The Relationships among Job Crafting, Knowledge Sharing Intention, and Job Performance: Focusing on the Conditional Direct Effect of Employee Engagement

Abstract: Many studies have been conducted because job performance (JP) was considered important as one of organizational outcomes (Dyer & Reeves, 1995; Judge & Bono, 2001). According to several studies, based on the job demands resources (JD-R) model, employee engagement (EE) is positively related to JP (Demerouti et al., 2001; Demerouti et al., 2010). Specifically, the results demonstrated that EE played a mediating role in the relationship between antecedents and JP.
However, from the perspective of the conservation resource theory, given that resources and EE tend to increase reciprocally, more research needs to be accumulated (Lee, 2010). In this light, some studies have been carried out to explore the moderating effect of EE (Heo & Cheon, 2016; Lee, 2016). These studies revealed that EE moderated the relationship between the antecedents (e.g. job stress) and outcomes (e.g. organizational trust).
Considering the potential moderating of EE, this study set job crafting (JC) as an independent variable and JP as a dependent variable and tried to investigate whether EE regulates the relationship. Because while the relationship between JC and JP was directly related in some studies (e.g. Moon et al., 2018), other studies only demonstrated an indirect relationship (e.g. Tims et al., 2014). And also the relationship between JC and EE is dynamic enough to consider the various aspects of the relationship (Bakker, 2011; Bakker et al., 2012).
On the other hand, knowledge sharing intention (KSI) was set as a variable mediating the relationship between JC and JP. Knowledge sharing (KS) is a process of creating, utilizing, and sharing knowledge that affects performance, and is positively related to JC and JP (Eom, 2014; Swanson et al., 2020; Yoon et al., 2022). By the theory of reasoned action, KSI can be expected to have a positive relationship with the JC and JP like KS (Bock et al., 2005; Yoon et al., 2022).
Therefore, the research questions of this study are as follows. First, is the positive effect of JC on JP statistically significant? Second, does KSI has a mediating effect on the relationship between JC and JP? Third, is the relationship between JC and JP moderated by the EE?
The subject of this study was composed of employees of a semiconductor company in Korea (N=226). As a convenience sampling approach, self-report questionnaires were distributed and collected from employees of the semiconductor company. JC was measured by 15 items from Lee (2017). KSI was measured by 5 items from Bock et al. (2005). JP was measured by 9 items from Goodman and Svyantek (1999). And EE was measured by 12 items from Shuck et al. (2017). The 5-point Likert scale was used for JC, KSI, and EE, and the 7-point Likert scale was used for JC. The Cronbach’s α of the measurements was .879 (JC), .902 (KSI), .939 (JP), and 921 (EE) each. All correlations were less than 0.85 (p < .001), thus it was confirmed that no multi-collinearity issue (Lei & Wu, 2007).
Since the research model of this study includes a conditional effect, first, the overall fit of the measurement model was evaluated, and then the mediating and moderating effects of the research model were examined (Cheung & Lau, 2017). Regarding the measurement model, the chi-square test was statistically significant (p <.05), but other overall fit indices like CFI (≥.90), TLI (≥.90), RMSEA (≤.10), and SRMR (≤.08) met the cutoff criteria (Bae, 2014; Browne & Cudeck, 1992). Thus, it could be assumed that the measurement model was suitable for the data. The conceptual and statistical research models are as follows.
The bootstrap estimate was performed to examine the mediating effect of KSI in the relationship between JC and JP. The result demonstrated that KSI played a significant mediator in the relationship between JC and JP (ab = 0.227, SE = 0.095, 95% CI [0.032, 0.403]). And also its moderated direct effect was statistically significant (c3′ = -0.446, 95% CI [-0.712, -0.179]).
This study revealed that JC had a significant and positive effect on KSI and JP. Furthermore, the mediating effect of KSI and the moderating effect of EE in the relationship between JC and JP were statistically significant. Based on these results, companies could consider how to increase and maintain KSI and EE of their employees, which ultimately leads to encouraging their employees’ JP. In particular, it suggests that company’s strategic intervention is effective for employees with low or middle EE levels.