Emerging adulthood is proposed to be a developmental period that can be distinguished from both adolescence and young adulthood, ranging from the age 18 to 25. Emerging adulthood differs from adolescence and young adulthood in terms of identity exploration, instability, self-focus, feeling in-between, and possibilities (Arnett, Reference Arnett2000). Foremost among these features is identity explorations in the areas of love, work, and worldviews. Following adolescence, emerging adults start to experience long-term committed relationships characterized by high level of intimacy and sharing (Arnett, Reference Arnett2015; Rhoades & Stanley, Reference Rhoades and Stanley2014). The achievement of intimacy in close relationships emerges as a central developmental task during emerging adulthood (Arnett, Reference Arnett2000). Yet, despite the theoretical contributions and empirical findings emphasizing the importance of intimate and committed relationships during emerging adulthood, some evidence suggests that dysfunctional interaction patterns experienced in romantic relationships pose a threat to mental health (e.g., Simon & Barrett, Reference Simon and Barrett2010). Moreover, some researchers argue that there have been many ups and downs in romantic relationships during the period of emerging adulthood (Arnett, Reference Arnett2015). Fincham and Cui (Reference Fincham and Cui2011) note that what is experienced in the romantic relationships of emerging adults is linked with their mental health outcomes. Although there are several studies (e.g., Demir, Reference Demir2008; Novak & Johnson, Reference Novak and Johnson2017; Sağkal & Özdemir, Reference Sağkal and Özdemir2020) examining the associations between relational (e.g., relationship quality, relationship satisfaction, relationship conflict) and individual processes (e.g., happiness, life satisfaction, depression, anxiety, stress) in premarital relationships of emerging adults, it is evident that the existing studies are mostly cross-sectional and longitudinal studies addressing the links between romantic relationship experiences and mental health outcomes are rather limited. Therefore, the present study aimed to examine the longitudinal links between romantic relationship conflict and psychological distress in premarital romantic relationships of emerging adults.
In this study, romantic relationship conflict was conceptualized on the basis of interparental conflict (Grych et al., Reference Grych, Seid and Fincham1992; Osborne & Fincham, Reference Osborne and Fincham1996). Romantic relationship conflict is defined as a stress factor that prevents partners from becoming effective couples. In general, scholars (e.g., Cui et al., Reference Cui, Fincham and Pasley2008) conceptualize romantic relationship conflict in three dimensions, namely, frequency, intensity, and resolution. Conflict frequency refers to the frequency with which couples have disagreements or conflicts; intensity refers to the degree of negative affect or hostility expressed and the occurrence of physical aggression between couples; and resolution refers to the perception that couples are able to constructively deal with conflict. As a second main variable of the research, psychological distress was defined as an emotional state characterized by symptoms of depression (e.g., loss of interest, sadness, hopelessness) and anxiety (e.g., restlessness, feeling of tension) (Kessler et al., Reference Kessler, Andrews, Colpe, Hiripi, Mroczek, Normand, Walters and Zaslavsky2002).
Previous research addressed the importance of romantic relationships during emerging adulthood (Fincham & Cui, Reference Fincham and Cui2011) and showed that romantic relationships during this period involve more physical and emotional intimacy compared to adolescence (Arnett, Reference Arnett2015). In addition to this feature, romantic relationships in emerging adulthood are characterized by instabilities (Arnett, Reference Arnett2015) and relationship conflict (Creasey & Hesson-McInnis, Reference Creasey and Hesson-McInnis2001). It is scholarly highlighted that dysfunctional interactional patterns have the potential to influence the mental health of emerging adults in the negative way (Collins & van Dulmen, Reference Collins, van Dulmen, Arnett and Tanner2006; Manning et al., Reference Manning, Giordano, Longmore, Hocevar, Fincham and Cui2010). Empirical findings indicate that relationship dynamics such as romantic relationship conflict have been linked with depression (Harper et al., Reference Harper, Dickson and Welsh2006) and anxiety (La Greca & Harrison, Reference La Greca and Harrison2005) among adolescents and emerging adults.
Although biological cause is considered as a significant factor underlying mental health problems such as depression and anxiety, some researchers identified interpersonal difficulties as a major risk factor for psychological distress (e.g., Kendler et al., Reference Kendler, Gardner and Prescott2006). For example, marital conflict has been found to be associated with depressive symptoms among married couples (Laurent et al., Reference Laurent, Kim and Capaldi2009). There are some theoretical explanations for the assumption that romantic relationship conflict may link with higher levels of psychological distress. For example, the stress and coping model (Davila, Reference Davila2008) proposes that romantic relationships have some challenging features and not having the skills to cope with these features is associated with psychological distress. In the stress and coping model, relational challenging experiences like break-ups and rejection are seen as an important risk factor for mental health problems (Davila et al., Reference Davila, Stroud, Starr, Gotlib and Hammen2014; Larson et al., Reference Larson, Clore, Wood, Furman, Brown and Feiring1999). This theoretical view was supported by the findings of some longitudinal studies (e.g., Davila et al., Reference Davila, Steinberg, Kachadourian, Cobb and Fincham2004; Joyner & Udry, Reference Joyner and Udry2000). For example, findings from these studies reveal that adolescents being in a romantic relationship report a higher level of depression over time.
Another theoretical view for explaining the effects of dyadic interactions on mental health outcomes comes from the model of marital discord model of depression (Beach et al., Reference Beach, Sandeen and O’Leary1990). The marital discord model of depression suggests that marital discord is an important risk factor in the development and maintenance of depression. This model suggests that the problems or dissatisfaction experienced in the marital relationships reduce the level of support and increase the level of stress and hostility among couples, and which in turn increases the risk for depression. Findings from the retrospective studies demonstrate that while marital dissatisfaction predicts depressive symptoms, depression does not strongly predict dissatisfaction in marriage (O’Leary et al., Reference O’Leary, Riso and Beach1990). Longitudinal studies (e.g., Beach et al., Reference Beach, Katz, Kim and Brody2003) also indicate that when the initial depressive symptoms were controlled, the marital discord predicts depressive symptoms. Findings from experimental studies indicate that although individuals receive therapy for symptoms of depression, it does not lead to reduction in the level of marital discord (Foley et al., Reference Foley, Rounsaville, Weissman, Sholomaskas and Chevron1989). However, Brown et al. (Reference Brown, Lemyre and Bifulco1992) report that improvement of marital quality is associated with a decrease in depressive symptoms. As a result, all these empirical findings highlight that the direction of link goes from relational problems to individual outcomes.
Although aforementioned theoretical and empirical results suggest a direction of link from relationship conflict to psychological distress, it is also possible that psychological distress may result in relationship conflict. For example, Hammen (Reference Hammen1991) proposed the stress generation hypothesis to explain causal mechanism from mental functioning to relationship experiences. According to this hypothesis, individuals with depression problems are more likely to behave in ways that lead to negative interactions in their romantic relationships. Individuals with depression are more likely to have negative attitudes, cognitive distortions, and some irrational beliefs in their relationships (Beck, Reference Beck1979; Tecuta et al., Reference Tecuta, Tomba, Lupetti and DiGiuseppe2019). It has been evidenced in previous studies that these cognitive distortions lead to some relational problems such as marital conflict or romantic relationship conflict (Fincham, Reference Fincham2003). In addition, anxious individuals have a greater tendency to do and say potentially hurtful things without thinking them (Darcy et al., Reference Darcy, Davila and Beck2005). These individuals are also emotionally more reactive and may engage in aggressive or angry outbursts (Eronen et al., Reference Eronen, Angermeyer and Schulze1998). Thus, the emotional states and behaviors of individuals suffering from psychological distress increase the likelihood that they might behave in a way that leads to stress and conflict in their relationships. The empirical findings show that depression is associated with higher levels of destructive conflict behaviors and lower levels of constructive conflict resolution tactics in adults’ romantic relationships (Whisman & Uebelacker, Reference Whisman and Uebelacker2009). In a study (Papp et al., Reference Papp, Goeke-Morey and Cummings2007) using the daily diary approach, psychological distress experienced by spouses was shown to be associated with high negative emotions and conflict in marital relationships. Results of studies addressing those associations among adolescents also indicate that depressive symptoms are linked with higher levels of romantic relationship problems (La Greca & Harrison, Reference La Greca and Harrison2005) as well as lower levels of romantic relationship quality (Rudolph et al., Reference Rudolph, Flynn, Abaied, Abela and Hankin2008). Taken together, some theoretical views and empirical findings also suggest that psychological distress may have an effect on increasing romantic relationship conflict of individuals.
Despite the some evidence indicating the link from relationship conflict to psychological distress as well as from psychological distress to relationship conflict, it would be also possible that the association between relationship conflict and psychological distress is bidirectional. Some theoretical approaches argue that there would be a reciprocal link between relationship functioning and mental health outcomes. For example, one theoretical explanation comes from the vulnerability-stress-adaptation model (VSA; Karney & Bradbury, Reference Karney and Bradbury1995). This model emphasizes the consideration of multiple dimensions in understanding the relationship functioning and stability. The VSA as an integrative theoretical framework posits that enduring vulnerabilities, experiences of stressful events as well as adaptive processes interact to contribute to relationship functioning. For example, enduring vulnerability (e.g., negative affectivity) and adaptive processes (e.g., conflict resolution strategies) importantly link with each other. Consistent with the VSA, empirical evidence (e.g., Hanzal & Segrin, Reference Hanzal and Segrin2009; Kurdek, Reference Kurdek1997) indicate that there is a significant association between negative affectivity and problematic relationship conflict resolution style. As a result, based on the VSA model, it would be also expected that relationship conflict and psychological distress are bidirectionally related to each other.
The Present Study
Even though previous research has established the associations between relationship dynamics and individual outcomes across adolescence (e.g., Anderson et al., Reference Anderson, Salk and Hyde2015; Chow et al., Reference Chow, Ruhl and Buhrmester2015), emerging adulthood (e.g., Demir, Reference Demir2008; Novak & Johnson, Reference Novak and Johnson2017), and adulthood (e.g., Pruchno et al., Reference Pruchno, Wilson-Genderson and Cartwright2009), the direction of these associations are not fully clear. To the best of our knowledge, no previous research has specifically examined the directions of links between romantic relationship conflict and psychological distress among a sample of emerging adults. Besides, scholars (e.g., Arnett, Reference Arnett2015; Fincham & Cui, Reference Fincham and Cui2011) argue that the developmental processes, meaning, and consequences of romantic relationships in emerging adulthood could be different from those in either adolescence or adulthood. It is suggested that researchers should pay close attention to the association between romantic and individual functioning among emerging adults, particularly with regard to how it develops and the course it follows (Gómez-López et al., Reference Gómez-López, Viejo and Ortega-Ruiz2019). Therefore, to fill the gap in the literature, the present study aimed to explore the directions of links between romantic relationship conflict and psychological distress in a two-wave longitudinal data of emerging adults. As there is limited evidence for the notion that romantic relationship conflict predicts psychological distress or vice versa (Beach et al., Reference Beach, Davey and Fincham1999), addressing the direction of links would have important implications for prevention and early intervention efforts. Based on previous literature, we posited three main hypotheses: (i) Romantic relationship conflict at Time 1 will be related to higher psychological distress at Time 2, (ii) psychological distress at Time 1 will be related to higher romantic relationship conflict at Time 2, and (iii) romantic relationship conflict and psychological distress will be bidirectionally related to each other.
Method
Participants and Procedure
We used the data from the Turkish Longitudinal Romantic Relationship Study (TLRRS; Sağkal & Özdemir, Reference Sağkal and Özdemir2020). Using convenience sampling, unmarried emerging adults who were in a romantic relationship were invited to participate in the study. For the baseline assessment, 511 subjects (Mage = 21.00; SDage = 1.66; 81.8% female) who are taking education and psychology courses voluntarily participated in the study. Of the initial sample, 64.4% was dropped out due to relationship break-up. A total of 182 subjects (M age = 21.23; SDage = 1.62; 85.16% female) provided data at both Time 1 and Time 2. The mean romantic relationship length of the final sample at Time 2 was 28.63 month (SD = 23.32). The results demonstrated that the final sample did not differ from those who dropped out from follow-up assessment based on conflict frequency (t = –.517, p = .605), conflict intensity (t = –1.069, p = .286), conflict resolution (t = –1.881, p = .061), and psychological distress (t = –.901, p = .368). The data of this study were collected at two time points in the spring of 2018: First week (Time 1) and the last week of the semester, Week 14 (Time 2). Paper-pencil and self-report questionnaires were collected anonymously by the researchers in the classroom. Informed consent of the participants was obtained before the administration. No monetary incentive was given to the participants. The research protocol was consistent with the American Psychological Association (APA) ethical guidelines.
Measures
Relationship conflict. Conflict properties factor of Turkish version of the Children’s Perception of Interparental Conflict Scale (CPIC; Grych et al., Reference Grych, Seid and Fincham1992; Ulu & Fışıloğlu, Reference Ulu and Fışıloğlu2004) was used to assess romantic relationship conflict. The conflict properties factor (17 items) of Turkish version of the CPIC is composed of frequency (6 items), intensity (5 items), and resolution (6 items) dimensions. Items are rated on a 3-point Likert scale ranging from true to false. In the present study, items were so reworded that participants were asked to evaluate the frequency, intensity, and resolution of conflicts regarding their own romantic relationships (e.g., “When we have an argument, we usually work it out”). Higher scores from the scale reflect greater levels of frequent, intense, and unresolved relationship conflict. Previous research (e.g., Sağkal & Özdemir, Reference Sağkal and Özdemir2019) indicated that the scale has good validity and reliability to assess relationship conflict in premarital relationships of emerging adults. In the present research, the Cronbach’s alpha coefficients were .69, .74 and .71 for frequency, intensity, and resolution dimensions at Time 1 and .62, .68, and .68 at Time 2, respectively. For the total scores, Cronbach’s alpha coefficients were .88 and .87 at Time 1 and Time 2, respectively.
Psychological distress . Psychological distress was assessed with the 10-item Kessler Psychological Distress Scale (K10; Kessler et al., Reference Kessler, Andrews, Colpe, Hiripi, Mroczek, Normand, Walters and Zaslavsky2002). As a self-report measure, the K10 consists of 10 items and assesses global psychological distress related with anxiety (e.g., “During the last 30 days, about how often did you feel nervous?”) and depressive (e.g., “During the last 30 days, about how often did you feel hopeless?”) symptoms. Respondents rate items on a 5-point Likert scale (1 = None of the time to 5 = All of the time). Higher scores from the scale represent higher levels of psychological distress. Previous research (e.g., Sağkal & Özdemir, Reference Sağkal and Özdemir2017; Sağkal, Özdemir, & Koruklu, Reference Sağkal, Özdemir and Koruklu2018) has demonstrated that Turkish version of the K10 has strong psychometric properties in adolescents as well as emerging adults. In the present study, the items of the K10 were parcelled into three group based on item-to-construct balance approach (Little et al., Reference Little, Cunningham, Shahar and Widaman2002). The Cronbach’s alpha coefficients were .81, .78, and .81 for three parcels at Time 1 and .83, .82, and .84 for three parcels at Time 2. For the total scores, Cronbach’s alpha coefficients were .93 both at Time 1 and Time 2.
Data Analysis
SPSS 24.0 and AMOS 24.0 were used to conduct statistical analyses of the data. Preliminary analyses demonstrated that the percentages of missing data ranged from 0.0% to 2.7% and missing data were completely at random (Little’s MCAR test: χ 2 = 1027,395, df = 997, p = .245). We therefore used the full information maximum likelihood (FIML) method to handle with missing data. Seven cases with standardized z-scores above ± 3.29 were removed from the data set, leaving a total of 175 cases for analyses (Tabachnick & Fidell, Reference Tabachnick and Fidell2007). We conducted independent sample’s t-tests to explore whether the final sample differ from drop-out sample in terms of relationship conflict and psychological distress. Additionally, we explored the effects of demographic variables (age, gender, and relationship duration) on Time 1 and Time 2 assessments of relationship conflict and psychological distress. The demographic variable found to be significant was included as a covariate in the model. Prior to testing main hypothesis, we computed the means, standard deviations, Cronbach’s alphas as well as intervariable correlations. In order to determine whether the observed variables loaded on their corresponding latent variables, we performed a confirmatory factor analysis. To test the main research hypothesis, a two-wave two variable cross-lagged autoregressive panel model was conducted (Selig & Little, Reference Selig, Little, Laursen, Little and Card2012). The following criteria were considered as indicators of acceptable model fit: (i) The ratio of chi-square to degrees of freedom less than 3.0, (ii) the Comparative Fit Index (CFI) value of .90 and above, and (iii) the Root Mean Square Error of Approximation (RMSEA) less than .08 (Kline, Reference Kline2011).
Results
Descriptive Statistics
The means, standard deviations, internal consistency coefficients, and bivariate correlations among research variables are shown in Table 1. The research variables were all significantly and positively associated with each other both within and across time points. Furthermore, romantic relationship conflict at Time 1 was positively correlated with relationship conflict at Time 2 and respectively, psychological distress at Time 1 was positively associated with the degree of psychological distress at Time 2. In terms of background variables, only relationship duration was significantly and negatively associated with psychological distress across time points (T1: r = –.15, p < .05; T2: r = –.21, p < .01). Therefore, we included relationship duration as a covariate in the cross-lagged model.
Descriptive Statistics, Internal Consistency Coefficients, and Bivariate Correlations
Note. T1 = Time 1; T2 = Time 2.
* p < .05.
** p < .01.
Measurement Model
We specified a measurement model including relationship conflict (T1 and T2) and psychological distress (T1 and T2) as latent variables with three observed variables each. While frequency, intensity, and resolution dimensions were used as the indicators of romantic relationship conflict, three parcels created from the K10 using the item-to-construct balance technique (Little et al., Reference Little, Cunningham, Shahar and Widaman2002) were used as the indicators of the psychological distress. The measurement model provided an acceptable fit to the data: χ2(47) = 81.377, p < .01, χ2/df = 1.73, CFI = .98, and RMSEA = .07 90% CI [.04, .09]. Standardized factor loadings ranged from .83 to .87 for relationship conflict T1, .86 to .91 for psychological distress T1, .80 to .89 for relationship conflict T2, and .89 to .93 for psychological distress T2. The results suggest that all the observed variables loaded strongly on their respective latent variables. Additionally, all the latent variables were positively and significantly linked with each other (p < .001).
Furthermore, we also tested the time invariance of the longitudinal measurement model. The configural model, in which factor loadings were allowed to vary freely across time, provided an adequate fit to the data: χ2(16) = 30.236, p < .05, χ2/df = 1.89, CFI = .990, and RMSEA = .051 90% CI [.02, .08]. The metric model, in which factor loadings were constrained to be equal across time, yielded an adequate fit to the data as well: χ2(22) = 38.954, p < .05, χ2/df = 1.771, CFI = .989, and RMSEA = .047 90% CI [.02, .07]. The change for the CFI and RMSEA (ΔCFI = .001 and ΔRMSEA = .004) values between the configural and metric model provided statistical evidence for metric invariance. As a result, the results clearly established the measurement invariance of constructs across waves.
Cross-lagged Autoregressive Panel Model
In the present research, we employed a cross-lagged autoregressive panel model to examine the links between relationship conflict and psychological distress over time in emerging adults. The cross-lagged model for relationship conflict and psychological distress is presented in Figure 1. The hypothesized longitudinal cross-lagged model included relationship conflict T1, psychological distress T1, relationship conflict T2, and psychological distress T2 as latent variables. As relationship duration was significantly and negatively associated with psychological distress at both Time 1 and Time 2, this variable was added as a covariate in the model. The latent variables of relationship conflict were represented by three subdimensions (frequency, intensity, and resolution) of conflict properties factor of the CPIC. The latent variables of psychological distress were represented by three parcels. The hypothesized model testing the cross-lagged associations between relationship conflict and psychological distress provided acceptable fit to the data: χ2(58) = 90.736, p < .01, χ2/df = 1.56, CFI = .98, and RMSEA = .06 90% CI [.03, .08]. Figure 1 depicts the standardized path coefficients. The research results indicated that relationship conflict (β = .86, p < .001) and psychological distress (β = .35, p < .001) were stable over time. In the cross-lagged model, the path from relationship conflict at Time 1 to psychological distress at Time 2 was significant (β = .19, p < .05); however, the path from psychological distress at Time 1 to relationship conflict at Time 2 was not significant (β = –.10, p > .05). These findings suggest that at the between-person level, frequent, intense, and unresolved relationship conflict at baseline associates with higher levels of psychological distress at follow-up assessment but not vice versa. The proposed model accounted for 66% of the variance in relationship conflict at Time 2 and 26% of the variance in psychological distress at Time 2.
A Cross-lagged Autoregressive Panel Model
Standardized path estimates are reported. Relationship duration was controlled as a covariate but omitted for clarity. P1 – P3 = three parcels of psychological distress. T1 = Time 1. T2 = Time 2.* p < .05. ** p < .01. *** p < .001.
Discussion
Establishing and maintaining healthy romantic relationships has been acknowledged as an important developmental task for emerging adulthood (Arnett, Reference Arnett2000). Although being in a romantic relationship has been generally positively associated with greater well-being in individuals (Umberson & Williams, Reference Umberson, Williams, Aneshensel and Phelan1999), some aspects of romantic relationships (e.g., destructive conflict behaviors) have been linked with symptoms of depression (Harper et al., Reference Harper, Dickson and Welsh2006) and anxiety (La Greca & Harrison, Reference La Greca and Harrison2005). Hovewer, much of current knowledge is based on cross-sectional studies and less is known about the directionality of the association between romantic relationship functioning and mental health in emerging adults. Therefore, the present study aimed to examine the direction of links between romantic relationship conflict and psychological distress using a 14-week interval, two-wave data from a sample of emerging adults. The results provided support for the view that the direction of the link goes from romantic relationship conflict to psychological distress in premarital relationships of emerging adults. Overall, this study extends previous literature by highlighting the directionality of the links between relationship conflict and psychological distress.
In the present study, we examined cross-lagged predictive relationships between romantic relationship conflict and psychological distress. Findings revealed that at the between-person level, romantic relationship conflict at Time 1 was linked to psychological distress at Time 2. In other words, emerging adults who reported relatively more frequent, intense, and unresolved conflict in their relationships were likely to report psychological distress. This finding is congruent with previous theoretical views (e.g., Beach et al., Reference Beach, Sandeen and O’Leary1990; Davila, Reference Davila2008) and empirical evidence (e.g., Beach et al., Reference Beach, Katz, Kim and Brody2003; Choi et al., Reference Choi, Weston and Temple2017; Kendler et al., Reference Kendler, Gardner and Prescott2006; Kouros et al., Reference Kouros, Papp and Cummings2008; Laurent et al., Reference Laurent, Kim and Capaldi2009; Pruchno et al., Reference Pruchno, Wilson-Genderson and Cartwright2009) highlighting the role of romantic relationship experiences on individuals’ mental health. The findings of the present study provide some evidence that the marital discord model of depression (Beach et al., Reference Beach, Sandeen and O’Leary1990) is also applicable to premarital romantic relationship. According to marital discord model of depression, decreased social support and increased hostility and stress in the relationships might serve as a predictor of depression. This framework suggests that marital discord is linked with increases in depressive symptoms by increasing hostile conflicts between partners (Beach et al., Reference Beach, Sandeen and O’Leary1990). In addition, the stress and coping model also proposed relationship conflict as a potential predictor of psychological distress (Davila, Reference Davila2008). Indeed, the present finding supports these two theoretical frameworks by highlighting the importance of relationship conflict in predicting psychological distress in premarital relationships of emerging adults. However, although the association between relationship functioning and mental health outcomes is proposed to be bidirectional, some scholars (e.g., Braithwaite & Holt-Lunstad, Reference Braithwaite and Holt-Lunstad2017; Whisman & Baucom, Reference Whisman and Baucom2012) argue that stronger effects are detected when relationship functioning is the predictor and mental health outcome is the outcome variable. That is, the direction of link goes more strongly from relationship functioning to mental health than vice versa. Besides, scholars (Whisman & Baucom, Reference Whisman and Baucom2012) posit that while improving relationship functioning decreases depressive symptoms, improvements in the psychopathology does not reliably result in improvements in relationship functioning. As a result, consistent with the above-mentioned theoretical and empirical evidence, the present research contributes to current literature by providing some evidence on the direction of link from romantic relationship conflict to psychological distress.
Furthermore, finding a non-significant link from psychological distress to romantic relationship conflict in cross-lagged panel model also merits discussion. While the marital discord model suggests marital discord as a salient antecedent of depression, some other theoretical frameworks assume psychological distress as an important predictor of relationship functioning. For example, according to the stress generation model (Hammen, Reference Hammen1991), individuals with psychological distress are more likely to experience romantic relationship conflict with their partners. However, contrary to this theoretical framework, in the present research, we found that baseline psychological distress was not significantly related to romantic relationship conflict over time. The statistically nonsignificant link between psychological distress at Time 1 and relationship conflict at Time 2 does not necessarily mean that an association does not exist. As a possible explanation, the lack of statistical significance might be attributed to the small sample size in the current study. In addition, the group participated in the study was not a clinical sample in terms of psychological distress. The predictive power of psychological distress on romantic relationship conflict may depend on the severity of psychological distress. In the clinical group, we might predict that severity of psychological distress might serve to more dramatical increases in romantic relationship conflict. Moreover, as the relationships were relatively new and short, participants would not have reflected their depressive experiences in order to preserve their relationships. Futhermore, longitudinal link from psychological distress to romantic relationship conflict may be influenced by some other mediating and/or moderating variables, such as personality factors, dyadic processes, external factors like family interactions (Peterson et al., Reference Peterson, Peugh, Loucks and Shaffer2018), and stressful life events (Proulx et al., Reference Proulx, Helms and Buehler2007). The stress generation model also posits that adaptive (e.g., coping skills, social support) and maladaptive processes (e.g., negative cognitive style, social skills deficts) would serve as mediating and moderating mechanisms, buffering the effects of psychological distress (Braithwaite & Holt-Lunstad, Reference Braithwaite and Holt-Lunstad2017). According to this model (Hammen, Reference Hammen1991), individuals with depressive symptoms are likely to experience more stressful life events that are due at least in part to their own characteristics and behaviors. For example, individuals with depressive symptoms would have negative perceptions about the self, world, and future, which in turn may lead to increased conflict tendency and interpersonal conflict (Keser et al., Reference Keser, Kahya and Akın2020). Additionally, receiving social support from close others would mitigate the effects of depressive symptoms on negative life events (Hammen, Reference Hammen2006). Furthermore, detecting a high stability coefficient in respondents’ level of romantic relationship conflict would also have an effect on the present findings. Although a strong autoregressive coefficient for romantic relationhip conflict is informative about the stability of the construct over time, this might left little variance to explain by psychological distress. Finally, the lack of a non-significant link from psychological distress to romantic relationship conflict could also be explained by the relatively short time lag between the baseline and follow-up assessments. The magnitude of link from psychological distress to relationship conflict would be dependent upon time-lag and the magnitude of association would increase over time. But, further research is needed to ensure whether the directions and magnitudes of the links vary with a longer time-lag.
Although the present findings further advance the understanding of the links between romantic relationship conflict and psychological distress in emerging adults, there are some limitations as well. First of all, the sample was recruited from university students and hence caution should be taken when generalizing findings to other populations. The links between romantic relationship experiences and mental health outcomes should be tested in diverse sample groups including emerging adults out of school. Secondly, since nature of associations may differ for males and females, given the relatively low levels of male participants, the findings should be interpreted in view of this limitation. As previous research indicated that quality of romantic relationship differentially associates with mental health outcomes across females and males (Palner & Mittelmark, Reference Palner and Mittelmark2002), future studies would regard sex as a moderator variable. Another important limitation of the study was that the data were based on self-report measures and retrieved only from one partner. Assessing romantic relationship conflict and psychological distress at the individual level may not allow true capturing of cross-partner effects (i.e., the effect of one partner’s conflict behaviors on the psychological distress of the other and vice versa). Romantic relationship conflict is a dyadic process and it is likely that one’s individual reactions would emerge as a response to behavior of his or her partner. Future studies would collect dyadic data and this would allow better test directional links between relational and individual factors in premarital relationships of emerging adults. Since relationship exploration is basic characteristic of emerging adulthood, future research may consider studying other relationship types (e.g., friends with benefits, cyclical relationships, or long distance relationships) to capture a more complete picture of directionality of links between relationship dynamics and mental health outcomes. Finally, in the present study, we examined the associations between romantic relationship conflict and psychological distress including a one-semester time lag. In future studies, the links between these two variables can be followed for longer periods of time.
Despite these limitations, the current study provides significant theoretical and clinical implications. From a theoretical perspective, detecting a direction of link from romantic relationship conflict to psychological distress supports the stress and coping model (Davila, Reference Davila2008) and the marital discord model of depression (Beach et al., Reference Beach, Sandeen and O’Leary1990). According to the stress and coping model, romantic relationships are challenging in nature and partners who do not have adequate coping resources are at increased risk for psychological distress. Besides, the marital discord model of depression posits that receiving less social support from the partner as well as experiencing higher levels of hostility in the relationship is a significant risk factor for psychological distress. Utilizing a two-wave, cross-lagged panel model, findings of the present research provide empirical support to the stress and coping model as well as the marital discord model of depression by revealing the temporal associations between romantic relationship experience and mental health outcome. From a practical perspective, the finding that psychological distress is predicted by relationship conflict has crucial implications for prevention and intervention efforts. As the present findings highlight the role of relationship conflict on psychological distress, individual and/or group therapy interventions would focus first on relational dynamics as an important antecedent of the mental health outcomes. We recommend couple therapists to train their clients about the ways of resolving relationship conflicts. Based on findings of the present research, it would be suggested that assisting spouses in their daily communication skills as well as conflict resolution skills would lower the risk of psychological distress. However, it should be also cautioned that romantic relationship conflict is a significant antecedent of psychological distress, though there could also be other factors (enduring vulnerabilities, stressful events, and maladaptive processes) at play (Karney & Bradbury, Reference Karney and Bradbury1995).
In conclusion, the present research advances our understanding of the association between relationship functioning and mental health outcome by examining the links over time using a cross-lagged panel model analysis. Although there are several studies indicating the links between relational and individual processes (e.g., Anderson et al., Reference Anderson, Salk and Hyde2015; Chow et al., Reference Chow, Ruhl and Buhrmester2015; Pruchno et al., Reference Pruchno, Wilson-Genderson and Cartwright2009), they are mostly based on cross-sectional data sets and conducted with either adolescent or married couples. In this regard, the present research extends previous literature by exploring the direction of links between romantic relationship conflict and psychological distress in a sample of emerging adults with a two-wave cross-lagged design. The research provides empirical evidence regarding the cross-lagged effects of romantic relationship conflict on psychological distress than vice versa. The present findings support the view that romantic relationship is a key component of emerging adulthood and seems to have a potential influence on mental health outcomes of emerging adults. Understanding the course of relationship functioning and mental health across the life span provides a crucial knowledge for researchers and practitioners to promote relational and individual functioning.
All procedures were in accordance with the ethical standards of APA as well as author’s Institutional Ethics Board.
