Mediating Effects of Family Stress on the Relationship between Self-Efﬁcacy and Midlife Crisis in Middle-Aged Men

: Middle-aged men bridge younger and older generations and are burdened with social roles and responsibilities that increase their risk of poor health. The aim of this study was to investigate whether family stress had a mediating effect on the relationship between self-efﬁcacy and midlife crisis in a sample of 198 middle-aged South Korean men. Mediation analysis was performed according to the Baron and Kenny method and using the Sobel test. This study conﬁrmed that midlife crisis decreased with increased self-efﬁcacy and increased with family stress level. Family stress was also a signiﬁcant partial mediator of the relationship between self-efﬁcacy and midlife crisis. As family stress inﬂuences all members of the family, interventions are required that engage all family members to alleviate stress. Strategies to boost self-efﬁcacy should be utilized, and family support constitutes the best means of coping.


Introduction
Middle adulthood, also called middle age, is a pivotal period in one's life. This stage generally spans ages 40 to 60, although some individuals have self-identified as middleaged until age 80 [1]. Because middle-aged adults experience diverse and complicated experiences, life during this period is characterized by different stresses and crises [2,3]. In South Korea, middle-aged men bridge the younger generation (20s and 30s) and older generations (age 65 and older) and are burdened with social roles and responsibilities that increase the risk of health problems [4]. Middle-aged men entering this transitional life stage are exposed to job stress and stressful psychological work environments [4,5]. However, men are not accustomed to dealing with their mental stress or psychological symptoms [6][7][8]. In addition, men tend not to seek help from their family or friends, which may have its origins in middle childhood. For example, Rose et al. [9] found that boys had trouble expressing their emotions and considered doing so to be strange or a waste of time.
A midlife crisis refers to physical, social, and emotional confusion and conflict that may be experienced as a middle-aged adult faces a developmental transition [10,11]. Moreover, a midlife crisis may be triggered by multiple life stressors or events, such as job demands, financial responsibilities for family, parental death, or children leaving home upon entering adulthood [12,13]. These factors can generate a family crisis [10], increase the difficulty of preparing for a successful postretirement life, and eventually hinder the transition into older adulthood [10,14]. A study by Freund and Ritter [13] reported that by middle age, new goals need to be set for internal and external changes, which are stressed by the central role of modifying old goals and setting new developmental goals. Although a midlife crisis is negative, there is a possibility that it can be a transformative experience that contributes to the positive development of one's life depending on how one accepts and responds to stress; people who suffer from it will try to adapt to greater life challenges as they get older.
Self-efficacy is a sense or belief that an individual can effectively control the abilities he or she needs in life [15]. In this sense, self-efficacy beliefs may contribute to how an Sustainability 2021, 13, 3761 2 of 12 individual adjusts to this life stage transition. Self-efficacy, which is a person's belief about the degree of confidence in oneself to achieve desired goals [15], has a positive effect on an individual's life satisfaction [16]. Studies on midlife crises in the South Korean population reported that perceived midlife crises decreased with increased self-efficacy [17,18].
The degree of adjustment to consequent changes in the family life cycle and family relationships due to role conflicts and burdens induces family stress [19][20][21][22]. Family stress is a pressure that disrupts the stability of the family and is the tension and pressure within the family system that individuals and families experience in the process of developing over time [23]. As a result, some middle-aged men enter into a "crisis," or a stressful period of balancing their new roles in the family. For example, they may need to re-establish relationships with their wives because they had been focused on the children who have now moved out of the family home. It is important to note, however, that family stress does not always cause problems or crises. Stress levels may differ depending on one's capacity to cope with stress and whether family support is provided. Moreover, individuals may even experience family stress as a result of expectations and in positive situations. Thus, making adjustments based on tension in these stressful situations could actually help resolve problems [24].
Overall, family stress is an important factor in determining one's well-being and maintaining good psychological and emotional health in middle-aged South Korean men. However, few studies have examined whether family stress might mediate the relationship between self-efficacy and midlife crisis, especially among middle-aged South Korean men. Therefore, this study aimed to examine whether family stress has a mediating effect on the relationship between self-efficacy and midlife crisis in middle-aged South Korean men.

Midlife Crisis
A midlife crisis is when middle-aged individuals experience physical, social, and emotional confusion and conflict in the face of a developmental transition. In particular, men who have been looking forward to social success without time to take care of their surroundings or their inner selves experience physical and social crises as they enter middle age and experience psychological crises, such as meaninglessness, emptiness, loneliness, and depression [11,25]. Middle-aged men begin to show signs of danger to their health along with individual physical weakness as they age. Given that their role as parents diminishes as their children grow up and their social and economic roles are limited, they face various difficulties and conflicts among generations and experience midlife crises, such as anxiety, stress, and depression [1,2]. The midlife crisis can vary depending on how individuals adapt to various changes, such as physical, professional, and sexual changes, in the process of changing and adapting to the middle-aged period. If people adapt well to these changes, they can lead a more mature life, but if they do not adapt well, it will not only negatively affect their mental health [26] but also lead to a family crisis [27]. A study by Chang [28] found that the higher the stress, the higher the severity of the midlife crisis, and that stress is a factor affecting the midlife crisis. Thus, a midlife crisis can vary in crisis level by the negative appraisal they give their changing situations. In other words, the level of midlife crisis may vary depending on how individuals cope with the situations that they experience. Therefore, it is necessary to identify factors that can control or lower the level of midlife crisis.

Self-Efficacy
Bandura has defined self-efficacy as the belief in one's ability to succeed in particular situations or accomplish a task [15]. In other words, self-efficacy is an important factor in dealing with or solving problems. Lazarus and Folkman [29] said that high self-efficacy can be better served in difficult situations, suggesting that self-efficacy is a variable that affects the ability to solve problems. Self-efficacy is reported to affect emotional stability by affecting the ability to manage negative emotions [30]. A study by Bauer and Bonanno [31] found that self-efficacy affects psychological health in crisis situations. It is reported that self-efficacy is lowered by perceived stress [32]. A study of middle-aged people in Korea found that the higher the self-efficacy, the lower the perceived midlife crisis in middle-aged people [18]. It is important to increase self-efficacy in middle age because middle age is a turning point in life, where one must prepare for retirement. The higher the self-efficacy, the more confident a person is and does not avoid threatening situations or conflicts [15]. Hence, it is an important factor in preparing for a successful retirement by continuously promoting and maintaining health promotion. In this regard, self-efficacy can be predicted as an influential factor in crisis situations and stress.

Family Stress
Family stress is the pressure to change the family system caused by any event or situation that changes the stable family status and causes tension [23]. Middle-aged men often develop stress in their family relationships, and their stress levels are high [21]. Lazarus and Folkman [29] define psychological stress in the cognitive-relational theory of stress as "a particular relationship between individuals and the environment that is appraised by the individuals as taxing or exceeding their own resources and endangering their wellbeing." Lazarus and Folkman [29] viewed stress as a complex and dynamic interaction between individuals and the environment. The individual not only mediates the stimuli and responses of the environment but also personal characteristics, such as perception, cognition, and the ability to cope with stress, that become part of the environment and affect the environment. That is, they emphasize the cognitive appraisals of individuals in mediating the effects of stressful situations that can affect the stress experience. Even if the stressful situation is the same, emotional and behavioral responses vary depending on the difference in cognitive appraisal. This appraisal goes through two cognitive mechanisms: primary and secondary appraisals. Primary appraisal is used to determine whether a situation will benefit the present or future. In the primary appraisal, the situation is evaluated as irrelevant, benign-positive, or stressful to the well-being of the individual. The situation is divided into stress assessments if it is perceived as harmful, lost, threatening, and challenging to the individual. Secondary appraisal is an assessment of whether an individual has the resources and abilities to cope with a situation. In that situation, what is a useful coping option is whether it can be efficiently performed and what the outcome of the coping method is evaluated as [29]. As such, stress is associated with an individual's subjective assessment of the situation, and the primary and secondary appraisals interact to determine the stress level, emotional response level, and content. Moreover, family stress has been confirmed to be closely related to mental health [27]. As family stress leads to changes in individual self-awareness and identity [33], it could represent another crisis in the developmental cycle. In this regard, family stress is a very important issue to monitor and investigate. In a study by Rayan [34], stress was significantly correlated with selfefficacy, while a study by Dos Santos [35] showed that family stress as an environmental factor plays an important role in reducing self-efficacy. According to a study of Koreans, family stress negatively affects psychological well-being in middle age [36]. These results allow us to predict that family stress perceived by middle-aged male individuals is an important factor in maintaining their well-being and psychological and emotional health. However, studies that have identified mediating effects to control the midlife crisis in middle-aged men are rare. Therefore, in this study, we tried to determine whether family stress had a mediating effect on the relationship between self-efficacy and midlife crisis in middle-aged men.

Study Design
This study is a descriptive correlational study aimed at identifying the mediating effect of family stress on the relationship between self-efficacy and midlife crisis in middle-aged South Korean men.

Participants
The participants were 198 middle-aged South Korean men who were employees of companies located in three cities in South Korea. They were between the ages of 38 and 61 years, employed in the three cities, had no communication problems, and provided informed consent to participate. Using G-power 3.1 software, the minimum sample size was calculated to be 107 with a significance level of 0.05, power of 0.95, and moderate effect size of 0.15. Accounting for potential withdrawals of 10% and incomplete responses, the survey was distributed to 200 men. Ten participants were excluded due to missing data and incomplete responses. Therefore, the final sample comprised 190 participants (response rate of 95%).

Self-Efficacy
Self-efficacy was measured using 17 items for general self-efficacy from the Self-Efficacy Scale originally developed by Sherer et al. [37] and modified and adapted by Hong [38]. Each item was rated on a five-point Likert scale from 1 ("strongly disagree") to 5 ("strongly agree"). Self-efficacy includes items such as, "Failure just makes me try harder, When I make plans, I am certain that I can make them work, and I am a self-reliant person." Eleven negatively worded items were reverse scored, and a higher score indicated a higher level of perceived self-efficacy. Cronbach's α was 0.86 in Hong's [38] study and 0.91 in our study, indicating high reliability.

Family Stress
Family stress was measured using an instrument based on the Family Inventory of Life Events and Changes developed by McCubbin et al. [39]. The instrument comprises 22 items about family relationships, family finances, family occupational relationships, family health, and family loss. Each item is rated on a five-point Likert scale ranging from 1 ("strongly disagree") to 5 ("strongly agree"), and a higher score indicates a higher level of family stress. Cronbach's α was 0.88, indicating high reliability.

Midlife Crisis
Midlife crisis was measured using a scale comprising 53 items and five subscales that we developed by modifying the instrument created by Kim and Yoon [40] and revised by Kim [41]. The subscales assess emotional crisis ("I feel so empty and numb"), individuation ("I spend a lot of time thinking, 'Who am I?' and 'What have I done in my life?"), loss of vitality ("I can feel it every day, the limit of my ability, and I can't work as much as I used to"), occupational dissatisfaction ("I haven't accomplished anything enough in my life"), marital dissatisfaction ("I often feel more disconnected from my children and wife than ever"), and dissatisfaction with the relationship with one's children ("My children don't need me or want to spend time with me anymore." Each item is rated on a five-point Likert scale ranging from one ("strongly disagree") to five ("strongly agree"), and a higher score indicates a higher level of crisis. Cronbach's α for the entire scale was 0.96 in Kim's [41] study on middle-aged adults and 0.97 in our study, indicating high reliability.

Data Collection
We collected data over a two-year period, from 1 April 2017 to 3 April 2019. A postal or online questionnaire was sent to 200 middle-aged men who consented to participate. Completing the survey took approximately 15 min. Participants answered self-report questionnaires on self-efficacy, family stress, and midlife crises.

Data Analysis
The collected data were analyzed using SPSS statistical software (version 23; IBM Corp, Armonk, New York, NY, USA) by applying the following statistical techniques. Demographic characteristics were analyzed using descriptive statistics, including frequency with percentage and mean with standard deviation. The relationships between self-efficacy, family stress, and midlife crisis were analyzed using Pearson's correlation coefficients.
To determine whether family stress has a mediating effect on the relationship between self-efficacy and midlife crisis, a three-step multiple regression analysis was performed using Baron and Kenny's [42] method of mediation. The first step examined the effect of the independent variable on the mediator. The second step examined the effect of the independent variable on the dependent variable. In the third step, the effect of the mediator and independent variable on the dependent variable were examined by controlling for the independent variable and mediator, respectively. We then determined whether the effect size of the independent variable on the dependent variable in the third step was smaller than that of the independent variable on the dependent variable in the second step. An effect of the independent variable on the dependent variable that was not significant would indicate complete mediation, whereas a significant effect would indicate partial mediation. We utilized simple mediation analysis for SPSS macro PROCESS (Mode 4) with 5000 bootstraps. The goal of the simple mediation model was to investigate the direct and indirect effects between self-efficacy, midlife stress, and family stress. The indirect effect predicts statistically significant results when the 95% bias-corrected confidence intervals (CIs) do not contain zero [43][44][45]. Figure 1 shows the statistical diagrams of this simple mediation model. report questionnaires on self-efficacy, family stress, and midlife crises.

Data Analysis
The collected data were analyzed using SPSS statistical software (version 23; IBM Corp, Armonk, New York, NY, USA) by applying the following statistical techniques. Demographic characteristics were analyzed using descriptive statistics, including frequency with percentage and mean with standard deviation. The relationships between self-efficacy, family stress, and midlife crisis were analyzed using Pearson's correlation coefficients. To determine whether family stress has a mediating effect on the relationship between self-efficacy and midlife crisis, a three-step multiple regression analysis was performed using Baron and Kenny's [42] method of mediation. The first step examined the effect of the independent variable on the mediator. The second step examined the effect of the independent variable on the dependent variable. In the third step, the effect of the mediator and independent variable on the dependent variable were examined by controlling for the independent variable and mediator, respectively. We then determined whether the effect size of the independent variable on the dependent variable in the third step was smaller than that of the independent variable on the dependent variable in the second step. An effect of the independent variable on the dependent variable that was not significant would indicate complete mediation, whereas a significant effect would indicate partial mediation. We utilized simple mediation analysis for SPSS macro PROCESS (Mode 4) with 5000 bootstraps. The goal of the simple mediation model was to investigate the direct and indirect effects between self-efficacy, midlife stress, and family stress. The indirect effect predicts statistically significant results when the 95% bias-corrected confidence intervals (CIs) do not contain zero [43][44][45]. Figure 1 shows the statistical diagrams of this simple mediation model.

Participants' Demographic Characteristics
Demographic characteristics of the participants are provided in Table 1. The mean age of the participants was 48.3 years (SD = 5.8) and ranged from 38 to 61 years. The majority of participants were college graduates or higher (n = 159, 80.3%) and were married (n = 185, 97.4%). One hundred and thirty (65.7%) were office workers or managers, and

Participants' Demographic Characteristics
Demographic characteristics of the participants are provided in Table 1. The mean age of the participants was 48.3 years (SD = 5.8) and ranged from 38 to 61 years. The majority of participants were college graduates or higher (n = 159, 80.3%) and were married (n = 185, 97.4%). One hundred and thirty (65.7%) were office workers or managers, and 131 (66.2%) were satisfied with their occupation. One hundred and ten (55.6%) were affiliated with a religious group. Seventy-eight (39.4%) had been married for 10-19 years, and another 66 (33.3%) had been married for 20-29 years. One hundred and twenty-six (63.6%) had two children.

Mediation Model
To verify the mediating effect of middle-aged men's family stress on the relationship between self-efficacy and midlife crisis, we performed a three-step regression analysis [41]. Prior to analyzing the mediating effect of family stress, we calculated the Durbin-Watson statistic, which was found to be 2054, confirming the absence of autocorrelation in the residuals. The tolerance limit was above 0.1 (0.807), and the variance inflation factor (VIF) was smaller than 10 (1239), confirming the absence of multicollinearity. In addition, residual analysis confirmed the linearity of the model and normality and equal variances of errors.
The following results were obtained in the three-step regression analysis (Table 4). In Step 1, self-efficacy (independent variable) had a statistically significant effect on family stress, which was the mediator (β = −0.326, p < 0.001). In Step 2, self-efficacy (independent variable) had a significant effect on midlife crisis (dependent variable: β = −0.392, p < 0.001). In Step 3, a regression analysis was performed to analyze the effects of family stress (mediator) on midlife crisis (dependent variable). Self-efficacy had a significant impact on midlife crisis (β = −0.205, p < 0.001) and showed a decrease from the β value (−0.392) in Step 2; family stress was confirmed to have partial mediation effect (β = 0.569, p < 0.001). A Sobel test verified the significance of the mediating effect, and the results confirmed that family stress was a significant partial mediator in the relationship between self-efficacy and midlife crisis (Z = −4.270, p < 0.001; Figure 1). We performed the PROCESS macro for SPSS. The total effect of self-efficacy on midlife crisis was significant (

Discussion
In this study, we investigated the mediating effect of family stress on the relationship between self-efficacy and midlife crisis in 190 middle-aged South Korean men. The total mean score for the midlife crisis scale was 2.4 (0.5), which is higher than the midlife crisis score in a sample of middle-aged men in a study by Seo et al. [45]. In this study, the mean scores for a midlife crisis were the second highest for loss of vitality, followed by occupational dissatisfaction. Most South Korean men experience a sense of accomplishment when they perform difficult life tasks, feeling happy and motivated to work harder and achieve goals. However, middle-aged men in South Korean society suffer from the stress of early retirement at work. In European countries, an individual's early retirement increases their level of satisfaction or well-being [46]. However, when middle-aged men in South Korea face an early retirement and they have not prepared a stable postretirement life, it exposes them to financial difficulties, physical illnesses, and various sources of stress [47,48]. It is considered that such situations increase the levels of dissatisfaction that employed middle-aged men experience in the workplace. Our participants would be exposed to the stress of a midlife crisis due to potential future challenges and the instability of their current employment, which leads to a loss of vitality. Social, financial, and family crises in middle adulthood can trigger mental problems, such as depression, as well as physical illnesses, thereby hindering a stable postretirement life [14,48]. Because this study only examined midlife crises, subsequent studies should also investigate mental problems (e.g., depression, anxiety) and physical illnesses in middle-aged men.
Less self-efficacy is associated with a deeper midlife crisis, although the direct effects of less self-efficacy may be partially mediated by family support, which accounted for the influence of self-efficacy. When self-efficacy directly influenced midlife crisis, the effects were negative as greater levels of self-efficacy were associated with a less severe midlife crisis. This finding may be explained by considering that middle-aged men who have mitigated a midlife crisis may be more likely to have self-efficacy and be supported by their families. In our study, midlife crisis decreased as self-efficacy increased, a finding that is consistent with previous studies [17,18]. People with higher self-efficacy are more likely to proactively engage in activities to overcome difficulties or a particular situation they face. People with low self-efficacy are more vulnerable to stressful situations and experience negative relationships with their families [49]. In addition, a nonsupportive family environment increases the risk for mental problems, such as depression [50].
Increased family stress is influenced by changes in family roles as a result of children's growth, family conflicts, and social and economic changes [21,22]. Our results also confirmed that both family stress and midlife crises increased with decreasing self-efficacy. The simple mediation analysis study found family stress to be an individual and environmental factor that affects midlife crisis, which mediated the relationship between self-efficacy and midlife crisis and shows more complicated family situations. Family stress is one of the strongest factors that worsen mental health, and the ability to cope with such stress varies across individuals. Therefore, family stress can cause one family member's experience to influence other members or even disrupt family stability. Moreover, family stress was reported to have an adverse impact on health [51]. Stress can be triggered by a variety of family problems, including relationships, finances, occupational relationships, health, and loss of family members. Family relationships can help an individual to cope with stress and lead to a higher level of well-being [26].
In Step 3 of the regression analysis, the relationship between self-efficacy and midlife crisis remained significant when family stress was evaluated for its effect. Such a partial mediating effect indicates that self-efficacy has both a direct impact on midlife crisis and an indirect impact through experiencing family stress. No studies have identified the mediating effects of family stress in the relationship between self-efficacy and midlife crisis. However, a study by Young and Chae [52] showed that the more stress middle-aged men get from family relationships, the deeper the midlife crisis will be. In several prior studies, stress has been identified as a factor affecting self-efficacy [53][54][55][56]. In Lee and Lee's study [53], factors affecting self-efficacy were identified as family support and family functioning. Additionally, the higher the support of the family and the more positive the family experienced, the higher the self-efficacy. Kim's findings [54] showed that stress has a mediating effect on the relationship between self-efficacy and suicidal ideation. Cho's study [55] showed that stress has a partial mediating effect on the relationship between family resilience and self-efficacy. We think that the results of our research are significant in terms of the aspect of experiencing the midlife crisis and the process of solving it, which may differ depending on the conditions that individuals are facing. There was no prior study to directly compare our findings. However, our findings, along with those previously presented, aimed at increasing self-efficacy levels and reducing the level of midlife crisis. We can interpret these results to mean that men with high self-efficacy are less likely to experience midlife crises, but family stress can have a negative effect on the strength of this relationship. Thus, family stress plays an important role in the influence of self-efficacy on midlife crises. As family stress has a direct adverse impact on adolescent children as well as middle-aged men [52], stress among family members provokes physical problems for the entire family, which, in turn, affects their immune system and increases the risk of depression [57]. Therefore, because family stress can affect all members of the family [27], receiving family support would be an effective method for reducing or alleviating family stress.

Limitations and Recommendations
One limitation of this study is the limited generalizability of the findings because the participants were recruited from only a few cities in South Korea. Subsequent studies should additionally examine the mediating effects of other potential mediators on the relationship between self-efficacy and midlife crisis in middle-aged men. Moreover, replication studies should be conducted to further examine the predictors of midlife crisis in middle-aged South Korean men.

Conclusions
This study aimed to investigate family stress as a mediator in the relationship between self-efficacy and midlife crisis in middle-aged South Korean men. Family stress was found to have a partial mediating effect on the negative relationship between self-efficacy and midlife crisis. The study participants demonstrated moderate levels of self-efficacy, midlife crisis, and family stress. Self-efficacy was negatively associated with both midlife crisis and family stress, while family stress was positively associated with a midlife crisis. This study is significant because it confirms the importance of both family stress and selfefficacy in relation to the experience of a midlife crisis in middle-aged South Korean men. Consequently, the results provide supportive evidence for developing interventions to mitigate family stress and increase family support for middle-aged men. Interventions