The Effect of Gender and Attachment Styles on the Relationship between Marital Adjustment and Psychological Symptoms

The aim of this study is to investigate the moderating effects of attachment styles and gender on the predictive strength of marital adjustment on psychological symptoms among Turkish married individuals. Correlational model was used and the sample consisted of 178 married individuals. The data was gathered through online survey. The Turkish form of Brief Symptom Inventory (BFI) was used to measure psychological symptoms. The Turkish form of Dyadic Adjustment Scale (DAS) was used to measure marital adjustment and the Turkish form of Relationship Scales Questionnaire (RSQ) was used to measure attachment styles. The hypothetical model was formed and tested through path analysis technique of structural equational modeling. Results showed that secure, dismissing and fearful attachment styles and gender moderated the predictive strength of marital adjustment on psychological symptoms. It was found that the relationship between marital adjustment and psychological symptom is stronger among individuals who have secure attachment style.


Introduction
The quality of marital relationships and the experiences in marriage affect couples' psychological health (Jabamelian, 2011).The marital problems, conflicts, adjustment issues and lack of communication may cause depression and anxiety disorders (Hafner & Spence 1988;Kim 2012;Whitton & Whisman 2010).The dyadic adjustment is an important factor related to the quality of marriage and psychological symptoms (Stack & Eslhman, 1998).The dyadic adjustment includes sharing the similar opinions about marriage in general and marital issues, it includes managing to solve problems as a couple, having marital satisfaction, feeling happiness of marriage together and evaluating positively about marriage (Erbek, Bestepe & Akar, 2005;Fisiloglu, 2001).Not all individuals experiencing marital dysfunction and maladjustment have psychological problems (Scot & Cordova, 2002).There are some psychological agencies which directly affect the perception of dyadic adjustment and indirectly affect couples' psychological health.
Attachment styles of married couples are one of these psychological agencies and affect dyadic adjustment in terms of how couples perceive marital experiences, how they evaluate their partners and what kind of strategies they use when they have problems.(Wayment & Campell, 2000).A married individual's evaluation of himself/herself and evaluation of his/her partner and the relationship schemes of his/her mind differentiate depending on the attachment styles of him/her formed in infancy and childhood and transferred into adolescent and adulthood (Gallo & Smith, 2001).Individuals who have insecure attachment styles have more depressive and anxiety symptoms and have more difficulties in their intimate relationships (Bowlby, 1980;Keskin & Cam, 2008).Attachment styles may affect the relationship between marital adjustment and psychological symptoms by functioning as a moderator variable in this relationship (Beach & O'Leary;Scott & Cordova, 2002).
Another possible moderator variable which may affect the relationship between marital adjustment and psychological symptoms is gender.Women and men behave differently in their romantic relationships and they perceive situations, problems, conflicts and many other issues related to the relationship differently considering the traditional sex roles the society imposes to them (Surra & Longstreth, 1990).
The aim of this study is to determine the moderating effects of attachment styles and gender on the predictive strength of marital adjustment on psychological symptoms among Turkish married couples.

Sub-problems
The research questions of the study were set as below; 1. What are the correlations between marital adjustment and gender, age, years of marriage, total income and the number of children? 2. What are the correlational relationships between psychological symptom and gender, age, years of marriage, total income and the number of children? 3. Do the attachment styles and gender moderate the predictive strength of marital adjustment on psychological symptoms?

Research Design
This research is a quantitative study, examining the moderating effects of attachment styles and gender on the predictive strength of marital adjustment on psychological symptoms among married individuals by using correlational model.The hypothetical model shown in Figure 1 was tested through path analysis technique of structural equational modeling by using IBM SPSS AMOS 21 program.First, confirmatory factor analysis was applied to the scales used in this study in order to see the factor structure.The independent variable of the model is determined as marital adjustment and the dependent variable is determined as psychological symptoms.The possible moderator variables are determined as secure, fearful, preoccupied, dismissing attachment styles and gender.Correlational models are the research models aiming to determine the existence and/or the degree of the correlation between two or more variables.Even though correlational model doesn't provide a real cause-effect relationship, it allows to predict a variable if the status of the other is known (Karasar, 2006).

Research Sample
The data was gathered from 178 married individuals (62 males, 116 females) through online survey.Structural equational models are Large or small sample sizes affect fit indices negatively (Kenny & McCoach, 2003).In structural equational models, sample size below 100 are defined as small, sample size between 100 to 300 are defined as medium, and sample size more than 300 are defined large (Kline, 1998(Kline, , 2005)).It is emphasized that the sample size should be ten times more than the parameters that are measured but it should not be less than 150 (Anderson & Gerbing, 1988;Bentler & Chou, 1987).In a study investigating the sample size for structural equational modeling (Dogan, 2015), it was found that for GFI, AGFI, CFI, NNFI, IFI, RMSEA and RMR fit indices 150 data and more is acceptable.Participants' age ranges from 19 to 63 and the mean age was found 32.3.The children participants have range from none to 4 and the mean children number was found 0.95.Gender, years of marriage and total income distribution of married individuals who have participated in the study are given in Table 1.As seen in Table 1, most of the married individuals' years of marriage fall into 0-5 years.These individuals can be named as newly married individuals when compared to others.Considering total income, most of the married individuals' total income falls into 2500-5000 Turkish liras group.It indicates that most participants belong to the middle-class in terms of socio-economic status.

Correlations
The correlations between marital adjustment, psychological symptoms, gender, age, years of marriage, total income and the number of children are given Table 2.As seen in Table 2, marital adjustment has a negative correlation with psychological symptoms in a significant level (r= -.36, p< .01).The increase in marital adjustment plays a preventive role against the emergence of psychological symptoms.
The number of children has a negative correlation with marital adjustment in a significant level (r= -.27, p< .01).It indicates that the more children couples have the less marital adjustment they are more likely to experience.
The total income has a negative correlation with psychological symptoms in a significant level (r= -.36, p< .01).It shows that the increase in socio-economic level promotes individuals' psychological well-being.
The years of marriage have a negative correlation with marital adjustment in a significant level (r= -.21, p< .01).Spending years together as a married couple has a negative effect on marital adjustment.
The age has a negative correlation with marital adjustment in a significant level (r= -.20, p< .01).Older married individuals explain less marital adjustment than younger married individuals.
The gender has a negative correlation with marital adjustment (r= -.18, p< .05)and has a positive correlation with psychological symptoms (r= .21,p< .01).It means that the married women are more likely to have less marital adjustment and more psychological symptoms.

Model Fit Indices
The hypothetical model was tested and the fit indices of the model confirmed the significant relationships between the variables.The fit indices of the model were given in Table 3.

Moderating Effects
The regression weights of marital adjustment on psychological symptoms for each attachment styles and gender were given in Table 4.

Discussion and Conclusion
This study was conducted to determine the moderating effects of attachment styles and gender on the predictive strength of marital adjustment on psychological symptoms among Turkish married individuals.The correlations between marital adjustment, psychological symptoms, gender, age, years of marriage, total income and the number of children were also analyzed within the study.
The women participating in the study have less marital adjustment and more psychological symptoms than men.This finding is consistent with previous studies (Sendil & Kizildag, 2003;Sendil & Korkut, 2012).In eastern cultures, married women are supposed to have more responsibilities of marriage than men in terms of housework and they are supposed to be passive rather than being active and equal to men in terms of directing and managing the family issues with the effect of traditional gender roles (Cheung, 2005).These conditions may cause women to complain more about their marriage, to have less marital satisfaction and more psychological symptoms especially depression and anxiety symptoms.
The total income has a positive correlation with married individuals' psychological health in a significant level.This finding is consistent with previous studies (Bradbury, Fincliam & Beach, 2000;Dakin & Wampler, 2008).The increase in total income promotes couple's life quality by enhancing participating in more social, educational and intellectual activities and having more physical means of living.Low-income and unemployment cause stress on couples related to getting along with living and this endanger their marital adjustment and psychological health.
The number of children has a negative correlation with marital adjustment in a significant level.It indicates that the increase in number of children diminish the quality of marital adjustment.Many studies demonstrating the negative correlation between the number of children and marital adjustment have been conducted (Belsky, 1990;Cummings, 1994;Grych & Fincham,1990;Sendil & Korkut, 2012).Couples have to face new economic issues related to caring of their baby when they have a child and new conflicts may occur.Couples also begin to spend less time together as a couple because growing up the child requires much more effort and time and these issues may reduce the quality of their marriage and marital adjustment.
The age and years of marriage have a negative correlation with marital adjustment in a significant level.Aging together as a married couple may reduce performing the behaviors related to love, intimacy, passion and compassion and this may cause marital adjustment to drop (Lauer & Lauer, 1997;Sendil & Korkut, 2012).
The predictive strength of marital adjustment on psychological symptoms was analyzed considering individuals' attachment styles and gender as moderators.Results indicated that secure, fearful and dismissing attachment styles and gender moderate this relationship in a significant level and secure attachment was found the strongest moderator.
Individuals with secure attachment style comfort with closeness and have low anxiety over their relationships.They view themselves and evaluate others positively.These people report greater marital satisfaction and adjustment (Banse, 2004;Egeci & Gencoz, 2006;Ozmen ve Atik, 2010;Scott & Cordova, 2002).In this study, it was found that the relationship between marital adjustment and psychological symptoms is more stronger among individuals who have secure attachment style.These individuals' marital adjustment predict their psychological symptoms negatively.
Results revealed that when secure adults have marital problems and low marital adjustment they get more prone to have psychological distress.It may be claimed that secure married individuals see interpersonal and intimate relationships as important factors promoting their psychological health and when they experience problems related to relationships like intimate relationship, their psychological health gets more likely to be affected negatively.Another possible explanation about secure adults is that these individuals are not used to experience relational problems because their past relational patterns include positive and healthy interactions and when they face relational problems, the lack of schemes and defense mechanisms to overcome these problems or conflicts cause them to have more psychological distress.
Individuals with fearful attachment style have high avoidance and anxiety over their relationships and they view themselves and others negatively.They are more likely to have less marital adjustment and more psychological symptoms (Banse, 2004;Gallo & Smith, 2001).Their negative expectations and interpretations over a situation, action, person or thought direct their relational patterns (Collins, 1996;Mikulincer, 1998).Fearful adults have serious doubts about their own lovability and they deny their need for others' love and mistrust others.These individual' past relational patterns consist of negative and unhealthy interactions with others.These repeated unhealthy interactions cause fearful adults to have low self-worth.It may be that when they encounter marital conflicts or problems their concerns about the stability and the quality of their partners are activated and because of high avoidance they perform they do not put an effort to solve their problems.The high level of anxiety they have, their doubts about their relationships and the feeling of low self-worth endanger their psychological health.
Individuals with dismissing attachment style have high avoidance and low anxiety over their relationships.They view themselves positively and evaluate others negatively (Bowlby, 1980).Avoidant behaviors in marriage are negatively correlated with marital adjustment and in some studies, were found the strongest predictor of marital adjustment (Noftle & Shaver, 2006).Their avoidant behaviors include self-protective and distancing strategies when faced with relational conflicts and problems (Simpson, Rholes, & Phillips).Dismissing adults remain emotionally distant from their romantic partners regardless of the quality of their marital adjustment (Scott & Cordova, 2002).They try to isolate themselves from being affected by problems that relational dysfunction creates.The findings of this study demonstrated that even though dismissing adults struggle not to be affected by marital problems by denying or ignoring them, their psychological health gets affected negatively but the effect size is smaller comparing to secure and fearful attachment styles.
It was determined that married women tend to have more psychological symptoms than men when they experience marital maladjustment or they tend to have less psychological symptoms than men when they experience high marital adjustment.The quality of women's romantic relationships is more associated with their psychological health than men.Many researches have shown that women struggle more to maintain their relationships and they consider their romantic relationships more important than men (Huston, Surra, Fitzgerald, & Cate, 1981;Surra & Longstreth, 1990).Traditional sex roles may cause women to take more responsibilities of romantic relationship or marriage in terms of holding family together and maintaining intimate interactions.When dysfunction occurs in marital relations, women may be more likely to have psychological distress because of these responsibilities.

Figure
Figure 5. Female

Consensus Satisfaction Expression Cohesion Adjustment Symptoms Depression Anxiety Negative self Somatization HostilityTable 1 .
Demographic variables

Table 2 .
Correlations between variables

Table 3 .
Model fit indices