Going the extra mile (or not): A moderated mediation analysis of job resources, abusive leadership, autonomous motivation, and extra-role performance

Parent-Lamarche, Annick; Fernet, Claude; Austin, Stéphanie Article Going the extra mile (or not): A moderated mediation analysis of job resources, abusive leadership, autonomous motivation, and extra-role performance Administrative Sciences Provided in Cooperation with: MDPI – Multidisciplinary Digital Publishing Institute, Basel Suggested Citation: Parent-Lamarche, Annick; Fernet, Claude; Austin, Stéphanie (2022) : Going the extra mile (or not): A moderated mediation analysis of job resources, abusive leadership, autonomous motivation, and extra-role performance, Administrative Sciences, ISSN 2076-3387, MDPI, Basel, Vol. 12, Iss. 2, pp. 1-12, https://doi.org/10.3390/admsci12020054 This Version is available at: https://hdl.handle.net/10419/275325 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Citation: Parent-Lamarche, Annick, Claude Fernet, and Stéphanie Austin. 2022. Going the Extra Mile (or Not): A Moderated Mediation Analysis of Job Resources, Abusive Leadership, Autonomous Motivation, and Extra-Role Performance. Administrative Sciences 12: 54. https://doi.org/10.3390/ admsci12020054 Received: 18 March 2022 Accepted: 25 April 2022 Published: 28 April 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). administrative sciences Article Going the Extra Mile (or Not): A Moderated Mediation Analysis of Job Resources, Abusive Leadership, Autonomous Motivation, and Extra-Role Performance Annick Parent-Lamarche * , Claude Fernet and Stéphanie Austin Department of Human Resources Management, Universitédu Québec àTrois-Rivières, 3225, Albert-Tessier, Trois-Rivières, QC G8Z 4M3, Canada; [email protected] (C.F.); [email protected] (S.A.) *Correspondence: [email protected] Abstract: Abusive leadership is particularly prevalent in nursing and it can have multiple adverse effects on performance at work. However, little research has examined whether and under what conditions abusive leadership may be detrimental to nurses’ extra-role performance. This crosssectional study explores whether abusive leadership intensifies the effects of emotional job resources on autonomous motivation, a psychological mechanism that could be responsible for extra-role performance. Data were collected from dyads of registered French-Canadian nurses and their immediate supervisors (n= 99 dyads). The models were tested with path analysis using Mplus. Our results show that extra-role performance is positively associated with nurses’ job emotional resources and autonomous motivation, but negatively associated with abusive leadership. Nurses’ cynicism is also negatively associated with autonomous motivation. Importantly, the indirect relation between emotional resources and extra-role performance through autonomous motivation is moderated by abusive leadership, providing support for a moderated mediation effect. These results add to those supporting a similar moderated mediation mechanism to explain employee attitudes and demonstrate the relevance of self-determination theory in a work context. These findings reinforce the need to focus on the quality of leadership practices as well as interventions aimed at promoting the performance of nurses at work. Keywords: abusive leadership; autonomous motivation; extra-role performance; emotional resources; job demands–resources model; self-determination theory; path analysis; moderated mediation 1. Introduction Abusive leadership, also defined as abusive supervision, relies on repeated negative behavior, such as anger, intimidation, yelling, and ridiculing (Tepper 2000). Data suggest that an estimated 13.5% of U.S. workers (Schat et al. 2006) are exposed to abusive leadership. In fact, in this country, it is estimated that corporations lose USD 23.8 billion annually in productivity as a result of abusive leadership. A study conducted among a random sample of 6500 registered nurses (Estes 2013) showed that the prevalence of abusive supervision is particularly high in a nursing context. Specifically, 46.6% of participants reported perceiving abusive supervision, among which 36.6% believed it to undermine their performance. Research has shown that abusive leadership has a negative impact on employees’ performance and an array of organizational elements (Tao et al. 2017). Employees’ perceptions of working under a supervisor’s duress is related to adverse personal (e.g., work stress, burnout, and psychological distress (Huang et al. 2020)) and organizational outcomes (e.g., turnover intention and workplace deviance; (Lyu et al. 2019)). For instance, Lavoie- Tremblay et al. (2016) found that it predicts a poorer quality of care in nurses towards patients beyond that explained by transformational leadership. This finding corroborates those of studies suggesting that the effects of negative supervisor–employee relationships are greater and last longer than those of more positive relationships (Baumeister et al. 2001). Adm. Sci. 2022,12, 54. https://doi.org/10.3390/admsci12020054 https://www.mdpi.com/journal/admsci Adm. Sci. 2022,12, 54 2 of 12 However, few studies have examined whether and under what conditions abusive leadership may be detrimental to extra-role performance. In contrast to job duties and task performance, extra-role performance, emphasizing discretionary behaviors. This type of behavior is usually not directly or explicitly recognized by an organization’s formal reward system (e.g., volunteering for non-mandatory roles and tasks, collaborating with other nurses to improve the quality of services). Nonetheless, discretionary behaviors are linked to efficiency and organizational effectiveness (Organ 1988). As such, discretionary behaviors appear to be particularly important in the current nursing context given the strict resources (shortages and the caring and empathetic nature of helping professions). 1.1. Purpose of the Study To address this gap, our cross-sectional study draws on recent developments in theory and research on abusive leadership (Jha 2019;Tepper et al. 2017), which suggest that employees may respond differently to the work environment as a function of the degree to which they perceive abusive behaviors from their supervisor. We explore whether abusive leadership intensifies the effects of job emotional resources (such as emotional support from patients, colleagues, and supervisors) on autonomous motivation (acting with volition and choice), a psychological mechanism that could be responsible for extrarole performance. As such, abusive leadership would moderate the effect of emotional resources on autonomous motivation, which entails conditions that direct and energize nurses’ discretionary behaviors. 1.2. Background Self-determination theory (Deci and Ryan 2000) provides a useful perspective for understanding the underlying reasons that motivate behaviors and the associated job and individual outcomes. To this end, SDT offers two main classifications of motivation that correspond to the quality of the efforts expended at task: autonomous and controlled motivation. For the first, it brings together the reasons linked to the inherent pleasure and satisfaction of carrying out a task (intrinsic motivation), as well as those that allow the achievement of goals considered important since they are in harmony with one’s personal values (identified regulation). For the second, controlled motivation reflects the efforts that are invested to either obey and conform to internal pressures (introjected regulation), such as avoiding feelings of guilt, or to respond to external pressures (external regulation), such as dodging punishment or obtaining rewards of a material or social nature. However, and despite various job constraints (e.g., regulatory actions and operations), nurses are liable to perform extra-role behaviors if they are autonomously motivated at work. This is likely to occur when they receive proper support and feedback, feel ownership of their work, and understand the meaning of their tasks (Deci et al. 2017). SDT-based research points to the importance of employee autonomous motivation in relation with attitudinal, affective, and behavioral outcomes (Gagnéand Deci 2005; for a recent meta-analysis, see Van den Broeck et al. 2021). In the nursing context, autonomous motivation has been positively related to organizational and occupational commitment (Fernet et al. 2017,2021) and in-role performance (Fernet et al. 2015), but negatively to emotional exhaustion (Fernet et al. 2020b), turnover intentions, and sickness absence (Austin et al. 2020). Despite their interest, these studies were based on self-report measures of job performance (increasing the possibility of self-evaluation bias) and did not directly address extra-role performance and its potential predictors. 1.2.1. The Mediating Role of Autonomous Motivation According to SDT, employee motivation largely depends on various “aspects of the social environment, including both aspects of the job and the work climate” (Gagnéand Deci 2005;Sandrin et al. 2021). Aside from autonomy support, as proposed by SDT, studies have investigated different job resources (e.g., decision latitude, recognition, and social support) that may facilitate or hinder employee motivation (Olafsen and Deci 2020). This Adm. Sci. 2022,12, 54 3 of 12 is in line with the motivational process assumption of the job demands–resources model (JD-R model), which proposes that adequate job resources foster employee motivation and desirable outcomes, including in-role and extra-role performance (Bakker et al. 2004). In this framework, job resources are defined as the psychosocial, physical, and organizational aspects that support employees in their actions and health. For example, social support, as an emotional resource, helps employees accomplish their tasks while contributing to their well-being (Bakker and Demerouti 2017). Recent studies (Fernet et al. 2015,2020b) have indicated that job resources (e.g., emotional, cognitive, and physical) are positively associated with nurses’ autonomous motivation. Of particular interest, a study by Fernet et al. (2015) found that job resources predict in-role performance through autonomous motivation. However, some studies have also showed that job demands and burnout (an affective strain reaction mainly characterized by exhaustion and cynicism (Schaufeli and Bakker 2004)) are involved in the motivational process (e.g., Bakker et al. 2004). Similar to a lack of job resources, too many job demands or a state of burnout can be unfavorable to employee performance because it would prevent them from the necessary mental and physical energy to engage in discretionary actions. The depletion of emotional resources—which is relevant for nurses because of the high emotional demands associated with their job—would be especially important for discretionary, self-initiated actions. On the basis of these theoretical and empirical considerations, we propose the following hypothesis: Hypothesis 1 (H1). Controlling for job emotional demands and burnout, emotional resources will be positively related to extra-role performance through autonomous motivation. 1.2.2. The Moderating Role of Abusive Leadership Tepper (2000) defines abusive leadership as sustained forms of hostility perpetrated against subordinates (e.g., nasty or demeaning comments, loud outbursts). Krasikova et al. (2013) describe that such behavior can take various forms that go against the values of the organization. Examples are the use of negative communication to intimidate, criticize, or to get the subordinate to comply or adhere to a specific course of action. In the workplace, perceptions of abusive leadership have been negatively associated with in-role performance, including formal and supervisory ratings (Moin et al. 2020), organizational citizenship behaviors (Kim et al. 2020), as well as a quality of patient care (Lavoie-Tremblay et al. 2016). In answer to abusive leadership, individuals tend to voluntarily engage in inappropriate or even deviant behaviors, including sabotage, lateness, theft, shirking, and working slower than usual (Tepper et al. 2017). Accordingly, they should be less keen to perform extra-role behaviors. From a motivational standpoint, abusive behaviors would inhibit employees’ extra-role behaviors insofar as they conflict with a sense of self-determination. In other words, employees are less inclined to invest their cognitive, emotional, and physical energy into their performance in the face of organizational hurdles, such as a mismatch between organizational expectations and personal values (Kahn 1990). This rationale suggests that the mediating role of autonomous motivation in the relation between emotional resources and extra-role performance may depend on the supervisor’s leadership behavior. To our knowledge, only one empirical study has examined the potentially moderating role of abusive leadership on extra-role performance. Jha (2019) found that telecom employees’ psychological empowerment (which is prototypically autonomous motivation) mediated the relation between psychological safety and intentions to stay with the organization, such that the indirect relation was stronger when abusive leadership was low. Expending on this proposal and based on the aforementioned theoretical background, the following hypothesis is proposed: Hypothesis 2 (H2). Controlling for job emotional demands and burnout, abusive leadership will moderate the indirect relation of emotional resources and extra-role performance through autonomous motivation, such that the mediated relation will be stronger when abusive leadership is low. See Figure 1. Hypothetical model. Adm. Sci. 2022,12, 54 4 of 12 Adm. Sci. 2022, 12, x FOR PEER REVIEW 4 of 13 such that the indirect relation was stronger when abusive leadership was low. Expending on this proposal and based on the aforementioned theoretical background, the following hypothesis is proposed: Hypothesis 2 (H2). Controlling for job emotional demands and burnout, abusive leadership will moderate the indirect relation of emotional resources and extra-role performance through autonomous motivation, such that the mediated relation will be stronger when abusive leadership is low. See Figure 1. Hypothetical model. Figure 1. Hypothetical model. 2. Methods 2.1. Procedure and Participants Data were collected from dyads of registered French-Canadian nurses and their immediate supervisor after ethics approval from the University of Quebec in Trois-Rivières ethical committee (CER-10-159-06.20) and participants’ informed consent was obtained. All participants were members of the Professional Association of Nurses of Quebec. After having read an invitation letter describing the objective of the study, they completed an informed consent form and an electronic questionnaire which asked them to provide the contact details of their immediate supervisor. Supervisors were then invited to complete an electronic questionnaire within a two-week period. Of the 541 nurse participants, 116 supervisors completed the questionnaire, for a response rate of 21%. We used listwise deletion to obtain a full dataset of nurses and supervisors of the same dyad who had both fully answered to the questionnaire (n = 99 dyads; the nurse and her/his supervisor). Most of the nurses were women (n = 92, 92.93%) with an average age of 27.19 years (SD = 7.01), as for the immediate supervisor (n = 77, 77.78%), but with an average age of 44.09 years (SD = 9.53). Most nurses worked in a non-ICU unit (81.97%) and held a permanent position (81.03%). 2.2. Measures All of the instruments used in the present study are already available in the public domain, and no authorization was required for the utilization of any of these measures for research purposes. All measures were administrated in French and selected based on previous studies supporting their validity and reliability (Bocéréan et al. 2019; Fan et al. 2020; Fernet et al. 2020a; Nasr et al. 2009). The means, standard deviations, and correlation coefficients of these measures are presented in Table 1. Figure 1. Hypothetical model. Note. See Table 3 for the indirect effects of emotional resources on extra-role job performance via autonomous motivation (Hypothesis 1). See Figure 2 and Table 3 for the moderated mediation effect of abusive leadership on the relationship between emotional resources and extra-role job performance. 2. Methods 2.1. Procedure and Participants Data were collected from dyads of registered French-Canadian nurses and their immediate supervisor after ethics approval from the University of Quebec in Trois-Rivières ethical committee (CER-10-159-06.20) and participants’ informed consent was obtained. All participants were members of the Professional Association of Nurses of Quebec. After having read an invitation letter describing the objective of the study, they completed an informed consent form and an electronic questionnaire which asked them to provide the contact details of their immediate supervisor. Supervisors were then invited to complete an electronic questionnaire within a two-week period. Of the 541 nurse participants, 116 supervisors completed the questionnaire, for a response rate of 21%. We used listwise deletion to obtain a full dataset of nurses and supervisors of the same dyad who had both fully answered to the questionnaire (n= 99 dyads; the nurse and her/his supervisor). Most of the nurses were women (n= 92, 92.93%) with an average age of 27.19 years (SD = 7.01), as for the immediate supervisor (n= 77, 77.78%), but with an average age of 44.09 years (SD = 9.53). Most nurses worked in a non-ICU unit (81.97%) and held a permanent position (81.03%). 2.2. Measures All of the instruments used in the present study are already available in the public domain, and no authorization was required for the utilization of any of these measures for research purposes. All measures were administrated in French and selected based on previous studies supporting their validity and reliability (Bocéréan et al. 2019;Fan et al. 2020;Fernet et al. 2020a;Nasr et al. 2009). The means, standard deviations, and correlation coefficients of these measures are presented in Table 1. Table 1. Descriptive and correlational statistics. M SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. Extra-role performance 5.66 0.77 – 2. Abusive leadership 1.35 0.45 −0.34 ** – 3. Emotionaldemands 4.37 1.29 −0.06 0.34 ** – 4. Emotional resources 5.28 1.13 0.25 * −0.44 ** −0.35 ** – 5. Supervisor Emotional exh 1.91 1.23 −0.07 −0.05 0.03 0.07 – 6. Supervisor cynicism 0.91 0.96 −0.10 −0.04 −0.05 0.07 0.79 ** – 7. Aut motivation 5.57 0.97 0.37 ** −0.25 * −0.27 ** 0.33 ** −0.05 0.09 – 8. Employee Emotional exh 2.61 1.44 −0.17 0.32 ** 0.58 ** −0.34 ** −0.02 −0.09 −0.38 ** – 9. Employee cynicism 1.39 1.35 −0.18 0.45 ** 0.46 ** −0.50 ** 0.02 −0.09 −0.62 ** 0.65 ** – Note A: * p ≤0.05 and ** p ≤0.01; exh = exhaustion; aut = autonomous. 2.2.1. Abusive Leadership Abusive leadership was measured with the 15-item scale developed by Tepper (2000). A sample item is: “My immediate supervisor blames me to save himself/herself embarrassment” ( α = 0.86). Each item was rated on a five-point scale from 1 (never) to 5 (almost always). Adm. Sci. 2022,12, 54 5 of 12 2.2.2. Extra-Role Performance Supervisor-rated nurses’ extra-role performance was measured with an eight-item scale adapted from Williams and Anderson (1991) (French by Nasr et al. 2009) (e.g., “This nurse goes out of way to help new employees”; α = 0.72). Each item was rated on a seven-point scale from 1 (do not agree at all) to 7 (very strongly agree). 2.2.3. Job Emotional Resources A four-item scale adapted from the DISC 2.0 questionnaire (Van de Ven et al. 2008) was used to assess nurses’ emotional resources. A sample item is: “I get emotional support from others (e.g., patients, colleagues, supervisors) when a tough situation occurs at work”; α= 0.84. Each item was rated on a five-point scale from 0 (never) to 6 (almost always). 2.2.4. Job Emotional Demands A four-item scale adapted from the DISC 2.0 questionnaire (Van de Ven et al. 2008) was used to assess nurses’ emotional demands. A sample item is: “I have to do a lot of emotionally draining work”; α = 0.79. Each item was rated on a five-point scale from 0 (never) to 6 (almost always). 2.2.5. Burnout Nurses’ and supervisors’ emotional exhaustion and cynicism were assessed with the Maslach burnout inventory-general survey (MBI-GS) (Schaufeli et al. 1996). Each dimension contains five items, such as “I feel emotionally drained from my work” (emotional exhaustion; α = 0.94 for nurses and α = 0.93 for supervisors) and “I doubt the significance of my work” (cynicism; α = 0.92 for nurses and α = 0.86 for supervisors). Each item was rated on a seven-point scale from 0 (never) to 6 (daily). 2.2.6. Autonomous Motivation Nurses’ autonomous motivation was assessed using two dimensions of the multidimensional work motivation scale initially developed in French (Gagnéet al. 2015): identified regulation (3 items, e.g., “because putting efforts in this job aligns with my personal values”) and intrinsic motivation (3 items, e.g., “because I have fun doing my job”). Nurses rated their key reasons for investing in their job on a seven-point scale from 1 (not at all for this reason) to 7 (exactly for this reason). The Cronbach’s alpha was calculated with the mean subscale scores (α= 0.88). 2.2.7. Control Variables We took supervisors’ burnout (emotional exhaustion and cynicism) as control variables because it has been suggested or showed to account for variance in job attitudes and behaviors, such as motivation and performance. In line with past studies on job performance (Parent-Lamarche et al. 2020,2021), we also adjusted the tested models for nurses’ and supervisors’ sex and age as well as nurses’ work schedule. 2.3. Data Analysis The proposed model was tested following the approach developed by Preacher et al. (2007). The moderated mediation effect corresponds to conditional indirect effects, meaning that the mediation effect is conditioned by the values of at least one moderator (Preacher et al. 2007). Our model explores whether abusive leadership moderates the indirect effect of emotional resources on extra-role performance through autonomous motivation. Our analyses involved several steps. First, we tested a model including the predictors (emotional resources, abusive leadership) and the control variables (nurses’ job demands and emotional exhaustion and cynicism) to estimate the main and interaction effect on nurses’ autonomous motivation and extra-role performance. Second, we looked at the significant variables for autonomous motivation that were indirectly (but significantly) related to extra-role performance via autonomous motivation. In the final step, we verified whether abusive Adm. Sci. 2022,12, 54 6 of 12 leadership moderates the indirect relation between emotional resources and extra-role performance through autonomous motivation. Path analysis was performed using Mplus (Muthén et al. 2017) with robust maximum likelihood estimator (MLR) to estimate all models. The goodness of fit was evaluated using the Tucker–Lewis index (TLI) and comparative fit index (CFI). Values over 0.90 and 0.95 are considered to be indicative of a satisfactory and excellent fit, respectively (Hoyle 1995). A two-tailed probability (p ≤ 0.05) for rejection of the null hypothesis was used to evaluate the significance of the combined contribution of the variables and each individual regression coefficient. 3. Results 3.1. Descriptive Analysis The descriptive statistics and correlations of the measured variables are presented in Table 1. 3.2. Multiple Regression Analyses The main and interaction effects on autonomous motivation and extra-role performance are presented in Table 2. Table 2. Main effects on autonomous motivation and objective job performance. Independent Variables Autonomous Motivation Extra-Role Performance Constant 0.911 4.404 ** AUTONOMOUS MOTIVATION Autonomous motivation 0.358 ** LEADERSHIP Abusive leadership −0.083 −0.626 ** JOB CHARACTERISTICS Emotional resources 0.027 0.094 NURSE BURNOUT Nurse emotional exh −0.028 −0.011 Nurse cynicism −0.435 ** 0.088 INTERACTION EFFECTS Emotional resources–abusive −0.470 ** −0.069 leadership ADJUSTMENTS CFI 1.00 1.00 TLI 1.00 1.00 χ2(df) 123.743 (27) ** 123.743 (27) ** Note A: ** p ≤ 0.01; exh = exhaustion; aut = autonomous. Note B: the following variables were controlled for in all models: nurse sex, nurse age, supervisor sex, supervisor age, supervisor emotional exhaustion, supervisor cynicism, nurse job demands, and nurse work schedule (unstandardized coefficients). Controlling for supervisors’ emotional exhaustion and cynicism and adjusting the model for nurses’ age, sex, and work schedule as well as supervisor’ age and sex, the results show that nurses’ extra-role performance is positively related to autonomous motivation (0.358 **) and negatively related to abusive leadership ( − 0.626 **). Nurses’ cynicism is also negatively related to autonomous motivation ( − 0.435 **). As presented in Table 3, only nurses’ cynicism (and not emotional resources) is indirectly related to extra-role performance through autonomous motivation ( − 0.158 **). These results did not support Hypothesis 1. Adm. Sci. 2022,12, 54 7 of 12 Table 3. Indirect effects on objective job performance. Independent variables Estimate SE LEADERSHIP Abusive leadership −0.030 0.063 JOB CHARACTERISTICS Emotional resources 0.010 0.025 NURSE BURNOUT Nurse emotional exh −0.010 0.025 Nurse cynicism −0.158 ** 0.049 MODERATED MEDIATION Emotional resources–abusive −0.170 * 0.084 leadership interaction Note A: * p ≤ 0.05 and ** p ≤ 0.01; exh = exhaustion; aut = autonomous. Note B: the following variables were controlled for in all models: nurse sex, nurse age, supervisor sex, supervisor age, supervisor emotional exhaustion, supervisor cynicism, nurse emotional demands, and nurse work schedule (unstandardized coefficients). 3.3. Moderated Mediation Analyses The significant interaction between abusive leadership and emotional resources ( − 0.470 **) (see Table 2) indicates that the indirect relation of emotional resources on extra-role performance through autonomous motivation is moderated by abusive leadership ( − 0.170 *) (see Table 3). These results are consistent with Hypothesis 2. Figure 2 shows the indirect relation at low and high levels of abusive leadership. This indicates that the conditional indirect effect of autonomous motivation is stronger and significant when abusive leadership is low, but weaker and significant when abusive leadership is high. Adm. Sci. 2022, 12, x FOR PEER REVIEW 8 of 13 Figure 2. Interaction effect of emotional resources and abusive leadership on nurse autonomous motivation. 4. Discussion In the aim of enriching our understanding of when abusive leadership undermines extra-role performance in nurses, we proposed and tested a moderated mediation model. One noteworthy finding is that abusive leadership qualifies the conditions under which emotional resources facilitate autonomous motivation and contribute to extra-role performance. Although unexpected, we also found that nurses’ cynicism is negatively associated with extra-role performance through autonomous motivation. These results have valuable theoretical and managerial implications. 4.1. Theoretical Contributions Our results highlight the need to rethink and expand our knowledge on the motivational significance of abusive leadership behaviors. While corroborating studies showing that abusive leadership can directly influence in-role (Tariq and Ding 2018) and extra-role (Huang et al. 2019) employee performance, our results emphasize the relevance of paying particular attention to the abusive leadership’s moderating role. By exhibiting abusive behaviors, immediate supervisors are likely to interfere with the motivational process underlying nurses’ discretionary behaviors, as their emotional resources have been shown to foster autonomous motivation and facilitate their extra-role performance. Focusing on work performance, these results contribute to those of Jha (2019), who showed a similar moderated mediation mechanism to explain employee attitudes (e.g., intention to stay with an organization). Our study also contributes evidence in favor of the JD-R model. Although this model explains leadership behaviors as resources for achieving objectives, our results indicate that they should not be considered “simple” resources (Nielsen et al. 2017), but taken in tandem with other resources given their synergistic interaction with emotional resources in the present study. Interestingly, our results reveal that emotional resources are favorable for extra-role performance because they promote nurses’ autonomous motivation, where they value their tasks and enjoy carrying them out. However, these same emotional resources may not be fully beneficial, or even harmful, to their autonomous motivation and discretionary behavior in contexts of high abusive supervision. Consistent with SDT (Deci et al. 2017), basic psychological needs (e.g., competence, autonomy, and relatedness) should be largely suppressed in situations of abusive leadership, which would lead nurses to alternative behaviors, often defensive or self-protective, such as the development of introjected or external regulations. These behaviors prevent them from experiencing a full sense of volition and choice in their actions (Deci et al. 2017). In this sense, emotional resources normally conducive to work (e.g., being able to express emotions, being listened 0 0.5 1 1.5 2 Low emotional resources High emotional resources Autonomous motivation Low abusive leadership High abusive leadership Figure 2. Interaction effect of emotional resources and abusive leadership on nurse autonomous motivation. 4. Discussion In the aim of enriching our understanding of when abusive leadership undermines extra-role performance in nurses, we proposed and tested a moderated mediation model. One noteworthy finding is that abusive leadership qualifies the conditions under which emotional resources facilitate autonomous motivation and contribute to extra-role performance. Although unexpected, we also found that nurses’ cynicism is negatively associated with extra-role performance through autonomous motivation. These results have valuable theoretical and managerial implications. 4.1. Theoretical Contributions Our results highlight the need to rethink and expand our knowledge on the motivational significance of abusive leadership behaviors. While corroborating studies showing that abusive leadership can directly influence in-role (Tariq and Ding 2018) and extra-role Adm. Sci. 2022,12, 54 8 of 12 (Huang et al. 2019) employee performance, our results emphasize the relevance of paying particular attention to the abusive leadership’s moderating role. By exhibiting abusive behaviors, immediate supervisors are likely to interfere with the motivational process underlying nurses’ discretionary behaviors, as their emotional resources have been shown to foster autonomous motivation and facilitate their extra-role performance. Focusing on work performance, these results contribute to those of Jha (2019), who showed a similar moderated mediation mechanism to explain employee attitudes (e.g., intention to stay with an organization). Our study also contributes evidence in favor of the JD-R model. Although this model explains leadership behaviors as resources for achieving objectives, our results indicate that they should not be considered “simple” resources (Nielsen et al. 2017), but taken in tandem with other resources given their synergistic interaction with emotional resources in the present study. Interestingly, our results reveal that emotional resources are favorable for extra-role performance because they promote nurses’ autonomous motivation, where they value their tasks and enjoy carrying them out. However, these same emotional resources may not be fully beneficial, or even harmful, to their autonomous motivation and discretionary behavior in contexts of high abusive supervision. Consistent with SDT (Deci et al. 2017), basic psychological needs (e.g., competence, autonomy, and relatedness) should be largely suppressed in situations of abusive leadership, which would lead nurses to alternative behaviors, often defensive or self-protective, such as the development of introjected or external regulations. These behaviors prevent them from experiencing a full sense of volition and choice in their actions (Deci et al. 2017). In this sense, emotional resources normally conducive to work (e.g., being able to express emotions, being listened to in a challenging situation) could arouse negative feelings, such as guilt, shame, or anxiety in environments of abusive supervision, thus worsening their sense of self-determination. These findings invite additional research on the characteristics of cognitive, emotional, and physical job resources, as they may differentially impact employee motivation and behavior in contexts of abusive supervision. Resources at other levels of the organization such as the psychological safety climate or HRM practices should then be considered. Finally, the current results contribute to the importance of SDT in the workplace (Deci et al. 2017). Although this perspective highlights the relevance of the interpersonal aspect of the work environment, such as the effect of the immediate supervisor’s interpersonal style and leadership practices on employee motivation (Fernet et al. 2015), our study brings new insights into the complex mechanisms that can explain this relation. To our knowledge, this is the first study assessing supervisor-rated extra-role performance. Moreover, our unexpected finding that cynicism plays a role in motivation and extra-role performance reveals the importance of considering employees’ psychological state to explain job outcomes (Bakker and Demerouti 2017). A cynical job attitude of withdrawal and detachment would not offer the motivation necessary to drive their discretionary behaviors. 4.2. Limitations and Future Directions This study contains limitations that pave the way to future research directions. First, self-report measures were used for most of the variables, except for extra-role performance. Such measures can increase the risk of social desirability and biased responses. Second, similar to prior studies (Tariq and Ding 2018), the abusive leadership of the immediate supervisor was assessed only in terms of the frequency of the perceived behaviors. It is possible that frequency and intensity do not have the same impact on employee functioning. As mentioned by Tariq and Ding (2018), some workers may react more quickly to silent treatment, while others may react to an explosive incident of verbal abuse. Thus, the effect of the intensity of abusive relationship should be considered in future studies. Third, the crosssectional results refrain from drawing conclusions about the directionality of associations, so we cannot rule out the possibility of inverse or reciprocal relations between the study variables. For example, it is conceivable that nurses with less autonomous motivation perceive their work environment more negatively and present higher burnout scores. It is