Parental marital quality and adolescent psychological well-being: A meta-analysis

Abstract Previous findings on the relationship between parental marital quality and adolescent psychological well-being are inconsistent; thus, we conducted a meta-analysis of 23 relevant studies. Results revealed significant associations between: (1) positive parental marital quality and high psychological well-being; (2) positive parental marital quality and low psychological wellbeing; (3) negative parental marital quality and high psychological wellbeing; (4) negative parental marital quality and low psychological well-being. Future research implications will be discussed in this article.

Hepi Wahyuningsih is currently lecturer and researcher in Departement of Psychology and Master of Professional Psychology at Faculty of Psychology and Socio-Cultural Studies, Universitas Islam Indonesia. She graduated from of Doctoral Program of Psychology, Universitas Gadjah Mada with dissertation focusing on marital quality. She teaches developmental psychology, child abnormal psychology, family psychology, statistics, and research methodology.Her current research interest includes marital quality, marital commitment, religiosity, educational psychology, family psychology, and measurement development.
Fitri Ayu Kusumaningrum is currently lecturer and researcher in Departement of Psychology at Faculty of Psychology and Socio-Cultural Studies, Universitas Islam Indonesia. Her research interest includes parenting in children and adolescents. She is active in many social activities involving training and workshop on parenting and other relevant works.
Resnia Novitasari is currently lecturer and researcher in Departement of Psychology at Faculty of Psychology and Socio-Cultural Studies, Universitas Islam Indonesia. She teaches developmental psychology, child abnormal psychology, and early childhood psychology. Her research interest includes well-being in children and adolescents, attachment and parenting, special needs children, and bullying.

PUBLIC INTEREST STATEMENT
This meta-analysis demonstrates the importance of marital quality in adolescent development. It is found that positive marital quality is most strongly related to high psychological wellbeing. But the weighted mean obtained was not highly convincing because of publication bias. Publication bias occurs because studies of marital quality in positive concepts related to psychological well-being in positive concepts are only three studies. This shows the importance of research on the relationship of marital quality with adolescent psychological well-being in a positive concept. In this meta-analysis study also found a sufficient correlation between the marital quality and psychological well-being in a negative concept with a low publication bias. This shows that adolescents who have low psychological well-being because they perceive the quality of parental relationships is not good. Results also may imply that therapy should not only focus on adolescent, but also on directly dealing with their parent.

Introduction
Adolescents are vulnerable to numerous changes and challenges, such as academic, physical changes, social relationships with family and peers, and changes in emotional maturity. This stage is also a period of increasing independence, which includes mental health aspects (World Health Organization, 2013& Saxena et al., 2013. As such, change (Kobak et al., 2017) and autonomy (Kocayörük et al., 2014) are important parts in the development of adolescent mental health. One indicator of adequate adolescent mental health is the achievement of psychological well-being.
The results of a study by a Lancet Commission on Adolescent Health and Well-Being showed that adolescents and young adults have until recently been overlooked in social policy and global health, adolescent have had fewer health gains with economic development than other age groups. The many threats in this global era, including unhealthy lifestyles, family instability, armed conflict, and environmental degradation, are important factors affecting adolescent welfare (Patton et al., 2016). Lu et al. (2018) also reported that an estimated 25% of mental disorders occur in the age group under 24 years in the global population (data from 132 countries). This indicates the urgency of the problem of mental health issues in the world's adolescent population. Slade (2010) stated that prevention efforts to improve mental health involved promoting wellbeing. Psychological well-being not only refers to a condition free from psychological problems, but its meaning is much broader, including the ability of individuals to perceive themselves positively related to others, with environmental mastery, independence, and life goals and emotions that lead to healthy development (Ryff & Singer, 2008).
Other researchers have defined psychological well-being with different variations. De Silva et al. (2017) suggested that psychological well-being in adolescents involves global self-esteem and lack of depressive disorders. On the other hand, Twenge et al. (2017) identified psychological well-being as a component of self-esteem, life satisfaction, self-satisfaction, and happiness. Lopez et al. (2018) stated that psychological well-being is composed of positive and negative affect in addition to the minimum symptoms of depression. Liddle and Carter (2015) defined psychological wellbeing in adolescents as a condition marked by the presence of positive emotions as well as a positive appearance. Tennant et al. (2007) formulated the concept of psychological well-being into 66. Mental well-being has been defined as an emotional intelligence, satisfaction in life, and positive affect and negative affect. Based on previous research, it appears that the construct of psychological well-being is still not well established and requires further study.
A high level of psychological well-being may predict good physical health in the future (Ryff et al., 2015). Low psychological well-being may have a negative impact on future health, such as depression (Yasmin et al., 2015), indicative of poor physical health (Mechanic & Hansell, 1987). Moreover, Huppert (2009) summarized various research results and reported that psychological well-being, marked by the presence of positive emotions, strengthens the immune, hormonal, and physiological systems. Accordingly, psychological well-being may be closely related to the quality of individuals' health, which further reinforces the importance of the role of psychological wellbeing in adolescent development.
However, well-being in general is always associated with psychological distress, both in the form of internalization and externalization. This is evident in some of the results of previous research. The tendency that arises is that when measuring well-being with psychological distress, the correlation value tends to be consistently high. Therefore, both are seen as overlapping constructs (Bartels et al., 2013). In addition, the result of research from Winefield et al. (2012) shows that psychological well-being has a negative relationship with psychological distress. It is also suggested from the results of the research that psychological well-being and psychological distress variables should be measured together. This is also confirmed a literature study which states that in the case of children and adolescents, psychological well-being can be characterized by the absence of symptoms of internalization and externalization (Moore et al., 2016). Based on this explanation, this research tends to use the terminology of high psychological well-being (e.g., life satisfaction, positive relationships with others) and low psychological well-being (e.g., internalization and externalization disorders).
One of the factors that influence adolescent psychological well-being is the role of family and parents. In general, parental marriage contributes to the well-being of children (Ribar, 2015) because of the financial support between husband and wife, social interaction between family members, and stability of the relationship between parents who are bound in marriage commitments. Shek et al. (2014) found that parental marital conditions, family functioning, relationships between children and parents, and communication patterns affect the level of psychological well-being of adolescents. The role of the family becomes very important in the context of adolescent development, especially for emotional (Meggiolaro & Ongaro, 2014) and psychological well-being (Keresteš et al., 2012;Stanescu & Romer, 2011;Tran & Richey, 1997). The study by Anuradha et al. (2012) found that some factors of the family that play a role are attachment, harmony, and family functioning. Conversely, family factors that also have the potential to reduce the quality of adolescent welfare include parents' marital problems, conflict, and divorce. Thus, the quality of parental marriage may also affect the quality of adolescent mental health.
Marital quality is a broad concept. However, researchers try to make a classification based on previous studies. It was stated that marital quality is a continuum of positive and negative dimensions (Leidy et al., 2009). Positive marital quality is characterized by the emergence of positive aspects in marriage and the low negative aspects, vice versa. This is in line with Fincham & Linfield' research that there are positive and negative dimensions in marital quality (Fincham & Linfield, 1997). Positive dimensions are characterized by positive perceptions and feelings on the quality of the couple, as well as positive emotions related to marriage, vice versa. Negative dimensions are also characterized by the couple's inability to adjust to marriage, e.g., marital discord/conflict. Positive parental marital conditions are associated with emotional well-being and low psychological stress in children (Fishman & Meyers, 2000). If parents are able to establish an adequate marital relationship, this will affect the attachment developed by adolescents, and this attachment to parents supports the formation of psychological well-being in adolescents (Žukauskienė, 2014).
On the other hand, a negative quality marriage also has an impact on adolescent psychological well-being. Various studies showed that when the quality of the relationship between parents is low or with high conflict, this tends to lead to psychological problems in adolescents, such as internalizing and externalizing disorders (Goldberg & Carlson, 2014;Ha et al., 2009;Schoppe-Sullivan et al., 2007;Stroud et al., 2015). This may be related to the level of support provided between partners and also to their children. The negative effect is stronger if the low-quality of marriage is accompanied by aggression between the partners, leading to psychological problems in adolescents such as depression and anxiety as well as emotional insecurity (Bergman et al., 2014). Gerard et al. (2006) reported that negative marriage quality, such as spouse's conflict will also cause conflicts between parents and adolescents, ultimately affecting their psychological well-being. Thus, it can be concluded that non-optimal parental marital quality may lead to low psychological well-being in adolescents.
However, the study of the relationship between the two variables is still limited with varying results (Gerard et al., 2006;Jeong & Chun, 2010;Schoppe-Sullivan et al., 2007). This suggests the need to conduct research through a meta-analysis to examine the results of previous research on these two variables. Therefore, this study aimed to examine the association between the quality of parental marriage and the psychological well-being of adolescents based on the results of previous research.There are several proposed hypothesis in this research: 1. There will be positive correlation between positive marriage quality and high psychological well-being among adolescents.
2. There will be negative correlation between positive marriage quality and low psychological well-being among adolescents.
3. There will be negative correlation between negative marriage quality and high psychological well-being among adolescents.
4. There will be positive correlation between negative marriage quality and low psychological well-being among adolescents.

Literature search
Research references on marital quality and psychological well-being are carried out through EBSCOHOST and PROQUEST that are subscribed to by Ministry of Research and Technology in Indonesia. The keywords for psychological well-being used are psychological wellbeing, wellbeing, happiness, self-esteem, depression, externalizing, internalizing, life satisfaction, purpose in life, optimism. These keywords were combined with the keywords for marital quality. The keywords for marital quality used are marital quality, marital satisfaction, marital adjustment, marital happiness. marital hostility, inter-parental conflict, parental conflict, marital conflict. For data updates, the references sought are limited from 2007 to 2019. The inclusion criteria of this study: (1) report the correlation coefficient (r); (2) reporting the number of research subjects; and (3) include children/ adolescents/students/parents as research subjects. Afterward, researchers have discussed to select the journals based on the inclusion criteria. Based on the inclusion criteria, 23 journals were obtained that met the criteria. The detailed journals can be seen in Table 1.

Meta-analysis procedure
As explained in the introduction, both the concept of marital quality and the concept of psychological well-being are broad concepts. Therefore, to see the potential for moderating variables due to the broad concept of each variable, the first step taken by the researcher is to make a grouping both for the concept of marital quality and for psychological well-being. There were four groups, namely, (1) positive marital quality; (2) negative marital quality; (3) positive psychological wellbeing; and (4) negative psychological well-being. The detailed four groups can be seen in Table 2.
Then measure heterogeneity using the homogeneity statistic Q for selecting fixed or random effects models. I squared index is also used to see the level of heterogeneity. According to Huedo-Medina et al. (2006) low heterogeneity is indicated by the value of I squared equivalent to 25%, medium heterogeneity is indicated by the value of I squared equivalent to 50%, and high heterogeneity is indicated by the number I squared is equivalent to 75%. Weighted average correlation coefficient (r) is calculated from each study, besides the researcher also calculates 95% confidence intervals. To see the publication bias also included the results of fail safe N. All analyzes were performed using CMA software.

Results
The results of the meta-analysis are provided in Table 3. Before analyzing the weighted mean for each hypothesis, the results of the heterogeneity test need to be considered. The heterogeneity test of the relationship of positive marital quality with positive psychological well-being was not significant (Q = 1.45, p >.01, I2 = 0%), which indicates that the null hypothesis is accepted, meaning that there is homogeneity in the effect size of the correlation between positive marital quality and high psychological well-being; so we used a fixed effects model. The weighted mean of the correlation between positive marital quality and high psychological well-being was significant ). This small fail-safe N value indicates that the weighted mean obtained was not highly convincing because of publication bias.
The heterogeneity test of the relationship of positive marital quality with low psychological wellbeing showed a significant result (Q = 79.13 p < .01, I2 = 86.10%), indicating no homogeneity (i.e., heterogeneity) in the effect size of the correlation between positive marital quality and low psychological well-being. The weighted mean of the correlation between positive marital quality and low psychological well-being was significant (r = −.092; p < .01; fail-safe N = 1080). This high fail-safe N value indicates that the weighted mean obtained was very convincing. Although the correlation between positive marital quality and low psychological well-being is influenced by moderator variables, the publication bias is low.
The heterogeneity test of the relationship of negative marital quality with high psychological well-being was not significant (Q = 19.30; p > .01, I2 = 42.99%), suggesting homogeneity in the effect size of the correlation between negative marital quality and high psychological well-being. The weighted mean of the correlation between negative marital quality and high psychological well-being was significant (r = −.173; p < .01; fail-safe N = 489). This shows that the weighted mean of this relationship was significant and convincing because it did not experience publication bias.
The heterogeneity test of the relationship of negative marital quality with low psychological well-being was significant (Q = 784.3 p < .01, I2 = 89.93%), suggesting no homogeneity in the effect size of the correlation between negative marital quality and low psychological well-being. The weighted mean of the correlation between negative marital quality and negative psychological well-being was significant (r = .186; p < .01; fail safe N = 6892).

Discussion
The first hypothesis was to examine the relationship between positive marital quality and high psychological well-being of adolescents. Positive marital quality was measured through marital quality and marital satisfaction. Positive psychological well-being of adolescents was measured through self-esteem. The results of the first meta-analysis indicated the weighted mean of the correlation between positive marital quality and high psychological well-being of adolescents was positive and significant. Moreover, the data were found to be homogeneous. This is consistent with the study by Shek (2000) emphasizing that the quality of parental marriage in general can be an indicator of adolescent psychological well-being. Adolescent self-esteem can be influenced by marital satisfaction of the mother and father (Mutchler, 1991). High adolescent self-esteem has an impact on other positive behaviors, with adolescents also displaying higher psychological wellbeing (Sarkova et al., 2014). Adolescent self-esteem was found to be influenced by marital quality of the parents (Jeong & Chun, 2010).
The second hypothesis was to analyze the direction of the relationship between positive marital quality and low psychological well-being of adolescents. Positive marital quality was measured as marital satisfaction, marital quality, marital warmth, and inter-parental positivity. Low psychological well-being was assessed as internalizing problems, externalizing problems, and depression. The weighted mean of the direction of the relationship between positive marital quality and low psychological well-being of adolescents was negative and significant. Merrilees et al. (2018) reported that the interaction that occurs between mother and father will affect the feelings of sadness and happiness of teenagers. If the interaction is positive, the emotions displayed by adolescents will be positive, and vice versa. Heterogeneous data was found in this analysis; accordingly, it is possible to have a moderator in the relationship between positive marital quality and low psychological well-being in adolescents. Baril et al. (2007) identified the role of parenting moderators between positive marital quality, namely, marital love with depressive and risky behavior in adolescents. In addition, the high heterogeneity is due to broad concepts that lead to the use of various measurements.
The third hypothesis was examined the direction of the relationship between negative marital quality and high psychological well-being of adolescents. Negative marital quality was measured as couple discord, inter-parental conflict, inter-parental hostility, and parental fighting/disagreement. Regarding high psychological well-being of adolescents was measured as positive affect, optimism, self-esteem, life satisfaction, and happiness. The weighted mean of the relationship between negative marital quality and high psychological well-being of adolescents was significant and negative. Subsequently, we found homogeneous data. This was confirmed by Hernandez et al. (2018), reporting that adolescent well-being could be defined by several psychological constructs, for example, self-esteem, life satisfaction, happiness, positive affect, optimism, life meaning, and purpose. Plunkett and Henry (1999) reported that the quality of parental marriage perceived by adolescents in the form of frequency of conflict, child rearing, role/finance conflict, and overt or covert conflict negatively contributed to the life satisfaction of all family members as perceived by adolescents. Xin et al. (2009) identified that positive emotions such as satisfaction, optimism, usefulness, and self-confidence were negatively correlated with parental marital conflict. Bradley and Corwyn (2004) emphasized that marital status and conflict in family settings negatively adolescents' life satisfaction. Lucas-Thompson et al. (2017) reported that adolescents perceiving threatening parental conflict expressed lower happiness levels. Adolescents' stressful situations may result from low marital satisfaction of mother and father (Fishman & Meyers, 2000;Mutchler, 1991). London-Johnson (2010) reported that couple discord has a negative influence on adolescents' positive emotions. Moreover, Chaudhry and Shabbir (2018) reported that optimism and inter-parental conflict are negatively correlated, while inter-parental conflict, aggression and pessimism are positively correlated.
The fourth hypothesis was to examine the correlation of the relationship between negative marital quality and low psychological well-being of adolescents. The weighted mean of the direction of the relationship between negative marital quality and low psychological well-being of adolescents was significant and positive, and also indicated heterogeneity. Amato and Sobolewski (2001) reported that marital discord was associated with lower levels of adolescent psychological well-being. Doyle and Markiewicz (2005) found that adolescent self-esteem was negatively correlated with marital conflict. Violence between partners (inter-parental violence) can affect children's self-esteem until adulthood (Shen, 2009). Children become unhappy when their parents experience a dispute (Uusitalo-Malmivaara & Lehto, 2013). Another study found a relationship between adolescents' perceived quality of the parental relationship and adolescent happiness (Luk et al., 2017). Xin et al. (2009) identified that negative emotions such as disappointment, confusion, feeling lonely, frustration, and pessimism were positively correlated with conflict between parents. Sadness is a negative emotion that also results from marital discord between parents (London-Johnson, 2010). Spouse conflict can contribute to adolescent depression (London-Johnson, 2010), and Bradford et al. (2008) found that overt and covert conflicts between parents have an impact on adolescent depression. The quality of marriage can be inferred from the presence of marital conflict, and Cummings et al. (2012) found that parental marital conflict increased the level of teen depression.
Regarding positive and negative marital quality related with low psychological well-being of adolescents, we found heterogeneous data. Based on this, it is possible to have a moderator in the relationship between negative marital quality and negative psychological well-being in adolescents. Cummings et al. (2012) reported the moderator role of parenting in the relationship of marital conflict with adjustment problems. Bradford et al. (2008) found that religiosity moderated the relationship of marital conflict with adolescent depression. Benson et al. (2008) found the role of moderators in the relationship of marital conflict with depression and internalizing problems. In addition, the results of high heterogeneity due to broad concepts that have an impact on the use of various measurements.
The relationship between positive marital quality and low psychological well-being is tend to be heterogonous. This is possible because of the contribution of moderator or mediator variables. From the results of previous studies, it is possible that other variables that influence include: the level of parenting stress (Robinson & Neece, 2015), gender and attachment (Nelson-Coffey, 2018), and parent-child involvement (Fishman & Meyers, 2000). Based on this, it appears that there are various factors that are possible to contribute in determining the relationship between the two variables.
The association between negative marital quality and low psychological well-being of adolescents has been summarized in previous studies. Cui and Conger (2008) have examined that parenting behaviors are moderating marital probes and adolescent maladjustment. Larsen et al. (2007) found that moderators in inter-parental conflict and maladjustment in adolescents are friendship quality. Furthermore, Lindsey et al. (2009) explain the role of mother-child emotional reciprocity from a correlation study between marital conflict and adolescent peer aggression.
From this study, it can be inferred some conclusions. First, the broad concept in both variables (marital quality and psychological well-being) impacted in variability of the results.
There was a positive correlation between positive marital quality and high psychological wellbeing. Contrary with that, there was a negative correlation between positive marital quality and low psychological well-being. Then, the correlation was also negative between negative marital quality and high psychological well-being. On the other side, there was a positive correlation between negative marital quality and negative psychological well-being was significant. Based on that, the hypotheses are relevant with the findings. Second impact of the broad concepts implied on heterogeneity and homogeneity. For second and fourth hypotheses, the variables were affected with moderator variables.

Limitations
The limitations of this study are several things. First, the construct of marital satisfaction is very broad, so that it impacts on the categorization of positive and negative concepts. The same thing with the construct of psychological wellbeing, conceptualized at low and high levels of wellbeing. Second, the terms variable are used widely so that various measuring tools appear. Third, in the journal selection process the inter-rater writers did not make an effort to the worthiness of the substance of the object of study of the article. Fourth, there are several sources of research methodology, namely observations and perceptions on the part of children and perceptions of partners regarding the quality of marriages. This will have different viewpoints in the interpretation of the actual data.

Future research
The broad concept of psychological well-being can be realized in the realm of affection and behavior. This will have an impact on the variety of measuring instruments used so that the results are possible heterogeneous. Future studies can focus on the psychological realm of the intended psychological well-being. Another thing is that various methods of the journal study object from the point of view of adolescents and parents can provide different interpretations, further research can focus on just one source of data collection. For more valid results, efforts are needed from the research team to provide an assessment of each of the quality of the journal.
Further researchers also need to put in a stricter effort in the process of reducing data from the various stages passed so that it truly meets the specified criteria.