Introduction

Feedback-seeking behavior refers to “(the) conscious devotion of effort toward determining the correctness and adequacy of behaviors for attaining valued end states” (Ashford, 1986, p 466). Previous studies have shown that feedback-seeking behavior is beneficial for feedback seekers (see a meta-analysis, Anseel et al., 2015), such as enhancing their performance (Lan et al., 2020), building positive images in the workplace (Chun et al., 2018), promoting individual learning and growth (Anseel, 2017), improving creativity (Sung & Choi, 2021), and reinforcing their identification as well as the fit with their jobs and organizations (Bauer et al., 2019; Young & Steelman, 2014).

However, few studies have focused on the potential influence of being asked for feedback on the feedback source (Ashford et al., 2016, for exceptions see Krasman, 2018; Krasman & Kotlyar, 2019). This lack of attention to feedback sources is surprising given that feedback episodes are dynamic and contain interpersonal interactions between the seekers and givers (i.e., the source) (Anseel et al., 2018). In this case, feedback-seeking behavior will have an impact on both sides (i.e., not only on the seekers but also the sources) involved in this dynamic process (Anseel et al., 2015; Ashford et al., 2016). For example, Minnikin et al. (2021) posited that responding to feedback seeking behavior from others consumed the time and energy of the feedback source. Considering the important role of feedback sources in determining the quality of feedback (Lechermeier & Fassnacht, 2018), an in-depth exploration on the reactions of sources to feedback-seeking behavior is needed.

Moreover, prior research has paid little attention to the instant reactions and subsequent work experience of the feedback source (Ashford et al., 2016), leaving it unclear whether being asked for feedback will enhance or destroy sources’ experience at work through their instant reactions. Indeed, one possible and important type of sources reactions is affective reactions, as affections largely derived from work events and determined an individual’s work state (Bledow et al., 2011; Weiss & Cropanzano, 1996). On the one hand, sources may enjoy the process of being asked for feedback, as they can gather valuable information in line with their own work goals from the interactive feedback process (Moss & Martinko, 1998). These positive reactions may then raise sources’ feelings and keep them in a good state at work. On the other hand, being asked and responding to feedback need sources to allocate resources such as time and energy (Minnikin et al., 2021), which may hinder their own goal progress (Koopman et al., 2016). This resource-consuming and interrupting process may disturb feedback sources and their following work experiences (Koopman et al., 2016). In this way, feedback seeking behavior may have a complex and mixed blessing (i.e., both positive and negative) on feedback sources’ work experiences. As such, we aim to apply a balanced perspective to explore the immediate affective reactions of sources after encountering feedback seeking behavior, as well as their subsequent work experiences.

To do so, we applied AET to examine the affective reactions of the feedback source after encountering feedback seeking events (Weiss & Cropanzano, 1996). We especially focused on leaders’ reactions to the feedback seeking behavior from subordinates, as leaders usually act as feedback sources while subordinates are usually seekers in the feedback seeking process (Ashford, 1993). Therefore, we conceptualized subordinates’ feedback seeking behavior as a type of work event (i.e., SFSE) that leaders encountered in their daily work, which then influence their affective reactions. Specifically, when leaders could obtain useful information from SFSE, such as subordinates’ work progress and problems, they may experience stronger positive affect (i.e., higher level positive affect and less negative affect). In contrast, when leaders experienced interruption and resource consumption that obstructed their own goals after encountering SFSE, they may experience stronger negative affect. The positive and negative affects leaders experienced through SFSE will further increase or decrease their daily work state, such as work engagement (i.e., a positive and motivational state that is characterized by vigor, dedication, and absorption, Schaufeli et al., 2002), respectively. The reason why we consider work engagement is that it is not only good for leaders’ mental health but also an important predictor for leader effectiveness (Lanaj et al., 2019; Qin et al., 2018).

Considering the central role of affect in AET, we further expect the moderating role of one emotion regulation strategy (i.e., emotion suppression) in the relationship between SFSE and affective reactions (Matta et al., 2014). Emotion suppression is a stable individual difference that refers to the tendencies of individuals to inhibit his/her true emotion expression (Diefendorff & Richard, 2003). Individuals with higher emotion suppression tendencies, are not only unable to remove negative emotions, but also put themselves into a poor-resource situation (Gross & John, 2003). In this vein, we posit that emotion suppression lightens the positive affect, while aggravates the negative affect elicited by SFSE.

To test our theoretical model of how and when SFSE influences leaders’ daily emotional and work experiences (see Fig. 1), we conducted a study applying experience sampling methodology. This research contributes to the current literature in several ways. First, by focusing on the sources’ perspective, we extend the literature on the outcomes of feedback seeking behavior. Second, by combining the positive and negative affective reactions into one theoretical model, we demonstrate the mixed roles of SFSE and provide a more comprehensive view to explain feedback seeking and responding process (i.e., the dynamic and interactive process). Finally, by including emotion suppression as the boundary condition, we further highlight the central roles of affects in AET, and offer a trajectory of leaders’ affective reactions and emotion regulation when dealing with daily work events.

Fig. 1
figure 1

The theoretical model

Full size image

Theory and hypotheses

Subordinates’ feedback seeking as an affective event

In the organizational context, AET contends that work events can elicit affective reactions (i.e., positive and negative affect) that influence subsequent work attitudes and behaviors (Weiss & Cropanzano, 1996). More specifically, positive work events lead to positive affect through facilitating goal attainment, and negative work events lead to negative affect through obstructing goals progress (Weiss & Cropanzano, 1996). Advancing the opposite valences about workplace events, recent studies demonstrated that certain events have mixed influences on affective reactions, especially when the events contained interactions between two or more parties (Liu et al., 2022). For example, research has indicated that workplace friendship and participating in organizational citizenship behavior (OCB) has both costs and benefits for employees (Koopman et al., 2016; Methot et al., 2016). Align with this, it is feasible for us to explore both the positive and negative affective reactions of feedback sources to SFSE, as this event involves interactions between leaders and subordinates and has the potentials in stimulating and hindering leaders’ daily goal progress.

Although AET was originally developed to infer the attitude and behavior consequences of affective reactions (Weiss & Cropanzano, 1996), a later review posited that affects aroused by work events also influence individual cognitive functions, such as flexibility, attentional focus, and motivational fit (Ashton-James & Ashkanasy, 2005). As work engagement refers to an affective and cognitive state (Schaufeli et al., 2002), and contains attitudinal, energetic and involvement components (Bledow et al., 2011), it can also be explained by AET. In summary, we aim to examine the dual effects of SFSE on leaders’ affective affections, and then their work engagement based on AET.

SFSE and leader’s emotional experiences

Positive affect refers to a pleasant emotional state in which individuals feel active, enthusiastic, and alert, whereas negative affect reflects the degree to which individuals experience tension, anxiety, and worry (Watson et al., 1988). Previous studies have shown that positive affect and negative affect are independent of each other (Bledow et al., 2011). In this vein, individuals could experience these two affects simultaneously. More specifically, individuals would perceive more positive affect and less negative affect when encountering events that align with their goals and would perceive more negative affect and less positive affect when encountering events that hindering their goals (Liu et al., 2022; Weiss & Cropanzano, 1996). As SFSE not only involves valuable information that advances the work goals of leaders and the entire team but also has the potential to disturb leaders’ work pace and leave behind their planned goals, it may determine leaders’ affective reactions.

First, we expect leaders to perceive positive affect based on the functions of SFSE on their own and team goals. In detail, one of the most important roles leaders need to play in their daily work is to direct and motivate their subordinates to achieve higher performance (Rosen et al., 2019). SFSE provides leaders with such opportunities to acknowledge the progress of their subordinates’ work and team work, as well as the chances to implement their guidance and influence as a leader (Heen & Stone, 2014). Besides, research has indicated that part of leaders’ job performance depends on their employees (Moss & Martinko, 1998). In this vein, SFSE works on facilitating the goal attainment of leaders and induces their positive affective reactions. Research has provided indirect support for this view, indicating that dealing with task-centered rather than task-independent emails from subordinates disturbs leaders less in their daily goal progress (Rosen et al., 2019).

Second, we assume a positive relation between SFSE and leaders’ negative affect, as SFSE may hinder leaders’ personal goals. Specifically, despite responding to subordinates’ needs, leaders also have other responsibilities (e.g., making personal decisions) to complete (Lanaj et al., 2019). While responding to and dealing with SFSE consume the limited time and attention of leaders, which would obstruct leaders’ goal progress and then evoke negative affect (Rosen et al., 2019). To support this view, Lanaj and Jennings (2020) found that being responsive to personal help requests from subordinates puts leaders in a bad mood on a daily basis. Accordingly, we propose the following hypotheses:

  • Hypothesis 1a: SFSE will be positively related to leaders’ positive affect.

  • Hypothesis 1b: SFSE will be positively related to leaders’ negative affect.

The mediating role of affective reactions

Work engagement is an active working state for leaders, which is characterized by vigor, dedication, and absorption (Schaufeli et al., 2002). According to AET and previous research, work engagement may also fluctuate with affective reactions (Bledow et al., 2011). While a systematic review of work engagement has concluded that job resources are “the most important predictors of work engagement (Bakker et al., 2014, p 393)”, and positive affects help broaden and build resources (Ouweneel et al., 2012), while negative affects narrow the resources allocation (Bledow et al., 2011), we expect a positive relationship between positive affect and work engagement and a negative relationship between negative affect and work engagement respectively. Indeed, previous studies have provided support for this proposition, with Ouweneel et al. (2012) found an increase in work engagement owing to positive affect and Bledow et al. (2011) found a decrease in work engagement owing to negative affect. Accordingly, we propose the following hypotheses:

  • Hypothesis 2a: Positive affect will be positively related to leaders’ daily work engagement.

  • Hypothesis 2b: Negative affect will be negatively related to leaders’ daily work engagement.

Altogether, combining Hypotheses 1a and 1b, which posited the positive relationship between SFSE and affective reactions (i.e., positive affect (Hypothesis 1a) and negative affect (Hypothesis 1b)), with Hypotheses 2a and 2b, which assumed the positive relationship between positive affect and work engagement (Hypothesis 2a), as well as the negative relationship between negative affect and work engagement (Hypothesis 2b), we argue that SFSE will influence leaders’ work engagement via increasing their positive or negative affect. Accordingly, we propose the following hypotheses:

  • Hypothesis 3a: SFSE will have a positive indirect effect on leaders’ daily work engagement via positive affect.

  • Hypothesis 3b: SFSE will have a negative indirect effect on leaders’ daily work engagement via negative affect.

Moderating effects of leaders’ emotion suppression

Emotion suppression reflects the extent to which individuals inhibit expressing his or her true feelings (Diefendorff & Richard, 2003). While emotion suppression is usually regarded as a stable individual difference, examining its moderating roles also aligns with the proposition of AET, which contends that there are individual differences in affective reactions to the same work events (Weiss & Cropanzano, 1996). Indeed, emotion suppression has been proven to be a resource-consuming emotion regulation strategy and is ineffective in alleviating negative affect (Gross, 2015; Gross & John, 2003). Therefore, we propose that emotion suppression may decrease the positive affect aroused by SFSE because of its consumption of resources, and may increase the negative affect aroused by SFSE owing to its nature in resource-consuming and its inability in lightening negative affect.

Specifically, leaders with high levels of emotion suppression tend to conceal their true feelings by inhibiting negative affect while displaying positive affect (Diefendorff & Richard, 2003). This emotion regulation process needs them to invest resources and leads to a poor cope capacity (Brotheridge & Grandey, 2002). Besides, while emotion suppression has little effect on releasing the negative affect (Gross & John, 2003), it accordingly fails to reduce the negative affect derived from SFSE. Therefore, the positive affect elicited by SFSE will be buffered, and the negative affect elicited by SFSE will be strengthened for this type leader.

In contrast, leaders with low levels of emotion suppression tend to express their feelings authentically (Diefendorff & Richard, 2003), and expressing emotion in this way does not cause additional drains on their own resources. Leaders of this type will react normally (i.e., perceive positive and negative affect at the original level) when experiencing SFSE. To support these propositions, Zhou et al. (2019) found that employees who comply with display rules (i.e., suppress negative emotions and express positive emotions) experience stronger negative affect from workplace incivility compared to employees who do not comply with the rules. Therefore, we propose the following hypotheses:

  • Hypothesis 4a: Emotion suppression will moderate the relationship between SFSE and positive affect, such that this relationship is weaker for leaders with higher levels of emotion suppression.

  • Hypothesis 4b: Emotion suppression will moderate the relationship between SFSE and negative affect, such that this relationship is stronger for leaders with higher levels of emotion suppression.

Method

Participants and procedures

We recruited part-time MBA students at a university in northern China as participants. We first introduced the background and purpose of our research project to MBA students and clarified the voluntary nature of this project. Participants had to meet the following criteria: (1) hold a leadership positionFootnote 1 and supervise at least one subordinate; (2) have a full-time job; (3) have no business trips within 5 working days of the daily investigation. To increase the response rate, we used a random lottery (with management books as prizes) to motivate participants to complete the surveys. In addition, participants can also obtain a research report at the end of the project.

We collected data in two phases, using the experience sampling methodology (ESM) via online surveys hosted by WeChat, a popular messaging app in China. In the first phase, we sent a baseline survey to participants, asking them to report their demographic information and emotion suppression. In the second phase, which was about a week after phase one, we collected data twice a day for 5 consecutive workdays. Specifically, we sent the first daily survey at 3 p.m. (Time 1), which was used to measure daily SFSE, positive affect and negative affect. We sent the second daily survey at 6 p.m. (Time 2), asking participants to report their daily work engagement. For each time, participants were given a one-hour window to respond to the survey.

Of the 119 participants that initially signed up for our research project, 13 of them quit the research project due to personal reasons. The remaining 106 individuals provided the final 504Footnote 2 day-level data points with a response rate of 84.7% (out of a possible 595 total data points). The final sample consisted of 106 leaders who were employed in a variety of industries, including technology, finance, internet, education, consulting, and sales. Within that sample, 53.8% of the participants were males, the average age was 33.81 years (SD = 5.09), and the average tenure of leadership was 4.86 years (SD = 3.59). Most of them (57.5%) held bachelor’s degrees, and had ten subordinates on average (SD = 9.72).

Measures

The measures we used were originally constructed in English. We performed a standard translation and back-translation procedure to ensure cross-cultural consistency (Brislin, 1980). All the scales were rated on a 7-point Likert scale and, unless otherwise indicated, 1 to 7 represented “strongly disagree” to “strongly agree”.

Between-individual measurement

Emotion suppression

We measured emotion suppression in the baseline survey using a four-item scale developed by Diefendorff and Richard (2003). Sample items are “I control my emotions and make sure they are appropriate” and “I conceal negative emotions about tasks and others”. The α of this scale was 0.80.

Within-individual measurement

SFSE

To measure the intensity of SFSE experienced by leaders in their daily work, we adopted the feedback-seeking behavior scale developed by Ashford and Tsui (1991). The scale contained three items, and a sample item is “Today my subordinates asked me for feedback about his or her performance”. Participants rated each item on a 7-point Likert scale, with 1 to 7 representing “Never” to “More than five times”. The average α across five days was 0.93.

Positive affect and negative affect

We used Watson et al.’s (1988) twelve-item PANAS scale to measure leaders’ positive affect and negative affect. Sample items are “interested” and “enthusiastic” for positive affect, and “worried” and “irritable” for negative affect. The average α across five days was 0.92 for positive affect, and 0.93 for negative affect.

Work engagement

We measured work engagement using a three-item scale adopted by Rich et al. (2010). Sample items are “Today I exerted a lot of energy on the job” and “Today I was absorbed by the job”. The average α across five days was 0.95.

Control variables

Following previous research (e.g., Vogel et al., 2022), we controlled the study day (coded with 1 to 5) to remove common methods bias.

Analytic strategy

Due to the nested structure of our data (i.e., multiple days nested within leaders, and the within-individual variance proportion of our study variables ranged from 37.1 to 52.7%, indicating the feasibility in using multilevel analysis, see Table 1), we conducted a multilevel path analysis using Mplus 8.3 to test our hypotheses. Specifically, the between-individual cross-level moderator (i.e., leaders’ emotion suppression) was modeled at level 2, and the within-individual variables (i.e., SFSE, leaders’ positive and negative affect, and leaders’ work engagement) were modeled at level 1 using random slopes. In addition, following Enders and Tofighi (2007), we group-mean centered the within-individual variables and grand-mean centered the between-individual variables to distinguish the variances between different levels.

Table 1 Percentage of within-individual variance among the daily variables
Full size table

We conducted two independent statistical models to test the hypotheses. Model 1 (M1) was used to test the mediating effect of positive and negative affect. Following Preacher et al. (2010), we ran two parallel mediating path analyses (i.e., 1–1 (1) -1 mediating model) at within- and between-individual levels, and examined the indirect effect by calculating the product of the path coefficients (i.e., the coefficients from the independent variable to the mediating variable, and from the mediating variable to the outcome variable) using Model Constraint command at the within-individual level. Based on Model 1, Model 2 (M2) included emotion suppression to test the cross-level moderating effect. Specifically, we set the random slopes between SFSE and leaders’ positive affect and negative affect as dependent variables and ran a linear regression on emotion suppression to examine the moderating role of emotion suppression.

Results

We conducted a multilevel confirmatory factor analysis (CFA) to examine the discrimination of the studied variables. The results in Table 2 demonstrated that the five factors model (SFSE, leaders’ positive affect, leaders’ negative affect, leaders’ work engagement, and leaders’ emotion suppression) had a better fit (χ2 = 875.12, df = 406, CFI = 0.94, TLI = 0.93, RMSEA = 0.05, SRMRwithin = 0.05, SRMRbetween =0.08) than the other three competitive models, indicating that variables in this study had good discriminant validity.

Table 2 Results of confirmatory factor analysis for studied variables
Full size table

Table 3 summarizes the means, standard deviations, internal consistency reliabilities, and correlations among the studied variables. The results indicated that SFSE was positively correlated with leaders’ positive affect (r = .10, p < .001) and negative affect (r = .12, p < .001); leaders’ positive affect (r = .11, p < .001) and negative affect (r = − .08, p < .001) were positively and negatively correlated with leaders’ work engagement respectively. These results provided initial support for Hypotheses 1a to 2b.

Table 3 Descriptive statistics and correlations among studied variables
Full size table

Tests of hypotheses

Within-individual hypotheses

M1 was used to test Hypothesis 1 to Hypothesis 3, which posited that SFSE had indirect effects on leaders’ work engagement via positive and negative affect. As shown in Table 4, SFSE was positively related to leaders’ positive affect (B = 0.11, SE = 0.03, p < .001), and was positively related to leaders’ negative affect (B = 0.14, SE = 0.04, p < .001) after controlling the demographic variables. Thus, Hypothesis 1a and Hypothesis 1b were supported. Additionally, leaders’ positive affect was positively related to work engagement (B = 0.22, SE = 0.05, p < .001), and leaders’ negative affect was negatively related to work engagement (B = − 0.13, SE = 0.04, p < .01), therefore Hypothesis 2a and Hypothesis 2b were supported.

Table 4 Multilevel path analysis results for mediation test (M1)
Full size table

To confirm mediating roles of positive and negative affect, we calculated the product of the path coefficients using the Model Constraint command in Mplus. Results indicated that the indirect effect of the relationship between SFSE and leaders’ work engagement through positive affect was 0.024 (SE = 0.01, p < .01, 95% CI [0.006, 0.042]), the indirect effect of the relationship between SFSE and leaders’ work engagement through negative affect was − 0.018 (SE = 0.01, p < .05, 95% CI [-0.034, -0.003]). Thus, Hypotheses 3a and 3b were also supported. To further explain the relationship between SFSE and work engagement, we calculated the difference between the positive indirect and negative indirect effects, and the total effect (i.e., two indirect effects plus the direct effect) of SFSE on work engagement. Results showed that the difference between the positive path (i.e., positive affect as the mediator) and the negative path (i.e., negative affect as the mediator) was 0.042 (SE = 0.01, p < .001, 95% CI [0.023, 0.062]); and the total effect was 0.087 (SE = 0.03, p < .01, 95% CI [0.025, 0.149]), indicating an offsetting effect of positive indirect effect on the negative indirect effect and a positive effect of SFSE on work engagement.

Between-individual hypotheses

Based on M1, we tested the cross-level interaction hypotheses (i.e., Hypotheses 4a and 4b) by including leaders’ emotion suppression into the statistical model (i.e., M2). As shown in Table 5, the interaction term of SFSE and leaders’ emotion suppression was negative and significant (B = − 0.05, SE = 0.02, p < .05), Thus, Hypothesis 4a was supported by the data. When taking negative affect as the dependent variable, the interaction term of SFSE and leaders’ emotion suppression was not significant (B = − 0.02, SE = 0.03, ns.), Hypothesis 4b, therefore, was not supported by the data.

Table 5 Multilevel path analysis results for moderation test (M2)
Full size table

To further explain the moderating effect of leaders’ emotion suppression, following Cohen and Cohen (1983), we conducted regression analyses at one standard deviation above and below the mean of emotion suppression and depicted the relationship between SFSE and leaders’ negative affect. As shown in Fig. 2, the relationship between SFSE and positive affect was significant and positive (simple slope = 0.09, t = 2.00, p < .05) when leaders’ emotion suppression was at lower levels; however, this relationship was not significant when leaders’ emotion suppression was at higher levels (simple slope = 0.01, t = 0.23, ns).

Fig. 2
figure 2

Moderating effect of emotion suppression on the relationship between SFSE and positive affect

Full size image

Supplemental analysis

To further expand the model, we examined whether emotion suppression can moderate the indirect effect of the relationship between SFSE on work engagement via positive affect (i.e., the moderated mediating effect). According to Edwards and Lambert (2007), we calculated the indirect effect of SFSE on leaders’ work engagement through positive affect at higher (+ 1 SD) and lower (-1 SD) levels of emotion suppression. The results showed that this indirect effect was marginally significant when emotion suppression was low (indirect effect = 0.022, SE = 0.01, p = .07, 95% CI [-0.027, 0.032]), but not significant when emotion suppression was high (indirect effect = 0.002, SE = 0.02, ns), and the difference between these two indirect effects was significant (indirect effect = − 0.019, SE = 0.01, p < .05, 95% CI [-0.037, − 0.001]), providing support for the moderated mediated effect.

Discussion

Drawing upon the AET (Weiss & Cropanzano, 1996), we developed and tested a model explaining how and when experiencing SFSE affects leaders’ emotional experiences and daily work engagement. Findings from a daily diary survey on 106 leaders for 5 workdays revealed that SFSE had a dual effect on leaders (i.e., feedback source). On the bright side, SFSE elicited leaders’ positive affect, which in turn increased their daily work engagement. On the other hand, SFSE also elicited negative affect, which in turn decreased leaders’ daily work engagement. Moreover, our research results supported the moderating role of leaders’ emotion suppression on the relationship between SFSE and positive affect, with higher levels of emotion suppression hindering the positive affect derived from SFSE.

Theoretical implications

Our study has important theoretical implications. First, we extend workplace feedback-seeking literature by shifting the predominant focus from the feedback seeker to the feedback source. Research has indicated that organizational phenomena (e.g., leadership style) which contain social interaction processes will impact both embedded sides (e.g., Lanaj et al., 2016; Liu et al., 2022). As feedback seeking and responding are interactive by nature (Ashford et al., 2016), a source-centric view of feedback seeking behavior is needed. Our findings suggest that SFSE elicits leasers’ affective reactions, which in turn influence their daily work state. As such, we highlight the source’s perspective in feedback seeking and responding process and enrich the outcomes of feedback seeking behavior.

Second, our research focuses on the paradox traits of SFSE, extending AET theory by examining the positive and negative affective reactions in one theoretical model. Prior research has indicated that certain work events, behaviors, or characteristics have both bright and dark effects on employees’ or team results (Koopman et al., 2016; Liu et al., 2022). We advance this research line by suggesting that SFSE has a balanced effect (i.e., both positive and negative) on leaders’ affective reactions. Besides, research on feedback seeking behavior has demonstrated the potential positive and negative impacts of this behavior on sources separately (Krasman, 2018; Krasman & Kotlyar, 2019), we contribute to these studies by offering a comprehensive perspective of the conflict influences.

Finally, we provide a new contingency view (i.e., emotion suppression) on the relationship between work events and the affective reactions of leaders. Previous studies have demonstrated that personal traits such as big-five personality (Lanaj et al., 2016) and self-control (Rosen et al., 2019) could influence leaders’ reactions to certain work events. However, few studies have focused on leaders’ emotion regulation in this relationship. By demonstrating the moderating role of emotion suppression on the relationship between SFSE and positive affect, we highlight the central roles of emotions in AET and provide a holistic picture involving both affective reactions and emotion regulation in the process of handling work events.

Practical implications

This study also has important practical implications. First, organizations should pay attention to leaders’ responses to SFSE and improve their abilities to deal with SFSE. While our findings suggest that SFSE has both positive and negative effects on leaders’ affective reactions and daily work experience, organizations can take steps to help leaders focus more on the positive aspects of SFSE, such as by encouraging leaders to engage in positive self-reflection (Lanaj et al., 2019), think deeply about the possible benefits of feedback-seeking behavior to subordinates, themselves and the team. Moreover, organizations could also provide relevant trainings aimed at developing leaders’ self-regulation and self-control (Rosen et al., 2019), eventually improving their abilities to handle SFSE.

Secondly, organizations should take measures to develop leaders’ emotion regulation abilities. Owing to the hindering moderating effect of emotion suppression, organizations should spend effort in judging and improving leaders’ emotion regulation orientation. This could be done through two ways. First, when organizations recruit or select leaders, it should involve candidates’ differences in emotion regulation in the assessment criteria. Second, organizations can provide courses related to emotion regulation, such as guiding leaders to express real emotions through deep acting or cognitive reappraisal (Alam & Singh, 2021; Matta et al., 2014).

Finally, our findings also shed light on the management of daily feedback seeking behavior. Specifically, due to the possible interrupting effects of SFSE on leaders’ goal progress and negative effect, organizations should guide this behavior, such as instructing employees to fully consider the motivations and purposes for the feedback seeking (Minnikin et al., 2021), or providing formal feedback seeking area and setting regularly time for feedback seeking.

Limitations and future directions

This study, inevitably, has some limitations. First, this study measured leaders’ positive and negative affects at fixed time points (i.e., interval-based sampling), which make us unable to capture leaders’ immediate emotional responses after SFSE. Future studies are encouraged to measure instant affective reactions by applying event-based measurement (e.g., Wijewardena et al., 2017).

Second, this study only focused on the frequency of SFSE, neglecting the content of feedback seeking. Based on our proposition, when the content of SFSE is more related to leaders’ goal achievement, it will arouse stronger positive rather than negative affect. In contrast, when the content is less related to leaders’ goal achievement, it will arouse stronger negative rather than positive affect (Lanaj & Jennings, 2020). In this vein, future research will benefit from distinguishing the content of SFSE, and then providing a more detailed examination between SFSE and affective reactions.

Third, this study only examined one possible boundary condition (i.e., emotion suppression) in the relationship between SFSE and affective reactions. Other factors such as perspective taking, and leadership experience will also play role in this process, as leaders with these traits have higher coping abilities (Chun et al., 2018; Ku et al., 2015). Notably, the moderating role of emotion suppression on the relationship between SFSE and negative affect was not supported by our data. One possible reason may exist in the inability of this emotion regulation in dealing with negative affect (Gross & John, 2003). The other emotion regulation strategy (i.e., reappraisal) would play a more important role in the process of negative affective reactions (Gross & John, 2003). Future studies thus are encouraged to explore more boundary conditions in work events and affective reactions relationship.

Finally, this study mainly focused on the effect of SFSE on leaders’ work engagement. Future research could move beyond the current focus on affective reactions and work engagement by investigating how SFSE will shape leadership behaviors such as daily transformational leadership (Rosen et al., 2019), moving its results to employee even team level.

Conclusion

The results from the current study indicate that SFSE has dual effects on leaders daily work engagement via the increased positive affect and negative affect; individual difference (i.e., emotion suppression) released the positive relationship between SFSE and positive affect. Our findings contribute to FSB literature on how FSB influences the emotional responses and attitudes of feedback sources.