How does power affect happiness and mental illness? The mediating role of proactive coping

Abstract

Previous research suggests that sense of power is associated with happiness and mental illness. However, the mechanisms underlying this relationship is relatively unexplored. This study aimed to examine whether proactive coping mediated the relationship between power and mental illness and between power and happiness. Specifically, we predicted that power—which is associated with goal-oriented tendencies, high-level construals, and positive characteristics—activates the use of proactive coping strategies, which in turn leads to greater happiness and less mental illness. By carrying out a survey (Study 1, N = 150) and an experiment (Study 2, N = 143), we found results that were consistent with our predictions. Overall, this study demonstrates the psychological mechanisms behind the influence of power on happiness and mental illness from the perspective of coping.

Keywords:

• sense of power
• proactive coping
• happiness
• mental illness

PUBLIC INTEREST STATEMENT

Power considerations are ubiquitous. According to power studies, higher sense of power is associated with greater happiness and less mental illness. However, it remains relatively as to how power affects the variables. In this study, the authors posit that the use of proactive coping strategy, defined as individuals’ efforts to cope with future stressors in advance (Aspinwall & Taylor, 1997) would be the psychological mechanism behind the link between power and happiness and between power and mental illness. Across the two studies, the authors found that individuals with higher sense of power tend to use more proactive coping strategy, which in turn, experience greater happiness and less mental illness. Overall, this study is meaningful in that it contributes to our understanding the relationship among power, coping, happiness, and mental illness, all of which are important to us.

1. Introduction

Given that power is central to social life (Keltner et al., 2003), it has a profound impact on individuals (Guinote, 2017). Particularly, previous studies suggest that possessing power is related to more perceived happiness. For example, power enhances subjective and affective well-being and life satisfaction (Adler et al., 2000; Berdahl & Martorana, 2006; Gonzaga et al., 2008; Kifer et al., 2013; Ku & Kim, 2020; Langner & Keltner, 2008; Park & Suh, 2018; Wang, 2015a, 2015b). Although less studies have focused on the relationship between power and mental illness, some suggest that power is negatively associated with mental illness (e.g., anxiety and depressive symptoms; Belmi & Pfeffer, 2016; Byrne et al., 2004; Hatcher et al., 2012; Malecki et al., 2015; Maner et al., 2012; Wetherall et al., 2019).

There have been some attempts to examine the mechanisms underlying the association between social power and happiness. For example, Wang (2015b) suggested that agency mediates the relationship between social power and happiness. Particularly, authenticity, which is the correspondence between an individual’s internal state and behavior (Deci & Ryan, 1985), has been examined as a possible mediator in the relationship between power and happiness (Datu & Reyes, 2015; Kifer et al., 2013; Wang, 2015a). However, findings on this mediation were inconsistent; Kifer et al. (2013) and Wang (2015a) found that power increased happiness via authenticity, whereas power decreased happiness via authenticity in Datu and Reyes (2015)’s study. Furthermore, a paucity of literature has examined the mechanism underlying the association between power and mental illness. Therefore, further research is needed to examine how power is linked to happiness and mental illness.

In this article, we examined these links by suggesting a new mediator, proactive coping; this construct is defined as individuals’ efforts to cope with future stressors in advance (Aspinwall & Taylor, 1997). According to the literature on power and proactive coping, we predicted that power—which is associated with goal-oriented tendencies, high-level construals, and positive characteristics—activates the use of proactive coping strategies, which in turn, affects happiness and mental illness. Specifically, we posit that power, whether dispositional or manipulated, predicts enhanced happiness and reduced mental illness via a greater use of proactive coping strategies.

1.1. Definition of power

Power can be defined at various levels, such as the macro (e.g., political and military power), the middle (e.g., gender and social class), the group (e.g., leadership roles and positions of power), and the micro level (e.g., relationships with friends and family; Guinote, 2017). Additionally, it is notable that power has been defined in several forms, such as social/relational/influence power (i.e., the ability to influence other people) or personal/agentic/autonomy power (i.e., the ability to act for oneself; Ku & Kim, 2020; Lammers et al., 2016, 2009; Leach et al., 2017).

In this study, we focused on social power at the micro level and regarding social/relational/influence power. Based on Anderson et al. (2012), social power was defined as one’s perceived and generalized ability to influence another person in a relationship. Moreover, although individuals’ sense of power is relationship-specific, it is also moderately consistent across relationships (e.g., parent, romantic partner, and supervisor; Anderson et al., 2012).

1.2. Definition of proactive coping

In our daily lives, we face a variety of stressful events and experience negative emotional states. However, experiencing stressors do not always produce negative results because the impact of stress varies by individuals’ stress coping ability (Aldwin & Revenson, 1987; Tachè & Selye, 1985). Coping has been defined as one’s cognitive and behavioral efforts to control, resist, and mitigate the internal and external demands of stressful events (Folkman & Lazarus, 1980). Because coping is a goal-oriented process, when coping with stress, individuals think and act in a way aimed at solving the fundamental causes of stress or at managing the emotional responses caused by it (Lazarus, 1993). Thus, coping serves as a potential influencing factor of various psychological adaptations and mental illnesses (Lazarus & Folkman, 1984). Accordingly, coping is one of the important concepts when examining the mental health of individuals faced with stressors (Wu et al., 2008).

Coping studies have traditionally assumed that the coping process begins with evaluations and responses to the potential loss of or threat to individual’s important goals (Greenglass & Fiksenbaum, 2009). Therefore, traditional coping studies have primarily evaluated coping as a reactive concept that manages stressful events that have already occurred in the past or present (Aspinwall & Taylor, 1997); based on this view, this construct was generally classified as problem-focused and emotion-focused coping. Problem-focused coping is the process of perceiving that the current stressor can be controlled and trying to solve the stressor, while emotion-focused coping is the process of mitigating the emotional pain caused by the stressor owing to perceptions that it cannot be controlled (Forsythe & Compas, 1987; Lazarus & Folkman, 1984, 1991).

However, humans are not constant; sometimes they react passively to changes in their internal and external conditions, and sometimes they react proactively and aimed at creating their own environment (Schwarzer & Taubert, 2002). Thus, an integrated view of coping requires a definition that encompasses not only reactive coping (i.e., response to already occurring stressors) but also proactive coping (i.e., predicting, discovering, and preparing for future stressors; Aspinwall & Taylor, 1997). Based on this awareness, some notable studies have proposed the concept of proactive coping, which refers to a future stressful event that has not yet occurred but that can be predicted, a definition that differs from the traditional coping concept that deals only with stressful events that have been addressed in the past or present (e.g., Aspinwall & Taylor, 1997; Greenglass et al., 1999). Thus, proactive coping is the process of minimizing the negative impact of future stressors, which means the following: challenging them without looking at them as threats; recognizing that there is a coping strategy available for future stressful events; and trying to successfully prevent and resolve stress in an attempt to use them as an opportunity for self-growth (Drach-Zahavy & Erez, 2002; Folkman & Lazarus, 1980; Schwarzer & Taubert, 2002).

1.3. Possible effect of power on proactive coping

To the best of our knowledge, power has not been associated with proactive coping in previous research. The reasons might be attributed to that power refers to a perception of one’s capacity (Galinsky et al., 2003) whereas coping means a cognitive and behavioral effort to address the stressors (Lazarus, 1993; Lazarus & Folkman, 1984; Miller & Green, 1985). However, some evidence suggest that sense of power may lead to more use of proactive coping strategies. First, power and proactive coping share the essential characteristics: perceived control. Higher levels of control in individuals’ relationships or locus of control is associated with higher sense of power (Anderson et al., 2012). In the same manner, proactive coping is accompanied by high levels of perceived control (Bagana et al., 2011; Greenglass & Fiksenbaum, 2009).

Second, previous studies have suggested that high power activates approach-related tendencies, whereas low power activates inhibition-related tendencies (Anderson & Berdahl, 2002; Keltner et al., 2003; Lammers et al., 2009; Smith & Bargh, 2008). Specifically, high power increases attention to rewards and opportunities; this, in turn, motivates individuals with high sense of power to pursue and obtain their current goals and desires using flexible strategies. Meanwhile, low power increases the perception of threat and punishment from others; this makes individuals with low sense of power inhibit their own behaviors.

Individuals using proactive coping strategies perceive stressing demands as challenges and goals, while individuals using reactive coping strategies regard these demands as harm, loss, and threats (Greenglass & Fiksenbaum, 2009; Schwarzer & Taubert, 2002). Furthermore, proactive coping refers to a goal management process that includes self-determined and autonomous goal setting, taking actions toward a goal (Bagana et al., 2011), and making practical and active efforts by assessing the detailed steps that need to be taken to achieve a goal (Schwartz et al., 1999; Sohl & Moyer, 2009). Given the goal-oriented nature of both power and proactive coping, we predicted that power would increase people’s tendency to use proactive coping strategies.

Third, based on the construal level theory, the focus of people with high sense of power is based on high-level construals that capture abstract and central features of social events, whereas people with less sense of power have a focus based on low-level construals that capture concrete and specific features (Trope & Liberman, 2003, 2010). For example, Smith and Trope (2006) found that power increased people’s ability to have a more holistic view, to plan ahead of time, and to monitor higher goals. Similarly, proactive coping is future-oriented and seeks to manage goals, whereas traditional coping focuses on stressful events that have harmed or are currently harming the person (Greenglass & Fiksenbaum, 2009; Schwarzer, 2001). Therefore, proactive coping is characterized by a multidimensional and forward-looking strategy (Greenglass et al., 1999; Sohl & Moyer, 2009). Furthermore, Miao and Gan (2019) suggested that proactive coping was associated with a meaning in life that constitutes a high-level construal compared with pleasure (Kim et al., 2014). Given the nature of proactive coping, it is possible to denote that it may be associated with high-level construals, rather than low-level construals. Thus, we predicted that people with high sense of power would be more likely to cope proactively when faced with stressors.

Fourth, power is associated with other positive characteristics, such as self-efficacy (Keltner et al., 2003) and optimism (Anderson & Galinsky, 2006), and both are also related to proactive coping. People with high optimism actively seek resources to cope with an expected stressor; particularly, optimistic people focus on proactive coping strategies that aim to minimize/prevent the negative effects of a stressful event before the occurrence of a burnout (Chang & Chan, 2015). Therefore, optimism can be seen as a positive influencing factor of proactive coping (Griva & Anagnostopoulos, 2010; Lopes & Cunha, 2008). Additionally, one study suggested that self-efficacy also increased proactive coping, which in turn increased life satisfaction (Stanojević et al., 2014). The connections between these positive characteristics and proactive coping lend support to our prediction that individuals with a high sense of power would be more likely to engage in proactive coping strategies, given that these positive characteristics are also related to power.

Fifth, D’cruz and Noronha (2018) suggested that people with a low sense of power (i.e., outcasts) had a relatively short resource availability to deal with anticipated stressful events. Therefore, they ultimately lacked a variety of proactive coping strategies, a finding that is consistent with our prediction. Overall, our reasoning and the scientific evidence lead to our prediction that power would be associated with an increased tendency to use proactive coping strategies.

1.4. Effect of proactive coping on happiness and mental illness

According to the Conservation of Resources Theory proposed by Hobfoll (1989), stress occurs when coping resources (e.g., health, property, knowledge and skills, social reputation) are already depleted, soon depleted, or consumed without further acquisition. Meanwhile, proactive people seek to enhance their abilities and accumulate resources to cope with future stressors (Uskul & Greenglass, 2005). Therefore, they are less affected by a the stressful event (Greenglass et al., 1999; Schwarzer & Taubert, 2002), and consume fewer resources to cope with it (Ouwehand et al., 2007). Furthermore, proactive people constantly strive to offset, eliminate, and reduce the effects of stress, and to improve quality of life (Greenglass et al., 1999). Accordingly, these efforts contribute to the reduction of depression (Greenglass et al., 1999; Pasikowski et al., 2002; Uskul & Greenglass, 2005) and anxiety (Griva & Anagnostopoulos, 2010), which are equivalent to stress emotions (Lazarus, 1999).

Proactive coping also fills in the function of accumulating coping resources that can enhance the value of life and explore its purposes, which can have a positive effect on various aspects of life (Schwarzer & Taubert, 2002). Additionally, the essential characteristics of proactive coping are high levels of perceived control and autonomy (Bagana et al., 2011; Greenglass & Fiksenbaum, 2009). Owing to these characteristics, proactive coping was found to be related to many positive outcomes, such as positive affect (Greenglass & Fiksenbaum, 2009; Stiglbauer & Batinic, 2015), engagement (Gan et al., 2007), social well-being (Zambianchi & Ricci Bitti, 2014), and life satisfaction (Prenda & Lachman, 2001; Uskul & Greenglass, 2005). For example, Sohl and Moyer (2009) found that proactive coping predicted happiness (i.e., affective well-being and life satisfaction).

1.5. Overview of the present study

This study aimed to examine whether proactive coping would mediate the relationship between social power and happiness or between social power and mental illness across two studies. In Study 1, we measured dispositional sense of power, tendency to use proactive coping strategy, trait happiness, and trait mental illness. In Study 2, we manipulated power and subsequently measured state proactive coping, state happiness, and state mental illness. In both, we predicted that sense of power, regardless of being dispositional or manipulated, would have an effect on proactive coping; the latter, in turn, would have an effect on happiness and mental illness. Figure 1 shows the conceptual framework of the present study.

Figure 1. Research model of the present study, controlling for age and gender

Note. The product of path coefficient A and B (A$\times$B) represents the indirect effect. The path coefficient C represents the total effect (The effect of predictor on outcomes). The path coefficient C′ represents the direct effect (The effect of predictor on outcomes, controlling for a mediator).

2. Study 1

Hypothesis 1. Dispositional sense of power predicts a) more happiness (i.e., life satisfaction and affective well-being), and b) less mental illness (i.e., trait depression and anxiety).

Hypothesis 2. Dispositional sense of power predicts a greater tendency to use proactive coping strategies.

Hypothesis 3. The use of proactive coping strategies mediates the relationship between sense of power and happiness, such that dispositional sense of power positively predicts proactive coping, which in turn positively predicts a) life satisfaction, and b) affective well-being.

Hypothesis 4. The use of proactive coping strategies mediates the relationship between sense of power and mental illness, such that dispositional sense of power positively predicts proactive coping, which in turn negatively predicts a) trait depression, and b) trait anxiety.

2.1. Methods

2.1.1. Design

Study 1 had a cross-sectional online survey design. Participants were recruited via advertisement on an online bulletin board in a Korean university, accessible only by undergraduate and graduate students and alumni of the Korean university. They were asked to report their dispositional sense of power, proactive coping, life satisfaction, affective well-being, trait depression, trait anxiety, and demographic variables (i.e., age and gender), and all assessments were obtained through an internet survey website, Qualtrics.

2.1.2. Participants

In total, 150 Korean adults (69 male, 81 female; mean age = 25.97 years, SD = 6.16) participated, and they received a mobile convenience coupon worth 2,000 Korean won (equivalent to United States 2 USD). All the procedures in this study complied with the principles of the Declaration of Helsinki, and all the participants signed an online voluntary-informed consent prior to study onset.

2.1.3. Measures

2.1.3.1. Dispositional sense of power

It was measured using the 8-item Sense of Power Scale (Anderson et al., 2012). Participants were asked to rate their sense of power in their general relationship with others. Sample items were “If I want to, I get to make the decisions,” and “My ideas and opinions are often ignored” (reverse-scored). All items were answered on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). The scale was translated according to general scale translation guidelines (e.g., Shim & Yoo, 2012) along with two researchers, one bilingual who is familiar with both American and Korean cultures, and one native speaker. The present study yielded a reliability coefficient of.85.

2.1.3.2. Proactive coping

It was measured with the 14-item Proactive Coping Inventory (PCI, Greenglass et al., 1999), and we adopted its Korean version (K. Kim, 2001); sample items were “I like challenges and beating the odds,” and “I try to let things work out on their own” (reverse-scored). Participants rated the extent to which each point represented their normal traits and behaviors. All items were rated on a scale ranging from 1 (not at all true) to 4 (completely true). The present study showed a reliability coefficient of .79.

2.1.3.3. Life satisfaction

It was measured using the 5-item Satisfaction With Life Scale (SWLS; Diener et al., 1985); we adopted its Korean version (Kim, 2007). Participants were asked to estimate the extent to which they agreed with each item on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). The present study showed a reliability coefficient of .84.

2.1.3.4. Affective well-being (Happiness)

It was assessed via the Positive and Negative Affect Schedule (PANAS; Watson et al., 1988). We adopted its Korean version (Lim et al., 2010). It comprises the following subscales: Positive affect (PA; e.g., “interested”; 10 items) and negative affect (NA; e.g., “upset”; 10 items). Participants were asked to rate the extent to which they generally experienced these emotions on a scale ranging from 1 (strongly disagree) to 5 (strongly agree). Affective well-being was calculated by subtracting the NA scores from PA scores (e.g., Park & Suh, 2018). The present study showed a reliability coefficient for the PA and NA subscales of .87 and .91, respectively.

2.1.3.5. Trait depression

It was assessed by the 10-item State-Trait Depression Scale (STDI; Spielberger, 1995). We adopted its Korean version (Lee et al., 1999). The present study showed a reliability coefficient of .88.

2.1.3.6. Trait anxiety

It was measured using the 20-item State-Trait Anxiety Inventory (STAI; Spielberger, 1983). We adopted its Korean version (Hahn et al., 1996). The present study showed a reliability coefficient of .91.

2.2. Results

The descriptive statistics and correlations are presented in Table 1. The results indicated that all variables were significantly correlated with each other, all ps < .001. Dispositional sense of power and proactive coping were positively correlated with happiness (i.e., life satisfaction and affective well-being), whereas they were negatively correlated with mental illness (i.e., trait depression and anxiety).

Table 1. Descriptive statistics and correlation matrix of variables in Study 1 (N = 150)

	M (SD)	1	2	3	4	5
1. Dispositional power	4.73 (0.85)	-
2. Proactive coping	2.72 (0.42)	.69***	-
3. Life satisfaction	4.36 (1.33)	.58***	.65***	-
4. Affective well-being	0.43 (1.10)	.54***	.55***	.60***	-
5. Trait depression	2.24 (0.58)	−.58***	−.59***	−.73***	−.79***	-
6. Trait anxiety	2.36 (0.42)	−.57***	−.56***	−.68***	−.80***	.88***

Note. Affective well-being was calculated by subtracting the NA scores from the PA scores.

*** p <.001.

To test our hypotheses, we used a SPSS macro designed to simultaneously examine the total (path C in Figure 1), direct (path C′ in Figure 1), and indirect effect (path A$\times$B in Figure 1) with bootstrap procedures (Preacher & Hayes, 2008). The parameter estimates and confidence intervals of the total, direct, and indirect effects were estimated based on 10,000 bootstrap resamples (Table 2).

Table 2. Indirect effect of power on happiness and mental illness via proactive coping controlling for age and gender in Study 1 (N = 150)

	Unstandardized estimates		Bootstrapping procedure
Mediation route	Total effect	Direct effect	Indirect effect	BCa 95% CI
dispositional power → proactive coping → life satisfaction	.91***	.37***	.53	0.30 ∼ 0.78
dispositional power → proactive coping → affective well-being	.69***	.39***	.30	0.15 ∼ 0.49
dispositional power → proactive coping → trait depression	−.39***	−.22***	−.17	−0.28 ∼ −0.08
dispositional power → proactive coping → trait anxiety	−.35***	−.21***	−.13	−0.32 ∼ −0.10

Note. Confidence intervals for the indirect effects were estimated based on 10,000 bootstrap resamples. CIs excluding zero are interpreted as significant.

** p <.05. *** p <.001.

First, we examined the relationship between dispositional sense of power, proactive coping, and life satisfaction. Hypothesis 1a was supported, such that dispositional sense of power was a significant positive predictor of life satisfaction (total effect), while controlling for age and gender, b = .90, SE = .11, p < .001, BCa CI = [0.70, 1.12]. Dispositional sense of power also predicted life satisfaction, while controlling for proactive coping, age, and gender (direct effect), b = .37, SE = .13, p = .006, BCa CI = [0.11, 0.64]. Moreover, hypotheses 2 and 3a were supported, such that proactive coping significantly mediated the relationship between dispositional sense of power and life satisfaction (indirect effect), b = .53, SE = .12, BCa CI = [0.30, 0.78]. Namely, dispositional sense of power was a significant predictor of proactive coping (supporting Hypothesis 2), b = .34, SE = .03, p < .001, BCa CI = [0.28, 0.40], and, in turn, proactive coping was a significant predictor of life satisfaction (supporting Hypothesis 3a), b = 1.57, SE = .27, p < .001, BCa CI = [1.03, 2.10].

Afterwards, we examined whether proactive coping mediated the relationship between dispositional sense of power and affective well-being. As Table 2 shows, the total, direct, and indirect effect of dispositional sense of power on affective well-being were all significant, which supported Hypothesis 1a, Hypothesis 2, and Hypothesis 3b. Specifically, dispositional sense of power was a significant positive predictor of proactive coping, and, in turn, proactive coping was a significant positive predictor of affective well-being, b = .88, SE = .24, p < .001, BCa CI = [0.41, 1.35].

Last, we investigated whether dispositional sense of power predicted mental illness via proactive coping. Consistent with our predictions, the total, direct, and indirect effect of dispositional sense of power on trait depression and anxiety were significant, which supported Hypothesis 1b, Hypothesis 2, and Hypothesis 4a-b. Namely, dispositional sense of power was a significant positive predictor of proactive coping, and, in turn, proactive coping was a significant negative predictor of trait depression, b = −.49, SE = .12, p < .001, BCa CI = [−0.74, −0.25], and trait anxiety, b = .40, SE = .11, p < .001, BCa CI = [−0.62, −0.17].

2.3. Discussion

Study 1 supported our predictions that proactive coping is one of the variables underlying the relationship between power and happiness and between power and mental illness. We found that power was associated with more happiness and less mental illness (Hypothesis 1a-b), a result consistent with the literature (e.g., Kifer et al., 2013; Wetherall et al., 2019). Moreover, proactive coping was positively associated with power (Hypothesis 2) and mediated the relationship between power and happiness and between power and mental illness (Hypothesis 3a-b and 4a-b).

Although the findings are compelling, Study 1 was not an experimental study, but a correlational one. Thus, these findings do not demonstrate the possible causal chain: that power enhances the use of proactive coping strategies, which in turn enhances happiness or decreases mental illness. To address this limitation, we conducted Study 2, which was an experimental study.

3. Study 2

Study 2 aimed to expand on the results of Study 1 by demonstrating the causal effect of power on happiness and mental illness via proactive coping. Accordingly, we manipulated power and examined how it predicted state happiness and mental illness via state proactive coping. We predicted that manipulated power would lead individuals to use more proactive coping strategies, which in turn would increase state happiness and decrease state mental illness. The specific hypotheses are described herein.

Hypothesis 1. Manipulated sense of power enhances a) state happiness (i.e., life satisfaction and affective well-being), and b) decreases state mental illness (i.e., state depression and anxiety).

Hypothesis 2. Manipulated sense of power increases state proactive coping.

Hypothesis 3. State proactive coping mediates the relationship between manipulated sense of power and state happiness, such that manipulated sense of power enhances the state proactive coping, which in turn increases a) state life satisfaction, and b) affective well-being.

Hypothesis 4. State proactive coping mediates the relationship between manipulated sense of power and state mental illness, such that manipulated sense of power enhances the state proactive coping, which in turn decreases a) state depression, and b) state anxiety.

3.1. Methods

3.1.1. Design

Study 2 had an experimental design. As in Study 1, we obtained informed consents and all measures through an internet survey website, Qualtrics. First, participants were randomly assigned to one of the two conditions (high-power vs. low-power; independent variable); then, we manipulated participants’ sense of power with an essay task, and subsequently measured their state proactive coping (mediator variable), life satisfaction, affective well-being, depression, and anxiety (dependent variables). Finally, participants were asked to report their demographic variables (i.e., age and gender).

3.1.2. Participants

Participants were recruited via the procedure described in Study 1. Consequently, 143 Korean adults (65 males, 78 females; mean age = 28.89 years, SD = 11.91) participated, and they received a mobile convenience store coupon worth 2,000 Korean won in return for their participation.

3.1.3. Measures

3.1.3.1. Manipulated power

Based on previous studies (e.g., Kifer et al., 2013; Lammers et al., 2016), we used an essay task to assess manipulated power. First, participants were assigned to one of the two conditions (high-power vs. low-power); in the high-power condition, participants (n = 72) were asked to recall a time in which they had power over another individual or individuals. In the low-power condition, participants (n = 71) were asked to recall a time when someone else had power over them or when they lacked power. Then, participants in both conditions were instructed to describe the situations they recalled in detail. For manipulation checks, all participants were asked to estimate the extent to which they felt power in the situations they recalled using the Sense of Power scale (Anderson et al., 2012). The present study showed a reliability coefficient of .87.

3.1.3.2. State proactive coping

It was assessed via the PCI (Greenglass et al., 1999), like in Study 1. However, to ensure that we captured state proactive coping, we asked participants to rate how much each item represented their selves “right now,” namely, when they were responding to the items. The present study showed a reliability coefficient of .82.

3.1.3.3. State life satisfaction

It was measured via the Korean version of SWLS (Diener et al., 1985; Kim, 2007), like in Study 1. Once more, we asked participants to rate the extent to which the items represented their selves “right now.” The present study showed a reliability coefficient of .84.

3.1.3.4. State affective well-being (Happiness)

It was measured via the Korean version of the PANAS (Lim et al., 2010; Watson et al., 1988), like in Study 1. Again, we asked participants to rate the extent to which they experienced the emotions “right now.” The present study showed a reliability coefficient for the PA and NA subscales of .85 and .91, respectively.

3.1.3.5. State depression

It was assessed by the Korean version of the STDI (Lee et al., 1999; Spielberger, 1995), like in Study 1. The present study showed a reliability coefficient of .86.

3.1.3.6. State anxiety

It was assessed using the Korean version of the STAI (Hahn et al., 1996; Spielberger, 1983). The present study showed a reliability coefficient of .93.

3.2. Results

The manipulation check revealed that the priming was successful. Participants in the high-power condition (M = 5.12, SD = .79) reported a higher sense of power than those in the low-power condition (M = 3.67, SD = 1.03), t(141) = 9.47, p < .001, d = 1.57. Furthermore, high-power participants (M = 2.84, SD = .42) reported higher levels of state proactive coping than low-power participants (M = 2.69, SD = .36), t(141) = 2.26, p = .025, d = .37. However, all dependent variables (i.e., state life satisfaction, affective well-being, anxiety, and depression) showed no between-group differences, all ps > .20 (see Table 3), which did not support Hypothesis 1a-b.

Table 3. Descriptive statistics and t-test results between groups in Study 2 (N = 143)

	High-power (n = 72)	Low-power (n = 71)	t	d
	M (SD)	M (SD)
Manipulated power	5.12 (0.79)	3.67 (1.03)	9.47***	1.57
State proactive coping	2.84 (0.42)	2.69 (0.36)	2.26***	0.37
State life satisfaction	4.33 (1.16)	4.34 (1.25)	−0.04***	−0.01
State affective well-being	0.64 (0.94)	0.43 (1.08)	1.26***	0.21
State depression	2.11 (0.46)	2.16 (0.60)	−0.63***	−0.10
State anxiety	2.15 (0.52)	2.22 (0.58)	−0.67***	−0.11

Note. Affective well-being was calculated by subtracting the NA scores from the PA scores.

* p <.05. *** p <.001.

Then, we used a SPSS macro (Preacher & Hayes, 2008) as in Study 1 to test our hypotheses in Study 2. In the model, condition (1 = high-power condition, 0 = low-power condition) was the predictor variable, proactive coping was the mediator, and age and gender were covariates. To estimate the confidence intervals, we used 10,000 bootstrap resamples.

The results indicated that contrary to Hypothesis 1a-b, all the total effects of manipulated power on the dependent variables were non-significant. Furthermore, all the direct effects were also non-significant (see Table 4). However, supporting Hypothesis 2, 3a-b, and 4a-b, all the indirect effects of manipulated power on the dependent variables via proactive coping were significant (for a detail for mediation without evidence of a total effect, see Hayes & Rockwood, 2017). Regarding happiness, manipulated power increased state proactive coping (supporting Hypothesis 2), b = .14, SE = .07, p = .035, BCa CI = [0.01, 0.27], which in turn led to increased state life satisfaction (supporting Hypothesis 3a), b = 1.69, SE = .22, p < .001, BCa CI = [1.26, 2.13], and state affective well-being (supporting Hypothesis 3b), b = 1.47, SE = .18, p < .001, BCa CI = [1.12, 1.83]. Regarding mental illness, manipulated power increased state proactive coping (supporting Hypothesis 2), which in turn decreased state depression (supporting Hypothesis 4a), b = −.65, SE = .10, p < .001, BCa CI = [−0.85, −0.45], and state anxiety (supporting Hypothesis 4b), b = −.56, SE = .11, p < .001, BCa CI = [−0.77, −0.34].

Table 4. Indirect effect of power on happiness and mental illness via proactive coping controlling for age and gender in Study 2 (N = 143)

	Unstandardized estimates		Bootstrapping procedure
Mediation route	Total effect	Direct effect	Indirect effect	BCa 95% CI
manipulated power → proactive coping → life satisfaction	−.03	−.26	.23	0.03 ∼ 0.49
manipulated power → proactive coping → affective well-being	.18	−.02	.20	0.03 ∼ 0.43
manipulated power → proactive coping → state depression	−.04	.05	−.09	−0.19 ∼ −0.02
manipulated power → proactive coping → state anxiety	−.05	.03	−.08	−0.18 ∼ −0.01

Note. Confidence intervals for the indirect effects were estimated based on 10,000 bootstrap resamples. CIs excluding zero are interpreted as significant.

3.3. Discussion

Overall, Study 2 indicated that proactive coping was one of the mechanisms underlying the link between manipulated power and mental illness and between manipulated power and happiness. Specifically, manipulated power increased life satisfaction and affective well-being via proactive coping, and decreased depression and anxiety via proactive coping (indirect effect, Hypothesis 2 and 3).

However, manipulated power did not evoke significant between-group differences in state happiness and mental illness (total effect, Hypothesis 1). Furthermore, the direct effect of manipulated power on happiness and mental illness was non-significant. The possible reasons are discussed in the general discussion section.

4. General discussion

This study aimed to investigate whether proactive coping mediates the relationship between power and happiness and between power and mental illness. Our findings were consistent with our predictions in both studies; in Study 1, dispositional sense of power predicted trait happiness and mental illness via proactive coping. In Study 2, manipulated sense of power increased state happiness while decreasing state mental illness via state proactive coping in the same pattern as in Study 1. Overall, our experimental and correlational findings strongly and consistently denote that more power is associated with greater happiness and less mental illness, and that this effect owes largely to proactive coping.

As stated in the introduction, past studies have focused on authenticity as a mediator in the relationship between power and happiness (Datu & Reyes, 2015; Kifer et al., 2013; Wang, 2015a). However, results for this variable across studies were not consistent, suggesting that other possible mediators needed to be examined. Furthermore, there is a paucity of research focused on the underlying mechanisms of the relationship between power and mental illness. By suggesting that proactive coping is an additional mediator between not only power and happiness but also between power and mental illness, the current study contributes to the theoretical understanding of how power affects happiness and mental illness.

Based on prior research (e.g., Cheng & Furnham, 2003; Wood et al., 2010), our findings that power decreased depression and anxiety might not be surprising, because mental illness is often associated with less happiness. However, Keyes (2005) has suggested that mental health (e.g., emotional and psychological well-being) and mental illness (e.g., major depressive episode and generalized anxiety) are not opposite ends of a single continuum; namely, the presence of mental health does not equal to the absence of mental illness, and vice versa. For example, Gillham et al. (2011) found that perceived social support predicted low depression, but was not associated with life satisfaction. Therefore, our findings expand the literature on how power affects individuals’ cognitive and affective states by showing that it can not only increases happiness but also decreases mental illness.

This study was the first, to the best of our knowledge, to demonstrate the relationship between power and proactive coping. However, some studies have examined what contributes to engagement in proactive coping; for example, Ouwehand et al. (2008) found that future temporal orientation was the most powerful predictor of proactive coping than other variables (e.g., goal orientation). Moreover, Straud et al. (2015) found that conscientiousness and openness to new experiences were predictors of proactive coping. However, these were personal variables, and there has been little research on the interpersonal predictors of engagement in proactive coping. Accordingly, this study broadens our understanding on the predictors of proactive coping by showing that power predicts people’s tendency to use proactive coping strategies.

Additionally, in Study 1, the total effect of power on happiness was significant; this result was in line with many previous research (e.g., Adler et al., 2000; Kifer et al., 2013; Ku & Kim, 2020; Wang, 2015a, 2015b). Meanwhile, Datu and Reyes (2015) found the opposite result; that dispositional sense of power negatively predicted trait happiness. They asserted that the negative association between power and happiness could be attributed to the culture of interdependence in their sample, which comprised Filipino college students; the Philippines have a collectivistic culture. However, we are not able to conclude on the relationship between power and happiness based on cultural difference because participants in this study and in Wang (2015a, 2015b)’s studies also came from collectivist societies; while three studies reported a positive relationship between power and happiness, one reported a negative relationship. Therefore, future studies are warranted to further examine the effect of culture on the relationship between power and happiness.

In Study 2, manipulating power, alone, did not predict state happiness and mental illness (i.e., total and direct effects). Rather, it indirectly increased state happiness and decreased mental illness via proactive coping (i.e., indirect effect). These findings are not consistent with previous studies (e.g., Kifer et al., 2013; Park & Suh, 2018). We speculated that the nature of power manipulations could have evoked the inconsistent results. There are several prior studies that found a non-significant relationship between manipulating power and affective happiness (e.g., Galinsky et al., 2003; Rucker & Galinsky, 2008; Weick & Guinote, 2008). For example, manipulating power was found to have no impact on negative affect (e.g., sad, gloomy, and negative) and positive affect (e.g., happy, upbeat, and pleasant) in Rucker and Galinsky (2008). Based on these findings, Galinsky et al. (2015) asserted that the relationship between manipulating power and subsequent affective states is inconclusive. Therefore, the finding that the total and direct effects of manipulated power on happiness and mental illness was non-significant may have derived from specific power manipulations. Future studies are warranted to further examine the effect of manipulating power on state happiness and mental illness with more relevance and various methods.

Although the results are compelling, there are also some limitations. One such limitation comes from the fact that power was categorized into the following distinct constructs: social power (or influence and relational power) and personal power (or autonomy and agentic power) (Ku & Kim, 2020; Lammers et al., 2016, 2009; Leach et al., 2017; Park & Suh, 2018). The former represents power over other people, whereas the latter stands for freedom from other people, both of which result in different happiness outcomes (Ku & Kim, 2020; Park & Suh, 2018). For example, in a Korean sample, social power enhanced life satisfaction, whereas personal power decreased affective well-being, and, in an American sample, social power was not associated with happiness, while personal power increased life satisfaction and affective well-being (Park & Suh, 2018). In the present study, we focused on social power, and even used the Sense of Power Scale, which measures social power (Anderson et al., 2012). Therefore, our results are limited to the relationship between social power, proactive coping, happiness, and mental illness. Future studies should extend the examination by analyzing these relationships with personal power.

Previous studies have also suggested that power predicts happiness via authenticity (Kifer et al., 2013; Wang, 2015a) and agency (Wang, 2015b). Accordingly, we cannot rule out that authenticity and agency have affected our results because we did not measure nor control for these variables. It would be interesting to compare the role of mediators (e.g., agency, authenticity, and proactive coping) in the relationships between power and happiness and between power and mental illness in future studies.

5. Conclusion

Overall, this research suggests that having more sense of power brings more happiness and less mental illness by helping the agent to more proactively cope with stressful events, thus making the stressor become a challenge/goal rather than a threat. This study is meaningful in that it contributes to broadening research on power, coping, happiness, and mental illness.

Additional information

Funding

This research was funded by the Hwarangdae Research Institute at the Korea Military Academy.

Notes on contributors

Seungju Hyun

Seungju Hyun is a Lecturer in the Department of Psychology at the Korea Military Academy (KMA). He holds MA in clinical and counseling psychology from Seoul National University. His current research interest includes proactive coping, coping flexibility, attention, and hypervigilance.

Xyle Ku

Xyle Ku is an Assistant Professor of Psychology at the KMA. His research areas include power, happiness, facial appearance, leadership, and mortality salience.