Individuals’ perceptions of calling before entering the workforce influence how they think and behave (Bunderson and Thompson, 2009; Hunter et al., 2010; Dobrow-Riza and Heller, 2015). In a large scale of surveys using more than 5,000 college students, >40% of them reported they have a calling (Duffy and Sedlacek, 2010) and students discerned their calling based on diverse sources such as guiding forces, talents, and dedicated efforts (Hunter et al., 2010). College students’ callings had a wide influence on their lives as well as careers (Hunter et al., 2010). Their calling plays a critical role in using and training one’s ability (Hunter et al., 2010) and pursuing one’s career (Bunderson and Thompson, 2009; Dobrow-Riza and Heller, 2015). Researchers note that the callings students had upon entry into a career domain influences later career pursuits by creating psychological conditions under which people focus on their own beliefs about the career domain and by shaping how they think about themselves, their work, and their environment (Dobrow-Riza and Heller, 2015). These findings suggest that the perception of calling upon organizational entry would influence job attitudes and outcomes over time.

For newcomers, reducing uncertainty is a primary goal during the socialization period (Ashford and Black, 1996). Perceiving a calling can reduce employees’ cynicism and sustain their engagement in times of uncertainty by promoting work meaningfulness (Bunderson and Thompson, 2009; Hirschi, 2012) and reaffirming the social influences of their work (Cartwright and Holmes, 2006). People with calling positively frame themselves, their work, and their situation (Dobrow-Riza and Heller, 2015; Schabram and Maitlis, 2017) and newcomer’s positive attitude was found to be effective in increasing job satisfaction (Ashford and Black, 1996). When endorsing a calling, people focus on the ideal and positive aspect of their work (Elangovan et al., 2010) and have a clear goal for their work (Hall and Chandler, 2005). These attitudes, in turn, can contribute to job satisfaction. Having a calling can buffer psychological distress during challenging and uncertain contexts (Schabram and Maitlis, 2017) like newcomer socialization. Although the link between perceiving a calling and job satisfaction has yet to be examined among newcomers, the robust positive relation between perceiving a calling and well-being has received consistent support across diverse occupations (Wrzesniewski et al., 1997; Duffy et al., 2012; Conway et al., 2015). Based on these findings, we expected that perceiving a calling upon organizational entry would positively relate to job satisfaction 2 years later.

In predicting newcomers’ performance and adaptive behaviors, a significant predictor is newcomers’ proactive behavior at work (Ashford and Black, 1996). Among diverse proactive tactics, such as feedback seeking and networking, self-focused proactive strategies such as positive reframing were positively related to self-rated job performance (Ashford and Black, 1996). An implicit assumption about calling is that people are called to a specific way of work that motivates a course of action to achieve one’s goals (Elangovan et al., 2010). Scholars note that people with a sense of calling demonstrate high job performance because they establish clear goals and exert greater efforts to accomplish them (Hall and Chandler, 2005; Wrzesniewski, 2012). Individuals with callings are more engaged in their work (Bunderson and Thompson, 2009) and confront situations more proactively not because of situational affordance, but due to their mindset (Schabram and Maitlis, 2017). Thus, we hypothesized that perception of a calling upon organizational entry would be positively related to job performance.

Job involvement is defined as a cognitive belief of psychological identification with one’s job (Kanungo, 1982). Job involvement represents the strength of one’s attachment to a work (Kanungo, 1982). Among diverse occupations of US workers, viewing work as a calling was positively related to one’s belief that work is one of the most important things in their life (Wrzesniewski et al., 1997). Qualitative research by Bunderson and Thompson (2009) on zookeepers found that a sense of calling enables them to maintain a high level of job involvement by infusing even trivial tasks with transcendent meaning and significance, and the authors assert that zookeepers’ callings provide a compelling basis for identification with the occupation.

Although a causal relation between perceiving a calling and job involvement has received little attention from researchers, we expect that perceiving a calling would promote one’s attachment to work and psychological identification with one’s work. Hackman and Oldham (1975) suggest that intrinsic perceptions of job characteristics, such as work meaningfulness, increase job involvement. Those with callings show increased dedication (Hunter et al., 2010; Praskova et al., 2014) and the continuous efforts, which result in increased work meaningfulness (Hirschi, 2012) and job involvement (Brown, 1996). Also, people with callings seek to live out their calling (Elangovan et al., 2010) and have a strong desire to fulfill it by engaging in job crafting activities (Berg et al., 2010). Such job crafting behaviors longitudinally contribute to work meaningfulness (Tims et al., 2016), which leads to increased job involvement (Hackman and Oldham, 1975). Findings that show that job involvement is predicted by positive traits such as self-esteem and a clear sense of self (Brown, 1996), which are closely related to calling (Hall and Chandler, 2005), also support the positive link between perceiving a calling and job involvement. Based on these findings, we hypothesized,

Increased job involvement may be one route by which perceiving a calling relates to job satisfaction and job performance. Scholars have noted that fostering identification with occupation serves a means by which callings create attitudes toward jobs (Bunderson and Thompson, 2009). For example, a sense of a calling has a strong positive relation with occupational identification (Bunderson and Thompson, 2009), and occupational identification and occupational identity mediated the relation between calling and well-being at work (Bunderson and Thompson, 2009; Hirschi, 2012). According to a theoretical model of job involvement (Brown, 1996), those who view themselves as competent are more likely to seek challenges to improve personal growth at work, and the greater growth needs motivate them to engage in job activities that result in increased job satisfaction (Brown, 1996). Other findings suggest that the positive self-views (Hirschi, 2012; Dobrow-Riza and Heller, 2015) and strong growth initiative of those with callings (Bott and Duffy, 2015; Schabram and Maitlis, 2017) render these individuals to feel more satisfied with their jobs via increased job involvement.

Also, the findings that career commitment, a construct largely overlapped with job involvement (Brown, 1996), fully mediates the relation between perceiving a calling and job satisfaction (Duffy et al., 2011) also lends support to the mediating role of job involvement on calling and job satisfaction. An article that suggests a theoretical model of work as a calling, the work as calling theory (WCT), underlines that perceiving a calling predicts greater career commitment, which in turn predicts job outcomes (Duffy et al., 2018). As the first formal attempt to establish a theoretical model of calling, the WCT argues that the level of commitment in career or work would mediate the link of perceiving a calling and positive job outcomes such as job satisfaction and job performance (Duffy et al., 2018).

Given that job involvement is a primary factor in predicting organizational effectiveness (Pfeffer, 1994) and individual motivation (Hackman and Lawler, 1971), the relation between perceiving a calling and job performance can be explained by job involvement. Scholars argue that the influences of perceiving a calling on job performance can be accounted for by the continued effort and motivation of those with callings (Hall and Chandler, 2005; Wrzesniewski, 2012), as these individuals overcome obstacles at work by enhancing their skills and accumulating knowledge (Praskova et al., 2014; Schabram and Maitlis, 2017). Those with callings are likely to be immersed in their work (Bunderson and Thompson, 2009) and the motivational processes influence job performance (Diefendorff et al., 2002). Although a mediating role of job involvement on the relation between perceiving a calling and job performance has yet to be examined, job involvement serves as a mediating mechanism that connects individual’s traits and job outcomes (Brown, 1996). Traits that are predisposed to job involvement include internal motivation (Conway et al., 2015) and the strength of growth need (Bott and Duffy, 2015; Schabram and Maitlis, 2017), which are possessed by those with high levels of callings. Based on these findings, we hypothesized that job involvement would be a route by which perceiving a calling relate to job satisfaction and job performance.

The effects of perceiving a calling on job involvement can be conditional depending on perceptions about job and organization; also, individual and contextual factors play an important role in understanding the effects of callings (Hall and Chandler, 2005; Duffy et al., 2017). In explaining the interactive effects, Duffy et al. (2017) suggest that calling can be understood through motivational theories such as Self-Determination Theory (Deci and Ryan, 2000). According to this theory, the effects of goal pursuit and attainment concern the degree to which people are able to satisfy their needs for competence, relatedness, and autonomy (Deci and Ryan, 2000). The psychological needs for different types of motivation vary depending on individual and contextual factors and human needs specify the necessary conditions for well-being and adaptive behaviors (Deci and Ryan, 2000). In framing calling from a goal-setting perspective (Hall and Chandler, 2005; Praskova et al., 2014), researchers found that young adults engage in more goal-oriented cognition, activities, and career strategies (Praskova et al., 2014). Focusing on the motivation concept from Self-Determination Theory, a study (Duffy et al., 2017) found that calling motivation strengthened the relation between perceiving a calling and well-being. Qualitative research has also found that the effects of having a calling on job attitudes and job behaviors are facilitated and constrained by individual factors such as goal orientation and organizational factors such as developmental relationships (Schabram and Maitlis, 2017). These findings suggest that people with callings are oriented toward accomplishing goals and individual and social factors that satisfy the basic psychological needs can facilitate the effects of perceiving a calling on job perceptions, attitudes, and behaviors.

Perceived organizational support refers to employees’ general perceptions of being valued and cared about by their organization, and is based on quality social interactions at work (Eisenberger et al., 1990). Self-Determination Theory suggests that supportive environment can be an important condition to satisfy basic psychological needs (Deci and Ryan, 2000). When organization was perceived as supportive, employees feel free to express themselves, feel more connected and perceived their roles as challenging, thus their needs for competence, relatedness, and autonomy are likely to be fulfilled (Deci and Ryan, 2000; Greguras and Diefendorff, 2009). People with callings have a strong desire to pursue their goals (Hall and Chandler, 2005), in an environment that satisfies basic psychological needs, they are more likely to involve in their job. For example, people with calling view their social relations as a special bonding and they exert more effort based on the strong relatedness (Bunderson and Thompson, 2009). However, when individuals did not feel they received enough support from the organization and feel threatened by depleting interpersonal relations, people with callings experienced a sense of defeat in pursuing their callings and guard themselves from their work (Schabram and Maitlis, 2017). The moderating role of perceived organizational support on the link between perceiving a calling on work commitment such as job involvement is also supported by the WCT (Duffy et al., 2018). The theoretical model suggests that those who perceive a calling and who experience high levels of organizational support are likely to feel their work environment fit well and experience positive outcomes (Duffy et al., 2018). The social capital resulting from supportive social environment may facilitate beneficial effects of perceiving a calling while the lack of organizational support is likely to be perceived as a barrier to live out their calling and to perform positive behaviors at work (Duffy et al., 2018). Based on these theoretical and empirical findings, we hypothesized that perceived organizational support strengthen the positive effects of perceiving a calling on job involvement.

Jobs that provide opportunities to express an individuals’ values and beliefs (Shamir, 1991) are more likely to facilitate the effects of calling on job involvement. When employee perceives their abilities and skills that match the requirements of the job, they feel their need for competence satisfied (Greguras and Diefendorff, 2009). People seek work that can express their authentic self (Shamir, 1991), and feel authentic and competent in situations that allow for value attainment (Schlegel et al., 2009). When people with callings perceive high levels of person-job fit, they are likely to find more meaningfulness in their work and become more involved in their work (May et al., 2004). However, when their job does not fit with their core values, people with callings experience regret and distress due to difficulties in pursuing their callings (Berg et al., 2010). Thus, we hypothesized that those with calling would be more involved in their job when they perceive higher person-job fit.

Our sample consisted of new employees of a company headquartered in South Korea. The company’s concerns ranged from electronics to financial investment. Surveys were administered at three time points during new employees’ first two working years: during the first week of the orientation program (Time 1; T1); 1 year (Time 2; T2); and 2 years (Time 3; T3) after organizational entry. Measures at T1 were obtained during the first week of a newcomer’s orientation, before they had met their supervisor, been assigned to a team, or begun undertaking tasks. An HR manager at the orientation informed employees that the survey was voluntary and irrelevant to the organization, and posted the survey link on an Internet community board used for the 3-week orientation program. Email addresses were collected when new employees signed and submitted their informed consent form. Employees were assigned to a team after completing the orientation program. At T2 and T3, survey links were sent via email by one of the authors, who is employed by a research institution affiliated with the company.

The T1 survey collected data on the newcomer’s calling along with a control variable, core self-evaluations, and demographic information. The T2 survey collected data on job involvement, perceived person-job fit, and perceived organizational support and at T3, newcomers’ job satisfaction was measured via survey and task performance data were obtained from the company. All surveys were administered in Korean, and all measures were translated into and validated in Korean.

Five hundred thirty participants completed the survey at T1, 246 at T2, and 145 at T3. We sent emails to all participants who did not complete surveys but had consented to receive survey emails. Two hundred twenty-one completed both T1 and T2 surveys; and 145 completed all three surveys. Of the participants who responded to the T1 survey, all were Korean, 74.3% were male, and the average age was 25.32 years (SD = 1.83). Educational level was measured using four categories – high school graduate, 2-year college or technical college graduate, 4-year university graduate, and graduate school – and the category was coded on a scale from1 to 4. Most participants (90.6%) had a least 4-year university degree.

In examining the hypotheses, we controlled for the effects of core self-evaluations. Core self-evaluations represent the basic and fundamental beliefs individuals have about themselves and their functioning in the world (Judge et al., 2003). Core self-evaluations consist of four personality traits: self-esteem, generalized self-efficacy, locus of control, and emotional stability (Judge et al., 2003). Core self-evaluations have been found to be a significant predictor of job satisfaction and job performance across diverse occupations of working adults (Judge and Bono, 2001; Judge et al., 2003). Also, on average, a positive moderate correlation between perceiving a calling and core self-evaluations was found among diverse groups of samples (Duffy et al., 2012; Hirschi and Herrmann, 2013). When investigating the relations between calling, job satisfaction, and job performance, it is important to account for the effects of a basic personality trait (Hirschi and Herrmann, 2013). Controlling for the effects of core self-evaluations on job satisfaction and job performance yields stronger inferences regarding the hypothesized relations and more precise understanding of the relations.

First, we performed confirmatory analysis (CFA) with maximum likelihood estimation using AMOS 18. As fit indices, three fit indices were used in addition to the Chi-square test – the Tucker–Lewis index (TLI), the comparative fit index (CFI), and the root-mean-square error of approximation (RMSEA) with the following guidelines: Values of 0.90 can be defined as acceptable to assess the fit of TLI and CFI (Hu and Bentler, 1999). Values of 0.5 for the RMSEA indicate close fit, values in the vicinity of 0.08 indicate fair fit, and values of 0.10 and larger indicate poor fit (Browne and Cudeck, 1993).

Parcels were created using a method proposed by Little et al. (2002). For a calling, the three subscales were used as observed indictors. For job involvement, job satisfaction, and perceived organizational support, which have more than five observed indicators, parcels were created with a balancing assignment (Little et al., 2002). Two 3-item parcels and one 4-item parcels for job involvement, two 3-item parcels for job satisfaction and three 3-item parcels for perceived organizational support were created according to the size of the factor loading. Although perceived person-job fit has five observed indicators, to create an interaction term using a matched-pair strategy (Marsh et al., 2004), we created two 2-item parcels and one observed indicator that match with the three parcels of calling. For job performance, one observed variable was used.

We used structural equation modeling to test our hypotheses. Maximum likelihood was used to estimate missing data. As with most longitudinal data, some participants were unavailable at one or more time points. We used full-information maximum-likelihood (FIML) estimation, which uses full information from all observations (Yuan and Bentler, 2000) but yields relatively unbiased results (Arbuckle, 1996). FIML has been found to be efficient for incomplete data (Schafer and Olsen, 1998), and is highly recommended for structural equation modeling analysis using incomplete data (Arbuckle, 1996). As normality assumptions are important for FIML estimation, we investigated the normality of each variable whether skewness had an absolute value of 2.0 or larger and standardized kurtosis had an absolute value of 7.0 or larger (West et al., 1995). All study variables were well-suited to the guidelines. We also conducted an attrition analysis by comparing those who completed surveys at all time points (N = 145) and those who completed the T1 survey only (N = 385). No differences in levels of initial calling, core self-evaluations, age, gender ratio, or education level were found. The results suggest that, overall, attrition was not systematic with regard to study variables.

To examine moderating effects, we followed a two-stage maximum-likelihood approach described by Ping (1996). In stage 1, we obtained estimates of factor loadings, factor covariances, and error variances from the structural equation model without interaction terms. In stage 2, we added interaction terms. To create latent construct interactions, we multiplied standardized indicators of independent and moderating variables (Ping, 1996; Marsh et al., 2004). Using a matched-pair strategy, we created interaction terms with three latent variables.

Table 1 shows the means, standard deviations, reliabilities, and correlations among the study variables. Calling positively correlated to supervisor-rated job performance (r = 0.42, p < 0.01), but did not correlated to job satisfaction (r = −0.01, ns). The correlation between calling and job involvement was also positive (r = 0.21, p < 0.01). In regard to moderators, calling related neither to perceived organizational support (r = 0.09, ns) nor perceived person-job fit (r = 0.06, ns). Job involvement had positive correlations with both perceived organizational support (r = 0.23, p < 0.01) and perceived person-job fit (r = 0.21, p < 0.01). Because education level was positively related to job performance, we controlled for it in the structural model analyses.

Before evaluating the structural models, we examined the fit indices of a measurement model and evaluated how well indicators loaded onto their factors. The measurement model had good fit to the data, χ²(df = 150) = 280.30, p < 0.01, CFI = 0.96, TLI = 0.95, and RMSEA = 0.04. Also, all indicators were loaded onto their respective latent variables at values of 0.71 or higher, showing strong relations between observed indicators and latent factors.

We then assessed the fit of our hypothesized model to the data. We specified a model in which perceiving a calling predicts job involvement, which in turn predicts job satisfaction and job performance. We specified a partial mediation model that included these relationships, together with direct effects of the independent variables on the dependent variables when controlling for core self-valuations and education level. This model showed a good fit, χ²(df = 71) = 139.62, p < 0.01; CFI = 0.97, TLI = 0.95, RMSEA = 0.04. Perceiving a calling was positively related to job involvement (β = 0.25, p < 0.01). The path from job involvement to job satisfaction was positive and significant (β = 0.42, p < 0.01). However, there were no statistically significant links from job involvement to job performance (β = 0.08, ns), and from perceiving a calling to job satisfaction (β = −0.12, ns). Due to a non-significant relation between job involvement and job performance, the mediating role of job involvement was not examined; thus, Hypothesis 5 was rejected.

To improve model fit indices, we excluded all nonsignificant paths that include a path from calling to job satisfaction and a path from job involvement to job performance. This model showed similar fit indices, χ²(df = 73) = 141.55, p < 0.01; CFI = 0.97, TLI = 0.95, RMSEA = 0.04, but the difference in Chi-square between the two model was significant, p < 0.05. Thus, we chose this model as a final model without interaction terms. Consistent with Hypothesis 3, the path from calling to job involvement was positive (β = 0.24, p < 0.01) and the link from job involvement to job satisfaction was significant and positive (β = 0.39, p < 0.01). The path from calling to job performance was significant (β = 0.27, p < 0.01), supporting Hypothesis 2.

We tested the significance of indirect effects using a Monte Carlo approach that constructs the appropriate confidence intervals (CIs) (Preacher and Selig, 2012). We entered standardized coefficients and standard errors on the relation between calling and job involvement (β = 0.24, SE = 0.09) and the relation between job involvement and job satisfaction (β = 0.47, SE = 0.11) into a web-based Monte Carlo calculator to compute a 95% CI based on 20,000 simulated draws from the distributions for the parameters (Preacher and Selig, 2012). The relationship between perceiving a calling and job satisfaction was fully mediated by job involvement (95% CI 0.02, 0.22), which supports Hypothesis 4.

All moderated effects were tested using the procedure recommended by Ping (1996). However, it should be noted that not all of the possible observed variable cross-product terms were used when constructing the latent interaction term. Jöreskog and Yang (1996) demonstrated that it is sufficient to construct the latent interaction terms using a matched-pair approach. To create an interaction term, we centered the independent variable (i.e., calling) and moderators (i.e., perceived organizational support, perceived person-job fit). Then, we multiplied standardized indicators of independent and moderating variables to create latent construct variables. We obtained estimates of factor loadings, factor covariances, and error variances from the fully mediated model without interaction terms. Then, we added two interaction terms along with the two moderators. When the interaction terms are included in the model, the focus is solely on the significance of the estimated effects of the latent interaction constructs on job involvement because the interaction is orthogonal to the main effects of the constructs (Marsh et al., 2004; Little et al., 2002).

We added interaction terms for perceived organizational support and perceived person-job fit. The model showed an acceptable fit, χ²(df = 287) = 689.64, p < 0.01, CFI = 0.90, TLI = 0.87, RMSEA = 0.05. The interaction of calling and perceived organizational support on job involvement was significant (β = 0.26, p < 0.01), while the interaction of calling and person-job fit was not significant (β = −0.14, ns), rejecting Hypothesis 7. The model without the person-job fit interaction showed better fit indices, χ²(df = 164) = 340.11, p < 0.01, CFI = 0.94, TLI = 0.93, RMSEA = 0.04. Path estimates for this model were shown in Figure 1. Consistent with Hypothesis 6, the interactive effect of perceived organizational support and calling on job involvement was significant (β = 0.16, p < 0.05).

To interpret the nature of the interaction, we plotted the relations using a procedure recommended by Aiken et al. (1991). We plotted the relation between perceiving a calling and job involvement that corresponds to the low (one standard deviation above the mean) and the high (one standardized deviation below the mean) values of perceived organizational support (Figure 2). Our findings show that, when newcomers perceived high levels of organizational support, their calling had a positive influence on job involvement (β = 0.37, t = 3.77, p < 0.05). However, when newcomers perceived low levels of organizational support, perceiving a calling was not statistically related to job involvement (β = 0.06, t = 0.56, ns). This suggests that under high levels of perceived organizational support, calling have a greater positive influence on job involvement than in situations in which they received little support from their organization.

The results of this study are subject to several limitations that can be addressed in future research. First, our findings may be restricted by the subject sample; whether the results are generalizable to other types of employees or employees in different cultural contexts warrants investigation. Also, cultural differences are present in newcomers’ behaviors and perceptions of organizational tactics (Morrison et al., 2004). Collectivistic cultures such as South Korea are more likely to assign weight to relatedness and organizational support than in America (Kitayama et al., 2000), and this may strengthen the relation between having a calling and perceived organizational support on job involvement. Future research that includes different contexts and diverse types of employees would be valuable.

Second, although the longitudinal nature of the research can alleviate the potential for method variance, all data except for job performance were collected from self-reports, and the study variables were measured only once. Measuring the constructs at more time points and examining reciprocal influences on the variables are necessary to substantiate and clarify the findings in this study. Given that positive views toward work and organization peaked during first few weeks after organizational entry (Boswell et al., 2009), our findings based on calling measured at the very first week of organizational entry should be interpreted with caution and more longitudinal research during the career transition period is needed to strengthen the results of this study.

Finally, although the mediator and moderators in this study were chosen based on prior findings, when and why calling upon organizational entry relates to job performance needs more future research with solid theoretical background. Job performance in this study was rated by a forced distribution approach and the approach has been criticized for rewarding people with high visibility by facilitating political game playing (Guralnik et al., 2004; Schleicher et al., 2009). In a competitive environment that fosters competition and facilitates political behaviors, how people with callings react to the environment and what factors mediate the link between calling and supervisor-rated job performance during the socialization period await future research. In particular, based on a career calling model (Hall and Chandler, 2005; Praskova et al., 2014) and Self-Determination Theory (Deci and Ryan, 2000), what type of goal-directed efforts lead to job performance and how satisfaction of different types of psychological needs influence the effects of calling on job performance would be an interesting area for future research.