What Leads Entrepreneurial Employees to Want to Quit, or Stay in, Their Job? Exploring Two Conflicting Mechanisms

Past turnover research has posited personality traits as important antecedents to quit intentions. Nevertheless, previous literature has not investigated the relationship between employees’ entrepreneurial tendencies—a constellation of domain specific traits—and turnover. Drawing on dispositional trait theory and attraction-selection-attrition theory, we propose engagement and intentions to start a business as mediators of the relationship between entrepreneurial tendencies and quit intentions. We test our predictions in a sample of full-time employees from the United Kingdom (N = 224). In line with our hypotheses, an inconsistent mediation is found, where both positive and negative links between entrepreneurial tendencies and turnover intentions were mediated by engagement and intentions to start a business respectively. Thus, entrepreneurial employees were more likely to be engaged, but at the same time also more likely to be considering starting their own business, leading to a conflicting relationship to turnover intentions. The current study informs the human resource management literature concerning the motivational mechanisms explaining turnover intentions among entrepreneurial employees. It also provides practical insights with regards to the effective management of this workforce.


INTRODUCTION
Employee turnover has been linked to a multitude of negative organizational outcomes, including decreased productivity (Hausknecht, Trevor, & Howard, 2009) and decreased competitive advantage (via human or social capital losses or trade secret theft; Agarwal, Ganco, & Ziedonis, 2009). For organizations, understanding the antecedents of quitting intentions among employees is, This hypothesized complex interplay between entrepreneurial tendencies and intentions to quit is likely to reveal valuable insights with regards to the management and retention of entrepreneurial employees within firms. Given the importance of entrepreneurial behavior within organizations (Rauch et al., 2009), such insights should be of considerable interest to managers and human resource professionals. However, to our knowledge, few studies have examined these relationships empirically. The aim of the current study was to take an initial step in this direction, and fill what seems to be an important gap in the literature. The next section describes the key variables examined in this study and outlines the theoretical foundations of the research. This is followed by a description of the data and methods. Results are then presented and discussed. We conclude with a summary of the key findings and their implications, together with the study's limitations and suggestions for future research.

Entrepreneurial Tendencies
Entrepreneurial tendencies have been defined as enduring psychological and behavioral tendencies, or traits, related to recognizing and exploiting opportunities, innovating, and creating change (Ahmetoglu & Chamorro-Premuzic, 2017;Shane & Venkataraman, 2000). Ahmetoglu and Chamorro-Premuzic (2017) draw a conceptual distinction between entrepreneurial tendencies and the tendencies of entrepreneurs, where the latter describes observed behaviors or psychological characteristics common to an occupational group (i.e., entrepreneurs), whereas the former describes a multidimensional personality construct, or a constellation of traits (Ahmetoglu & Chamorro-Premuzic, 2017). Grounded in dispositional trait theory (Funder, 2001), this individual differences framework views entrepreneurial tendencies as stable patterns in thoughts, feelings, and behaviors. Accordingly, how entrepreneurial a person is, or their "trait standing" (Fleeson & Gallagher, 2009), is a function of the frequency with which entrepreneurial thoughts, feelings, and behaviors are manifested by that individual (Ahmetoglu & Chamorro-Premuzic, 2017). For instance, a person who frequently recognizes and pursues commercial opportunities, and innovates and creates value (i.e., manifests trait related behaviours; Fleeson & Gallagher, 2009), would be considered highly entrepreneurial, whereas a person who engages in these behaviors (and thoughts/feelings) less frequently would be considered less entrepreneurial. Within this framework, therefore, a manager (or employee) may be considered highly entrepreneurial (i.e., opportunistic and innovative), whilst an entrepreneur (i.e., a founder of a company) may not be very entrepreneurial at all (e.g., displaying no desire for change or value creation; Ahmetoglu & Chamorro-Premuzic, 2017;Shane, 2008). 1 Ahmetoglu and Chamorro-Premuzic (2017) also distinguish between general entrepreneurial tendencies, contextual behavior, and outputs. General tendencies are broad psychological and behavioral patterns, whereas contextual behaviors are manifestations of general tendencies in specific situations (Fleeson & Gallagher, 2009). Output is the consequence of behaviors (e.g., doing research and drafting are behaviors that may or may not result in "tangible" outputs like a prototype for a product, or a filed patent; Bird, Schoedjt, & Baum, 2012). Thus, in this conceptual framework, concepts such as intrapreneurship would refer to contextual behavior and outputs, but not necessarily to (general) entrepreneurial tendencies, because the behavioral manifestation (and the output) of general tendencies are partially determined by context (cf. Ahmetoglu & Chamorro-Premuzic, 2017 for a detailed distinction between these concepts). For instance, a highly entrepreneurial employee (in terms of general tendencies) who finds her/himself in a constrained work environment (e.g., a highly standardized factory job), may not engage in much intrapreneurship, and fail to produce any relevant (intrapreneurial) outputs. We focus on entrepreneurial tendencies, as we are interested in the reasons why these valuable (entrepreneurial) employees may choose to leave, or stay, in their incumbent organization, rather than when, or under what contexts, they engage in intrapreneurship.
Past research has shown that entrepreneurial tendencies are empirically distinct from the Big Five personality traits (Leutner, Ahmetoglu, & Chamorro-Premuzic, 2014), as well as a number of other established psychological constructs (see . These tendencies have also been found to incrementally predict both context-specific entrepreneurial behavior (e.g., seeking investment for an invention, initiating events), as well as output (e.g., patents registered, inventions sold, number of new products and services introduced, etc.) across a variety of organizational contexts (e.g., Ahmetoglu, Leutner, & Chamorro-Premuzic, 2011;Almeida et al., 2014). Importantly, entrepreneurial tendencies have been positively related to various behavioral and performance metrics within organizations, including employee intrapreneurship (e.g., implementing new methods or changing organizational procedures to accomplish tasks), social capital, salary attainment, and job performance (Ahmetoglu et al., 2019;Ahmetoglu, Scarlett, Codreanu, & Chamorro-Premuzic, in press;Akhtar & Kang, 2016;Almeida et al., 2014;Leutner et al., 2014). This line of evidence demonstrates that entrepreneurial employees can be highly valuable for organizations.
The notion that entrepreneurial employees can benefit organizational competitiveness is both theoretically and empirically substantive (Ahmetoglu et al., 2019;Gawke et al., 2017;Rauch et al., 2009). However, there is reason to expect these incumbents to also be more likely to want to leave, or quit, their current job than their less entrepreneurial colleagues (Cromie & Hayes, 1991), especially if they are disengaged. In this respect, we propose that there is likely to be a complex and significant relationship between entrepreneurial tendencies and intentions to quit, with two conflicting motivational mechanisms at play: employee engagement and intentions to start a business. We argue that for managers and human resource professionals to be able to manage entrepreneurial employees in an effective way, understanding the influence of, and relationship between, these conflicting mechanisms is critical. We elaborate on this complex interplay below.

The Mechanisms Relating Entrepreneurial Tendencies and Quit Intentions
An accumulating number of studies in the past 10 years have demonstrated that engagement is one of the most important antecedents of employee turnover (Harter et al., 2002). Schaufeli and colleagues (2002) defined engagement as a "positive, fulfilling, work-related state of mind" (p. 74) that can be viewed as a combination of vigor, dedication and absorption; thus, workers who are more engaged are more energetic at work, more dedicated to their job, more absorbed in the tasks of the job. These motivational states, in turn, make employees more likely to want to preserve their positive state and, consequently, far less likely to want to part with their job. Employee engagement, therefore, has been recognized as an important goal for organizations, and widely studied in the human resources and psychology literature (Bailey, Madden, Alfes, & Fletcher, 2017;Shuck, 2011).
We argue, however, that engagement can also shed light on an intriguing nature of entrepreneurial employees and their existence within established organizations. We outline two reasons for this. First, there is both a theoretical and empirical literature to implicate a positive relationship between entrepreneurial tendencies and work engagement. For instance, entrepreneurial individuals have been found to have a more positive and optimistic mindset, be more proactive and absorbed at work, and have a higher self-regard (cf. Ahmetoglu et al., 2011;Akhtar, Ahmetoglu, & Chamorro-Premuzic, 2012). Employees who are self-efficacious and have a positive and optimistic mindset, in turn, are found to be "most likely to experience high levels of work engagement" (Xanthopoulou, Bakker, Demerouti, & Schaufeli, 2009, p. 241). Supporting these theoretical links, recent research has found a positive and moderately strong relationship between entrepreneurial tendencies and engagement at work (Ahmetoglu et al., , 2018. Second, it has been hypothesized that employees engaging in intrapreneurship increase their personal resources over time, which in turn foster higher levels of engagement at work (Bakker & Demerouti, 2017). In line with this theoretical perspective, employee intrapreneurship has been shown to lead to increases in work engagement (Gawke et al., 2017). Given that entrepreneurial employees are more likely to engage in intrapreneurship, there may be a positive feedback loop whereby intrapreneurship among entrepreneurial employees also leads to them feeling more engaged. Thus, a positive link between entrepreneurial tendencies and work engagement is expected, despite the likely existence of moderators that may attenuate this relationship (e.g., Gawke, Gorgievski, & Bakker, 2018). Given that employee engagement is one of the most important antecedents of turnover (Harter et al., 2002), it would be sensible to expect, therefore, a negative relationship between entrepreneurial tendencies and intentions to quit, mediated by employee engagement. That is, the positive relationship between entrepreneurial tendencies and work engagement should decrease entrepreneurial employees' intentions to quit their current job.
Moreover, there is theoretical reason to conjecture that entrepreneurial employees have a stronger desire to start their own business than less entrepreneurial incumbents. Theories of vocational choice indicate that personal dispositions influence which environments people prefer and tend to inhabit (Armstrong et al., 2011). Such "niche-picking" behaviors are crucial for understanding career choices and their associations to individual differences. For instance, attraction-selection-attrition theory (ASA; Schneider, 1987) suggests that individuals gravitate toward working environments (occupations and jobs) that are congruent with their personalities and are more likely to exit those environments that are not. This is relevant for understanding motivational mechanisms of entrepreneurial employees. Research demonstrates that self-employment is perceived as an occupation that offers autonomy to pursue one's vision, to innovate, and exploit opportunities (Baum, Frese, & Baron, 2007). One could argue, therefore, that the perceived job and environmental characteristics of self-employment are congruent with the manifestation of entrepreneurial tendencies, and that entrepreneurial employees are, consequently, more likely to "gravitate" toward self-employment than less entrepreneurial incumbents. Given that selfemployment is almost always associated with quitting one's job, however, this means that an entrepreneurial employee who is attracted towards selfemployment is, by definition, more likely to consider leaving their current job.
In line with the above arguments and previous research, therefore, one may expect conflicting links between entrepreneurial tendencies and intentions to quit. On the one hand, entrepreneurial tendencies are expected to correlate positively with levels of engagement, which, in turn, are predicted to correlate negatively with intentions to quit. On the other hand, entrepreneurial tendencies are expected to correlate positively with one's intentions to start a business, which are in turn expected to correlate positively with intentions to quit. Therefore, one may expect both negative and positive links between entrepreneurial tendencies and intentions to quit, with the negative link mediated by engagement, and the positive link mediated by intentions to start a business.
Despite the importance of understanding these relationships, however, few studies have examined them. Most previous research has focused on investigating individual traits that are likely to enhance employees' intentions to want to venture out on their own (Douglas & Fitzsimmons, 2013;Zhao, Seibert, & Lumpkin, 2010), or traits that relate to turnover intentions (Griffeth, Hom, & Gaertner, 2000;Zimmerman, 2008). Yet, neither literature directly addresses the question of what causes entrepreneurial employees to want to leave, or remain in, their jobs. For instance, meta-analytic studies have established clear links between the Big Five personality traits and both start-up intentions as well as turnover intentions (Zhao et al., 2010;Zimmerman, 2008). However, the Big Five assess very broad psychological and behavioral tendencies (i.e., traits); deducing domain-specific tendencies-such as a person's predisposition to recognize and exploit opportunities-from this research, is both theoretically and empirically problematic (Ahmetoglu & Chamorro-Premuzic, 2017). Research investigating the relationship between these domain-specific traits and turnover intentions is lacking from the literature. Furthermore, existing research did not investigate the mechanisms through which entrepreneurial tendencies and quit intentions are linked. Yet, we argue that for managers and human resource professionals to be able to manage entrepreneurial employees in an effective way, understanding these mechanisms is critical because this relationship may be complex and mediated by conflicting motivational mechanisms.
Thus, there seems to be a critical gap in the human resource management literature in terms of the potential motivational causes for turnover amongst entrepreneurial employees. We feel that such research is needed in order to bridge gaps in evidence and theory concerning the retention of these valuable employees within incumbent organizations (Rauch et al., 2009). Accordingly, the present study aims to empirically investigate the relationship between individual differences in entrepreneurial tendencies, employee engagement, intentions to start a business, and their influence on intentions to quit one's job. Based on the arguments above, it is hypothesized that: Hypothesis 1: Entrepreneurial tendencies will be significantly and positively correlated with engagement. Hypothesis 2: Engagement will be significantly and negatively correlated with intentions to quit.
Hypothesis 3: Entrepreneurial tendencies will be significantly and positively correlated with intentions to start a business.
Hypothesis 4: Intentions to start a business will be significantly and positively correlated with intentions to quit.
Hypothesis 5: There will be both a positive and a negative indirect link between entrepreneurial tendencies and intentions to quit, with the negative link being mediated by engagement, and the positive link being mediated by intentions to start a business.
We summarize our proposed theoretical model in Figure 1. We do not hypothesize a direct relationship between entrepreneurial tendencies and intentions to quit, as the inconsistent mediation effect is likely to result in a non-significant overall relationship between these two variables (MacKinnon, Fairchild, & Fritz, 2007).

METHOD participants
The research team contacted employees from a range of organizations in the UK and Europe, based on both personal and university contacts. In total, 224 participants (109 male, 115 female) agreed to take part in the study. The FIGURE 1. Hypothesized theoretical model mean age was 35.9 (SD = 9.1), ranging from 20 to 64 years. Participants were mostly from the UK (75.4%) or other European countries (14.3%); 85.7 percent of the participants were employed full-time (the rest were employed part-time), with 42.4 percent of them in employee roles, and 53.1 percent in managerial roles. Participants came from a wide range of industries including business (21.4%), technology (17%), banking (10%), and creative industry (7.1%). The mean income of the sample ranged from £40,000 to £60,000 pa before tax deduction.

Measures
Utrecht Work Engagement Scale-Short Version (UWES; Schaufeli, Bakker, Salanova, 2006). The Utrecht Work Engagement Scale-Short Version (UWES; Schaufeli, Bakker, & Salanova, 2006) is a 9-item self-report scale that assesses three aspects of work engagement: Vigor (3 items; e.g., "At work, I feel full of energy"), Dedication (3 items; "I am enthusiastic about my job"), and Absorption (3 items; "I feel happy when I am working intensely"). Participants responded using a 7-point Likert scale ranging from 0 (Never had this feeling) to 6 (Always). The scale is an accepted short version of UWES-17, which has shown good reliability and factorial validity (Schaufeli, Martínez, Marques-Pinto, Salanova, & Bakker, 2002), as well as cross-cultural validity (Shimazu, Schaufeli, Kosugi et al., 2008). Ahmetoglu et al., 2011). Entrepreneurial tendencies were assessed though a selfreport inventory, the Measure of Entrepreneurial Tendencies and Abilities (META; Ahmetoglu et al., 2011), consisting of 65 items that measure the four dimensions of entrepreneurial tendencies: the tendency to recognize opportunities: Opportunism (11 items, e.g., "I try to take advantage of every profitable opportunity I see"); the tendency to exploit opportunities: Proactivity (17 items, e.g., "I am quick to spot profitable opportunities"); the tendency to innovate: Creativity (16 items, e.g., "In groups, I usually have the most innovative ideas"); and the desire to create value and change: Vision (21 items, e.g., "I want to make a difference in the world"). The four dimensions were derived from a content analysis of themes on the dimensionality of entrepreneurship that were the most recurring in the literature and able to be integrated within an individual differences framework (cf. . Participants are instructed to rate themselves using a 5-point Likert scale that ranges from "Completely Disagree" (1) to "Completely Agree" (5). Given the inventory showed good measurement properties (see Tables 1 and 2   Intentions to Start a Business. This scale comprised three items that measured the extent to which participants were determined to create a new venture in the future, how much effort they would make to start and run their own business, and whether entrepreneurship was their professional goal. Participants used a 5-point Likert scale to respond to the statements, ranging from 1 (Total disagreement) to 5 (Total agreement). Similar items have been used in previous research to measure intentions to start a new business (e.g., Liñán & Chen, 2009).

Intentions to Quit.
Intentions to quit the current workplace were measured by a 3-item scale, adapted from previous research (Bozeman & Perrewe, 2001;Poon, 2004). Items included "I feel like quitting my job," "I have been looking for another job recently," and "I am reluctant to change my job." Participants responded using a 5-point Likert scale, ranging from "Total disagreement" (1) to "Total agreement" (5). The 3-item scale used to measure intentions to quit has shown good reliability and validity in the past (e.g., Poon, 2004). procedure Participants completed the survey online, through a website advertised on various social media websites. Participants first completed the work engagement scale, followed by the intentions to start a business and intentions to quit scales; they then completed META. On average, the survey took 10-15 minutes to complete.

REsULTs
Descriptive statistics, internal consistencies, and bivariate correlations for all measures are shown in Table 1. All scales that were used in the study demonstrated good internal consistencies (Cronbach's alpha values above 0.7 are considered appropriate; George & Mallery, 2003). The correlation between META and intentions to quit was non-significant. There was a significant positive correlation between META and engagement. The correlation between engagement and intentions to quit was significant and negative. META also significantly and positively correlated with intentions to start a business. Finally, new business intentions correlated significantly and positively with intentions to quit. Although, for full mediation, there should be a significant correlation between META and intentions to quit (Baron & Kenny, 1986), a test of inconsistent mediation (a mediation when at least one mediated effect has a different sign than the other mediated or direct effect in the model, resulting in a non-significant relationship between the initial and the outcome variable; MacKinnon et al., 2007) may still be tested. Accordingly, this was done using structural equation modeling.

structural Equation Modeling (sEM)
Structural equation modeling was carried out using AMOS 5.0 (Arbuckle, 2003). The choice of ordering is rarely straightforward in SEM (Davis, 1985;Kenny, 1979;Pearl, 2000), and a predictive rather than causal model was tested, primarily to provide a general picture of the relationship between target variables.
The 12 observed variables included in the hypothesized model (META dimensions, engagement factors, intentions to start a business, and intentions to quit) were modeled as shown in Figure 2. META and engagement were modeled as latent variables (given intercorrelations between the META sub-dimensions and engagement sub-dimensions), whereas intentions to start a business and intentions to quit were treated as observed variables through the use of item parcelling. We used item parcelling given the unidimensionality of the measurement instruments for these two variables (Kishton & Widaman, 1994;Sass & Smith, 2006), and in order to increase the parsimony of the model. The loadings of observed variables onto latent variables are shown in Table 2. Finally, a correlational path between engagement and intentions to start a business was specified. The model's goodness of fit was assessed via the χ 2 statistic (Bollen, 1989; tests the hypothesis that an unconstrained model fits the covariance or correlation matrix as well as the given model; ideally values should not be significant); the goodness of fit index (GFI; Tanaka & Huba, 1985; a measure of fitness, where values close to 1 are acceptable); the comparative fit index (CFI; compares the fit of a target model to the fit of an independent model-a model in which the variables are assumed to be uncorrelated; values greater than 0.95 indicate a very good fit; Bentler, 1990); and the root mean square residual (RMSEA; Browne & Cudeck, 1993; values of 0.08 or below indicate reasonable fit for the model).
In the first model we tested, we controlled for the effects of age, gender, managerial status, and industry on the mediators (employee engagement and intentions to start a business) and dependent variable (intentions to quit). Whereas this model fit the data moderately well, χ 2 = (153 df, p < .001) 339.67, GFI = 0.89, CFI = 0.86, RMSEA = 0.074, none of the coefficients associated with the control variables was significant. The second model thus excluded the control variables, and included five paths between exogenous and endogenous variables. This model fit the data well: χ 2 = (60 df, p < .01) 116.43, GFI = 0.93, CFI = 0.97, RMSEA = 0.065 (0.05, −0.08). The indirect effect of META on intentions to quit was tested using a percentile bootstrap estimation approach with 5,000 samples (Hayes, 2018).

DIsCUssION
The current study fills an important gap in the human resource management literature concerning the motivational mechanisms for turnover intentions among entrepreneurial employees. Although past research has shown that psychological traits influence quit intentions (Griffeth et al., 2000), few studies have investigated whether and in what way employee's entrepreneurial tendencies relate to such intentions. Given the importance of entrepreneurial employees for organizational performance and intrapreneurship (Ahmetoglu & Chamorro-Premuzic, 2017), this research is a timely addition to the literature. Based on previous theory and research, we hypothesized an inconsistent mediation, with entrepreneurial tendencies being both positively and negatively linked to turnover intentions through two motivational mechanisms, namely engagement and intentions to start a business. The results supported the hypotheses of the study.
In line with our predictions (Hypothesis 1), entrepreneurial tendencies were found to be positively related to employee engagement, indicating that more entrepreneurial employees are also more engaged at their current workplace. In some respects, this finding may seem counterintuitive, given popular perception that entrepreneurial employees are difficult to manage (Sibel, 2012). Yet, we believe there is both theoretical and empirical reason to expect such results. First, it is well known that personality traits are intricately linked with affective and motivational constructs, such as subjective wellbeing (Steel, Schmidt, & Schultz, 2008) and job satisfaction (Judge, Heller, & Mount, 2002), stemming from individual differences in dispositional levels of affect and information processing (Diener, Oishi, & Lucas, 2003). In a similar vein, engagement has been found to be, to a sizeable extent, a function of a person's dispositional traits (Halbesleben, 2010). Given that previous studies have shown entrepreneurial tendencies to be associated with personality traits related to engagement (e.g., core self-evaluations, emotional stability, extraversion; Ahmetoglu et al., 2011;Leutner et al., 2014), theoretically it is reasonable to expect entrepreneurial employees to be "dispositionally" more engaged at work. Few studies have, however, empirically investigated this theoretical supposition. Our results thus contribute valuable information to the literature on the intrinsic work motivation of entrepreneurial employees, which may often be overlooked, or misconceived (Sibel, 2012).
Consistent with the second hypothesis of the study (Hypothesis 2), more entrepreneurial employees also reported having significantly higher intentions to start a business. That is, opportunistic, proactive, creative, and visionary (i.e., entrepreneurial) tendencies fuel employees' desires to venture out and start their own business. This finding is in line with ASA theory (Schneider, 1987) and supports the notion that self-employment is perceived as a desirable, or congruent, occupational environment for employees with higher entrepreneurial tendencies (Shane, 2008). In that sense, it is also consistent with the common perception that, on an aggregate level, incumbent organizations may not be perceived as work environments conducive to the manifestation of trait-related entrepreneurial behaviors (such as innovating, exploiting opportunities, and pursuing one's vision), at least not as much as selfemployment is perceived to be (Baum et al., 2007). Our findings also build on past personality-start-up intentions literature (Lee, Wong, Foo, & Leung, 2011;Zhao et al., 2010), demonstrating a direct relationship between employees' entrepreneurial tendencies, conceptualized as a constellation of individual traits (Ahmetoglu & Chamorro-Premuzic, 2017), and start-up intentions.
Finally, we found full support for the main hypothesis of the study (Hypothesis 5). First, in line with previous research, both engagement (positively; Hypothesis 3) and intentions to start a business (negatively; Hypothesis 4) were found to relate to turnover intentions. Critically, a significant and inconsistent mediation was found, where both positive and negative links between entrepreneurial tendencies and turnover intentions were mediated by engagement and intentions to start a business. Thus, our findings suggest that entrepreneurial employees are simultaneously residing with both motivational pull and push factors with regards to turnover intentions within organizations. Moreover, they are more likely to be engaged and, consequently, want to remain with their organization. Yet, at the same time, they are also more likely to consider starting their own business and, therefore, to leave their current jobs. These results shed light on an intriguing relationship between entrepreneurial tendencies and turnover intentions, which are likely to have important implications for human resource management research and practice.

Implications
Turnover is associated with numerous negative organizational outcomes, particularly when it relates to employees who provide a competitive advantage to the organization (e.g., Hausknecht et al., 2009). It is of critical importance, therefore, to understand the factors that positively or negatively influence turnover intentions among entrepreneurial employees. Indeed, the intricate relationships between dispositional tendencies and resulting motivations found in our study, provide a far more nuanced understanding of the nature of entrepreneurial employees and, therefore, the strategies that may or may not be effective for managing and retaining them.
First, our results suggest that entrepreneurial employees are likely to benefit organizations not only directly, through intrapreneurship (Akhtar et al., 2012;Almeida et.al., 2014) and job performance (Ahmetoglu et al., 2019, but also indirectly, by being more energetic, dedicated, and absorbed at work (Leutner et al., 2014;Schaufeli & Bakker, 2004). This suggests that organizations hiring entrepreneurial employees can gain a competitive edge associated with the motivational and behavioral manifestations of this workforce. Paradoxically, however, entrepreneurial tendencies may have a detrimental impact on organizational functioning because they fuel employees' intentions to start their own business, thereby increasing the "flight risk" of employees with these personal tendencies.
It can be conjectured from these results, therefore, that critical tasks for human resource professionals include understanding, predicting, and managing the direction of the impact (i.e., whether positive or negative) of entrepreneurial tendencies on these motivational outcomes. The results of the current study show that the negative co-variation between the two mediators is of key importance. Specifically, our findings suggest that entrepreneurial employees who are more engaged within organizations are less likely to want to venture out on their own and therefore leave their job. Conversely, those who are less engaged are more likely to consider starting their own business and eventually want to leave their job.
From a practical perspective, these results indicate that selecting entrepreneurial employees without appropriate management of their engagement levels may be unwise. Thus, it may be necessary to consider organizational processes and structures that influence engagement levels within the company, before selecting or promoting entrepreneurial employees. Of particular interest would be those processes and structures that may negatively impact engagement levels of entrepreneurial employees-given that these employees tend to have higher "baseline levels" of engagement. This is also a potent area for future research. Although the literature on the antecedents of engagement is abundant (Saks, 2006), less is known about the factors that may moderate the relationship between entrepreneurial tendencies and engagement.
Another practical consideration involves the environmental cues that organizations present to entrepreneurial employees. Employees can only express their dispositional tendencies and preferences if relevant situational cues within the organization are present, and forces that prohibit these tendencies are absent (Tett & Gutternman, 2000). Accordingly, designing environments in a way that allows and enables entrepreneurial employees to enact their dispositional tendencies is also likely to minimize their intentions to venture out to other environments (such as self-employment), provided that their natural inclinations are satisfied at work. The literature on innovation ecosystems offers some suggestions in terms of designing organizational environments promoting innovative performance (Oh, Phillips, Park, & Lee, 2016), but research on the interaction between entrepreneurial tendencies and innovation ecosystems is still in its infancy (see Ahmetoglu et al., 2018). This may not be an easy task, but consideration of such factors is warranted, should retaining entrepreneurial employees be a strategic organizational goal.

Limitations and Future Research
One limitation of the present study was the cross-sectional nature of the data. Future research should use longitudinal data, as this would clarify the causal direction of the relationship between engagement and intentions to start a business, and its relationship to intentions to quit. However, from a theoretical standpoint, it is more likely that engagement precedes intentions to quit than intentions to quit causing engagement. Similarly, entrepreneurial tendencies are more likely to precede intentions to starts a business rather than the other way around, given the dispositional nature, stability, and cross-situational consistency of personality traits. Second, the current study did not account for external and situational factors that may influence engagement, new business intentions, and intentions to quit. It has been shown that occupational choice is not only determined by motivational factors (such as engagement), but also "reality factors" such as unemployment and capital (Shane, 2008); therefore, some control over these factors should be exerted in future investigations.

Conclusion
The current study contributes to the applied psychology literature by highlighting two conflicting motivational mechanisms explaining the relationship between entrepreneurial tendencies and turnover intentions. Moreover, entrepreneurial employees "naturally" experience higher levels of job engagement and, consequently, desire to remain within the incumbent organization. At the same time, they are more likely to want to start their own business and, therefore, more likely to want to leave their current jobs. These results reflect a paradoxical, yet intriguing nature of entrepreneurial employees-a workforce that can be likened to a double-edged sword within organizations. In particular, the current study emphasizes the importance for human resource professionals to consider appropriate management strategies that reflect these conflicting internal motives of entrepreneurial employees.