Abstract
Do individuals with callings perform better than those without? Why or why not? There are not clear answers to these questions in the literature. Using a social exchange framework, we posit an intervening process between callings and job performance, focusing on the role of organizational commitment and ideological contract fulfillment – the degree to which organizations live up to their ideological promises. Specifically, individuals with callings will be more committed to their organization, and this commitment, in turn, leads to job performance. Further, this relationship of calling to job performance through commitment will be attenuated when employees perceive under-fulfillment of ideological contract. We found support for these hypotheses across three studies that utilized self- or supervisor-rated performance data from a non-profit organization and multiple for-profit organizations. Interestingly, while the relationship between commitment and performance did depend on fulfillment of the ideological psychological contract, contrary to our prediction, the calling-commitment relationship was not attenuated by under-fulfillment of ideological contract. Our findings deepen our understanding of the organizational implications of callings from a social exchange-based perspective. This study further informs practitioners as to hiring and motivating individuals with a calling.
Keywords
Individuals are increasingly turning to work to find meaningfulness in their lives (Colby et al., 2001; Tams and Marshall, 2011). For individuals pursuing meaningfulness in their employment, working merely for compensation or a promotion is insufficient. Instead, people look for opportunities in order to pursue a calling, the ‘place in the occupational division of labor in society that one feels destined to fill by virtue of particular gifts, talents, and/or idiosyncratic life opportunities’ (Bunderson and Thompson, 2009: 37). In line with this societal trend, the concept of callings has gained a great deal of attention from researchers (e.g., Berg et al., 2010; Dobrow, 2013). Extant research shows that callings have usually been associated with beneficial individual outcomes such as enhanced levels of work and life satisfaction (Duffy et al., 2011; Wrzesniewski et al., 1997), zest (Peterson et al., 2009) and well-being (Conway et al., 2015).
In light of the growing trend of pursuing a calling, it has also become important to understand the organizational implications of hiring individuals with a calling. Yet, the impact of callings on organizational outcomes, such as job performance, remains unclear in this nascent field of research. An association between one’s calling and job performance is commonly anticipated to be positive owing to calling-oriented employees’ devotion to their work (e.g., Miller-Perrin and Mancuso, 2015). Studies utilizing self-reported performance have begun to support this relationship (Lobene and Meade, 2013; Rawat and Nadavulakere, 2015). Nevertheless, other scholars doubt whether calling-oriented employees make superior performers, speculating that these employees may prioritize the objectives mandated by their occupationally-based calling over organizational objectives (Dobrow, 2004) or organizational community involvement (Pratt and Ashforth, 2003). Indeed, a recent study reported that the hypothesized positive association was not found between callings and objective performance (Park et al., 2016), leaving the nature of this focal relationship still uncertain.
Even less is known about the mechanisms through which callings relate to job performance (for an exception, see Park and colleagues (2016), who identified occupational self-efficacy as a mechanism). Further, scholarly efforts to explain the impact of callings on individual outcomes have typically focused on work- or occupation-related mediators such as work meaningfulness or occupational identity (Bunderson and Thompson, 2009; Duffy et al., 2011; Hirschi, 2012), relying on the assumption that callings transcend any organizational context (e.g., Hall and Chandler, 2005). However, research has hinted that there could be additional pathways through which callings relate to workplace behaviors, requiring a broader understanding of the underlying mechanisms. In particular, Cardador and colleagues (2011) questioned the focus on occupational variables in favor of also including organizational characteristics, such as organizational instrumentality. In addition, a recent qualitative study of calling-oriented animal shelter workers illustrated that employees with a similar passion and purpose exhibited vastly different behaviors as they navigated the challenge of callings in their varying organizational contexts (Schabram and Maitlis, 2016), suggesting organizationally embedded mechanisms underlying the relationship between callings and workplace behaviors. As such, there is a need to better understand how the organizational context impacts the calling-performance relationship.
Our objective in this article is thus to explore an intervening process between callings and performance. Drawing from social exchange theory, we argue that the calling-performance relationship is mediated by the quality of the social exchange relationship between employees and their organization, as indicated by affective organizational commitment (Meyer and Allen, 1984). Additionally, we theorize that the mediated relationship is contingent on ideological contract fulfillment (ICF) – the extent to which an employee believes his/her organization fulfills its implicit obligations toward the pursuit of valued principles or causes (Thompson and Bunderson, 2003). Given that the ideological contract is based on the pursuit of values beyond self-interest, its fulfillment likely becomes a critical context that shapes the workplace experiences of calling-oriented employees.
We conducted three independent studies in very different contexts to test our hypotheses. In Study 1, we utilized self-rated in-role performance involving 200 staff members from a US megachurch. The follow-up studies extend Study 1 by utilizing a two-phased survey (Study 2) and supervisor-rated in-role performance (Study 3) in for-profit organizations. Across the three studies, we found that affective commitment serves as a mediator between callings and performance. Further, and perhaps counterintuitively, ICF does not play a moderating role in the relationship between callings and affective commitment. However, in all three samples, it does play an important moderating role in the relationship between affective commitment and performance, indicating that individuals who have callings and high affective commitment may still have low levels of performance if the organization does not live up to its ideological promises. Ultimately, we unpack the calling-job performance relationship by offering an organizationally embedded explanation of how callings relate to in-role performance.
Callings and job performance
According to Bunderson and Thompson (2009), a calling implicates (although it is distinct from) individuals’ strong identification with their line of work, that is, occupation and their occupational community. Thus, those with callings come to accept the occupational beliefs and ideologies about the significance of their work to society. As a result, they derive heightened meaningfulness and motivation to contribute to society from their work. Furthermore, although the authors’ conceptualization of a calling does not rely on religious beliefs, it still resonates with the term’s classical origins, such as ‘God’s commandment to the individual to work for the divine glory’ (Weber, 1930: 106), and emphasizes one’s moral obligation to contribute to society (e.g. prosocial orientation) (Dik and Duffy, 2009). According to the job characteristics model (Hackman and Oldham, 1975) and prosocial motivation theory (Grant and Berg, 2011), work meaningfulness and a prosocial orientation are associated with higher levels of persistence and job performance (e.g. Grant, 2008; Grant et al., 2007; Naff and Crum, 1999). In line with this reasoning, prior studies provide preliminary evidence that calling-oriented employees exhibit superior performance in various occupations (e.g. police officers, teachers, ministers). These individuals are so devoted to their work that they tend to take on higher workloads voluntarily (Chen, 2010), spend their non-paid time on their jobs (Clinton et al., 2017; Serow, 1994), and evaluate their performance highly (Lobene and Meade, 2013). Hence, we propose (Figure 1):

Moderated mediated relationship among calling, ICF, affective commitment and in-role performance.
Hypothesis 1: Calling is positively related to in-role performance.
Affective commitment as a mediator of the calling-performance relationship
Social exchange (Blau, 1964) has been frequently adopted as a theoretical framework to understand the relationship between employees and their organizations (Coyle-Shapiro and Conway, 2005). A social exchange relationship refers to an enduring interaction pattern, entailing unspecified, trustful mutual obligations (Cropanzano and Mitchell, 2005). Generally speaking, the social exchange process is initiated when one party perceives a benefit from the other party, which may be returned in kind based on the norm of reciprocity (Blau, 1964; Gouldner, 1960). A series of successful exchanges contribute to the formation of a quality social exchange relationship, where the parties perceive personal feelings and obligations toward each other (Cropanzano et al., 2017). Owing to this emphasis on the attachments that one party feels toward the exchange partner, past research has utilized affective commitment – an employee’s emotional attachment to his or her organization (Meyer and Allen, 1984) – as an indicator of the quality of the social exchange relationship (Cropanzano et al., 2003; Lavelle et al., 2009).
Prior studies have shown that employees with a calling tend to develop quality social exchange relationships with their organizations (as indicated by affective commitment, for example) (e.g. Duffy et al., 2011). Cardador et al. (2011) drew upon theories of role investment and argued that individuals’ fundamental devotion to their line of work through their callings may be extended to their organizations because they may view their organizations as instrumental in achieving the valued goals associated with their callings. Given the perceived instrumentality of their organization, these employees with a calling become willing to invest in the relationship with it (Labroo and Kim, 2009). This argument is consistent with a core tenet of social exchange theory: individuals form social exchange relationships to the extent that they receive worthwhile benefits (Cropanzano and Mitchell, 2005).
The social exchange relationship with an organization provides a proximal cause for employees’ workplace behaviors (Cropanzano et al., 2017). Indeed, prior research suggests that employees reciprocate a quality social exchange relationship with an organization (as indicated by affective commitment) through their behaviors to help organizations reach their goals (Cropanzano and Mitchell, 2005; Cropanzano et al., 2003). In addition, as affectively committed employees come to internalize the goals of the organization and identify with its well-being (O’Reilly and Chatman, 1986), they are motivated to engage in behaviors that benefit the organization. Meta-analytic studies provide further support for the relationship between affective commitment and performance (Meyer et al., 2002; Riketta, 2002). All together, we suggest:
Hypothesis 2: Affective commitment mediates the positive relationship between calling and in-role performance.
ICF as a key moderator
In order to investigate how the organizational context influences this proposed mediated relationship, we look to research on ICF. The psychological contract refers to ‘individual beliefs in a reciprocal obligation between the individual and the organization’ (Rousseau, 1989: 121). As such, psychological contracts are predicated on a social exchange framework, with an emphasis on promises made and kept by a partner, rather than the benefits provided by the partner (Coyle-Shapiro and Conway, 2005; Wayne et al., 1997). Psychological contract research traditionally focused on economic (e.g. pay, benefit) and socio-emotional (e.g. long-term growth, security) currencies exchanged between employees and their employers (Robinson and Morrison, 1995; Robinson and Rousseau, 1994) until Thompson and Bunderson (2003) proposed ideological currency as a third component of the psychological contract. An ideological contract entails employees’ implicit expectations for their organizations to demonstrate a credible dedication to valued principles that transcend instrumental interests and benefit some constituency (e.g. employees, the community, or broader society) (Bingham et al., 2013). The ideological contract also entails employees’ own obligations to participate in the organizations’ pursuit of these principles (Thompson and Bunderson, 2003). Studies have since demonstrated the distinctiveness of ideological currency from the other two currencies in for-profit and non-profit sectors (Bal and Vink, 2011; Bingham, 2005; Vantilborgh et al., 2013, 2014).
We focus on an organization’s core values as the central content of the ideological contract (e.g. Scheel and Mohr, 2012), as they stand for an organization’s beliefs regarding desirable, motivational goals, which serve as guiding principles to those inside the organization (Van Lee et al., 2002). Furthermore, organizational members are likely to develop the belief that their organization is obligated to demonstrate a credible commitment to its core values, as they are formally endorsed and espoused by an organization to its members in order to create shared interpretations and commitments between organizations and members (Cable and Parsons, 2001; Gruys et al., 2008). As a result, core values are reflected in members’ understanding of the core character of their organization (Ashforth and Mael, 1996; Gioia et al., 2000), and they react to the extent to which the organization behaves in line with its core values (Cha and Edmondson, 2006; Lencioni, 2002). Employees with a calling, in particular, are likely to be drawn to the core values of the organization as they actively search to work for organizations that uphold values represented in their calling (Hagmaier and Abele, 2012; Hirschi and Herrmann, 2013).
There are two possible ways in which ICF may impact the relationship between callings and job performance. First, ICF may strengthen the relationship between callings and affective commitment. Employees make evaluations of their organization before they ‘invest in’ social exchange relationships, characterized by affective commitment (Cropanzano et al., 1997; Randall et al., 1999). However, such evaluations are not necessarily based on all expectations, but those that are of high personal importance (Morrison and Robinson, 1997; Wanous et al., 1992). Expectations of high personal importance are those that are value-based (Bunderson, 2001; Thompson and Bunderson, 2003) and rooted in ‘pre-employment schemas [which] provide a lens through which workers view employment experiences and the obligations these create’ (Rousseau, 2001: 515). Thus, for employees with a calling, their affective commitment depends on the extent to which the organization fulfills the promises concerning its core values.
There is reason to believe that employees with high and low levels of callings will deem different types of expectations as critical. For instance, materialistic concerns (e.g. extrinsic rewards and security) are related to organizational attachment only for employees with a low level of calling (Cardador et al., 2011). Instead, owing to their personal emphasis on prosocial goals and the transcendent meaning derived from their work (Wrzesniewski, 2003), employees with a high level of calling prioritize ideological expectations relating to social values that go beyond economic interests (Thompson and Bunderson, 2003); therefore, these employees are likely to be more responsive to an organizational climate where values such as humanism are emphasized (Vandenberghe, 2011). As a result, when employees with a high level of calling gauge their organization and decide how much to invest in a social exchange relationship with it, they examine the extent to which their organization lives up to its commitment to its espoused principles (i.e. ICF). In contrast, ICF should have less influence on social exchange development among those with a low level of calling, owing to less emphasis on ideological concerns:
Hypothesis 3: ICF moderates the positive relationship between calling and affective commitment such that the relationship is stronger when ICF is higher.
Second, we also propose that ICF strengthens the relationship between affective commitment and job performance. The basic tenet of social exchange theory is that benefits from one party oblige the other party to reciprocate on the basis of the norm of reciprocity (Blau, 1964; Gouldner, 1960). However, as Blau (1964) also noted, the reciprocation is not solely based on the benefits or services already rendered; when an individual does a favor for someone, there is an expectation of some future return. The estimation of a high probability of an organization to carry out its obligations over the long run (Holmes, 1981) would prompt employees to be more willing to balance their future benefits from the organization with present contributions to the exchange relationship (Coyle-Shapiro and Kessler, 2002). In support of the role of future return in reciprocation, a prior study showed that the less wary employees were about the organization’s future reciprocation, the more they reciprocated quality relationships (Lynch et al., 1999).
Linking the expectation of future return and reciprocation (Blau, 1964), we argue that employees are more likely to act upon their affective commitment to generate job performance under high ICF, as it may signal a high probability that their organization will reciprocate future benefits to them. An organization’s discretionary actions to pursue prosocial values – such as fulfilling its ideological promises – would likely be interpreted by employees as indications that the organization generally values social relationships (Rupp et al., 2006), or even that the organization has moral character (i.e. it does the right thing) (Bauman and Skitka, 2012; see Lange and Washburn, 2012). Such favorable pro-relational or moral judgments of one’s organization would mitigate employees’ inherent concerns in social exchange relationships about being exploited by their exchange partner (Abele and Wojciszke, 2007; Wojciszke, 2005). As a result, once committed employees derive the necessary confidence about their organization from a fulfilled ideological contract, they are likely to engage in behaviors that are beneficial to their organizations. In contrast, when the organization falls short of fulfilling its ideological promises, employees, while still compelled by the values claimed by the organization, may become less confident that their organization would honor its exchange obligations over the long run; thus, employees may reciprocate by engaging in fewer beneficial behaviors for their organizations.
Additionally, ICF may affect the skill development of calling-oriented employees, and thus their performance. For instance, Schabram and Maitlis (2016: 594) illustrated how calling-oriented animal-shelter workers responded to workplace challenges such as ‘moral injustices committed against animals by the staff tasked to serve them’, which may be construed as under-fulfillment of the ideological contract, and thus may evoke negative emotions. Workers who felt intensely negative emotions from such challenges tended to work alone or to take on leadership roles. In both cases, they were less likely to develop their skills because in the first case, they already felt they were uniquely gifted and in the second case, they were no longer focused on their primary veterinary tasks. In contrast, workers who felt less intense negative emotions approached the same situations as opportunities to develop their skills and capacity. Accordingly, we predict:
Hypothesis 4: ICF moderates the positive relationship between affective commitment and in-role performance such that the relationship is stronger when ICF is higher.
In combining Hypotheses 2 through 4, we propose the following:
Hypothesis 5: ICF moderates the mediated relationship between calling and in-role performance through affective commitment such that the relationship is stronger when ICF is higher.
The empirical studies
Study 1
Participants and procedure
The data for Study 1 were collected among the staff members of a Presbyterian megachurch in the Eastern United States. The salience of callings and values in a church setting makes it a suitable research setting to examine callings and the ideological contract (Conway et al., 2015; Scheel and Mohr, 2012). Studies of volunteers can also provide valid insights that are applicable to paid-workers (Rodell, 2013). The megachurch has over 30,000 members, and 1000 staff members (100 paid staff members and 900 non-paid volunteers). After receiving permission from the pastor, staff completed confidential questionnaires in a large auditorium. Among our study variables, ICF and demographic variables were collected at Time 1, whereas calling, affective commitment, and in-role performance were collected at Time 2 after a five-month interval. Respondents created their own personal codes at Time 1 in order for the researchers to match their responses at Time 2. At Time 1, 609 individuals, or 60.9% of the entire staff population, completed the survey. At Time 2, we received 614 completed surveys, 200 of which were successfully matched to the Time 1 surveys (final response rate = 20%). 1 In the final sample, 24% were men, and 92% were volunteers (vs paid employees). The respondents had a mean age of 48.77 years (SD = 14.39).
Measures
Calling was measured with Bunderson and Thompson’s (2009) six-item scale (α = .93), which asks participants to indicate the degree to which they believe their current occupations are consonant with their callings (e.g. ‘I was meant to do the work I do’) on a seven-point scale (1 = strongly disagree, 7 = strongly agree). We measured ICF by asking participants to rate the extent to which their church fulfilled each of eight, formally endorsed organizational values (the Bible, integrity, discipline/self-control, relationships, leadership, superior service, stewardship and a good reputation) since they had become staff members (α = .83) on a five-point scale (1 = much less than expected, 5 = much more than expected). Our approach is consistent with prior studies (Bunderson, 2001; Scheel and Mohr, 2012) in that participants evaluated their organization against a set of an organization’s core values. We measured affective commitment with Meyer and Allen’s (1984) eight-item scale (e.g. ‘I really feel as if this organization’s problems are my own’) (α = .88), using a seven-point scale (1 = strongly disagree, 7 = strongly agree). We measured in-role performance using Ashford and Black’s (1996) five-item scale (α = .94), which has been used to capture self-rated performance (e.g. Briscoe et al., 2012). Participants were asked to rate how their supervisors would evaluate, for example, their ‘quality of performance,’ relative to that of their co-workers on a seven-point scale (1 = very poor, 7 = excellent). We controlled for age (Ng and Feldman, 2010), tenure (years) (Mathieu and Zajac, 1990), gender (0 = male, 1 = female) (Taylor, 2006), worker status (0 = paid, 1 = non-paid) (Vandenberghe and Tremblay, 2008), and employment status (0 = full time, 1 = part time) (Conway and Briner, 2002).
Results
To examine discriminant validity, we conducted a confirmatory factor analysis on the four self-reported study variables. The four-factor model showed a good fit to the data (χ2 [318, N = 197] = 549.72; Comparative Fit Index (CFI) = .94; Root Mean Square Error of Approximation (RMSEA) = .06), and a significantly better fit (p < .001) than any other more parsimonious model. 2 Table 1a summarizes the means, standard deviations, and correlations for the study variables. Among the planned control variables, we did not include age in further analyses in light of its high missing rate (20%) and the non-significant correlations between age and affective commitment, and age and performance among the remaining participants (Becker, 2005). In subsequent analyses, all substantive variables were mean-centered to compute the interaction terms for the moderation analyses (Aiken and West, 1991) and to facilitate the plotting of moderated mediation (Edwards and Lambert, 2007).
Means, standard deviations and correlations.
(a) Study 1.
Max N = 200. M: mean; SD: standard deviation; ICF: ideological contract fulfillment. Organizational tenure: 1 = less than 1 year, 2 = 1–5 years, 3 = 5–10 years, 4 = 10–15 years, 5 = over 15 years. Gender: 0 = male, 1 = female. Worker status: 0 = paid, 1 = non-pad. Employment status: 0 = full time, 1 = part time.
p < .01.
(b) Study 2.
Max N = 304. M: mean; SD: standard deviation; ICF: ideological contract fulfillment; T: time. Organizational tenure: 1= less than 1 year, 2 = 1–5 years, 3 = 5–10 years, 4 = 10–15 years, 5 = over 15 years. Gender: 0 = male, 1 = female.
p ≤ .05; ** p < .01.
(c) Study 3.
Max N = 201. M: mean; SD: standard deviation; ICF: ideological contract fulfillment. Gender: 0 = male, 1 = female. Employment status: 1 = full time, 2 = part time.
p ≤ .05; ** p < .01.
Lending support to Hypothesis 1, the results of the hierarchical regression shown in Table 2a indicate that calling is significantly and positively related to in-role performance (B = .23, t = 5.38, p < .001). Hypothesis 2 predicted a mediated relationship between calling and in-role performance via affective commitment. As shown in Table 2b, calling is positively related to affective commitment (B = .14, t = 3.95, p < .001) and affective commitment is, in turn, positively related to in-role performance (B = .20, t = 2.23, p < .05). We formally tested the mediation with PROCESS macro created by Hayes (2013), which employs a bootstrapping procedure to determine bias-corrected 95% confidence intervals (CIs) around the estimates of mediation effects based on 1000 bootstrapped samples. Bootstrapping provides evidence of mediation if the calculated CI excludes zero. The produced CIs for the mediated effect of calling on in-role performance (.028, 95% CI = .002, .083) provided support for Hypothesis 2.
Regression analyses predicting affective commitment and in-role performance – Study 1
(a) Coefficients estimates for the direct model for in-role performance.
(b) Coefficients estimates for the moderated mediation model for in-role performance.
N = 181. Numbers are unstandardized regression coefficients. DV = dependent variable; SE = standard error; ICF: ideological contract fulfillment. Organizational tenure: 1= less than 1 year, 2 = 1–5 years, 3 = 5–10 years, 4 = 10–15 years, 5 = over 15 years. Gender: 0 = male, 1 = female. Worker status: 0 = paid, 1= non-paid. Employment status: 0 = full time, 1 = part time.
p ≤ .05; ** p < .01; *** p < .001.
Hypotheses 3 through 5 concerned the moderating role of ICF. We examined Hypotheses 3 and 4 using a moderated multiple regression procedure (Aiken and West, 1991). Results in Table 2b indicate that the interaction term of calling and the moderator did not predict affective commitment (B = −.04, t = −.92, NS), thus failing to support Hypothesis 3. In contrast, the interaction term of affective commitment and the moderator significantly predicted in-role performance (B = .17, t = 1.96, p = .05). We plotted the interaction effect and calculated simple slopes at high (1 SD above the mean) and low (1 SD below the mean) levels of ICF (Dawson, 2014) (Figure 2a). The positive relationship between affective commitment and in-role performance was stronger when ICF was high (B = .41, t = 2.91, p < .01) versus low (B = .15, t = 1.62, NS). Thus, these results provide support for Hypothesis 4. Hypothesis 5 predicted a moderated mediation of calling on in-role performance through affective commitment. Moderated mediation occurs when the strength of the mediation effect of calling on in-role performance depends on the level of ICF (Preacher et al., 2007). The bootstrapping procedure utilizing PROCESS macro created by Hayes (2013) indicated that the mediation effect of calling on in-role performance was significant and positive at a high level of ICF (.046, 95% CI =.013, .121), but decreased to non-significance at a low level of ICF (.027, 95% CI = −.005, .116). To facilitate the interpretation, we plotted the moderated mediation effect in Figure 2d (Edwards and Lambert, 2007). These analyses provide support for Hypothesis 5.

Moderating role of ICF.
Discussion
Study 1 provided support for the direct relationship between calling and in-role performance, as well as the mediated relationship through affective commitment. Further, the mediated relationship was stronger when ICF was high (vs low), because the relationship between affective commitment and in-role performance was attenuated by low ICF. However, Study 1 is limited in at least two ways. Our results might have been valid only within a church context, and the simultaneous collection of the calling and performance variables from respondents may bias our findings 3 (Podsakoff et al., 2003a, 2003b). Hence, we conducted a follow-up study, utilizing a two-phased survey from employees in the for-profit sector.
Study 2
Participants and procedure
Participants were a part of a US-based survey response panel run by QualtricsTM. The panel members are paid with points, which they can redeem for merchandise from online merchants. We took procedural controls against potential common method bias by means of collecting perceptual variables one-month apart: calling and ICF at Time 1, and commitment and performance at Time 2 (Podsakoff et al., 2003a). Consistent with existing research that has used online survey organizations, we calculated the response rate based on the number of participants who opened the survey invitation (Brown and Robinson, 2011; Dumas et al., 2013; Long et al., 2011). At Time 1, 2587 out of 5497 individuals who opened the survey completed it (Time 1 response rate = 47.1%). A review of these completions as for the completion time resulted in a total of 2226 final respondents for Time 1, who were contacted through the email addresses they provided at Time 1. At Time 2, 304 out of 369 individuals who opened the survey completed it, representing our final sample (Time 2 response rate = 82.4%, or 13.7% of all invitations sent out). 4 Fifty-eight percent was female, and the average age was 39.9 years (SD = 11.2). More than 19 industries (e.g., finance, healthcare) were represented.
Measures
Participants rated their agreement using a seven-point scale (1 = strongly disagree, 7 = strongly agree). As in Study 1, calling was measured with Bunderson and Thompson’s (2009) scale (α = .94). We measured ICF by adapting Robinson and Morrison’s (2000) five-item measure of psychological contract fulfillment to capture how true employees perceive their companies are to their core values (α = .86). 5 We specified the content of ‘promises’ in the original items to be ‘core values’ (e.g. ‘So far my company has done an excellent job of fulfilling its core values since I joined the staff’), in light of our focus on core values as the central content of ideological contracts. Affective commitment was measured with Meyer et al.’s (1993) six-item scale (α = .86). In-role performance was measured with Williams and Anderson’s (1991) seven-item scale (e.g. ‘adequately completes assigned duties,’ α = .81). We controlled for age, gender (0 = male, 1 = female), and organizational tenure, as in Study 1. Employment status was not used, as all respondents reported full-time status.
Results
The hypothesized four-factor measurement model demonstrated a reasonable fit to the data, with one residual covariance added between two reverse-coded items of in-role performance (Bachman and O’Malley, 1986) (χ2 [245, N = 304] = 695.70; CFI = .93; RMSEA = .08), and was significantly better (p < .001) than any other more parsimonious model. The mean scores, standard deviations, and correlations are presented in Table 1b. We tested our hypotheses with the same procedure as in Study 1. As shown in Table 3a, calling was not a significant predictor of T2 in-role performance (B = −.03, t = −.64, NS). Hence, Hypothesis 1 was not supported. Despite the non-finding of the direct effect in Hypothesis 1, we proceeded to test a mediation relationship in Hypothesis 2, as unlike the widely used, traditional approach recommended by Baron and Kenny (1986), a more recent approach does not require a significant direct relationship between an independent and a dependent variable as a precondition to examine the mediation between the two (MacKinnon et al., 2002; Rucker et al., 2011; Shrout and Bolger, 2002). As shown in Table 3b, calling is positively related to T2 commitment (B = .31, t = 7.12, p < .001), and T2 commitment is, in turn, positively related to T2 performance (B = .20, t = 3.39, p < .001). The bootstrapping procedure provided formal support for Hypothesis 2, the mediated effect of calling on T2 in-role performance (.063, 95% CI = .019, .116).
Regression analyses predicting affective commitment and in-role performance – Study 2
(a) Coefficients estimates for the direct model for in-role performance.
(b) Coefficients estimates for the moderated mediation model for in-role performance.
N = 191. Numbers are unstandardized regression coefficients. DV = dependent variable; SE = standard error; ICF: ideological contract fulfillment. Organizational tenure: 1= less than 1 year, 2 = 1–5 years, 3 = 5–10 years, 4 = 10–15 years, 5 = over 15 years. Gender: 0 = male, 1 = female.
p ≤ .05; ** p < .01; *** p < .001.
In Table 3b, the interaction term of calling and ICF was not significant in predicting T2 affective commitment (B = –.01, t = –.58, NS), thereby failing to provide support for Hypothesis 3. However, the interaction term of T2 affective commitment and ICF was a significant predictor of T2 in-role performance (B = .09, t = 2.52, p < .05). Figure 2b shows that the relationship between T2 affective commitment and T2 in-role performance was positive and stronger at a high level of ICF (B = .36, t = 4.30, p < .05) than at a low level of ICF (B = .13, t = 1.93, NS), thereby supporting Hypothesis 4. The bootstrapping procedure indicated that the mediation effect of calling on T2 in-role performance was stronger at a high level of ICF (.106, 95% CI =.051, .194) than at a low level of ICF (.041, 95% CI = −.011, .110). We plotted the moderated mediation effect in Figure 2e. These analyses indicate that Hypothesis 5 is supported.
Discussion
Study 2 replicated the findings of Study 1, with the only exception being that the data did not support the direct effect of calling on in-role performance. Such replication demonstrates the generalizability of our findings, as Study 2 utilized a two-phased dataset collected from individuals in various for-profit organizations. Yet, Study 2 is still limited by the use of a self-rated performance measure, although such a measure has been found to correlate with other-rated or actual performance (Harris and Schaubroeck, 1988; Harrison et al., 2011; Pearce and Porter, 1986). Hence, a follow-up study was conducted with a supervisor-rated performance measure, which has been rarely utilized in calling research.
Study 3
Participants and procedure
The participants were all employed in service teams of six companies operating banking, telecommunications or electricity in Eastern Canada. Owing to the prevalence of conflicts between organizational and occupational values such as the organizational focus on service efficiency versus the occupational concern with service quality (i.e. occupational concerns) (Frei, 2006; Zeithaml et al., 1988), these service teams provide an excellent research setting to examine the tension between organizational and occupational concerns (Bowen and Ford, 2002). Matched field data were collected from French-speaking employees and their supervisors. Employees completed a questionnaire including calling, ICF, and commitment, while supervisors rated employee performance in a separate questionnaire. Questionnaires were coded so as to allow matching the data. All respondents were assured of confidentiality. In total, 201 usable responses were obtained from employees (response rate = 79%). In the final employee sample, 52% were men, and 80% worked full time. Employees had a mean age of 33.30 years (SD = 7.00), a mean tenure with the supervisor of 1.98 years (SD = 1.78), and a mean organizational tenure of 2.82 years (SD = 2.80). Employees had high school (9%) or at least a two-year college (91%) education. Supervisors had a mean age of 44.33 years (SD = 3.20), with 9.67 years (SD = 5.24) of tenure. Four of the six supervisors were male.
Measures
Since the participants were French-speaking employees, all survey items, unless they were already available in French, were translated into French (Brislin, 1980). Participants rated their agreement using a seven-point scale (1 = strongly disagree, 7 = strongly agree) for calling and ICF, and a five-point scale (1 = strongly disagree, 5 = strongly agree) for affective commitment and in-role performance. As in Studies 1 and 2, calling was measured with Bunderson and Thompson’s (2009) scale (α = .95). The same scales as in Study 2 were used for the remaining variables: ICF with the adapted measure of Robinson and Morrison’s (2000) (α = .78), affective commitment with the French version (Vandenberghe et al., 2001) of Meyer et al.’s (1993) scale (α = .94), and in-role performance with Williams and Anderson’s (1991) scale (α = .93). We controlled for age, gender (0 = male, 1 = female), organizational tenure (years), and employment status (1 = full time, 2 = part time).
Results
The hypothesized four-factor measurement model demonstrated a good fit to the data (χ2 [246, N = 198] = 440.49; CFI = .96; RMSEA = .06), and a significantly better fit (p < .001) than other more parsimonious model. The descriptive statistics and correlations are shown in Table 1c. The results of the hierarchical regression 6 indicate that calling is significantly related to in-role performance (B = .10, t = 2.71, p < .05), thus supporting Hypothesis 1 (Table 4a). Calling is positively related to affective commitment (B = .16, t = 4.13, p < .001), and affective commitment is, in turn, positively related to in-role performance (B = .38, t = 5.75, p < .001) (Table 4b). A CI produced by the PROCESS procedure indicated a significant mediated effect of calling on in-role performance (.062, 95% CI = .028, .109), supporting Hypothesis 2. Results in Table 4b indicate that the interaction term of calling and ICF did not predict affective commitment (B = .00, t =.14, NS), thus failing to support Hypothesis 3. However, the interaction term of affective commitment and ICF was a significant predictor of in-role performance (B = .09, t = 2.11, p < .05). The plot of the interaction effect (Figure 2c) shows that the positive relationship between commitment and performance was stronger when ICF was high (B = .52, t = 5.53, p < .001) than low (B = .28, t = 3.46, p < .01). Therefore, Hypothesis 4 was supported. Hypothesis 5 predicted a moderated mediation of calling on in-role performance. The bootstrapping procedure indicated that the mediation effect of calling on in-role performance was stronger when ICF was high (.088, 95% CI =.028, .166) versus low (.044, 95% CI = .004, .112) (see Figure 2f for the plot of the moderated mediation effect). Hence, Hypothesis 5 was supported.
Regression analyses predicting affective commitment and in-role performance – Study 3
(a) Coefficients estimates for the direct model for in-role performance.
(b) Coefficients estimates for the moderated mediation model for in-role performance.
N = 191. Numbers are unstandardized regression coefficients. DV = dependent variable; SE = standard error; ICF: ideological contract fulfillment. Gender: 0 = male, 1 = female. Employment status: 1 = full time, 2 = part time.
p ≤ .05; ** p < .01; *** p < .001.
Discussion
Study 3, which incorporated supervisor-rated performance variables, provided consistent results with previous studies. As with Study 1, Study 3 replicated a positive direct effect of calling on performance. As with both Studies 1 and 2, Study 3 demonstrated the mediated effect of calling on performance via commitment, and the moderating effect of ICF on the commitment-performance link, and hence, the moderated mediated effect, while it failed to support the moderating effect of ICF on the calling-commitment link.
General discussion
Despite increasing attention to the concept of callings (Berg et al., 2010; Bunderson and Thompson, 2009; Cardador et al., 2011; Elangovan et al., 2010), it is not clear how individuals with a calling perform in their work roles. We addressed this question by unpacking the intervening process between callings and in-role performance. Three studies provided convergent evidence for our moderated mediated model of callings and performance. Affective commitment serves as a mediator of the relationship between callings and performance, and ICF is a moderator of the mediated relationship. Taken together, our findings deepen our understanding of the mechanisms that lead those with callings to be high performers; moreover, our results highlight the organizational conditions that explain why employees with a similar level of calling exhibit varied performance behaviors.
Theoretical contributions
First, our findings extend previous research by introducing a novel framework to understand the organizational implications of callings in the workplace. Departing from the previous focus in the literature on work- and occupation-centered explanations for the effect of callings, our study investigates organizationally embedded mechanisms between callings and job performance, drawing on the social exchange perspective. Extending the pioneering works demonstrating that calling-oriented employees tend to forge high-quality exchange relationships with their organization (Cardador et al., 2011; Duffy et al., 2011), our findings suggest that this quality exchange relationship becomes the mechanism that drives their superior job performance. Our findings further suggest that when employees perceive their organization as upholding its ideological obligations, their affective commitment acts as a stronger driver of their performance. Thus, social exchange, with an emphasis on ideological currency, provides a useful conceptual lens for examining how callings may shape employees’ attitudes and behaviors.
Second, our findings contribute to resolving divergent views on the performance implications of callings by suggesting that higher levels of calling and the resulting higher levels of affective commitment may be associated with increased job performance to the extent that employees perceive their organization as living up to its ideological promises. Our findings thus underscore the importance of the immediate context for the enacting process of callings. Although scholars tend to believe that calling-oriented individuals are largely guided and motivated by an ‘internal compass’ (Hall and Chandler, 2005) and ‘the purpose and meaningfulness of activities’ (Dik and Duffy, 2009: 429), the present findings suggest that their workplace behaviors are also heavily affected by their ongoing assessments of their organizational context. In that sense, our findings direct future research toward examining the interaction between individuals with a calling and their immediate context (Conway et al., 2015; Dobrow, 2013).
Third, our study sheds fresh light on employment relationships for calling-oriented employees by indicating that, counter to our expectations, the relationship between callings and affective commitment does not depend on ICF. Calling-oriented individuals have been thought to hold a more self-directed, protean approach toward their careers, which also implies high mobility and fragile organizational attachment (Hall, 2004; Hall and Chandler, 2005). As a result, they are often assumed to be misfits in their organization when its values are not aligned with their callings (e.g. teaching-oriented faculty at a research-oriented institution (Serow et al., 2002)), to the point that Elangovan et al. (2010: 437) speculated that a calling ‘could be a mixed blessing for the organization.’ Yet, this null finding, obtained across all three studies, indicates that an exchange relationship between calling-oriented employees and their organization, after all, may not be as fragile as previously anticipated, even when these employees admit under-fulfillment of the ideological contract on the part of the organization. Calling-oriented people may be less susceptible to violation when they encounter a breach, as compared to people with job- or career-orientations (Thompson and Bunderson, 2003). Such individuals may develop a distinctive coping strategy in which they stay committed in hopes of a closer alignment with their calling-based values in the future, but also decrease their behavioral input into the organization. Future research should further investigate the contingencies that determine when employees with callings will be affectively committed to their organizations.
Finally, our findings bear implications for our understanding on the complementarity, rather than the competition, between attachment to occupations and organizations (Noordegraaf, 2011). Previous studies have shown that employees can develop compatible attachments to their occupations and organizations (Russo, 1998; Vough, 2012; Wallace, 1993, 1995). We also find this compatibility and indicate that organizations may be seen as a vehicle through which callings can be enacted, but that the relationship goes further than that. Employees actually become attached to their organization because of the prospect of it addressing their goals. As such, the interplay between organizational and occupational attachment is not straightforward, but instead, reciprocal or synergistic. In order to fully explicate the interplay, future research should delve into this relationship in more depth, particularly for those employees with callings.
Practical implications
Our research has important practical implications. From the perspective of individuals with a calling, the results of our study point out the importance of their agency in finding the right organizational context if they desire to fulfill their callings and make personally and socially meaningful contributions in their work roles. Despite the good intentions and organizational loyalty on the part of calling-oriented individuals, organizations that fail to meet their ideological expectations could inhibit them from enacting their callings in their work roles. Thus, they need to select organizations that not only emphasize transcendent values, but are also capable of acting consistently with those values. In contrast, individuals with a calling who are already affiliated with organizations that tend not to fulfill their ideological promises should strive to make an impact on their organization in order to make it a better fit with their callings.
There are HR implications with respect to hiring practices, as well. Traditional practices of hiring based on skill sets alone may limit an organization’s ability to hire individuals with a calling, who are likely to be self-motivated and ready to expend effort for good performance. Indeed, some changing trends to hiring practices are noticeable. Managers are increasingly advised to base their hiring decisions on passion (e.g. Fisher, 2006; Sundheim, 2014), a defining component of a calling (Dobrow and Tosti-Kharas, 2011; Hunter et al., 2010), even more so than on experience or degrees (e.g. Finnigan, 2014). Yet, managers should be aware that the performance benefits from calling-oriented employees may be attainable only when employees feel that their ideological contract is upheld by their organizations. Therefore, managers should be careful about supporting an occupational ideology associated with employees’ callings. Managers should also be mindful of employee perceptions. Even when managers believe they are acting in accordance with organizational core values, their actions may be perceived otherwise. Managers need to communicate the intentions behind their actions in situations that are subject to potential misunderstanding. Along these same lines, the core values need to be clearly articulated in order to reduce differences in interpretation (Cha and Edmondson, 2006).
Limitations and future research directions
Our datasets across three studies do not permit us to test the direction of causality directly. Although our findings suggest that callings contribute to better job performance, the reverse causal direction cannot be ruled out. For example, good performance and resulting self-efficacy could lead to a strong sense of calling (Dobrow Riza and Heller, 2015). Research using experimental or longitudinal designs is necessary to clarify the issue of causality. There is no precedent use of our measure of ICF used in Studies 2 and 3, a modified scale of an extant psychological contract. The supplementary analysis of Study 2 data and a separate pilot study support the validity of this measure (see endnote 5). Our measure of ICF diverges from two trends in measuring psychological contracts: the separation of promise and delivery (Turnley and Feldman, 2000) and the use of polynomial regression (Lambert et al., 2003). Despite the merits in these approaches, our focus lies in participants’ conscious evaluation of ICF, rather than the effects of over- or under-fulfillment.
There are several venues to extend our study. Future research may investigate additional, organizationally embedded mechanisms that are complementary to social exchange relationships, especially considering the predominant focus on work or occupational explanations so far in the field. Besides, future research may investigate the conditions under which callings and employee attitudes and behaviors can vary in order to further clarify the organizational implications of callings. In particular, it would be beneficial to examine the variables that represent a more proximal context to one’s job, since work itself is important to these individuals. Further, future research may look into other consequences of callings than in-role performance, a positive outcome of callings from the employer’s point-of-view. As callings may have a darker side such as potential labor exploitation (Berkelaar and Buzzanell, 2015; Duffy and Dik, 2013), researchers need a balanced approach in investing the implications of callings.
Conclusion
Our study suggests that employees with callings are inclined to be strong contributors to their organization through their in-role performance. This relationship makes sense from a social exchange perspective, which takes into consideration the conveyance of ideological resources between the organization and the employee. Employees with callings are strongly committed to their organization and want to see the organization succeed. However, when organizations do not live up to their value-based promises, employees with callings may withhold their contributions in order to maintain parity in the exchange relationship. As such, we uncover important new insights for organizations employing those with callings.
Footnotes
Acknowledgements
We would like to thank Associate Editor Mathew Sheep and two anonymous reviewers for their comments on our manuscript through the revision process. We would also like to thank the participants of the Management Department Brown Bag meeting at the Daniels School of Business, University of Denver for their comments on earlier stages of this article.
Funding
Data collection for Study 2 and Study 3 was funded by Social Sciences and Humanities Research Council (grant number: 435-2016-1688) awarded to Patricia Hewlin and Canada Research Chairs Program (grant number: 950-219143) awarded to Christian Vandenberghe, respectively.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
