-being (Dooley et al., 2000;Galambos et al., 2006;Thönnissen et al., 2006Thönnissen et al., , 2008)), we anticipated that the perceived economic insecurity would be associated with lower levels of self-esteem and higher levels of depressive mood.Also, we expected that self-esteem and depression would be connected to one's ability to cope with challenges within developmental tasks (cf.Schulenberg and Zarrett, 2006), and therefore, connected to more difficulties in the separation-individuation process.Considering that the financial support from the family plays a major role on the attainment of economic independence in Southern European countries, we used the perception of one's and the parents' economic situation as an indicator of perceived economic insecurity.

Participants
Participants were 387 Portuguese emerging adults (136 male and 251 female; mean age = 22.32 years; SD = 3.47; range = 18-30).Participants were university students, attending various courses at the University of Porto (76 per cent; n = 244) and students with a non-university background, attending professional and technological schools in Porto (24 per cent; n = 76).
All participants were in a romantic relationship-an inclusion criterion with a mean length of 38.85 months (SD = 33.67,min = 1 month, max = 203).The majority of participants were living with at least one of the parents (74 per cent); 15 per cent were living with their romantic partner/children and 11 per cent were living alone, with friends or in another situation.Only 34 per cent of the participants had a remunerated activity or job, working, on average, 28.70 hours per week (SD = 14.54).

Procedures
Data were collected in university settings and professional/technological schools during class periods.Questionnaires were anonymous and filled out voluntarily after being informed about the purpose and confidentially of the study.The response rate to the survey was 100 per cent.Participants were not offered any incentives or compensation.

Economic Insecurity
The current financial situation and parent's current financial situation (Schwarz et al., 1997) assesses the perception of the general experience of financial hardship and the difficulties to correspond to monthly financial demands.Participants responded to their current financial situation (four items; a = 0.71; e.g., 'I worry very often if I will be able to pay my bills') and their parent's current financial situation (four items; a = 0.81; e.g., reversed item 'My parents have enough money for all they need').It is rated on a 4-point Likert-type scale (1 = never; 4 = always).

Self-esteem
The Rosenberg Self-esteem Scale (Rosenberg, 1965) was developed to capture subjects' global perception of their own worth.Several studies have found a twofactor structure to the scale (self-confidence and self-depreciation) (Farruggia et al., 2004;Owens, 1993;Sheasby et al., 2000).It is rated on a 4-point Likert-type scale (1 = totally disagree to 4 = totally agree).In this study, we used four items for selfdepreciation (a = 0.73; e.g., 'I feel I do not have much to be proud of') and three items for self-confidence (a = 0.67; e.g., 'I like myself how I am').

Depressive Mood
The Center for Epidemiologic Studies Depression Scale (CES-D scale; Radloff, 1977) designed to measure symptoms associated with depression experienced in the past week.It comprises six scales reflecting major dimensions of depression (e.g., depressed mood, feelings of guilt and worthlessness and sleep disturbance).It is scored on a 4-point scale ranging from 1 (less than 1 day) to 4 (5-7 days).In this study, we used eight items (a = 0.84; e.g., 'During last week I felt lonely').

Difficulties in Separation-individuation in Relation to the Romantic Partner
Separation-individuation was measured with the Munich Individuation Test for Adolescents (MITA; Walper et al., 1996) version for the romantic partner.It assesses specific problems and successful individuation towards the romantic partner, within a 4-point Likert-type scale ranging from 1 (not true) to 4 (really true).We assessed the difficulties in separation-individuation through the following subscales: ambivalence (four items; a = 0.80; e.g., 'I would like to do more things with him/her, but I am afraid to be annoying'); fear of love withdrawal (four items; a = 0.78; e.g., 'Sometimes I am afraid he/she prefers to be with other people rather than with me') and engulfment anxiety (six items; a = 0.87; e.g., 'I have the feeling he/she is controlling me all the time').
As part of the larger YAGISSP project, a questionnaire was used to collect socio-demographic data (e.g., age, marital and residential status, current employment status).

Data Analysis
Missing data were imputed using expectation maximization (EM) which is an adequate method when data are missing at random and none of the items has more than 5 per cent of missing values (Tabachnick and Fidell, 2007).Descriptive analysis and Pearson correlations were conducted to examine the relationships among study variables.Correlations between variables and the variance inflation factor (VIF) were checked in order to examine multicollinearity.
To test our model, we used Structural Equation Modeling (SEM; AMOS 21) and the Maximum Likelihood Estimation Method.A two-step approach was followed as suggested in Mâroco (2010).First a measurement model composed by three latent variables was tested, namely (1) perceived economic insecurity (represented by two manifest variables namely own's financial situation and parent's financial situation), (2) psychological well-being (represented by three manifest variables namely selfdepreciation, self-confidence and depression) and (3) separation-individuation difficulties (represented by other three manifest variables namely engulfment anxiety, fear of love withdrawal and ambivalence).In the first step, all latent variables were allowed to freely intercorrelate to evaluate the quality of the measurement model.This allows, if necessary, to improve model fit, in order to ensure that a misfit of the structural model to the data is not because of the lack of quality of the measurement model (Mâroco, 2010).
Once the measurement model presented a good fit to the data, a structural model was tested to analyse the proposed mediational model.Model fit evaluation was based on the following indicators: the chi-square/df statistic (> 0.2), the Bentler comparative fit index (CFI), the goodness of fit index (GFI) (both > 0.90) and the root mean square error of approximation (RMSEA) (< 0.07) (Hooper et al., 2008).Mediation was tested and evaluated through bootstrap resampling procedure according to Preacher and Hayes (2008) guidelines (bootstrap estimates based on 5,000 bootstrap samples with 95 per cent bias-corrected confidence interval [CI] for the unstandardized effects).

Descriptive Statistics and Correlation Analysis
Means, standard deviations and Pearson correlations of variables are presented in Table 1.All variables presented low-to-moderate correlations and VIF values less than 10 suggesting an absence of multicollinearity (Hair et al., 1998;Kline, 2005).

Mediation Analysis
The measurement model evidenced very good fit to data (χ²/df = 1.41;CFI = 0.99; GFI = 0.98; RMSEA = 0.03, 90 per cent CI (0.00, 0.06), p close > 0.05).All factor loadings were > 0.40 and statistically significant (p < 0.001) indicating that the latent variables were adequately measured by their manifest variables (see Table 2).No modifications were made.Since a good measurement model was developed, a structural model testing the hypothesized mediation model in which economic insecurity would be associated with separation-individuation difficulties towards romantic partners in the presence of low levels of well-being (less self-esteem and more depressive symptoms) was tested.In this structural model, correlations were replaced by causal paths (see Figure 1).The model fit remains the same since no modifications were made on the measurement model and the number of degrees of freedom were equal (χ²/df = 1.41;CFI = 0.99; GFI = 0.98; RMSEA = 0.03, 90 per cent CI (0.00, 0.06), p close > 0.05).This model explained 20 per cent of the total variance of the separation-individuation difficulties. 1Results from bootstrap procedure showed a significant indirect effect of economic insecurity (in the presence of low levels of well-being) on separation-individuation difficulties (β = 0.11, p < 0.01, 95 per cent CI (0.05, 0.20).The direct effect of economic insecurity on separationindividuation difficulties was not statistically significant (β = 0.09, ns).
Since our data are cross-sectional an alternative model was tested (i.e., economic insecurity being a predictor of lower levels of well-being, with separationindividuation difficulties as a mediator).This model fitted well to the data and explained 23 per cent of the total variance of levels of well-being (see Figure 2).Results from bootstrap procedure showed a significant indirect effect of economic insecurity (in the presence of more separation-individuation difficulties) on levels of well-being (β = 0.08, p < 0.01, 95% CI (0.03, 0.15).However, the direct effect of economic insecurity on levels of well-being remained significant, indicating only a partial mediation (β = 0.19, p < 0.01), which supports our proposed model.

Discussion
This study investigated the process of separation-individuation towards the romantic partner and its association with the perception of one's economic insecurity and dimensions of well-being.We intended to contribute to the exploration of personal and contextual variables that have an impact in the resolution of individuation towards the romantic partner in emerging adulthood.Results suggest that economic insecurity has an impact on the difficulties of individuation towards the romantic partner.Those who express greater concerns with their own financial situation and the parents' economic conditions report higher levels of engulfment anxiety, ambivalence and fear of love withdrawal.These results are consistent with other studies on the impact of economic instability and hardship on individuation difficulties towards parents (Thönnissen et al., 2008) and on the quality of romantic relationship (Conger et al., 1999;Chen et al., 2006).The results of this study contribute to the body of research that sustains that the sudden decline of institutional structure and support that characterizes this transition has contributed to the growth of social inequalities and exclusion in latemodern societies, and can be particularly demanding for those less equipped with social, psychological and financial resources (Côté and Bynner, 2008;European Commission, 2005;Moreno et al., 2012).Results suggest that the perception of economic insecurity is associated with difficulties in the resolution of developmental tasks central to emerging adulthood.For this reason, findings reinforce the importance of considering contextual variables when offering adequate help and support during the transition to adulthood.As outlined by Dakin and Wampler (2008), therapists should be aware that economic insecurities are associated with higher levels of psychological distress when outlining their interventions, and making realistic assessments of the possible changes.
It is possible to consider that the household's economic situation and social economic status seems to impact the construction of autonomy and differentiation necessary to the resolution of developmental tasks (such as separation-individuation, future orientation and the capacity to establish love relationships; Dias and Fontaine, 2002;Meier and Allen, 2008).Young people that perceive difficulties in meeting financial demands experience a lower sense of autonomy, independency and agency.This can be understood as a source of stress and pressure that can create a lower feeling of support and involvement with the partner (Carlson et al., 2004;Dakin and Wampler, 2008;Davis and Mantler, 2004;Karney and Bradbury 2005).Precariousness and   instability, by reducing the possibility to project oneself into the future, can be associated with a decrease in self-esteem, as a sense of trust in one's capacities to deal with challenges and difficulties.The increase in depression levels can then be seen as an adaptive reaction, triggering self-exploration and adjustment of goals to available opportunities (Schulenberg and Zarrett, 2006).These concerns with the financial situation can also be associated with situations of unemployment or underemployment, which are related to a decrease in well-being and with depression (Dooley et al., 2000;Kuwabara et al., 2007;O'connor et al., 2011).
The involvement with a romantic partner and the capacity to develop a sense of autonomous self within the relationship appear as one of the most important challenges in this transition.Results suggested that it is not the financial stress per se that is associated with the difficulties in developing a sense of self inside a significant relationship, but that this variable seems to have an impact on psychological well-being that can interfere with the separation-individuation process.Several studies evidenced the association between dimensions of well-being, in general, and self-esteem and depression, in particular, with romantic relationships and intimacy (e.g., Johnson et al., 2012;Joyner and Udry, 2000).In line with identity capital theory (Côté, 2002(Côté, , 2006) and empirical research with Portuguese emerging adults (Oliveira et al., 2014), we propose that self-esteem and absence of depressive mood are psychological resources that are likely to be associated with one's agentic ability to manage the difficulties of the autonomy and individuation process.
The alternative model we tested (with economic insecurity as a predictor of lower levels of well-being and separation-individuation difficulties as a mediator) suggested that it is relevant to consider the two-fold association between wellbeing and the process of separation-individuation.Higher levels of self-esteem and absence of depression seem to contribute to a healthy resolution of the separationindividuation process towards the romantic partner, but also, less difficulties in the separation-individuation process seems to contribute to higher levels of well-being.This two-fold association is supported by prior research (e.g., Chow and Ruhl, 2014;Vujeva and Furman, 2011;Shulman et al., 2014).However, in this model difficulties in the separation-individuation only partially mediated the relationship between economic insecurity and well-being, emphasizing the direct associations between economic insecurity and feelings of well-being.Our model seems therefore to more accurately describe the mediating role of well-being on important developmental tasks of emerging adults.
Finally, we wish to address an interesting result that conveys some important research questions.Although the mean age of our participants is 22 years, the average relationship length reported was of 38 months.This means that many young people are engaged in long-term relationships since late adolescence.There is no current data and statistics on this subject in Portugal.This result can be explained by the existence of a more linear transition pattern in the transition to adulthood coupled with the presence of familistic values (rooted in the Catholic heritage) that reinforce stability and tradition (Guerreiro and Abrantes, 2007;Vogel, 2002).Also, it is possible that young people tend to invest in long-term romantic relationships, since they are related to the capacity to project oneself into the future and with the transition out of the parental home (Lanz and Tagliabue, 2007;Crocetti and Tagliabue, 2016).
This study presents several limitations that call for caution in the generalization of the results.The cross-sectional design does not permit to infer causal relationships.Also, as self-esteem and depression are not stable across time (Galambos et al., 2006;Joyner and Udry, 2000;Schulenberg and Zarrett, 2006), a longitudinal design would bring a more comprehensive view of causal relationships and effects between variables.The sample was a convenience sample, constituted mainly by female participants and students.It is important to note that, although previous studies (Conger et al., 1993;Conger et al., 1999) have showed gender differences in the response to financial stress, in our study gender did not have an impact on the relationship between the perception of the economic situation and separationindividuation difficulties towards the romantic partner.Future research should strive for a more age, sex and educational background diverse and balanced sample that could bring a wider comprehension on the evolution on developmental processes across emerging adulthood.
Also, more detailed criteria for the definition of a 'stable romantic relationship' are advisable.In our study, the participants were the ones who reported as 'currently being in a stable relationship'.It is possible that relationships with different lengths and nature bring about different dynamics for individuation.Finally, like any other studies using self-report measures these results have also some limitations given the responses' spontaneous nature and reliability, especially due to the effect of social desirability and idealization of romantic relationships.
Our results suggested that economic resources can boost or hinder the resolution of developmental tasks by influencing dimensions of well-being.Future research should consider other relevant variables, including from the partner and the dyad (e.g., autonomy, identity, intimacy, conflict resolution strategies) that can function as predictors of separation-individuation towards the romantic partner.Also, more research is necessary in the comprehension of socio-economic variables that have an impact on developmental tasks.
This study has contributed to shed some light on the dynamic interaction between psychological processes and social conditions in the transition to adulthood in a Portuguese sample.It suggests that becoming an adult is far from being a universal and categorical process, and more research is necessary to bring in a comprehensive view of regional and national specificities in the selection of pathways in this transition.Psychological intervention should consider the influence of contextual variables in the developmental processes in order to provide an integrated and personalized care.

Table 1 .
Means, Standard Deviations and Correlations Coefficients

Table 2 .
Factor Loadings for the Measurement Model Source: Authors' own.