A Track from Attachment to Resilience During the Covid 19 Pandemic: Romantic Relationship Tendency

Abstract

[...]of structural equation modelling, anxious and avoidant attachment styles had no effect on the purpose of social media usage and the purpose of social media usage had no effect on psychological resilience. [...]in a study conducted with individuals with earthquake experience, which can be considered as a disaster that interrupts life such as a pandemic, it has been revealed that there is a positive relationship between secure attachment styles and psychological resilience (Karaırmak and Güloǧlu 2014). In romantic relationships, individual attachment styles seem to match a specific style of emotion regulation and coping (Pietromonaco and Overall 2021). [...]examining the relationship between romantic relationship tendencies and attachment styles in this period can provide important information. In this study, not the attitudes of the participants in their romantic relationship, but their romantic relationship tendencies in general were measured. [...]in this study, it was tried to measure the tendency of the participants towards the romantic relationship rather than their attitudes and behaviors in their current romantic relationships.

Headnote

ÖZ

Covid 19 salgını fizyolojik etkilerinin yanı sıra birtakım psikososyal sıkıntıları da beraberinde getirmiştir. Psikolojik saǧlamlık ise bu sıkıntılarla baş edebilmek üzere önemli bir özellik olarak karşımıza çıkmaktadır. Bu araştırmada, Covid 19 pandemi sürecinde deǧişen yaşam koşulları baǧlamında, bireylerin romantik ilişki eǧilimleri, baǧlanma stilleri, sosyal medya kullanımları ve psikolojik saǧlamlıkları arasındaki ilişkilerin incelenmesi yoluyla bir model ortaya koymak hedeflenmektedir. Araştırma örneklemini en az altı aylık romantik ilişki geçmişi bulunan ve evli olmayan 289 kişi oluşturmaktadır. Katılımcıların yaş ortalaması 27'dir. Yapılan yapısal eşitlik modellemesinin sonucunda, kaygılı ve kaçıngan baǧlanma stillerinin sosyal medya kullanım amacı üzerinde ve sosyal medya kullanım amacının da psikolojik saǧlamlık üzerinde etkisinin olmadıǧı bulunmuştur. Kaygılı ve kaçıngan baǧlanma stillerinin, psikolojik saǧlamlık üzerinde etkisinin olduǧu ve bu etkinin tamamıyla romantik ilişki eǧiliminin aracılıǧıyla gerçekleştiǧi bulunmuştur. Bulgular, psikolojik saǧlamlıǧı artırmaya yönelik çalışmalarda, kişilerin romantik ilişki eǧilimlerinin göz önünde bulundurulması gerekliliǧini ortaya koymaktadır.

Anahtar sözcükler: psikolojik saǧlamlık, baǧlanma stilleri, romantik ilişki eǧilimleri, covid 19, sosyal medya

ABSTRACT

The Covid 19 pandemic involves some psychosocial problems in addition to physical risks, and resilience appears to be a critical feature to cope with these problems. This study aims to present a model examining the relationships between the psychological tendencies associated with romantic relationships, attachment styles, social media usage, and resilience during the Covid 19 pandemic. These relationships were investigated in a sample of unmarried adults (n = 289) with a romantic relationship history of at least six months. The average age of the participants was 27 years. As a result of structural equation modelling, anxious and avoidant attachment styles had no effect on the purpose of social media usage and the purpose of social media usage had no effect on psychological resilience. Also, anxious and avoidant attachment styles affected resilience, and this effect was fully mediated by psychological tendencies associated with romantic relationships. The findings reveal the necessity to consider the psychological tendencies associated with romantic relationships in studies aimed at increasing resilience.

Keywords: resilience, attachment styles, romantic relationship tendency, covid-19, social media

Introduction

The current outbreak of Covid 19, which is one of the greatest disasters in human history, has brought many economic and psychosocial difficulties as well as its physiological effects. One of these psychosocial difficulties is the unmet need to relate to others. As a source of social connection, romantic relationships are affected by many of the stressors associated with the Covid 19 pandemic. Investigating how the various external stress factors and current vulnerabilities (e.g. various emotional difficulties, such as insecure attachment styles, rigid personality traits, or depression) brought about by the pandemic shape tendencies in romantic relationships may reveal which individuals are more at risk for negative relationships. In recent years, various models and conceptual relationships have been proposed and investigated (Vowels and Carnelley 2020, Overall et al. 2022) or existing models have been adapted (Pietromonaco and Overall 2021). For example, Pietromonaco and Overall (2021) adapted the Vulnerability-Stress-Adaptation Model (Karney and Bradbury 1995) to the conditions of this period and suggested that nonadaptive processes (e.g. hostility, withdrawal, less support) in coping with the external stresses related to the pandemic would weaken the relationship quality of couples. They also stated that this negative effect could be exacerbated by various demographic characteristics and individual vulnerabilities. It has been observed that studies on the reflections and consequences of the Covid 19 period on romantic relationships in Turkey are also quite limited. However, a missing aspect in the studies conducted in this period is how the romantic relationship tendencies of individuals affect various factors that affect romantic relationships or how these factors shape the relationship. In this study, a model of the relationship between negative attachment styles which is an individual vulnerability factor of adults in Turkey, and resilience which is evaluated as a protective factor, romantic relationship tendencies, and social media use was proposed and examined. The relevant literature on the variables in the model is discussed below, and then the model is introduced.

Resilience is a personal source of resistance and self-recovery that enables an individual to successfully adapt despite facing a significant threat, compelling event, or significant negativity that will disrupt periods of development (Luthar et al. 2000). As a matter of fact, in a study conducted with individuals with earthquake experience, which can be considered as a disaster that interrupts life such as a pandemic, it has been revealed that there is a positive relationship between secure attachment styles and psychological resilience (Karaırmak and Güloǧlu 2014). One of the important ways to remain resilient during the pandemic is to continue social and interpersonal relations despite risks (Chen and Bonanno 2020). Similarly, some studies show that the level of resilience increases in direct proportion to perceived social support (Wilks 2008). Perceiving the romantic relationship as a support is related to the quality of the relationship, individual attachment styles, and perceptions and tendencies about the relationship (Hazan and Shaver 1994).

Attachment styles are shaped by the infant's perceived sensitivity from caregivers and internalization of the relationship. Individuals turn these first relationships into schemas that determine how they perceive both themselves, others and relationships (Ainsworth et al. 1978, Bowlby 1982). These early relational schemas, in other words internal working models, are thought to be one of the key predictors of adult romantic relationships (Mikulincer and Shaver 2007). Moreover, attachment styles can also be considered to be a psychological structure reflected in romantic relationships; that is, emotion, thought and action observed in romantic relationships. In romantic relationships, individual attachment styles seem to match a specific style of emotion regulation and coping (Pietromonaco and Overall 2021). Therefore, examining the relationship between romantic relationship tendencies and attachment styles in this period can provide important information.

The use of social media and the internet is taking up more and more space in our lives (Mander 2020). The need for relationship, socialization and intimacy began to be met on these platforms when the restrictions were at their strictest (Saud et al. 2020). In this study, not the attitudes of the participants in their romantic relationship, but their romantic relationship tendencies in general were measured. Therefore, in this study, it was tried to measure the tendency of the participants towards the romantic relationship rather than their attitudes and behaviors in their current romantic relationships.

In the Covid 19 pandemic, social media has undoubtedly become one of the most important communication tools, as individuals are forced to abandon their daily routines. Indeed, while the functions of social media such as fundraising, informing, soothing and communicating can help people to overcome events in a healthier way (Keim and Noji 2011), they may negatively affect psychological resilience due to providing false information or excessive exposure to negative content (Reuter and Spielhofer 2017). Social media is an environment that can lay the groundwork for emotional intimacy in an online platform. Studies show that there is a significant increase in social media usage during Covid 19 pandemic (Mander 2020). In this sense, the question of how romantic relationships are affected by this increase has become one of the important issues of the present day. Studies revealed that excessive usage of social media has negative effects on romantic relationships such as decreases in romantic relationship quality (Abbasi and Alghamdi 2017), relationship dissatisfaction, low commitment (Abbasi 2019), separation and divorce risk (Valenzuela et al. 2014).

In summary, it is possible to say that how human relationships, more specifically how romantic relationships are affected by the context and new normal conditions, is a comprehensive research subject. Studies conducted on romantic relationships during the Covid 19 period show that there is an increase in lower level of relational satisfaction, psychological well-being, stress-related conflicts, and negative emotions (Candel and Jitaru 2021). It is suggested that these negative outcomes are associated with various factors both at the external and individual level. In this study, it is hypothesized that rapid changes such as the increase in the use of the internet and social media in interpersonal relationships due to external stress factors caused by the pandemic may have also affected individuals' approaches to romantic relationships. Relationships between insecure attachment styles which is a fragility factor that can be triggered in such conditions, and resilience which is a protective factor, can be shaped through romantic relationship tendencies and social media use. In this research, the predictive effects of insecure attachment style, tendencies in romantic relationship and purpose of using social networks on resilience, which is an important resource for individuals in the recent period, were examined. It is believed that the results of this study can provide an up to date perspective in explaining the predictors of resilience. The hypothetical model established is given in Figure 1:

Method

This research is a prediction study with descriptive method and relational screening model (Karasar, 1998). The relationships between insecure attachment style, romantic relationship tendency, social media usage purpose, and resilience were examined by structural equation modeling (SEM). Insecure attachment style is exogenous, resilience is endogenous, romantic relationship tendency and social media usage purpose are included in the research model as mediator variables. The process of reaching the participants and collecting data was carried out online.

Participants

The participants consisted of 289 adults, 220 females (76.1%) and 60 males (23.9%), aged 18-47 years (M = 27.13 SD = 5:52) living in Turkey. The G·Power 3.1.9.2 program was used to determine the sample size. In the power analysis, the significance level is a = .05, the power value is ß = .95, and the effect width is | p | = 0.1 is determined. The number of participants to be reached in this way was calculated as 220. Participants of the study were determined by using convenient sampling method due to pandemic limitations. The convenient sampling method is a non-probability-based method in which the researcher includes volunteers who have easy access (in the immediate environment) to the study (Stratton 2021). None of the participants were diagnosed with Covid 19; however, there are participants who have been diagnosed with Covid 19 in their relatives. Because being married can have many confounding effects, the sampling included couples who experienced physical and social isolation during the pandemic rather than married couples. In addition, while the partners focus on their similarities and differences around the first six months of romantic relationships, they focus on the extent to which their mutual needs can be met at the end of the six months and shape their relationships accordingly (Kerckhoff & Davis 1962). For this reason, it was determined as a prerequisite for the participants to have a relationship of at least six months in the past and/or in the current situation. Participants reported their relationship status with one or more options. Information about the participants is given in Table 1.

Data Collection Tools

The Multidimensional Relationship Questionnaire

The Multidimensional Relationship Questionnaire (MRQ), developed by Snell et al. (2002), measures various psychological tendencies in relation to close relationships. It consists of 8 factors and 53 items: extreme focus on relationship, relationship satisfaction, fear of relationship/relational anxiety, relational monitoring, relational esteem, external relational control, relational assertiveness, and internal relational control. The scale is scored on a five-point Likert type scale between 1 = "not suitable for me" and 5 = "very suitable for me". While the minimum score that can be obtained from the scale is 53, the maximum score is 265. Separately score could obtain from subscale as well as a total score could obtained. The Turkish adaptation of scale was carried out by Büyükşahin (2005). The Cronbach alpha coefficient for the internal consistency of the scale was found to be .81. Testretest reliability level was determined to be .80. In this study, the Cronbach alpha coefficient for extreme focus on relationship, relationship satisfaction, fear of relationship/relational anxiety, relational monitoring, relational esteem, external relational control, relational assertiveness, and internal relational control dimensions of the MRQ were .92, .91, .82, .89, .86, .78, .89 and .67, respectively.

Experience in Close Relationship Scale- Short Form

The Experiences in Close Relationships scale Short Form (ECRS-SF), developed by Wei et al. (2007), used to measure individual attachment style, comprises two sub-scales: anxiety and avoidance. This scale, which has a 7-point Likert-type rating between "totally disagree" and "totally agree", consists of 12 items in total. The lowest 12 points and the highest 84 points can be obtained on the scale. The internal consistency and reliability coefficients for the anxiety sub-scale of the scale are between .77 and .86; and for the avoidance sub-scale, it was determined to range from .78 to .88. Test-retest reliability coefficients, performed three weeks apart, were found to be .82 for the anxiety sub-scale and .89 for the avoidance sub-scale. The Turkish adaptation of the ECRS-SF was carried out by Savcı and Aysan (2016). In the adaptation study, scale items were graded with a five-point Likert rating (1 = Strongly disagree, 5 = strongly agree). The two-factor structure of this scale, which shows a twofactor structure in exploratory factor analysis, was confirmed as a result of confirmatory factor analysis. Internal consistency and reliability coefficients of the ECRS-SF were .90 for the anxiety sub-scale, .90 for the avoidance sub-scale, and .94 for the whole scale. In this study, the Cronbach alpha coefficients for the anxiety sub-scale, avoidant sub-scale and the whole scale were .68, .69, and .69, respectively.

Brief Resilience Scale

The Brief Resilience Scale (BRS), developed by Smith et al. (2008) to measure the individual's resilience, consists of a single dimension and 6 items. The scale is scored on a five-point Likerttype scale between 1 = "not at all suitable" and 5 = "completely suitable". While the minimum score that can be obtained from the scale is 6, the maximum score is 30. The scores obtained from the scale indicate high resilience. The internal consistency and reliability coefficient is between .80 and .91; the test-retest reliability coefficient was found between .62 and .69. The Turkish adaptation of BRS was carried out by Doǧan (2015). As a result of the exploratory factor analysis, this one-factor model of the BRS, which shows a one-factor structure, was confirmed by the confirmatory factor analysis. Related to the BRS, the internal consistency coefficient was found .83. In this study, the Cronbach alpha coefficient of BRS was found to be .91.

The Usage Purposes Scale of Social Networks

The Usage Purposes Scale of Social Networks (UPSSN), developed by Usluel et al. (2014), used to determine the usage purpose of social networks, consists of seven sub-dimensions and 26 items. With sub-dimensions of research, collaboration, initiating communication, establishing communication, maintaining communication, sharing content and entertainment, UPSSN is scored on a 7-point Likert-type rating between 1 = "totally disagree" and 7 = totally agree ". The lowest 26 points and the highest 182 points can be obtained on the scale. High scores obtained indicate the intensity of the intended use of the relevant social network. The internal consistency and reliability coefficient of the scale was found to be .92. In this study, the Cronbach alpha coefficient for the research, collaboration, initiating communication, establishing communication, maintaining communication, sharing content and entertainment sub-dimensions of UPSSN were .70, .80, .56, .90, .81, .81, and .66, respectively.

Procedure

In order to evaluate the ethical suitability of the research, an application was made to the Ege University Scientific Research and Publication Ethics Committee (EGEBAYEK), and ethical approval (29/12/2021, Meeting/Decision Number: 16/19, Protocol No: 1235) was obtained. The data of the study were collected online via Google Form due to the restrictions during the pandemic process. Demographic information and scales were arranged as a total of five pages. In the Google Form, instructions about the purpose, scope, and content of the research were presented. Participants were reached by promoting the research on social media platforms such as Facebook, Instagram, and WhatsApp. While the option to leave the scale items blank was not provided, the option to leave the demographic information blank was provided in the Google Form, which consists of 109 questions and takes approximately 20 minutes to answer. The E-mail information of all participants was checked to prevent the same participant from participating in the study more than once.

Statistical Analysis

A total of 324 participants were reached during the data collection process. 22 participants who filled out the questionnaires and reported that they were married or had a romantic relationship of fewer than six months were excluded. In addition, 13 participants who were determined to fill the questionnaires incompletely and carelessly were excluded from the data set. First of all, descriptive statistics and correlation analysis were calculated with the SPSS 25 program. Before correlation analysis, mean scores were calculated for each variable, and outliers were determined. The data of the three participants who created outliers were removed and the analyzes were continued. Second, structural equation modeling analysis with maximum likelihood estimation was used to estimate the hypothetical path model with AMOS 21. First, a validated measurement model for the structural model was examined, and then the structural model, which identifies direct and indirect paths with latent variables, was tested (Kline, 2015). The bootstrapping method was used to evaluate the significance of the mediation relationship. Bootstrapping analysis, it is aimed to examine the parameter distributions according to each of these samples produced by creating multiple sub-samples from an original data set (Byrne 2016). Significance levels of direct and indirect effects were calculated. The fit index values of ki %2 /df, GFI, RMSEA, RMR, CFI, and NFI were used to evaluate the model fit (MacCallum et al. 1996, Hu and Bentler 1999, SchermellehEngel et al. 2003, Dehon et al. 2005, Hooper et al. et al. 2008, Simon et al. 2010).

Results

Preliminary Analysis

Before starting the analyses, some preliminary analyses were carried out. It was observed that the data had multivariate normality (c.r. = 6.598 <10) (Kline 2015). The presence of a multicollinearity problem was examined. The VIF (variance increase factor) values of the variables were less than 10. Tolerance values were found to be greater than .10 and there was no multicollinearity problem (Tabachnick & Fidel 2007). In addition, the correlation values between observed variables ranged from .002 to .633 Correlation values between all observed variables are given in Table 2.

Note: 1. Anxious Attachment; 2. Avoidant Attachment; 3. Resilience; 4. Focus on Relationship Extremely; 5. Relationship Satisfaction; 6. Fear of Relationship/Relational Anxiety; 7. Relational Monitoring; 8. Relational Esteem; 9. External Relational Control; 10. Relational Assertiveness; 11. Internal Relational Control; 12. Research; 13. Collaboration; 14. Initiating Communication; 15. Establishing Communication; 16. Maintaining Communication; 17. Sharing Content; 18. Entertainment

Measurement Model

In structural equation modeling, it is often recommended to follow two stages, consisting of the measurement model and the structural model (Kline, 2015). The measurement model comprises four latent factors (insecure attachment, resilience, romantic relationship tendency, and usage purpose of social media) and twenty-three observed variables. In the first analyses, all of the standardized path coefficients in the measurement model were determined to be significant (p <.001). However, the sub-dimensions of the high level of focus on the relationship and internal relationship control had low path coefficients among the romantic relationship tendencies (<.10). These dimensions were removed from the measurement model. Goodness of fit indexes for the measurement model were: %2/df = 608.224 / 183 = 3.324, GFI = .82, SRMR = .093 CFI = .84 (Dehon et al. 2005, Schermelleh-Engel et al. 2003, Simon et al. 2010). These goodness of fit indexes for the measurement model were at acceptable levels. RMSEA indicates mediocre compatibility with a value of .090 (MacCallum et al. 1996). The NFI value (.79) was not at an acceptable level (Hooper et al. 2008, Simon et al. 2010).

Modifications were made with covariance established between the observed variables, one (among the items "I can recover quickly after difficult times" and "it takes a long time to recover from the effects of negativities in my life·") for resilience, two (between relational monitoring and relational esteem indicators and relational esteem and relational assertiveness indicators) for romantic relationship tendencies, and two (between indicators of cooperation with research and indicators of maintaining communication with research) for usage purpose of social media. After the modifications were made, the standardized path coefficients in the measurement model ranged from .31 to .87. In addition, all the factor loadings of the observed variables, representing the latent variables, were significant (p < .001). This measurement model had acceptable values for goodness of fit (x2/df = 498.831 / 178 = 2.802, GFI = .86, SRMR = .090, RMSEA = .079, CFI = .88, NFI = .82) (Dehon et al. 2005, Hooper et al. 2008, Hu and Bentler 1999, Simon et al. 2010). As a result, the measurement model was confirmed and compatible with the data.

Structural Model

A structural model, which constitutes the hypothetical model of the research and was created with latent variables, was tested. The path from anxious and avoidant types of insecure attachment style to the usage purpose of social media (p = .376) and the path from usage purpose of social media to resilience (p = .579) was not significant (p > .001). These findings show that the expected effects in the research model related to the purpose of social media use do not exist. The modeling process continued by excluding the intended use of social media. At this stage, the measurement model formed by the variables for anxious and avoidant types of insecure attachment style, romantic relationship tendency, and resilience was tested.

All of the standardized path coefficients in this measurement model were significant (p <.001). With the values of %2/df = 320.605 / 74 = 4.333, GFI = .86, SRMR = .090, CFI = .88 and NFI = .85, the measurement model showed an acceptable level of agreement with the data (Dehon et al. 2005, Hooper et al. 2008, Schermelleh-Engel et al. 2003, Simon et al. 2010). However, the RMSEA value (.108) was at the weak level of fit (McCallum et al., 1996). Modifications were made with the covariance established between the observed variables, one (among the items "I can recover quickly after difficult times" and "it takes a long time to recover from the effects of negativities in my life·") for resilience and two (between relational monitoring and relational esteem indicators and relational esteem and relational assertiveness indicators) for romantic relationship tendency. After the modifications, the standardized path coefficients in the measurement model ranged from .38 to .87. All of the standardized path coefficients at this stage of the measurement model were significant (p < .001). This measurement model has acceptable goodness of fit values (%2/df = 252,830 / 71 = 3.561 /, GFI = .89, SRMR = .084, CFI = .91, NFI = .88) (Dehon et al. 2005, Hooper et al. 2008, Schermelleh-Engel et al. 2003, Simon et al. 2010). RMSEA indicates mediocre fit with a value of .094 (McCallum et al., 1996).

The path from anxious and avoidant types of insecure attachment styles to resilience is negative and significant (p <.001). The path from anxious and avoidant types of insecure attachment styles to romantic relationship tendency is also negative and significant (p <.001). In addition, the path from romantic relationship tendency to resilience is positive and significant (p < .001).

Mediation Test

The mediation relationship was tested to examine whether romantic relationship tendency has a mediating role in the relationship between anxious and avoidant insecure attachment styles and resilience. The significance of mediation relations was examined by Bootstrapping Analysis. Bootstrapping Analysis was performed through 5000 resamplings (95% CI) to examine whether the indirect paths were significant. The path from anxious and avoidant insecure attachment styles to romantic relationship tendency was found to be negative and significant (p < .001). The path from romantic relationship tendency to resilience was found to be positive and significant (p < .001). However, the path from anxious and avoidant insecure attachment styles to resilience (p = .939) was found to be insignificant (p > .001). This finding indicates that romantic relationship tendency plays a full mediating role in the relationship between anxious and avoidant insecure attachment styles and resilience. By removing the path from anxious and avoidant insecure attachment styles to resilience, the model in which the only way in this relationship is provided through romantic relationship tendency was tested. All paths in this model were found to be significant (p < .001). The important parameter values of the final model are given in Table 3.

The model for the relationship between anxious and avoidant type of insecure attachment styles and resilience achieved through romantic relationship tendency had %2/df = 320.681 / 75 = 4.276, GFI = .86, SRMR = .090, CFI = .88 and NFI = .85. It had an acceptable level of goodness of fit to the values (Dehon et al. 2005, Hooper et al. 2008, Schermelleh-Engel et al. 2003, Simon et al. 2010). However, RMSEA was not at an acceptable level of fit (.107) (McCallum et al. 1996). Modifications were made with the covariance established between the observed variables, one (among the items "I can recover quickly after difficult times" and "it takes a long time to recover from the effects of negativities in my life·") for resilience and two (between relational monitoring and relational esteem indicators and relational esteem and relational assertiveness indicators) for romantic relationship tendencies. After the modifications, the model had acceptable goodness of fit values (%2/df = 252.835 / 72 = 3.512, GFI = .89, SRMR = .084, CFI = .91, NFI = .88) (Dehon et al. 2005, Hooper et al. 2008, Schermelleh-Engel et al. 2003, Simon et al. 2010). RMSEA had mediocre agreement with the value of .093 (McCallum et al. 1996).

As a result, the final model in the study confirmed that the effects of anxious and avoidant insecure attachment styles on resilience were entirely mediated by romantic relationship tendency. In this final model, the romantic relationship tendency has a dominant effect (= .39) on resilience. With this model, 15% of the total variance regarding psychological resilience is explained. The path diagram for the research model is given in Figure 2.

Discussion

The purpose of this study was to investigate the relationships between romantic relationship tendency, attachment style, social media usage and resilience of adult individuals during the Covid 19 pandemic.

Analyses indicated that anxious and avoidant attachment styles are negatively related to resilience. Rasmussen et al. (2018) analysed the results of 33 resilience studies and concluded that there is a positive relationship between the quality of attachment and psychological resilience, and that secure attachment may be one of the main factors in positive adaptation, which is a feature of psychological resilience. Results of other relevant studies show that there is a positive relationship between resilience and secure attachment (Bender and Ingram 2018, Karaırmak and Güloǧlu 2014, Simeona et al. 2007) and anxious and avoidant attachment style is significantly associated with low resilience (Caldwell and Shaver 2012). Therefore, this finding of the current study is consistent with the results of related studies in the literature.

Another finding indicated that anxious and avoidant attachment style have a negative relationship with relationship satisfaction, relational esteem, relational assertiveness, and internal relational control as well as positive relationship with fear of relationship/ relational anxiety, relational monitoring, and external relational control. Attachment experiences in early childhood shape the way individuals establish relationships in adulthood through internalized schemas (Mikulincer and Shaver 2007) and influence their approaches towards romantic relationships (Shaver and Hazan 1988). Therefore, attachment styles are also significantly related to the tendencies of individuals in their romantic relationships. The internal working models of self and others are explained with anxiety and avoidance dimensions (Griffin and Bartholomew 1994).

While anxious attachment style is characterized by a strong desire for closeness, intense worries, and insecurities about the existence of the partner's love and value perceived by the partner; avoidant attachment style is described as involving discomfort about intimacy and attachment to the partner, maintaining emotional distance, insecurity, and preferring only self-reliance (Mikulincer and Shaver 2007, Simpson 1990). Individuals with different attachment styles possibly have different romantic relationship tendencies considering the differences in the internal working models (Simpson 1990). On the other hand, Güloǧlu and Karaırmak (2017) found that extreme focus on relationship, relationship assertiveness and external relationship control tendencies positively correlated with secure attachment style, while fearful attachment style is related with relationship fear and relationship assertiveness. They also concluded that fear of relationship and an extreme focus tendency predicted the dismissive avoidant attachment style significantly. In another study, fearful attachment style positively predicted relationship fear and relationship anxiety (Teeruthroy and Bhwon 2012). Hence, the effect of anxious and avoidant attachment styles on romantic relationship tendencies is also supported by the literature.

Another important finding in the study is the positive relationship between romantic relationship tendency and resilience. Three of the romantic relationship tendencies, which are relational esteem, fear of relationship, and relationship satisfaction, explained the total variance at a higher level than the other relationship tendencies (.51 and above). Although attachment styles directly affect romantic relationships, people may not establish the same type of relationships with every romantic partner, and the tendencies exhibited in romantic relationships may vary. According to relational cultural theory, relationships that are convenient for meeting mutual needs, maintaining authenticity, and feeling competent can be positively related to resilience (Jordan 2017, Miller and Stiver 1997). Therefore, Hartling (2008) also suggests that resilience can be strengthened by being involved in relationships that develop a sense of value, competence, empowerment, and most importantly, connection. People who have relational esteem evaluate their competence to meet expectations positively, and relationship satisfaction means enjoying close relationships, and feeling intimacy needs are met within the current relationship (Snell et al. 1996). Consequently, these two romantic relationship tendencies seem in parallel with relational characteristics that improve resilience.

Studies show that resilience increases marital satisfaction (Bradley and Hojjat 2017), relationship satisfaction mediates the negative relationship between resilience and depressive symptoms (Hou and Ng 2014) and relationship satisfaction has a protective role against psychological distress in the face of difficulties (Manne and Badr 2008, Weitlauf et al. 2014). In conclusion, the finding that relational esteem and relationship satisfaction have positive effects on resilience is also supported by related studies.

In romantic relationships, fear of relationship refers to the anxiety and fear of establishing emotional intimacy with others (Snell et al. 1996). People who are anxious and fearful about romantic relationships may not rely on their ability to cope with possible negative situations they encounter in their relationships, in contradiction with the definition of resilience. In addition, considering that the roots of resilience are based on relational connections (Miller and Stiver 1997), the effect of fear of relationships on resilience, which can prevent the people from being involved in romantic relationships, is consistent with the literature.

The study shows that anxious and avoidant attachment styles have no relationship between the purpose of using social media, also there is no significant relationship between the purpose of using social media and resilience. It is noteworthy that there is no consensus in studies about the relationship between attachment styles and social media usage. According to studies, attachment styles do not predict social media usage (Blackwell et al. 2017), and social media usage differs according to attachment styles. Those with fearful attachment styles use social media to interact and spend more time on these platforms (Baek et al. 2014). The purpose of using social media is associated with indifferent and obsessive attachment styles (Kaplan 2019). The contrasting findings in the literature suggest other variables may affect the relationship between social media and attachment styles. During the Covid 19 pandemic, people spend more time on social media (Mander 2020) and started using these platforms to provide social support and information (Saud et al. 2020). Also, during the Covid 19 pandemic, it is thought that people's social media usage purposes and frequencies have become more similar, and this could be the reason the effect on other variables of purpose of using social media is reduced.

The most featured finding of the current study is that romantic relationship tendencies fully mediate the relationship between anxious and avoidant attachment styles and resilience. This finding indicates that the effects of anxious and avoidant attachment styles on resilience entirely occur through romantic relationship tendency. Romantic relationship tendencies are the reflection of attachment styles in adult romantic relationships, and attachment styles are known to affect many dimensions from choosing a romantic partner to conducting the relationship. Therefore, it is thought that attachment styles that are associated with resilience (Bender and Ingram, 2018) may contribute to the formation of either risk factors or protective factors for resilience through romantic relationship tendencies.

There are many studies indicating that close relationships and social ties are important sources of resilience (Afifi et al. 2016, Jordan 2004, Zautra 2014). Additionally, some studies show that social support has a mediating role in the relationship between avoidant attachment style and resilience, individuals with avoidant attachment style have less perceived social support, and as the perceived level of social support increases the level of resilience also increases (Dodd et al. 2015). Studies indicate that connection, support, approval, and encouragement in romantic relationships contribute positively to the psychological resilience of youth (Joly and Connoly 2019), and that supportive romantic relationships during adolescence positively affect psychological resilience and are a protective factor despite negative experiences with early caregivers (Szwedo et al. 2016).

In addition to all these, there are some studies showing that differences in attachment styles affect social support seeking behaviour and perceived social support (Collins and Feeney 2004, Florian et al. 1995). From this point of view, it may be more difficult for people with avoidant or anxious attachment style to use environmental and relational resources appropriately when they need them. People with avoidant attachment style especially may avoid asking for help, people with anxious attachment style may be perceived as demanding or fussy because of excessively expressing their needs and desires, and as a result, individuals with both attachment styles may have more difficulty in coping with stress (Mikulincer and Shaver 2012). The attachment system is basically related to the ability to perform emotion regulation by using appropriate resources in stressful and challenging situations (Bowlby 1982) and it is thought that close relationships directly affected by the attachment system are an important resource for resilience. In this regard, an individual's style of coping with stress and their capacity to adapt to difficult situations is mainly determined by attachment style and affects their resilience through romantic relationships.

The current study has some limitations. First, the lack of a balanced distribution in terms of gender of the participants (220 women, 69 men) and the lack of information about gender identity and ethnic origins among demographic information can be seen as limitations of the research sample. Second, due to the pandemic restrictions, collecting data online makes it impossible to report respondents' rate of return. Participants who agreed to participate in the study during the current pandemic conditions and quarantine implementation were included in the study. Therefore, the sample size of this study is limited to the participant group that can be reached under the current conditions. Additionally, in the planning phase of the research, it was observed that the scales adapted to Turkish regarding social media use were limited. Considering that the expressions in the romantic relationship tendency scale are aimed at heterosexual individuals, this may have negatively affected the motivation of non-heterosexual individuals to participate in the study. Also, while generalizing and interpreting the results, it should be considered that the validity and reliability studies for the romantic relationship tendency scale were conducted with heterosexual groups. Finally, 104 participants in the sample stated that they are not in any current romantic relationship. The lack of relationship experiences of the participants during the pandemic may also have affected the results.

Conclusion

The results from this study highlighted the role of romantic relationship tendencies on the relationship between anxious and avoidant attachment styles and resilience among Turkish adults in Covid 19 pandemic. There are many research findings in the literature showing that attachment styles are an important factor in romantic relationships. In this respect, we expected a relationship between insecure attachment styles and romantic relationship tendencies. In addition, the study showed that insecure attachment styles were also associated with psychological resilience. In the established model, it was revealed that romantic relationship tendencies mediate the relationship between attachment styles and resilience. This is a notable finding because it can provide some information about how variables that are closely linked to romantic relationships exhibit correlations at a time when psychological resilience becomes important, such as the Covid 19 pandemic. Accordingly, it can be said that the effect of anxious or avoidant attachment style on resilience is shaped by tendency in romantic relationships. Especially in this period, one of the important sources on the psychological resilience of individuals is romantic relationship support. When considered in the opposite direction, it can adversely affect psychological resilience. Insecure attachment styles are one of the factors that has a negative impact on romantic relationships. Thereby, practitioners who want to work on clients' insecure attachment styles may think they can manifest in some romantic relationship tendencies. Thus, it may be possible to plan more concrete interventions.

Since the application of the romantic relationship scales to one of the partners may provide missing or one-sided data about the relationship, in future studies participation of partners in the research together may be useful. In addition, while positive relationship experiences affect resilience positively, it can be thought that negative relationship experiences can have the opposite effect. When viewed from this aspect, investigating the relationship experiences of the participants in similar future studies can provide additional information. It may be useful to investigate whether resilience levels vary according to different romantic relationship tendencies. Finally, in similar future studies with Turkish samples, it would be helpful to include information such as ethnicity and sexual orientation.

In terms of counselling practices, it is thought that the relationship problems of the client may impact their psychological resilience. Especially when working with clients with anxious or avoidant attachment styles, it can be considered that the negative effects of insecure attachment style on resilience may be reduced through interventions aimed at increasing individual satisfaction in romantic relationships and self-confidence in relationships.

Sidebar

Address for Correspondence: Saadet Zümbül, Afyon Kocatepe University, Afyon, Turkey

E-mail: [email protected] Received: 27.06.2022 Accepted: 24.10.2022

ORCID ID: 0000-0002-3812-3482

References

References

Abbasi IS, Alghamdi NG (2017) When flirting turns into infidelity: the facebook dilemma. Am J Fam Ther, 45:1-14.

Abbasi IS (2019). Social media and committed relationships: what factors make our romantic relationship vulnerable? Soc Sci Comput Rev, 37:425434.

Afifi TD, Merrill AF, Davis S (2016) The theory of resilience and relational load. Pers Relatsh, 23:663-683

Ainsworth MDS, Blehar MC, Waters E, Wall S (1978) Patterns of Attachment: A Psychological Study of the Strange Situation. Mahmwah, NJ, Lawrence Erlbaum Associates.

Baek YM, Cho Y, Kim H (2014) Attachment style and its influence on the activities, motives, and consequences of SNS use. J Broadcast Electron Media, 58:522-541.

Baron RM, Kenny DA (1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol, 51:1173-1182.

Bender A, Ingram R (2018) Connecting attachment style to resilience: contributions of self-care and self-efficacy. Pers Individ Dif, 130:18-20.

Blackwell D, Leaman C, Tramposch R, Osborne C, Liss M (2017) Extraversion, neuroticism, attachment style and fear of missing out as predictors of social media use and addiction. Pers Individ Dif, 116:69-72.

Bowlby JM (1982) Attachment and loss: retrospect and prospect. Am J Orthopsychiatry, 52:664-678.

Bradley JM, Hojjat M (2017) A model of resilience and marital satisfaction. J Soc Psychol, 157:588-601.

Büyükşahin A (2005) Çok boyutlu ilişki ölçeǧi: geçerlilik ve güvenilirlik çalışması. Türk Psikiyatri Derg, 16:97-105.

Byrne BM (2016) Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming, 3rd ed. New York, NY, Routledge.

Caldwell JG, Shaver PR (2012) Exploring the cognitive-emotional pathways between adult attachment and ego-resiliency. Individ Differ Res, 10:141-152.

Candel OS, Jitaru M (2021) COVID-19 and romantic relationships. Encyclopedia, 1:1038-1046.

Chen S, Bonanno GA (2020) Psychological adjustment during the global outbreak of COVID-19: A resilience perspective. Psychol Trauma, 12:S51.

Collins NL, Feeney BC (2004) Working models of attachment shape perceptions of social support: evidence from experimental and observational studies. J Pers Soc Psychol, 87:363-383.

Dodd Z, Driver S, Warren A, Riggs S, Clark M (2015) Effects of adult romantic attachment and social support on resilience and depression in individuals with spinal cord injuries. Top Spinal Cord Inj Rehabil, 21:156-165.

Doǧan T (2015) Kısa psikolojik saǧlamlık ölçeǧinin Türkçe uyarlaması: geçerlik ve güvenirlik çalışması. The Journal of Happiness & Well-Being, 3:93-102.

Dehon C, Weems CF, Stickle TR, Costa NM, Berman SL (2005) A crosssectional evaluation of the factorial invariance of anxiety sensitivity in adolescents and young adults. Behav Res Ther, 43:799-810.

Florian V, Mikulincer M, Bucholtz I (1995). Effects of adult attachment style on the perception and search for social support. J Psychol, 129:665-676.

Griffin DW, Bartholomew K (1994) Models of the self and other: fundamental dimensions underlying measures of adult attachment. J Pers Soc Psychol, 67:430-445.

Güloǧlu B, Karaırmak Ö (2017) Erken çocuklukta baba kaybında baǧlanma biçimleri ve yakın ilişkilerdeki psikolojik eǧilimler. Türk Psikolojik Danışma ve Rehberlik Dergisi, 7:99-115.

Hartling LM (2008) Strengthening resilience in a risky world: it's all about relationships. Women Ther, 31:51-70.

Hazan C, Shaver PR (1994) Attachment as an organizational framework for research on close relationships. Psychol Inq, 5:1-22.

Hooper D, Coughlan J, Mullen M (2008). Evaluating model fit: a synthesis of the structural equation modelling literature. In 7th European Conference on research methodology for business and management studies. 195-200.

Hou WK, Ng SM (2014) Emotion-focused positive rumination and relationship satisfaction as the underlying mechanisms between resilience and psychiatric symptoms. Pers Individ Dif, 71:159-164.

Hu L, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Modeling, 6:1-55.

Joly LE, Connolly J (2019). It can be beautiful or destructive: Street-involved youth's perceptions of their romantic relationships and resilience. J Adolesc, 70:43-52.

Jordan JV (2004) Relational resilience. In The Complexity of Connection: Writings from the Stone Center's Jean Baker Miller Training Institute. (Eds JV Jordan, M Walker & LM Hartling):28-46. New York, Guilford Press.

Jordan JV (2017) Relational-cultural therapy. In Handbook of Counseling Women (Eds M Kopala, MA Keitel):340-351. Thousand Oaks, CA, Sage.

Kaplan M (2019) Üniversite öǧrencilerinin sosyal medya baǧımlılık düzeyleri, sosyal medya kullanım amaçları ve baǧlanma stilleri arasındaki ilişkinin incelenmesi (Yüksek lisans tezi). Gaziantep, Hasan Kalyoncu Üniversitesi.

Karaırmak Ö, Güloǧlu B (2014) Deprem deneyimi yaşamış yetişkinlerde baǧlanma modeline göre psikolojik saǧlamlıǧın açıklanması. Çukurova Üniversitesi Eǧitim Fakültesi Dergisi, 43:1-18.

Karasar N (1998). Bilimsel Araştırma Yöntemi. Ankara, Nobel Yayınevi.

Karney BR, Bradbury TN (1995) The longitudinal course of marital quality and stability: a review of theory, methods, and research. Psychol Bull, 118:13.

Keim ME, Noji E (2011) Emergent use of social media: a new age of opportunity for disaster resilience. Am J Disaster Med, 6:47-54.

Kerckhoff AC, Davis KE (1962) Value consensus and need complementarity in mate selection. Am Sociol Rev, 27:295-303.

Kline RB (2015) Principles and Practice of Structural Equation Modeling. New York, Guilford Press.

Luthar SS, Cicchetti D, Becker B (2000) The construct of resilience: a critical evaluation and guidelines for future work. Child Dev, 71:543-562.

MacCallum RC, Browne MW, Sugawara HM (1996) Power analysis and determination of sample size for covariance structure modeling. Psychol methods, 1:130-149.

Mander J (2020) Coronavirus: how consumers are actually reacting. https:// blog.globalwebindex.com/trends/coronavirusand-consumers/ (Accessed 1.12.2021)

Manne S, Badr H (2008) Intimacy and relationship processes in couples' psychosocial adaptation to cancer. Cancer, 112:2541-2555.

Mikulincer M, Shaver PR (2007) Attachment in Adulthood: Structure, Dynamics, and Change. New York, Guilford Press.

Mikulincer M, Shaver PR (2012) An attachment perspective on psychopathology. World Psychiatry, 11:11-15.

Miller JB, Stiver IP (1997) The Healing Connection: How Women Form Relationships in Therapy and in Life. Boston, MA, Beacon Press.

Overall NC, Chang VT, Pietromonaco PR, Low RST, Henderson AME (2022) Partners' attachment insecurity and stress predict poorer relationship functioning during covid-19 quarantines. Soc Psychol Personal Sci, 13:285298.

Pietromonaco PR, Overall NC (2021) Applying relationship science to evaluate how the COVID-19 pandemic may impact couples' relationships. Am Psychol, 76:1-13.

Rasmussen P, Storebø OJ, Løkkeholt T, Voss LG, Shmueli-Goetz Y, Bojesen AB et al. (2019) Attachment as a core feature of resilience: A systematic review and meta-analysis. Psychol Rep, 122:1259-1296.

Reuter C, Spielhofer T (2017) Towards social resilience: a quantitative and qualitative survey on citizens' perception of social media in emergencies in Europe. Technol Forecast Soc Change, 121:168-180.

Savcı M, Aysan F (2016) Yakın ilişkilerde yaşantılar ölçeǧi kısa formu'nun Türkçe uyarlaması: geçerlik ve güvenirlik çalışması. 8. Uluslararası Eǧitim Araştırmaları Kongresi içinde (s. 837-846). Çanakkale, Çanakkale Onsekiz Mart Üniversitesi.

Saud M, Mashud MI, Ida R (2020) Usage of social media during the pandemic: Seeking support and awareness about COVID-19 through social media platforms. J Public Aff, 20:e02417.

Schermelleh-Engel K, Moosbrugger H, Müller H (2003) Evaluating the fit of structural equation models: tests of significance and descriptive goodnessof-fit measures. Methods of Psychological Research, 8:23-74.

Simeona D, Yehudaa R, Cunilla R, Knutelska M, Putnamc FW, Smitha LM (2007) Factors associated with resilience in healthy adults. Psychoneuroendocrinology, 32:1149-1152.

Simon D, Kriston L, Loh A, Spies C, Scheibler F, Wills C et al. (2010) Confirmatory factor analysis and recommendations for improvement of the autonomy-preference-index (API). Health Expect, 13:234-243.

Simpson JA (1990) Influence of attachment styles on romantic relationships. J Pers Soc Psychol, 59:971-980.

Smith BW, Dalen J, Wiggins K, Tooley E, Christopher P, Bernard J (2008) The brief resilience scale: assessing the ability to bounce back. Int J Behav Med, 15:194-200.

Snell WE, Schicke M, Arbeiter T (2002) The multidimensional relationship questionnaire: psychological dispositions associated with intimate relations. In New Directions in The Psychology of Intimate Relations: Research and Theory. (Ed. WE Snell, Jr.). Cape Girardeau, MO, Snell Publications.

Stratton SJ (2021) Population research: convenience sampling strategies. Prehosp Disaster Med, 36:373-374.

Szwedo DE, Hessel ET, Allen JP (2016) Supportive romantic relationships as predictors of resilience against early adolescent maternal negativity. J Youth Adolesc, 46:454-465.

Tabachnick BG, Fidell LS, Ullman JB (2007) Using Multivariate Statistics, 5th edition. New York, Allyn and Bacon.

Teeruthroy VT, Bhowon U (2012) Romantic relationships among young adults: an attachment perspective. International Journal of Humanities and Social Science, 2:145-155.

Usluel YK, Demir Ö, Çınar M (2014) Sosyal aǧların kullanım amaçları ölçeǧi. Eǧitim Teknolojileri Araştırma Dergisi, 5:1-18.

Valenzuela S, Halpern D, Katz JE (2014) Social network sites, marriage wellbeing and divorce: survey and state-level evidence from the United States. Comput Human Behav, 36:94-101.

Vowels LM, Carnelley KB (2020) Attachment styles, negotiation of goal conflict, and perceived partner support during COVID-19. Pers Individ Dif, 171:1-6.

Wei M, Russell DW, Mallinckrodt B, Vogel DL (2007) The experiences in close relationship scale (ECR)-short form: reliability, validity, and factor structure. J Pers Assess, 88:187-204.

Weitlauf AS, Vehorn AC, Taylor JL, Warren ZE (2014) Relationship satisfaction, parenting stress, and depression in mothers of children with autism. Autism, 18:194-198.

Wilks SE (2008) Resilience and academic stress: the moderating impact of social support among social work students. Adv Soc Work, 9:106-125.

Zautra AJ (2014) Resilience is social, after all. In The Resilience Handbook: Approaches to Stress and Trauma. (Eds M Kent, MC Davis, JW Reich):185-196. New York, NY, Routledge.