Abstract
Research on self-compassion across various disciplines has consistently demonstrated numerous self-regulatory benefits associated with the construct. Despite the increasing interest, theory-driven research on self-compassion in the workplace has only begun to emerge recently. In the present research, we introduce the construct of work self-compassion (WSC). Building on Neff’s definition of self-compassion, we submit that WSC is comprised of work self-kindness, common work challenges, and work-specific mindfulness. Across two studies, we develop a scale to measure WSC and test its place within the larger nomological network of organizational constructs. Specifically, by integrating COR theory with prior research on self-compassion, in Study 1, we test the incremental validity of WSC beyond general self-compassion in predicting job performance. In Study 2, we further demonstrate that WSC is an important mediating mechanism that bridges the association between honesty-humility, namely the H-factor, and job performance and burnout, respectively. We discuss theoretical and practical implications of our findings and conclude with limitations and future research directions.
Keywords
In today’s ever-changing business environment, employees are exposed to various day-to-day challenges. For example, a recent report from Gallup (2022) indicated that employees around the globe have experienced all-time high stress level, with approximately 44% among 68,000 respondents experiencing high daily stress at work. Despite the increasingly uncertain nature of modern jobs, employees are nonetheless encouraged to, and are further expected to, carry out their optimal job performance. We contend that these conditions are not a prescription for enhanced performance and well-being, but rather a recipe for burnout and hampered performance, which ultimately has negative implications for organizations in the form of impaired employee commitment, turnover, and organizational performance (eg Ostroff, 1992; Taris & Schreurs, 2009). To combat this issue, it is timelier than ever before for organizational scholars to explore the inward-facing, rather than outward-facing, antecedents of employee job performance and psychological well-being. Indeed, in the midst of turbulence, some employees seem to have better positioned themselves against burnout and thrive by exhibiting high quality performance. Therefore, our driving research question is: What person-driven factor helps employees navigate their path to optimal functioning at work?
Building on Conservation of Resources theory (COR; Hobfoll, 1989, 2001), we posit that self-compassion is an important resource that enables desirable work outcomes. Drawn from Buddhist traditions, self-compassion broadly entails (a) being kind and understanding, rather than being judgmental, towards one’s own suffering, (b) acknowledging that pain and failures are part of a larger human experience, and c) being mindful, rather than over-identifying, about the painful feelings one experiences (Barnard & Curry, 2011; Neff, 2003a). While the construct has been widely examined in broader positive psychology space, to date, self-compassion research in organizational literature remains relatively scant (Schabram & Heng, 2022). Furthermore, initial evidence of self-compassion in the workplace has primarily focused on attitudes and well-being domains (eg Babenko, Mosewich, Lee, & Koppula, 2019; Schabram & Heng, 2022; Shepherd & Cardon, 2009; van der Meulen, Valentin, Bogels, & de Bruin, 2021), with its effects on performance remaining underexplored (exception: Reizer, 2019). Therefore, in the current research, we seek to examine the effects of self-compassion on job performance, in addition to burnout.
By integrating extant research on self-compassion and Conservation of Resources (COR) theory, we propose that self-compassion is a competitive personal resource that fuels performance and buffers against burnout. Specifically, self-compassion is a restorative approach to enhancing one’s self-regulatory and coping capacity, as well as to combating against depletion (Allen & Leary, 2010; Leary, Tate, Adams, Allen, & Hancock, 2007; Terry & Leary, 2011). For example, Allen and Leary (2010) theorized self-compassion as a cognitive coping resource, such that self-compassionate individuals engage in cognitive restructuring to “construe negative events in less dire terms.” (p. 109). Applied to the workplace, self-compassion is likely to protect employees against small and big work-related challenges, such as project failures, and drive their motivation to learn from the failure rather than to ponder upon it (Shepherd & Cardon, 2009).
Further building on one of the core tenets of COR theory, which suggests that resources travel in packs rather than individually (ie resource caravans; Hobfoll, Halbesleben, Neveu, & Westman, 2018, p. 106), we contend that self-compassion and honesty-humility, namely the H-factor, operate in a cumulative manner. Conceptually, some of the key strengths of individuals high in H-factor are that they remain truthful, fair, and impartial towards others (Ashton, Lee, & de Vries, 2014). We posit that the tendency to be truthful and fair towards others and oneself operate collectively, and the chain-like association between H-factor and self-compassion, in turn, prevents employees against depletion and fosters job performance.
We test our hypotheses across two studies. In Study 1, we refine self-compassion for organizational research and develop a short, work-specific scale to measure Work Self-Compassion (WSC). Drawing from theories suggesting that context-specific measures show superior predictive validity than non-contextualized measures (eg Allport, 1961; Schmit, Ryan, Stierwalt, & Powell, 1995), we validate the new WSC scale and test its incremental validity beyond self-compassion in predicting job performance. In Study 2, we further expand our research model and test the indirect association between H-factor and work outcomes, via WSC. Findings from our studies consistently indicate that WSC offers practical utility for predicting job performance and burnout.
Our research offers several contributions to the current self-compassion literature. First, while self-compassion research in organizations has rapidly emerged in recent few years, prior studies have relied on measures that reflect self-compassion in broader life domains, rather than in workspace (eg Neff, 2003b). Despite the increasing interest, to our knowledge, there have been few notable attempts to refine self-compassion exhibited at work. Meanwhile, theories suggest that domain-specific traits have greater influence on outcomes within a specific context (Allport, 1961; Wright & Mischel, 1987) and researchers have previously shown that this, indeed, is the case (eg Pathki, Kluemper, Meuser, & McLarty, 2022; Shaffer & Postlethwaite, 2012; Wallace & Chen, 2005). To address this, therefore, we introduce the WSC scale that more effectively captures work-specific self-compassion by making work context more salient to respondents. Further, we clarify self-compassion’s place in the larger organizational nomological network and show that WSC scale has greater utility in predicting job-related outcomes beyond self-compassion scale. Second, we help advance the literature by showing the conceptual and empirical link between H-factor and WSC. Specifically, we do so by expanding the current understanding of H-factor in relation to the ‘self,’ not just related to others, in work contexts. Recently, Dodson and Heng (2022:186) suggested that “identifying the dynamic features of self-compassion becomes a vital next step” and that “processual conceptualization of self-compassion also allows for greater precision in our examination of antecedents.” We argue that we addressed this point by examining the antecedent role of H-factor in predicting WSC. Third, as an extension of our second contribution, we offer initial evidence suggesting that WSC is a key mediating mechanism that explains how H-factor impacts work-related outcomes and present a new research avenue in a relatively underexplored area. Finally, by demonstrating the chain-like association between H-factor and WSC under the COR lens, we expand the understanding of “resource caravan” (Hobfoll, 2001, p. 349), such that inherently other-oriented and self-oriented individual traits can collectively foster positive work-related outcomes.
Theoretical Background
Prior Research on Self-Compassion across Various Disciplines.
Note. The list of studies is not exhaustive.
Given that much effort has been undertaken to validate self-compassion in various disciplines (eg Bluth & Blaton, 2014; Cunha, Xavier, & Castilho, 2016; Leary et al., 2007; Neff & McGehee, 2010), and given its remarkable potential for understanding human behaviors, the relative lack of self-compassion research in organizational contexts is rather alarming. More often, organizational scholars have focused on one of its subdimensions, mindfulness (eg Good et al., 2016) or compassion in organizations (eg Dutton, Workman, & Hardin, 2014; Kanov et al., 2004; Rynes, Bartunek, Dutton, & Margolis, 2012). As a result, most prior organizational research has examined compassion as part of organizational culture and the “caring social network” that activates and mobilizes the expression of compassion by and between colleagues, or managers to their subordinates (Atkins & Parker, 2012; Dutton, Lilius, & Kanov, 2007; Dutton, Frost, Worline, Lilius, & Kanov, 2002). Only recently, organizational scholars examined the role of self-compassion, primarily within the well-being domain (also see Table 1).
Nonetheless, initial evidence warrants a critical need to further investigate what it means to be self-compassionate at work, particularly given prevalent, and sometimes unavoidable, challenges that employees (Babenko et al., 2019; Shepherd & Cardon, 2009; van der Meulen et al., 2021) and leaders (Lanaj et al., 2022) experience (for a review, see Dodson & Heng, 2022). This offers an opportunity of research concerning 1) the conceptualization of self-compassion in the workplace and 2) potentially surfacing other relevant variables that might collectively contribute to important work-related outcomes, such as job performance and burnout, alongside work-specific self-compassion. To this end, we conduct two studies. In Study 1, we examine how work self-compassion (WSC) relates to general self-compassion and test its incremental validity beyond self-compassion in predicting job performance. In Study 2, we examine how WSC fits into the larger nomological network by testing the mediating properties of WSC between H-factor and job performance, and H-factor and burnout, respectively.
Study 1: Work Self-Compassion and Job Performance
The conditional model of dispositions suggests that traits are situation-dependent, such that the effects of traits may be more pronounced in some environmental contexts than others (Wright & Mischel, 1987). Indeed, researchers have previously discussed that domain-specific trait measurements tend to show greater influence over life-domain measurements (Allport, 1961; 1966; Schmit et al., 1995), and such an effect has been empirically and meta-analytically shown by several scholars (eg Bing, Whanger, Davison, & VanHook, 2004; Button, Mathieu, & Zajac, 1996; Chen, Goddard, & Casper, 2004; Pathki et al., 2022; Shaffer & Postlethwaite, 2012; Thatcher, Wright, Sun, Zagenczyk, & Klein, 2018; Wallace & Chen, 2005).
Applied to the current study, we posit that WSC will show greater predictive validity beyond non-contextualized self-compassion measurement. For example, employees who are typically self-compassionate in the workplace may not exhibit self-compassionate tendencies at home to the same extent. Conversely, organizational conditions may trigger employees to be less self-compassionate at work. For example, workplace has many unique factors that challenge employees in which they are demanded to manage professional successes, failures, and frustrations (Kyeong, 2013). Given that trait measurements can be susceptible to cultural factors, such as demand characteristics (Heller, Watson, Komar, Min, & Perunovic, 2007), employees may become less self-compassionate due to organizational norms associated with hard work (eg not taking vacations and suppressing mental well-being concerns). Under such an organizational culture, employees may refrain from having self-compassionate thoughts and feelings to avoid being labeled as a problem employee who is deviant from the norm, while they may also rely on self-compassion as a personal resource to cope against these demands. Therefore, while we acknowledge that individuals who are in the habit of practicing self-compassion in their personal lives are naturally predisposed to be self-compassionate at work, we posit that a domain-specific measure is required to understand the effects of self-compassion in organizations more precisely.
WSC is a manifestation and expression of self-compassion in the workplace. Mirroring self-compassion, WSC is comprised of three components: (a) work self-kindness (WSK), (b) common work challenges (CWC), and (c) work-specific mindfulness (WM). WSK is defined as handing oneself kindness when one faces work challenges, such as failing to meet goals and expectations. As such, employees high in WSK tend to extend oneself unconditional empathy regarding their own feelings, thoughts, and impulses. CWC pertains to acknowledging that the currently experienced work-induced pain is a common phenomenon observed across numerous employees, not just the self, thus employees high in CWC tend to be tolerant towards one’s work-related shortcomings. Finally, WM refers to observing, but not necessarily reacting to, thoughts and emotions experienced upon failing to execute desirable levels of performance at work. Considering the theoretical and construct domain overlap between self-compassion and WSC, we expect that the two are highly and positively related. Thus,
WSC is positively related to self-compassion. From a COR standpoint, WSC is a relatively reliable resource that facilitates employees’ future gains. We posit that WSC enables enhanced job performance. As briefly stated above, most prior self-compassion research in organizational contexts have primarily focused on employee well-being and attitude domains (eg Barnard & Curry, 2012; Gerber & Anaki, 2021; Prudenzi et al., 2021; Schbram & Heng, 2022; Vaillancourt & Wasylkiw, 2020). Nonetheless, initial empirical evidence suggests a positive link between self-compassion and job performance. Most notably, in a study of service-sector employees, Reizer (2019) demonstrated that self-compassion is positively associated with performance and citizenship behavior. Other findings further provide ancillary support for this relationship. For example, in occupational contexts, self-compassion was positively related to work engagement (Babenko et al., 2019) and innovation (Lefebvre, Montani, Courcy, & Dagenais-Desmarais, 2021), which are often examined as significant predictors of performance outcomes. Evidence associated with self-regulation provides further support. For example, researchers have discussed that when employees experience significant challenges at work, such as project failures, they are more likely to learn from the bitter experience and advance from the current state as they are compassionate towards themselves (Shepherd & Cardon, 2009). More broadly, Neff and Knox (2017) noted that self-compassion promotes individuals’ tendencies to “find hope and inner strength when faced with the difficulties of life,” (p. 2) which helps them overcome fear of failure. Empirical findings from educational contexts lend further support for the self-regulatory benefits of practicing self-compassion, in terms of motivation (eg Breines & Chen, 2012; Chamberlain & Haaga, 2001; Neff et al., 2005) and protection against demands (eg Leary et al., 2007). Another stream of researchers (eg Barnard & Curry, 2011; Lefebvre, Montani, & Francois-Courcy, 2020) theorized a positive link between self-compassion and resilience, which is commonly described as the capacity to “bounce back” from hardships, suggesting that self-compassionate individuals tend to have “resilient self-appraisals allowing for a more accurate rating of their abilities” (Barnard & Curry, 2011, p. 297). Supporting this, self-compassionate workers in healthcare contexts exhibited improved resilience (eg Kemper, Mo, & Khayat, 2015; Franco & Christie, 2021). Put together, self-compassion preserves individuals’ cognitive resources through improved self-regulation, which may promote emotional (eg Breen, Kashdan, Lenser, & Fincham, 2010; Neff, Rude, & Kirkpatrick, 2007; Neff & Germer, 2017), attitudinal (eg Hefferman, Griffin, McNulty, & Fitzpatrick, 2010), adaptive (eg Barnarad & Curry, 2011; Kelly, Zuroff, Foa, & Gilbert, 2010), and cognitive (eg Barnard & Curry, 2011; Lefebvre et al., 2020; Shepherd & Cardon, 2009) benefits. Therefore, we posit that WSC is an important nutriment for enhanced job performance. Employees who practice WSC are better positioned to take a balanced approach to work stressors, such that they respond positively to work setbacks and move forward, rather than being stagnated. Thus, we predict the following:
WSC is positively associated with job performance. Building on theories (eg Allport, 1961; Wright & Mischel, 1987) and evidence (eg Bing et al., 2004; Schmit et al., 1995; Shaffer & Postlethwaite, 2012) suggesting greater validity of context-specific trait measures, as opposed to non-contextualized measures, we narrow the bandwidth of self-compassion to be specific to the workplace. Specifically, we predict that WSC will have an incremental validity in predicting performance beyond self-compassion. Thus,
WSC explains additional variance in job performance beyond general self-compassion.
Method
Sample and Procedure
We collected data from Amazon’s Mechanical Turk (MTurk). MTurk is a widely used data collection platform that enables organizational researchers to access participants from diverse backgrounds (Buhrmester, Talaifer, & Gosling, 2018). Recent findings suggest that MTurk participants are reasonably equivalent to participants recruited via traditional methods (Paolacci & Chandler, 2014), and MTurk data is valid and reliable (Sheehan, 2018).
A final sample of 461 participants self-reported their self-compassion, WSC, and job performance. We also measured control variables that are known to influence job performance: gender, age, race, job status, and job function. Ages ranged from 19 to 75 with the mean age of 38 (SD = 12.20). Approximately 59% were female. Majority of participants were Caucasian (77%), and the rest were in the following order: Hispanic (6%), African American (10%), Asian (5%), Native American (1%), and other (1%). Most of the participants were employed on a full-time basis (76%), followed by part-time (21%) and unemployed (3%). For those participants who were employed, job functions broadly varied, including: Services, Sales, Information Technology, General Administrative, Executive/Upper Management, Research and Development, Engineering, Human Resources, Accounting/Finance, and Operations.
Measures
Self-Compassion
Self-Compassion was measured using Neff’s (2003b) 26-item Self-Compassion Scale. Using a 5-point Likert type scale (1 = almost never to 5 = almost always), participants were asked to indicate how often they behaved in the manner stated in the question. An example item is: When I’m feeling down, I tend to obsess and fixate on everything that’s wrong.
Work Self-Compassion
As noted earlier, we created and validated a measure that captures work-specific self-compassion. Researchers have previously documented the conceptual and empirical distinctions between self-compassion and other self-relevant constructs, such as self-esteem (eg Barnard & Curry, 2011; Neff & Vonk, 2009; Zhang et al., 2020), mindfulness (Neff & Dahm, 2015; van der Meulen et al., 2021), basic psychological need satisfaction (Neff, 2003b), and self-forgiveness (Dodson & Heng, 2022). Therefore, we proceeded with item generation for the new WSC scale.
Standardized loadings from Work Self-Compassion Scale.
Job Performance
Job performance was measured using 3-items from Welbourne, Johnson, & Erez’s (1998) task performance measure. Participants provided ratings they believed their supervisors would provide them regarding quantity, quality, and accuracy of work using a 5-point Likert type scale (1 = needs much improvement to 5 = excellent).
Results
Prior to testing our hypotheses, we verified the validity of the WSC measure and reduced its length to create a more parsimonious measure. Confirmatory factor analysis (CFA) was conducted on the 30 items written by the authors using MPlus, because we had strong theory driving our development and structure of WSC. The initial CFA that was conducted on the 30-item WSC was a 3-factor model with a single higher order factor of self-compassion. This model had poor fit. Examination of the modification indices and the factor loadings suggested that we could reduce our scale substantially to gain a valid and parsimonious scale. Consequently, we reduced our scale to 10 items, 3 for WSK, 3 for CWC, and 4 for WM. Using the shortened scale, the model was retested. As shown in Table 2, all standardized factor loadings were above .60 and the fit of the model was acceptable. These results support the notion that the construct of WSC consists of WSK, WM, CWC (ie first order factors) and all comprise a higher factor of WSC (ie second order factor). In addition, the internal consistency of each dimension and overall construct exceeded .70. Mirroring Neff’s (2003b) original conceptualization and operationalization of general self-compassion construct, the current measure indicated a higher order construct of WSC.
Means, SDs, and Correlations for all variables in Study 1.
Note. Correlations .09 or greater are significant at the .05 level; n = 461. Reliability estimates are presented on the diagonal.
Hierarchical regression predicting job performance in Study 1.
Note. *p < .05. n = 461.
We further conducted dominance analysis (Azen & Budescu, 2003) and found that WSC dominated the prediction of job performance with a relative importance of 77.04%, whereas self-compassion accounted for 22.96%. Together, findings from Study 1 provide initial evidence that WSC is indeed a more effective predictor of job performance than general self-compassion scale.
Study 2: Mediating Role of WSC between H-factor and Work Outcomes
Though Study 1 set the stage for the validity and practicality of WSC, it did not address the question of where WSC might fit within the larger nomological network in predicting job performance. In addition, though recent findings have consistently suggested a positive link between self-compassion and employee well-being, an empirical investigation is required to understand whether such a relational pattern is exhibited for WSC. Therefore, as an extension of Study 1, we designed Study 2 to evaluate how WSC impacts employee job performance and burnout, in conjunction with H-factor.
Often considered the sixth factor of the HEXACO model (Ashton & Lee, 2007), H-factor extends the five-factor model of personality (eg Costa & McCrae, 1992; Goldberg, 1990) in meaningful ways. For example, by broadening the lexical studies from English-only to various other languages, H-factor has repeatedly emerged as a reshuffling of variance from the Big Five (eg Saucier, 2009). Further, recent evidence has consistently indicated its validity beyond the Big Five in predicting several important organizational outcomes (eg Johnson, Rowatt, & Petrini, 2011; Pletzer, Bentvelzen, Oostrom, & de Vries, 2019).
Broadly defined in lexical terms such as being sincere, honest, fair-minded, and unassuming, H-factor encompasses four facets; that is, sincerity, fairness, greed-avoidance, and modesty (Ashton & Lee, 2007). Therefore, an individual who is high in H-factor is likely to acknowledge and appreciate one’s and others’ strengths and weaknesses, seek new information for growth and development, and be open and cooperative in social settings, rather than being entitled or manipulative (Ashton & Lee, 2007). In this way, though H-factor is commonly examined in relation to interpersonal domains (eg Barford, Zhao, & Smillie, 2015; Wang, Dunlop, Parker, Griffin, & Gachunga, 2022), it is both an intrapersonal and social construct. Given the theoretical thrust behind H-factor, it is a robust individual difference that may predispose some individuals to be more self-compassionate at work, which then elicits further benefits including enhanced performance and reduced burnout. Under the premises that COR theory offers, we posit that H-factor is an important precursor of WSC and that the two resources collectively work to enhance positive work-related outcomes.
H-factor → Job Performance and Burnout
In recent decades, H-factor has received an increasing attention as a unique predictor of job performance (eg Johnson et al., 2011; Lee, Berry, & Gonzalez-Mule, 2019; Oh et al., 2014). For example, researchers have conceptually laid out the link between H-factor and individual- and unit-level performance (Owens, Rowatt, & Wilkins, 2011), and organizational performance (Vera & Rodriguez-Lopez, 2004), respectively. Relevant to the current study, at the individual level, findings have consistently indicated that H-factor is predictive of contextual performance (Owens, Johnson, & Mitchell, 2013), counterproductive work behavior (CWB) (Marcus, Lee, & Ashton, 2007), and workplace deviance (Pletzer et al., 2019) and delinquency (Johnson et al., 2011; Lee, Ashton, & de Vries, 2005). Meta-analytic findings have further supported the associations between H-factor and core performance constructs (ie task performance, citizenship behavior, and CWB) (Lee et al., 2019), and the incremental validity of H-factor in predicting contextual performance beyond various other individual differences (Oh et al., 2014). While the strengths of these associations varied, evidence consistently suggests that H-factor is an important predictor of job performance.
As opposed to behavioral domains, the link between H-factor and well-being are relatively less explored. Nonetheless, extant literature suggests that honest individuals tend to report increased psychological well-being and are less likely to distort self-views (Colvin, Block, & Funder, 1995; Fredrickson, 2001; Vaillant, 1992). As such, employees who are relatively low, as opposed to high, in H-factor experienced higher exhaustion in response to job demands (Ceschi, Sartori, Dickert, & Costantini, 2016). We expect to replicate these findings as an important first step in developing our research model. Further, we predict that H-factor is negatively associated with burnout, which is defined as “prolonged response to chronic emotional and interpersonal stressors on the job” (Maslach, Schaufeli, & Leiter, 2001, p. 397). A strong prosocial orientation associated with H-factor is likely to manifest itself at work and, theoretically, this has been noted as the primary driver for reduced burnout (eg Hilbig & Zettle, 2009; Marcus et al., 2007) and enhanced performance (eg Grant & Wrzesniewski, 2010). In sum, employees high in H-factor are better equipped to recognize their own strengths and weaknesses in broader organizational network and can effectively allocate resources available for them (Deffler, Leary, & Hoyle, 2016; Hobfoll, 1989; Owens, Baker, Sumpter, & Cameron, 2016), which ultimately promotes growth and learning.
H-factor is positively associated with job performance.
H-factor is negatively associated with burnout.
H-factor → WSC
As COR theory suggests, resources often travel in packs rather than operate individually (Hobfoll et al., 2018). Together, the “resource caravan” (Hobfoll et al., 2018, p. 106) fosters future gains. Related to the current study, researchers have recently noted that certain personality traits are likely to precede self-compassionate responses (Dodson & Heng, 2022). Considering that individuals high in H-factor value honesty, fairness, and sincerity, we posit that this typically outward-facing tendency will fuel a subsequent resource that operates in a similar manner but directed towards oneself. In a work environment, specifically, possessing an enduring personality that promotes willingness to have fair perceptions of colleagues is likely to encourage employees to have a fair view of themselves. As such, H-factor functions as a cornerstone to being willing to view oneself accurately, to not distort one’s current situation in a particularly negatively biased way, and to focus on the current state of mind (Exline & Geyer, 2004; Exline & Hill, 2012; Davis et al., 2011). While not extensively examined, some initial empirical evidence supports this view (eg Pyszkowska, 2020; Thurackal, Coveleyn, & Dezutter, 2016). Thus,
H-factor is positively associated with WSC.
WSC → Burnout
In high pressure and service work contexts, such as in a healthcare setting, employees may be at risks of various dysfunctional stress responses including compassion fatigue, which can lead employees to experience physical, cognitive, and emotional exhaustion (Germer & Neff, 2013; Hashem & Zeinoun, 2020; Leiberg, Klimecki, & Singer, 2011). In such situations, being able to acknowledge and forgive oneself buffers against resource-depleting mechanisms (eg compassion fatigue, anxiety, off-tasking, stress; Neff et al., 2007; Thurackal et al., 2016), and allows individuals to preserve more resources to focus on personal growth and learning. Physio-psychologically, self-compassion allows individuals to self-soothe, reduce stress, and be better prepared for what future holds, which results in reduced cortisol levels and increased heart-rate variability (Ma, Liu, Bao, & Gu, 2022; Neff & Germer, 2017; Rockliff, Gilbert, McEwan, Lightman, & Glover, 2008). This buffering effect enhances cognitive resources through improved self-regulation, such as a positive mindset (Neff & Germer, 2017) and resilience (Neff & McGehee, 2010), which are particularly relevant for high-pressure jobs (Alliger, Cerasoli, Tannenbaum, & Vessey, 2015; Harker, Pidgeon, Klaassen, & King, 2016; Kossek & Perrigino, 2016).
Particularly related to burnout, employees who are resourceful with self-compassion are less likely to be vulnerable against emotional exhaustion, cynicism, and inefficacy (Maslach & Jackson, 1981a; Maslach, Schaufeli, & Leiter, 2001). In fact, the elements of burnout seem to be antithetical to the dimensions of WSC, such that “self-kindness” discourages negative evaluations of oneself, “common work challenges” reduces cynicism toward work and colleagues, and “work mindfulness” prevents distancing oneself emotionally and cognitively from work. Such a protective effect of self-compassion has been consistently demonstrated in various occupational contexts (eg Barnard & Curry, 2012; Dev, Fernando III, Lim, & Consedine, 2018; Schabram & Heng, 2022; Vaillancourt & Wasylkiw, 2020), and we anticipate that WSC will help alleviate burnout. Additionally, as hypothesized in Study 1, we expect that WSC is positively predictive of job performance.
WSC is negatively associated with burnout.
H-factor → WSC → Work Outcomes
As stated above, under the premise of COR theory, we posit that a personality that is tailored to seek a balanced understanding in others travel with the tendency to do so towards oneself. Together, these resources are likely to protect employees against burnout and help enhance job performance as those high in H-factor and WSC are better able to reserve their regulatory resources. Provided the theoretical and empirical link between these variables, we posit that H-factor promotes WSC and, in turn, job performance and burnout, respectively.
WSC mediates (a) the positive association between H-factor and job performance and (b) the negative association between H-factor and burnout.
Method
Sample and Procedure
Participants were healthcare professionals solicited from a major safety net medical center. Among 1000 individuals who were initially recruited, 723 participants completed the survey (72.3% response rate). All participants were matched to their immediate supervisor, and we sent an email containing a link to the survey to complete performance evaluation of their employees in terms of their service and citizenship. Upon receipt of supervisors’ completed survey, we further obtained employee performance data (ie competencies and goal attainment) from the organization’s PeopleSoft system. After matching employee and supervisor surveys, we were left with 682 complete data points for analysis.
Majority of the participants were female (84.3%). Approximately 26.8% were in the age group of 30 to 39, followed by 40–49 (24.9%) and 50–59 (22.0%). African American (53.5%) and Caucasian (31.4%) comprised most of the participants’ race. Majority of the participants were employed on a full-time basis (73.4%), followed by part-time (18.8%). In terms of job functions, most participants were nurses (42.7%). Regardless of the job currently held, the majority were individual contributors as opposed to those in leadership or management roles (64.1%). Finally, the largest tenure group were from 0-4 years in the current job (37.5%), followed by 10–14 years (20.5%) and 5–9 years (19.9%).
Measures
H-factor
H-factor was measured using 16 items from the 96 item HEXACO scale from AOE Science, a proprietary measure used in personnel selection. Using a 5-point Likert type scale (1 = strongly disagree to 5 = strongly agree), participants were asked to indicate the extent to which they agreed or disagreed with each statement. An example item is: I am a sincere person to those that I work with.
WSC
WSC was measured using the 10-item WSC scale developed and validated in Study 1. Participants were asked to record their responses using a 5-point Likert type scale (1 = almost never to 5 = almost always).
Burnout
Burnout was measured using the 22-item Maslach Burnout Inventory (MBI; Maslach & Jackson, 1981b). Using a 7-point Likert type scale (0 = never to 6 = everyday), participants were asked to rate each statement about personal feelings or attitudes. An example item is: I feel emotionally drained from my work.
Work Performance
Managers (N = 7) were interviewed to determine the most critical role for clinical care workers (ie nurses, physicians, technicians, and nursing assistants) and it was confirmed that employees have three primary performance roles: task performance, citizenship behavior, and customer service. Task performance was measured using the participants’ score in the organization’s annual performance evaluation, which included performance ratings based on competencies and goal attainment on a 5-point Liker type scale (1 = far below expectations to 5 = far exceeds expectations). Citizenship performance was measured using Welborune et al.’s (1998) four items. Participants’ supervisors rated participants’ citizenship performance (1 = needs much improvement to 5 = excellent). Finally, customer service performance was measured using Chen & Klimoski’s (2003) four items. As with citizenship performance, participants’ supervisors were asked to rate participants’ customer service performance (1 = needs much improvement to 5 = excellent). An example item is: The employee interacts professionally with customers/patients. We aggregated three performance measures in our modeling because (1) our SMEs suggested employee performance in this context is a combination of service, citizenship, competencies, and goal achievement, and (2) the very high correlations among our performance variables suggested a single construct.
Control Variables
Because conscientiousness is a well-examined predictor of job performance (eg Barrick & Mount, 1991), we decided to control for it. Conscientiousness was measured using 16 items of the 96 item HEXACO scale from AOE Science (1 = strongly disagree to 5 = strongly agree). We also controlled for race, age, job function, employment status, job level, and gender.
Results
Means, SDs, and Correlations for all variables in Study 2.
Note. Correlations .08 or greater are significant at the .05 level; n = 682. Reliability estimates are presented on the diagonal.
At the bivariate level, we found initial support for Hypotheses 4 and 5, such that H-factor was significantly and positively related to job performance (r = .21, p < .05) and negatively related to burnout (r = −.15, p < .05). H-factor was also positively related to WSC at the bivariate level (r = .22, p < .05), providing support for Hypothesis 6. Hypothesis 2 was, once again, supported, albeit with other source ratings of job performance, such that the correlations between WSC and performance variables were as follows: competencies (r = .25, p < .05), goal achievement (r = .22, p < .05), service performance, (r = .29, p < .05), citizenship behaviors (r = .21, p < .05), and overall job performance (r = .27, p < .05). WSC was also negatively related to burnout (r = −.16, p < .05), supporting Hypothesis 7.
Regression and indirect tests predicting Overall Work Performance in Study 2.
Note. *p < .05. n = 682.
Regression and indirect tests predicting Burnout in Study 2.
Note. *p < .05. n = 682.
Indirect effects were tested using PROCESS Macro (Hayes, 2017). We found an indirect association between H-factor and job performance via WSC (indirect effect = .09, 95% CI = .05 to .14). The direct effect of H-factor on job performance was .28 (95% CI = .15 to .41). Similarly, we found an indirect association between H-factor and burnout via WSC (indirect effect = −.05, 95% CI = −.09 to −.02). The direct effect of H-factor on burnout was −.21 (95% CI = −.35 to −.07). Together, these results suggest that WSC mediates the effect of H-factor on job performance and burnout, but partially. Therefore, while we found support for Hypotheses 8 (a) and 8 (b), there is statistical room for other mediators to potentially explain these relationships.
Finally, as with Study 1, we conducted dominance analysis and found that for predicting job performance, WSC had relative importance of 51.14%, followed by H-factor (27.61%) and conscientiousness (21.24%). For predicting burnout, WSC had a relative importance of 52.85%, followed by H-factor (43.09%) and conscientiousness (4.07%).
Discussion
Interest in self-compassion has rapidly grown over the past few decades in positive psychology space, yet only recently have organizational scholars conducted theory-driven research on self-compassion in the workplace. To contribute to the growing literature, the primary purpose of this research was to “surface an unknown work construct,” namely WSC. Our results from confirmatory factor analyses suggested that WSC has a strong construct validity. In addition, in line with the principles of COR theory (Hobfoll, 1989; 2001; Hobfoll et al., 2018), our findings suggested that WSC predicts job performance, and it explains additional variance in job performance beyond the original Self-Compassion Scale (SCS) developed by Neff (2003b). Findings from Study 2 further clarified how WSC fits within the broader nomological network in workspace, such that WSC mediates the association between H-factor and work-related outcomes. As predicted, our findings indicated that both WSC and H-factor serve individually as personal resources, and that the two could work in tandem to function as a “resource caravan” (Hobfoll et al., 2018, p. 106) to prevent burnout and enhance overall job performance. Taken together, our findings from two studies consistently suggest that WSC is a psychometrically sound construct, and it possesses predictive utility in the workplace beyond the effect of non-contextualized self-compassion.
Theoretical Contributions
One important contribution of this research is that we highlighted the need to focus on context when studying workplace human behaviors. Consistent with prior work suggesting the importance of studying organizational behaviors within their specific contexts (Allport, 1961; Shaffer & Postelthwaite, 2012; Wallace & Chen, 2005), we refined the definition of WSC and introduced a measure that more precisely conveys work context to respondents. By conceptualizing WSC as an important personal resource, our findings across two studies consistently demonstrated that WSC explains incremental variance in job performance and burnout beyond the non-contextualized measure of self-compassion.
In addition, our findings contribute to the current literature by examining a unique individual difference character as an antecedent of WSC. Consistent with a recent suggestion that theory-driven research can help advance the understanding of antecedents of self-compassion (Dodson & Heng, 2022), we drew from COR theory to demonstrate a process by which H-factor predicts enhanced WSC and, in turn, results in enhanced job performance and reduced burnout. Particularly, given that most prior research has focused on the antecedent role of the Big Five (eg Neff et al., 2007) and emotional intelligence (Di Fabio & Saklofske, 2021), our results shed light on a relatively underexplored relationship, such that H-factor promotes self-compassion.
Our findings further provide empirical evidence that inherently outward- and inward-facing characteristics can operate together to induce positive work-related outcomes. Specifically, while both H-factor and self-compassion entail being truthful and fair (eg Ashton et al., 2014; Neff, 2003b), for H-factor, the target of fair treatment is typically considered other individuals or the situation, whereas for self-compassion, the target is the self. Thus, our findings contribute to the theory by showing that resources that share some overlaps in conceptual terms but directed at different targets can nonetheless have a chain-like relationship to promote desirable outcomes. However, it remains unclear how H-factor promotes WSC, and we call for future studies to explore variables that bridge the two constructs.
Practical Implications
Our research has applied implications for managerial practices. Specifically, we provide empirical evidence that self-compassion predicts job performance and burnout. Relevant to our findings, we recommend managers and individuals in leadership positions to develop strategies to foster employees’ self-compassion. For example, managers may reconsider the concept of self-compassion and reflect on the current organizational climate to ensure that employees are surrounded by work environment that is supportive of practicing self-compassion. Furthermore, given prior research suggesting the developable nature of self-compassion, we contend that organizational leaders could implement relatively quick yet effective intervention programs to help enhance employees’ self-compassion. For example, the Self-Compassion Break (Gerner & Neff, 2013; Neff & Gerner, 2017) is a short pause an individual can provide him/herself where they take a deep breath to recognize that this is a moment of failure (mindfulness), acknowledge that failure is a part of work (common work experience), and repeat: “May I be kind to myself and may I live with ease” (self-kindness). Such a short exercise is shown to increase self-compassionate responses in the face of various challenges, and we envision that doing so may contribute to enhanced self-compassion at work.
Additionally, there are structured programs that could be implemented in a more systematic fashion in organizations. For example, the Mindful Self-Compassion Training Program (Gerner & Neff, 2013; Neff & Gerner, 2017) has been shown to not only be highly effective at increasing self-compassion, but also compassion for others and life satisfaction while decreasing negative consequences, such as anxiety and stress. Similarly, the Mindful2Work program was shown to reduce perceived employee stress and fatigue both directly and indirectly via self-compassion (van der Meulen et al., 2021). Considering that participants in the present research consisted of healthcare workers, who by all accounts work in a highly stress-inducing environment, such workplace self-compassion training programs could be particularly helpful to them and others working in high-stress environment.
Limitations and Future Directions
While this research is a positive step forward, it is not without limitations. First, and likely most obvious, we used an online panel of employees with self-report performance data in Study 1. As a stand-alone study, Study 1 may not meet the rigor to allow for strong inferences. However, our primary goal for Study 1 was to demonstrate that we are “in the ballpark” when it comes to WSC. We attempted to ensure that WSC is 1) important to work performance, 2) highly related to general self-compassion, and 3) more informative to work performance than general self-compassion. As such, this study allowed us to build a valid assessment of WSC and set the basis for a subsequent study. Coupling Study 1 with a more rigorous Study 2 made us more confident in making inferences from Study 1 results.
We also relied on concurrent data, even though it was temporally ordered to predict performance and burnout in Study 2. Therefore, results from this research should be taken with caution, particularly due to issues associated with common method variance (Podsakoff, MacKenzie, & Podsakoff, 2012). However, because our model was designed to examine the effects of a stable personality characteristic (ie H-factor) on a context-specific variable (ie WSC), we envision that the results are relatively reliable. Nonetheless, future studies could implement longitudinal designs to further extrapolate how this process unfolds over time. For example, researchers could examine possible intra-individual variations (eg within-person analysis) to find where WSC stands in broader positive organizational scholarship (POS) literature. In doing so, scholars could further examine how internally and externally focused compassion collectively function. Specifically, it is possible that being compassionate towards oneself might be a critical first step prior to extending genuine compassion for others, and that this relationship may be cyclical such that compassion for oneself feeds compassion for others, or vice-versa, and a positive cyclical relationship may develop over time. As such, examination of positive ripple effects may help advance the current literature. For example, by being self-compassionate, employees may become compassionate toward others and thus, can ward off dysfunctional behaviors such as workplace aggression (eg social undermining, see Duffy, Ganster, & Pagon, 2002; Hershcovis, 2011).
In addition, our WSC measure is developed based on Neff’s (2003b) original SCS. While the original scale has received some criticism associated with its psychometric validity, it is one of the most widely accepted measurement tools that is theoretically and empirically valid (Neff, 2016, 2020; Neff et al., 2019). Further, more recent conceptualizations of self-compassion in work contexts (eg Dodson & Heng, 2022; Gilbert et al., 2017) continue to retain Neff’s (2003b) core components, including mindfulness, common humanity, and self-kindness, as its theoretical basis. Nonetheless, the current WSC measure includes items that may convey double-barreled statements, and we encourage scholars to further test the scale’s validity and reliability across various work samples. Finally, future research could examine potential mediators and moderators of the association between WSC and work-related outcomes. For example, Dodson and Heng (2022) recently called for the processual approaches of examining work-related self-compassion. While we have incorporated the mediating role of self-compassion that links an individual factor and intrapersonal outcomes to show how self-compassion manifests at work, we have not addressed the processual effect of self-compassion on interpersonal outcomes. Future studies can explore this area to further expand the recent conceptualization of self-compassion at work.
Indeed, Whetten (1989) has noted that understanding features including what, how, for whom, and when are some of the key building blocks of theory development. As an initial attempt, in the current research, we found that WSC has positive implications for performance and well-being, yet we have not explored how such effects occur and for whom the effects may stand out. Particularly given that self-compassion is a competitive self-regulatory resource, we suggest researchers to explore how such benefits are enabled. Some scholars have provided initial theoretical and empirical grounds (eg Ma et al., 2022; Schabram & Heng, 2022; Zhang et al., 2020). We argue that it is an excellent starting point and more research on processual mechanism of WSC can help advance the theory.
Conclusion
Today’s workplace can be a demanding environment and inevitably, each employee, at one time or another, experiences various forms of failures, including failing at a task, losing initiatives, ruminating on what should/could have been done, and questioning their capabilities. Advancing forward upon overcoming setbacks can boost positive work-related outcomes, such as enhanced job performance, and can buffer against negative ones, such as reduced burnout. Self-compassion has been credited as one such mechanism (Neff et al., 2017) that affords enhanced self-regulatory benefits. In this research, we introduced WSC, which is a work-context specific construct. Building on the principles of COR theory, we explored the role of WSC as an important personal resource and, in doing so, demonstrated that individual resources, H-factor and WSC, can collectively elicit desirable work-related outcomes.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Associate Editor: Thomas Zagenczyk
