دو واسطه قدرت در بهزیستن ذهنی در چین

Abstract Power may enhance well-being via authenticity, but it can also lead to misery, which arises as a consequence of assertiveness. To address inconsistencies in the results of previous studies, we took the literature regarding well-being and personality of agency into account. Consequently, we formed a hypothesis in which the complex effects of power on subjective well-being could be explained via agency and unmitigated agency. Because power increases agency, which contributes to subjective well-being, it could enhance subjective well-being by encouraging people to become more agentic. However, when agency is not mitigated by communion that is referred to as unmitigated agency, it reduces subjective well-being due to dissatisfaction with relationships. Therefore, we hypothesized that power would enhance subjective well-being via greater agency and reduced unmitigated agency. Three surveys completed by 202 Chinese participants showed consistent evidence that power, both dispositional and role-specific, was positively related to subjective well-being and role satisfaction via agency and unmitigated agency. Both agency and unmitigated agency mediated the effects of power on subjective well-being. These results elucidate the complex psychological mechanisms underlying the influence of power on subjective well-being from the perspective of personality and provide a basis for future research.

. Results 5.1. Descriptive statistics and relationships between variables The descriptive statistics and bivariate correlations for all variables are presented in Table 1 and Table 2. As expected, power, whether dispositional or contextual, was positively correlated with agency and SWB (or role satisfaction). However, unmitigated agency was negatively correlated with all other variables. Table 1. Descriptive statistics for variables (n = 202). General survey Work survey Romance survey Mean SD Mean SD Mean SD SWB (or role satisfaction) .03 2.23 18.56 3.54 25.64 2.88 Power 30.24 4.77 30.29 5.26 33.02 3.60 Agency 30.60 4.75 – – – – Unmitigated agency 15.97 4.51 – – – – Note: SWB = subjective well-being. Table options Table 2. Correlations among variables. General survey Work survey Romance survey 1 2 3 1 2 1 2 1. SWB (or role satisfaction) – – – 2. Power .72⁎⁎⁎ – .66⁎⁎⁎ – .65⁎⁎⁎ – 3. Agency .76⁎⁎⁎ .66⁎⁎⁎ .60⁎⁎⁎ .60⁎⁎⁎ .39⁎⁎⁎ .30⁎⁎⁎ 4. Unmitigated agency −.34⁎⁎⁎ −.24⁎⁎⁎ −.14⁎ −.27⁎⁎ −.22⁎⁎ −.33⁎⁎⁎ −.33⁎⁎⁎ Note: SWB = subjective well-being. ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. Table options To further assess how power, agency, and unmitigated agency would contribute to SWB or role satisfaction, we employed a series of hierarchical multiple regression analyses. The results were consistent with our predictions that power emerged as a significant positive predictor of SWB after controlling for age and gender. Analysing the two role surveys confirmed our predictions that the positive effect of power could be generalized to a specific context (Table 3). Table 3. Results of stepwise regressions predicting SWB and role satisfaction. Predictor General survey (DV = SWB) Work survey (DV = word satisfaction) Romance survey (DV = romance satisfaction) Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Gender .01 .06 .11 .13⁎ −.09 −.07 Age .06 .07 .01 .02 .05 .05 Power .72⁎⁎⁎ .33⁎⁎⁎ .67⁎⁎⁎ .44⁎⁎⁎ .66⁎⁎⁎ .56⁎⁎⁎ Agency .52⁎⁎⁎ .32⁎⁎⁎ .20⁎⁎⁎ Unmitigated agency −.19⁎⁎⁎ −.12⁎ −.12⁎ R2 .52 .70 .45 .53 .44 .49 Note: SWB = subjective well-being. For each survey, model 1 tested the direct relationship between gender, age, power, relationship harmony, and SWB or role satisfaction; model 2 tested whether including agency and unmitigated agency in the model 1 reduced the effect of role power on SWB or role satisfaction. The table presents standardized regression coefficients. DV = dependent variable. ⁎ p < .05. ⁎⁎⁎ p < .001. Table options We also found that dispositional power positively predicted agency (β = .66, SE = .05, p < .0001) and negatively predicted unmitigated agency (β = −.23, SE = .07, p = .001) after controlling for gender and age. Furthermore, role power was a positive predictor of agency in work (β = .55, SE = .05, p < .0001) and romantic relationships (β = .41, SE = .09, p < .0001) after controlling for gender and age. However, role power was a negative predictor of unmitigated agency in work (b = −.19, SE = .06, p = .001) and romantic relationships (β = −.41, SE = .08, p < .0001) after controlling for gender and age. 5.2. Both agency and unmitigated agency mediate the effects of dispositional power on SWB We then used an SPSS macro, designed to assess multiple mediation models (Preacher & Hayes, 2008), to examine whether agency and unmitigated agency would both mediate the effects of dispositional power on SWB. As recommended by Preacher and Hayes (2008), assessment of the multiple mediation models involved an analysis of the total and specific indirect effects. The parameter estimates and confidence intervals of the total and specific indirect effects were based on 5000 random samples. As seen in Fig. 1, the total effect of general power on SWB was significant, c = .34, p < .001 with an overall model R2 of .69. However, after adjusting for the indirect effects of the mediators, the direct effect of general power on SWB was reduced significantly, c′ = .16, p < .001. Specifically, agency was a significant mediator, such that general power was positively related to agency (β = .65), which was positively related to SWB (β = .24). Additionally, unmitigated agency was related to both general power (β = −.23) and SWB (β = −.09). Table 4 contains the parameter estimates for the total and specific indirect effects on the association between power and SWB, as mediated by agency and unmitigated agency. A multiple mediation model of the association between power and SWB via agency ... Fig. 1. A multiple mediation model of the association between power and SWB via agency and unmitigated agency (n = 202). Note: SWB = subjective well-being. Path estimates are standardized. ∗p < .01, ∗∗∗p < .001. Figure options Table 4. Indirect effects of general power on SWB. Mediator Parameter estimate SE BCa 95%CI Total .18⁎ .03 .12–.25 Agency .16⁎ .03 .10–.22 Unmitigated agency .02⁎ .01 .01–.05 Note: SWB = subjective well-being. All bootstrapping procedures were based on 5000 random samples with replacement. The bias-corrected and accelerated (BCa) confidence intervals (CIs) include correction for median bias and skew. CIs excluding zero are interpreted as being significant. ⁎ p < .05. Table options 5.3. Both agency and unmitigated agency mediate the effects of role power on role satisfaction Finally, we examined whether agency and unmitigated agency would both mediate the effects of role-specific power on role satisfaction. As Fig. 2 shows, the total effect of work power on work satisfaction was significant, c = .45, p < .001 with an overall model R2 of .51. However, after adjusting for the indirect effects of the mediators, the direct effect of work power on work satisfaction was reduced significantly, c′ = .30, p < .001 ( Table 5). In addition, the total effect of romantic-relationship power on romantic-relationship satisfaction was significant, c = .51, p < .001 with an overall model R2 of .48. However, after adjusting for the indirect effects of the mediators, the direct effect of romantic-relationship power on romantic-relationship satisfaction was reduced significantly, c′ = .44, p < .001 ( Fig. 3 and Table 6). These results suggest a multiple mediation scenario in which the association between role-specific power and role satisfaction could be explained by both agency and unmitigated agency. A multiple mediation model of the association between work power and work ... Fig. 2. A multiple mediation model of the association between work power and work satisfaction via agency and unmitigated agency (n = 202). Note: Path estimates are standardized. ∗p < .01, ∗∗∗p < .001. Figure options Table 5. Indirect effects of work power on work satisfaction. Mediator Parameter estimate SE BCa 95%CI Total .14⁎ .04 .08–.23 Agency .12⁎ .04 .06–.20 Unmitigated agency .02⁎ .01 .01–.05 Note: All bootstrapping procedures were based on 5000 random samples with replacement. The bias-corrected and accelerated (BCa) confidence intervals (CIs) include correction for median bias and skew. CIs excluding zero are interpreted as being significant. ⁎ p < .05. Table options A multiple mediation model of the association between romance power and romance ... Fig. 3. A multiple mediation model of the association between romance power and romance satisfaction via agency and unmitigated agency (n = 202). Note: Path estimates are standardized. ∗p < .01, ∗∗∗p < .001. Figure options Table 6. Indirect effects of romance power on romance satisfaction. Mediator Parameter estimate SE BCa 95%CI Total .08⁎ .02 .04–.13 Agency .05⁎ .02 .02–.09 Unmitigated agency .03⁎ .02 .01–.07 Note: All bootstrapping procedures were based on 5000 random samples with replacement. The bias-corrected and accelerated (BCa) confidence intervals (CIs) include correction for median bias and skew. CIs excluding zero are interpreted as being significant. ⁎ p < .05.