ISSN 0439-755X
CN 11-1911/B

Acta Psychologica Sinica ›› 2024, Vol. 56 ›› Issue (5): 612-629.doi: 10.3724/SP.J.1041.2024.00612

• Reports of Empirical Studies • Previous Articles     Next Articles

Trapped by family or compensated from work? The influence of daily negative family events on daily effective leadership behaviors

LIU Depeng1, LI Juexing2(), ZHANG Shengjun3, PANG Xuhong3, WANG Zheng4   

  1. 1School of Management, Shandong University, Jinan 250100, China
    2School of Economics and Management, Yunnan Normal University, Kunming, 650500, China
    3School of Business, Shanghai University of Finance and Economics, Shanghai 200433, China
    4School of Management, Zhejiang University, Hangzhou 310058, China
  • Published:2024-05-25 Online:2024-03-06
  • Contact: LI Juexing E-mail:lijuexing@139.com

Abstract:

Negative family events and experiences have been major social problems in recent years due to the intersection and influence of technical, economic, and epidemic shocks. Negative family events influence leaders’ family domain and also have a cross-domain effect on leadership behaviors at work. However, there are two contradictory views on the relationship between negative family events and effective leadership behaviors. Some scholars claim a negative effect because of the depletion of leaders’ self-control resources, called the “trapped-by-family effect.” Others propose a positive effect owing to cross-domain compensation, called the “compensation effect.”

The inconsistency in existing literature prompts us to reconcile it using the theory of cross-domain leader identity. We argue that the influence of daily negative family events on leader identity and effective leadership behaviors depends on the leaders’ extraversion levels. Specifically, when leaders have higher levels of extraversion, the compensation effect will come into play; that is, daily negative family events will be positively associated with daily effective leadership behaviors by promoting daily leader identity. In contrast, when leaders have lower levels of extraversion, the trapped-by-family effect will play a role; that is, daily negative family events will be negatively associated with daily effective leadership behaviors by reducing daily leader identity. To capture within-person variance and test our model, we conducted two experience sampling studies of middle managers across 10 consecutive workdays.612-629/img_1.png429.0118.05612-629/img_2.png429.0118.05

In Study 1, participants were middle managers from three merchant banks in three cities. Before initiating the daily survey, participants were asked to complete a basic survey containing demographic questions and an extraversion personality inventory. After matching procedures, 461 observations from 67 managers were included in our final sample. In this group of 67 participants, the average age is 34.12 years (SD = 6.44). 68.66% are male, and the average department tenure is 3.88 years (SD = 4.43).

We used Mplus 8.3 (Muthén & Muthén, 2010) for multilevel confirmatory factor analysis in study 1 to assess the validity of the key variables in the model. As shown in Table 1, the results indicated that the five-factor model provided the best fit: χ2 = 1230.06, df = 534, χ2/df = 2.30 (< 5), RMSEA = 0.05 (< 0.08), CFI = 0.91 (> 0.9), TLI = 0.90 (> 0.9), SRMRwithin = 0.03 (< 0.08), and SRMRbetween = 0.07 (< 0.08).

As shown in Table 2, the percentage of within-person variances for negative family events, leader identity, initial structure, transformational leadership, positive affect, negative affect, positive family events, and sleep quality were 26.26%, 34.33%, 36.50%, 44.64%, 38.13%, 33.33%, 61.00%, and 41.40%, respectively. With a sufficient number of within-person variances, multi-level path analysis can be conducted in Study 1.

Table 3 illustrates the descriptive statistics and correlation coefficients of variables in Study 1. At the within-person level, a significant negative correlation was observed between negative family events and initial structure (r = −0.11, p = 0.02). Furthermore, leader identity exhibited a significant positive correlation with both initial structure (r = 0.40, p = 0.00) and transformational leadership (r = 0.32, p = 0.00).

Table 4 reports the results of the multi-level path analysis. As shown in models 2 and 3 in Table 4, negative family events had no significant impact on leader identity (β = −0.06, ns), while extraversion had a positive moderating effect on the relationship between negative family events and leader identity (β = 0.25, p = 0.00). The moderating effect of extraversion is shown in Figure 1. Simple slope analysis showed that the relationship between negative family events experienced by high extraversion leaders and leader identity was significantly positive (slope = 0.18, p = 0.02). On the contrary, the relationship between negative family events experienced by low extraversion leaders and leader identity was significantly negative (slope = −0.33, p = 0.00). The difference was significant (d = 0.52, p = 0.00). Hypothesis 1 was supported.

As shown in Models 5 and 7 in Table 4, leader identity had significant positive effect on initial structure (β = 0.19, p = 0.00) and transformational leadership (β = 0.21, p = 0.01). Hypothesis 2a and 2b were supported.

Table 5 reports the moderated mediation effects. When extraversion was low, the indirect effect of negative family events on initial structure through leader identity was significantly negative, and the 95% confidence interval was [−0.12, −0.02]. When extraversion was high, the indirect effect of negative family events on initial structure through leader identity was significantly positive, and the 95% confidence interval was [0.0002, 0.08]. The difference was significant, and the 95% confidence interval was [0.03, 0.18]. Therefore, hypothesis 3a was supported. When extraversion was low, the indirect effect of negative family events on transformational leadership through leader identity was significantly negative, and 95% confidence interval was [−0.14, −0.01]. When extraversion was high, the indirect effect of negative family events on transformational leadership through leader identity was significantly positive, and 95% confidence interval was [0.0001, 0.09]. The difference was significant, and 95% confidence interval was [0.03, 0.21]. Therefore, hypothesis 3b was supported.

We conducted a supplement analysis for the moderation of extraversion in long term. We speculate that a leader's personal resources will continue to be consumed in a high degree of negative family events for a long time, leading to resource exhaustion, even leaders with high extraversion would find it difficult to have enough resources for leader identity. Therefore, in the long run, the moderating effect of extraversion will not be significant. The results supported our speculation.

In Study 2, we collected data from participants from different regions and industries, and the final sample included 307 observations from 42 managers. In this group of 42 participants, the average age is 35.54 years (SD = 6.33). 80.95% are male, and the average tenure of 42 participants is 8.00 years (SD = 6.87).

As shown in Table 6, in Study 2, multilevel CFA results indicated that the six-factor model provided the best fit: χ2 = 1797.64, df = 915, χ2/df = 1.96, RMSEA = 0.06, CFI = 0.91, TLI =0.89, SRMRwithin = 0.04 (< 0.08), SRMRbetween = 0.09.

As shown in Table 7, the percentage of within-person variances for negative family events, ego depletion, compensation, leader identity, initial structure, and transformational leadership in Study 2 were 31.36%, 43.41%, 30.20%, 36.99%, 24.50%, and 30.99%, respectively.

The descriptive statistics and correlation coefficients of variables in study 2 are shown in Table 8. At within-person level, negative family events were positively correlated with ego depletion (r = 0.38, p < 0.001) and compensation (r = 0.31, p < 0.001), compensation was positively correlated with leader identity (r = 0.60, p < 0.001). Leader identity was positively correlated with initial structure (r = 0.26, p < 0.001) and transformational leadership (r = 0.25, p < 0.001).

The multi-level path analysis results of study 2 are shown in Table 9. The results showed that negative family events had a positive effect on leaders' ego depletion (β = 0.30, p = 0.02) and leaders' compensation for family deficiencies (β = 0.39, p = 0.04). Leaders' ego depletion had a negative effect on leader identity (β = −0.10, p = 0.04), whereas leaders' compensation was positively related with leader identity (β = 0.49, p < 0.001). The two mechanisms of ego depletion and compensation played opposite mediating roles between negative family events and leader identity, and the results also showed that negative family events had no significant effect on leader identity (β = 0.21, ns), indicating the existence of the trapped effect and the compensatory effect. Leader identity had significant positive effects on initial structure (β = 0.18, p = 0.04) and transformational leadership (β = 0.17, p= 0.05).

In study 2, we examined the mediating effects of ego depletion and compensation between negative family events and leader identity. As shown in Table 10, negative family events had negative influence on leader identity through ego depletion, and 95% confidence interval was [−0.08, −0.0002]; negative family events had positive influence on leader identity via compensation, and 95% confidence interval was [0.01, 0.39].

In sum, our data analysis results in two studies showed that negative family events did have both a trapped-by-family effect on leader identity and effective leadership behavior through ego-depletion and a compensation effect on leader identity and effective leadership behavior through compensation.

The theoretical contributions of this paper are fourfold. First, we integrate the inconsistent ideas of the relationship between negative family events and effective leadership behavior using the theory of cross-domain leader identity. We find that the levels of leaders’ extraversion play a vital role in deciding whether negative family events will have a trapped-by-family effect or a compensation effect on effective leadership behavior via leader identity. Second, unlike existing empirical studies, our findings suggest that negative family events will not always lead to negative leadership behaviors. At the within-person level, leaders with high levels of extraversion will exhibit more effective leadership behaviors at work after experiencing negative family events. Third, we extend the current research to further explore the effect of personality on leadership behaviors. Prior studies have suggested that extraversion assists leaders in handling the challenges of work, while we find that extraversion will also promote leaders to actively respond to negative family events by engaging in effective leadership behaviors. Fourth, we also contribute to leader identity studies by shifting its antecedents from work domain to family domain and by exploring the interactive effect of personal and situational factors on leader identity. The present study also provides practical guidance for organizations and leaders to cope with the challenge of negative family events and promote its potential positive effects.

Key words: negative family events, effective leadership behavior, initial structure, transformational leadership, leader identity, extraversion