From job crafting to job quitting? Testing a wise proactivity perspective

Purpose – This study investigates when and for whom job crafting may turn into job quitting. The authors hypothesize that approach job crafting relates more positively to turnover intentions and subsequent voluntary job changes among employees with (a) high (vs low) need for career challenges and (b) those with high (vs low) self-esteem. Design/methodology/approach – Data were collected from 575 employees of a large public organization in the Netherlands with two measurement moments three months apart. Hypotheses were tested using cross-lagged regression analyses and path modeling. Findings – Supporting the hypotheses, approach crafting related positively to an increase in turnover intentions only among employees with high need for challenge or high self-esteem. Moreover, via turnover intentions at Time 1, approach crafting related positively to the voluntary job change at Time 2 for employees with (a) high need for challenge, as wellas those with (b) high self-esteem.These findings heldafter controlling for avoidance crafting. Research limitations/implications – This study has been conducted in a relatively homogenous sample. Futureresearchmaytestthepredictionsinamoreheterogeneoussample,includingparticipantsfromdifferent culturalandeconomiccontexts. Practicalimplications – The authorsadvisehuman resource(HR)professionalstofacilitatethe jobcrafting efforts of employees with a high need for challenge and those with high self-esteem because these groups are particularly at risk of voluntarily quitting their jobs. Adopting insights from the wise proactivity model may help ensure that job crafting benefits both employees and employers. Originality/value – This study brings clarity to the inconsistent relationships between job crafting and job quitting by using the wise proactivity model as an

From job crafting to job quitting?Testing a wise proactivity perspective Job crafting refers to employees' self-initiated changes in the task, relational and cognitive boundaries of work (Tims and Bakker, 2010;Wrzesniewski and Dutton, 2001).Scholars have identified various behaviors through which employees craft their jobs (Bruning and Campion, 2018;Kooij et al., 2017;Tims et al., 2012;Zhang and Parker, 2019).These include approachoriented behaviors, such as engaging in professional development activities, but also avoidanceoriented behaviors through which employees try to ensure that their work is less demanding.In the past ten years, there has been a large uptake in empirical research on job crafting.Several meta-analyses have been published on this topic (Lazazzara et al., 2020;Lichtenthaler and Fischbach, 2019;Oprea et al., 2019;Rudolph et al., 2017), indicating that approach-oriented (but not avoidance-oriented) job crafting behaviors are positively associated with organizationally relevant outcomes, such as job performance and employee well-being.
Comparably less research exists on the relationships between job crafting and job quitting (e.g.turnover intentions), albeit staff retention is another important criterion for human resource management (HRM).The relative absence of empirical research on the links between job crafting and job quitting is surprising, given that approach-oriented job crafting is often portrayed as a "panacea" to various undesirable job situations.For example, job crafting has been proposed to be a helpful strategy for employees who experience person-job misfit (Vogel et al., 2016), overqualification (Debus et al., 2020), job insecurity (Wang et al., 2018), as well as aging-related challenges (Kooij et al., 2015(Kooij et al., , 2022)).In recent years, job crafting has also increasingly been studied as a proactive strategy that allows employees to advance in their careers (Akkermans and Tims, 2017;Janssen et al., 2021).While the relation between job crafting and turnover intentions has been explored in the past, this research has been inconclusive.This may be due to methodological (e.g.cross-sectional design) and conceptual limitations (e.g.lack of moderators), as well as due to the specific study contexts.
A meta-analysis found that quit intentions were positively associated with avoidance job crafting but not consistently related to approach job crafting (Rudolph et al., 2017).Some more recent studies found a negative association between approach crafting and quit intentions (e.g.Zhang and Li, 2020), while other studies found that some dimensions of approach crafting (e.g.role expansion) are positively associated with quit intentions (e.g.Bruning and Campion, 2018).Additionally, some studies reported non-significant relationships (e.g.Debus et al., 2020).What limits existing studies is that they base conclusions on cross-sectional analyses, which makes it hard to rule out alternative explanations.To clarify the relationship between approach crafting and quit intentions, we adopt a cross-lagged design, which allows for investigating prospective effects and can overcome several methodological artefacts (Kenny, 1975;Orth et al., 2022).
Besides methodological limitations, the inconsistent relationships between approach crafting and retention outcomes may point towards potential moderators.Here, we aim to clarify relationships based on the wise proactivity framework (Parker et al., 2019;Parker and Liao, 2016).This framework calls for examining how motivational and contextual factors moderate the effects of proactive behaviors, such as approach crafting (Zhang and Parker, 2019).On the one hand, approach crafting may help employees to alleviate undesirable situations that may cause them to quit their jobs (e.g.Debus et al., 2020).In this case, approach job crafting is expected to decrease turnover intention.On the other hand, it is conceivable that approach job crafting is motivated by a desire to advance in one's career (Janssen et al., 2021).In this scenario, approach-oriented job crafting may increase turnover intentions and may be used as a strategy to prepare for future job changes (Bruning and Campion, 2018).
In the current manuscript, we focus on the role of (a) career needs and (b) self-esteem as potential moderators to clarify the relationships between approach job crafting and job quitting.Career needs and self-esteem can offer insights to the motivational processes that Job crafting and job quitting may determine whether approach crafting translates into job quitting.By doing so, the present study makes three contributions to the literature.First, we examine prospective effects of job crafting on retention outcomes, namely quit intentions and voluntary job changes.The cross-lagged design solidifies existing cross-sectional findings on the role of avoidance crafting in the job quitting process (Rudolph et al., 2017) and clarifies the role of approach crafting in this process (Bruning and Campion, 2018).Second, we illuminate who, in terms of career needs and self-esteem, is most likely to transition from approach crafting to job quitting.Thereby, we contribute to advancing the wise proactivity perspective on job crafting (Parker and Liao, 2016).Finally, we discuss the possibility that job quitting may be wise for job crafters who wish to accumulate social capital and increase their employability (see also Ng and Feldman, 2010).As such, the present study paints a less daunting view on job quitting from the employee perspective.In contrast, from the employer perspective, we challenge the assumption of job crafting scholars who suggest that approach crafting primarily brings positive outcomes for organizations (e.g.financial returns; Oprea et al., 2019).Hence, this study reveals a potential dark side of approach crafting with regards to employee retention outcomes.Our findings may offer inspiration and preliminary insights on the role of job crafting in more recently evoked phenomena, such as the "great resignation" and "quite quitting" (e.g. Lee et al., 2023), highlighting the need for further research on this topic.

Theoretical background
A wise proactivity perspective on job crafting and job quitting Parker and Liao (2016) first introduced the concept of wise proactivity in their discussion of how employees can take initiative to build personalized career trajectories.Subsequently, Parker et al. (2019) refined the notion of wise proactivity in an in-depth review of the literature on proactive behavior and applied psychological theories of wisdom in their analyses.While proactivity refers to self-starting and future-focused action to change aspects of the self and/ or the (work) environment (Grant and Ashford, 2008;Parker and Collins, 2010), wisdom is about effectively managing the tensions, delicates, or paradoxes within a situation (e.g.Sternberg, 2004).Applying the concept of wisdom, Parker et al. (2019) defined wise proactivity as personal initiatives that take into account (a) one's own interests, (b) the interests of others, as well as (c) the wider task and strategic organizational context.We contend that the concept of wise proactivity may be useful in the analyses of approach job crafting in the voluntary job exit process (Klotz et al., 2021).
First, employee turnover scholars have emphasized that the decision to leave one's job is commonly preceded by several tensions, requiring 'wise' considerations of various perspectives (Bolt et al., 2022).For instance, the unfolding model of voluntary turnover (Lee et al., 1999) highlights that employees go through several phases until they decide to quit their jobs.The importance of balanced consideration is also highlighted by the job embeddedness model (Mitchell et al., 2001), which proposes that employees' connections to their job, coworkers, the organization and their community makes it more difficult for them to leave.The notion of dialects and tensions is also clearly visible in the early theory of Mobley et al. (1979), who introduced various factors that either "pull" individuals to remain in their job or "push" them to find alternative employment.Thus, decisions on whether to continue job crafting or to initiate job quitting require employees to take into account various interests, contexts and considerations (i.e.wisdom; Sternberg, 2004).
Second, we contend that the notion of wise proactivity may be useful to further advance research on job crafting and to clarify the link with job quitting.Given the potential impact that job change decisions may have on one's life, it may not be surprising to observe that job quitting is also commonly preceded by job crafting.For example, an employee working in a marketing agency may initially engage in job crafting to pursue a personal interest in corporate social responsibility (CSR).Eventually, this employee may realize that their current job as a marketing manager does not offer many opportunities to be engaged in CSR activities.Noticing the limits of job crafting, this employee may eventually decide to quit their job as a marketing manager to pursue their passion for CSR.Hence, approach crafting may (1) create events that can make employees reflect on job quitting and (2) serve as preparation for future job moves.Hence, by engaging in prior crafting, employees may cultivate the wisdom needed to guide them in their decision to leave (vs stay) in their current job.
Our proposed wise proactivity perspective may help to advance the emerging literature on job crafting (e.g.Tims et al., 2022;Zhang and Parker, 2019), as well as more long-standing employee turnover theory (e.g.Bolt et al., 2022;Hom et al., 2017).Job crafting scholars increasingly recognize that more attention must be paid to the contexts in which crafting takes place (e.g.Tims and Parker, 2020), as well as the characteristics of the job crafter and their initiatives (e.g.Fong et al., 2021).The wise proactivity perspective addresses this call because it requires scholars to explicate where and by whom crafting is initiated.Furthermore, integrating the job crafting literature with turnover theory can bring important insights to how "employees craft their exit transitions" (Klotz et al., 2021, p. 137).In what follows, we introduce the context of the present study before developing our hypotheses.This allows us to explicate how approach crafting may relate to retention outcomes (Rousseau and Fried, 2001).

Context of the present study
The present study was initiated in the context of the organization's growing awareness that excessive job embeddedness may demotivate civil servants from investing in continued human capital development (see also Ng and Feldman, 2010).Specifically, this study investigates the relations between approach crafting and job quitting among employees of a large public organization in The Netherlands.Employees in this organization may be considered as more embedded than the average Dutch worker because they receive various benefits as civil servants.Furthermore, the sample is characterized by relatively high levels of job security and high job tenure.Hence, the organizational context may "pull" employees to stay in their current jobs rather than seeking out alternative employment opportunities.We apply the wise proactivity framework (Parker et al., 2019), to understand for whom job crafting may translate into job quitting.By sampling employees from the same organizational setting (i.e.Dutch public service institute), we may hold task, strategic and relational considerations relatively constant.Hence, most information to clarify the link between approach crafting and job quitting in the present study context may be gained by what Parker et al. (2019) call self-regulatory considerations.Under this umbrella term they included studies that examined constructs such as career goal importance (Creed et al., 2017), image enhancement motives (Dahling et al., 2015) or reward sensitivity (Gawke et al., 2018) as moderators of proactivity.Here we investigate how career needs and self-esteem moderate the effect of job crafting.
The moderating role of need for career challenge Psychological needs were initially introduced as important predictors of job crafting (i.e. the need for positive self-image, human connection and meaningful work; Wrzesniewski and Dutton, 2001).Likewise, more recent job crafting theorists have treated needs as antecedents of proactive activities (e.g. de Bloom et al., 2020).However, the wise proactivity framework (Parker et al., 2019) suggests that psychological needs could also function as moderators because they influence the underlying motives (i.e.why) and the way in which employees Job crafting and job quitting implement their crafting efforts (i.e.how).Hence, psychological needs provide information on the goals employees try to achieve through job crafting, which could result in different outcomes for the employee and/or the organization (see also Zhang et al., 2021).Here, we examine relations between job crafting and job quitting in a sample of employees with relatively high job security, job embeddedness and relatively long job tenure.In this sample, employees who value career success and continuous development may feel that they are "locked at their job" (Feenstra-Verschure et al., 2023).To overcome this situation, these employees may engage in approach-oriented job crafting to enhance their employability and prepare for future job changes.In other words, for individuals who have a strong desire to advance in their career, approach crafting may eventually turn into job quitting.
That individuals take control of their careers to satisfy their needs, values and priorities has been already described in early theories of non-linear career patterns, such as the protean career model or the boundaryless career model (Greenhaus et al., 2008;Hall, 2004;McArdle et al., 2007;Sullivan and Baruch, 2009).This idea has been developed further in the Kaleidoscope Career Model (KCM), which proposes that individuals navigate their careers in accordance with three psychological needs, namely: the need for career authenticity, balance and challenge (Cabrera, 2007;Mainiero and Sullivan, 2005;Sullivan et al., 2005).Out of the three needs proposed by the KCM, we expect that particularly the need for career challenge may influence the relation between approach crafting and job quitting.Mainiero and Sullivan (2005) and more recently Mainiero and Gibson (2018) describe individuals with a high need for challenge as those who are driven to advance quickly in their career.Employees with a high need for challenge desire responsibility and stimulating work to get ahead on the labor market.Hence, these employees may engage in approach-oriented job crafting with the motive to advance in their careers (cf.Zhang et al., 2021).When approach crafting is motivated by the need for career challenges it may relate more positively to quit intentions because career-oriented individuals may react more quickly to a potential discrepancy between their current and their desired job (Direnzo and Greenhaus, 2011).As such, job crafters with a high need for career challenge may be more likely to transition from approach crafting to job quitting.
Support for this reasoning is evidenced in studies that investigated the links between career commitment and turnover intentions (Lansing, 1999;Vandenberghe and Basak Ok, 2013).Career commitment bears strong similarity with the need for career challenge because both constructs indicate a personal tendency to focus on extrinsic career success and professional advancement (Cabrera, 2007;Mainiero and Gibson, 2018;Plomp et al., 2016).It has been found that career commitment strengthens the relationship between proactivity (cf.approach crafting) and voluntary job change, because proactive individuals are more likely to turn their career commitment into action by looking for alternative employment opportunities (Vandenberghe and Basak Ok, 2013).In contrast, individuals who engage in job crafting with the desire for more meaningful work (e.g.need for authenticity) and improved work-life balance (e.g.need for balance) may persist in their crafting efforts rather than switching jobs.Thus, when approach crafting is motivated by career-advancement motives (e.g.need for challenge), this proactive behavior may be accompanied by the consideration of alternative job opportunities.In consequence, individuals with a high (vs low) need for career challenges may be more (vs less) inclined to consider leaving their current job when they engage in approach crafting: H1.Approach job crafting at Time 1 relates more positively to quit intentions at Time 2 among employees with a high (vs low) need for challenge.
The moderating role of self-esteem Psychological resources such as self-efficacy and optimism have often been studied as outcomes of job crafting (e.g.van den Heuvel et al., 2015;Vogt et al., 2016).More recently, intervention studies found that combining strategies that aim to develop psychological resources with job crafting can yield outcomes that are stronger than the effects of either intervention alone (Van Wingerden et al., 2017).Likewise, the wise proactivity model (Parker et al., 2019) suggests that employees who posit high (vs low) levels of psychological resources are better equipped to select appropriate proactive goals (e.g.Fang et al., 2011).Thus, building on more recent findings and the logic of the wise proactivity model we consider psychological resources as moderators in the relationship between job crafting and job quitting.Specifically, we focus on global self-esteem, which refers to individuals' overall evaluation of their self-worth (Rosenberg, 1965).Given the centrality of self-esteem in various psychological theories and the career literature (Lau and Shaffer, 1999), it is surprising that there exists no empirical study on how self-esteem influences the effects of job crafting until this date.
In their original job crafting theory, Wrzesniewski and Dutton (2001) propose that employees use job crafting to "enable a more positive sense of self to be expressed and confirmed by others" (p.183).Building on this proposition, it can be argued that employees with high (vs low) self-esteem are likely to set themselves more challenging job crafting goals to verify their self-image (cf.self-verification theory; Swann, 2012).In addition, people with high self-esteem tend to attribute failure to external causes (Tice, 1991).Hence, employees with high self-esteem may be inclined to 'look outside' when they experience job-related frustrations, which may cause them to engage in job crafting and to consider job quitting.Supporting this line of reasoning, Cai et al. (2015) found that self-esteem strengthens the link between proactivity and career exploration.They theorized that this is because individuals with high self-esteem strive to verify their positive self-image and feel more in control of their career development process compared to individuals with low self-esteem.
Earlier research suggested that individuals with high self-esteem tend to engage in more intense job search behavior (Ellis and Taylor, 1983), but that they are simultaneously less inclined to accept offers that are below their own standards (McArdle et al., 2007;Shamir, 1986).On the one hand, self-esteem may provide the psychological resources that individuals need to actively take charge of their careers beyond the boundaries of their current jobs (Parker and Liao, 2016).On the other hand, it has been suggested that individuals with high self-esteem are less likely to intend quitting their jobs because they see themselves as more competent organizational members (Pierce and Gardner, 2004).Here, we follow the wise proactivity perspective of Parker and Liao (2016) in arguing that self-esteem strengthens the link between approach crafting and quit intentions.This is because individuals with high self-esteem may have (a) higher expectations for what their job has to offer them, (b) attribute difficult job situations to external causes, which may lead them to (c) explore alternative job opportunities to maintain their positive self-image.Based on this reasoning, we expect: H2.Approach job crafting at Time 1 relates more positively to quit intentions at Time 2 among employees with a high (vs low) self-esteem.

From quit intentions to actual voluntary job changes
The intention to quit one's current job also referred to as turnover intentions is arguably the most commonly studied antecedent of actual voluntary job changes (Bolt et al., 2022;Hom et al., 2017).A review of five early meta-analyses suggested that quit intentions account for about 9%-25% of the variance in actual turnover rates (Dalton et al., 1999).A more recent cross-cultural meta-analysis indicated that the link between quit intentions and actual voluntary job change is stronger in countries that are characterized by high individualism, high power distance and low masculinity (Wong and Cheng, 2020).With regards to the present study, we may therefore expect to find a relatively strong link between intentions and actual job change.The Netherlands score relatively low on masculinity and are one of the Job crafting and job quitting most individualistic countries in Europe, as well as comparable to the USA in both individualism and power distance (Hofstede, 1983).Based on the Wong and Cheng (2020), we therefore expect: H3. Quit intentions at Time 2 relate positively to voluntary job change at Time 2.
Taken together, we adopt the wise proactivity framework (Parker et al., 2019;Parker and Liao, 2016) to gain a better understanding of the moderating role of need for career challenge and self-esteem in the relationship between approach crafting, turnover intention and actual voluntary job change.When engaging in approach crafting, employees proactively acquire skills, social capital and work experience that can help them to enhance their employability (Akkermans and Tims, 2017;Janssen et al., 2021;Plomp et al., 2016).Job crafters who have a high need for career challenge, as well as those with high self-esteem are more likely to develop turnover intentions and quit their jobs because switching to another job may allow them (a) to advance in their careers (need for challenge) and (b) to verify and maintain their self-image (self-esteem).Thus, Hypotheses 1 to 3 sum up to the following moderatedmediation model: H4a. Approach job crafting at Time 1 relates more positively voluntary job change at Time 2 among employees with high (vs low) levels of need for challenge via quit intentions at Time 1 (i.e.moderated mediation).
H4b. Approach job crafting at Time 1 relates more positively voluntary job change at Time 2 among employees with high (vs low) levels of self-esteem via quit intentions at Time 1 (i.e.moderated mediation).

Procedure and participants
We collected our data in 2016 using an online questionnaire at two measurement moments, with a time interval of 12 weeks in between.The sample is composed of 575 employees of a large public organization in the Netherlands.On average, participants were 47.03 years old (SD 5 9.56) and 60.50% of them identified as men.As concerns education levels, 11.1% had finished vocational school, 37.2% had finished a bachelor's degree and 51.7% had finished a master's degree or higher.Almost all participants (94.3%) had a permanent contract at the organization.The mean job tenure was 5.23 years (SD 5 5.39).Participants worked in a variety of different job positions in the public organization with most of them working in advisory (30.4%) or higher-level management positions (15.3%), followed by other managerial and operational positions, including policy development (9.2%), line management (6.1%),project management (5.7%) and research (5.2%).All questions were provided in Dutch and adapted to the 12-week time interval, mentioning that answers should be based on the past twelve weeks.

Measures
Career needs.The measures of the career needs were based on the scales of Sullivan et al. (2009).We reformulated the items to better reflect career needs by introducing them with the question "How important are the following things for you?" and letting participants rate each statement on a scale from 1 (not at all) to 5 (very important).All items with Dutch and English translations are provided in the online supplement together with factor loadings (https://osf.io/2cxeb/?view_only59cbe441d547d400b9397b32e4d6c373c).Need for challenge was measured with four items (α 5 0.77), while need for balance (α 5 0.79) and need for authenticity (α 5 0.72) were measured with five items each.To test whether the three needs CDI 28,6/7 are empirically distinct, we conducted CFAs and compared the fit indices of different models based on the conventional standards (i.e.CFI (Confirmatory Fit Index) and TLI (Tucker-Lewis Index) >0.90 and RMSEA (Root Mean Square Error of Approximation) < 0.08 indicate acceptable model fit; Marsh et al., 2004).A model with the items of the three needs loading on their respective latent factor provided an acceptable fit the data: χ 2 5 238.861, df 5 74, p < 0.001, CFI 5 0.93, TLI 5 0.91 and RMSEA 5 0.06.Additionally, the three-factor model fitted the data significantly better than any two-factor or one-factor model (Δχ 2 ≥ 376.78,Δdf ≤ 2, p < 0.001).
Self-esteem.We measured self-esteem with the five positively keyed items (α 5 0.79) of the Dutch Rosenberg Self-Esteem Scale (Franck et al., 2008).Items such as "On the whole, I am satisfied with myself" were rated on a scale from 1 (fully disagree) to 4 (fully agree).
Approach-oriented job crafting.We measured approach-oriented job crafting as a higherorder construct based on the items of the original Job Crafting Scale (JCS) of Tims et al. (2012), which consists of 15 items capturing proactive job crafting behaviors.Specifically, the JCS captures seeking social resources (e.g."I ask my colleagues for advice"), seeking structural resources (e.g."I try to develop myself professionally") and seeking challenges (e.g."When there is not much to do at work, I see it as a chance to start new projects") with five items each.Responses are given from 1 (never) to 5 (very often).We use the scale means as indicators of approach-oriented job crafting, as has been done in similar previous job crafting studies (e.g.Bakker et al., 2012;Plomp et al., 2016).Cronbach's alpha reliabilities of the approach-oriented JCS with the three subscale scores as indicators was α 5 0.76, hence acceptable.
Avoidance-oriented job crafting.We included avoidance-oriented job crafting (reducing hindrance demands) as a control variable in the analyses to partial out the variance that is unique to approach-oriented job crafting (LeBreton et al., 2007;Rudolph et al., 2017;Zhang and Parker, 2019).Avoidance crafting was measured with the six items (α 5 0.71) of the original scale of Tims et al. (2012; e.g."I make sure that my work is mentally less intense").
Quit intentions.We measured turnover intentions with three-items based on Ostroff and Kozlowski (1992).Specifically, we asked participants how frequently they (a) "Thought about looking for another job", (b) "Talked to colleagues about finding another job" and (c) "Took action to find another job", from 1 (never) to 5 (very often).Cronbach's alpha reliabilities were α 5 0.89 at Time 1 and α 5 0.89 at Time 2.
Voluntary job changes.At Time 2, we included a single-item to measure voluntary job changes ("Have you voluntarily changed your job position within the past 3 months?"),coded with 1 (yes) and 0 (no).In total, 40 participants indicated that they had voluntarily changed their job position since the Time 1 measurement.Unfortunately, it was not possible to verify whether participants changed their job within the same organization (internal turnover) or voluntarily left to start a new job at a different organization (external turnover).

Analyses strategy
We test the hypotheses in three steps.First, we test Hypothesis 1 and 2 cross-sectionally using the Time 1 data.Subsequently, we control for the autocorrelation path in predicting quit intentions at Time 2. This allows us to establish prospective effects of job crafting on quit intentions and to rule out several methodological artefacts (Orth et al., 2022).Finally, we test Hypotheses 3 (voluntary job change) using logistic regression and Hypotheses 4 in a path model to test the conditional indirect effect of approach crafting at Time 1 on voluntary job changes at Time 2 via quit intentions at Time 1 (i.e.moderated mediation).

Results
Table 1 displays means, standard deviations and correlations of the study variables.Results of cross-sectional and cross-lagged regression analyses are shown in Tables 2-4, respectively.Hypothesis 1 stated that approach job crafting relates more (vs less) positively to quit intentions among employees with high (vs low) need for challenge.The interaction term was significant (p < 0.01) in the cross-lagged analysis (see Table 3), while it was not significant in the cross-sectional analyses (see Table 2).We plotted the significant interaction of approach crafting and need for challenge from the cross-lagged model (see Figure 1), which revealed a pattern that aligns with Hypothesis 1. Specifically, the relationship between approach crafting and (changes in) quit intentions was positive and significant (B 5 0.21, t 5 2.98, p 5 0.003) among employees with a high (þ1 SD) need for challenge, while it was not significant among employees with a low (À1 SD) need for challenge (B 5 À0.030, t 5 À0.426, p 5 0.670).Thus, Hypotheses 1 was supported because among employees with a high (vs low) need for challenge, approach crafting related to an increase in turnover intentions over time (see Figure 1).
Hypothesis 2 stated that approach job crafting relates more (vs less) positively to quit intentions among employees with high (vs low) self-esteem.The interaction term was significant (p < 0.05) in both cross-sectional (see Table 2) and cross-lagged analyses (Table 3).Given that the cross-lagged analysis helps to rule out several methodological artifacts, we plotted the significant interaction from the cross-lagged model shown in Figure 2.This revealed that approach crafting had a positive significant relationship with (changes in) quit intentions at high (þ1 SD) self-esteem (B 5 0.18, t 5 2.55, p 5 0.011), while there was no relationship at low (À 1 SD) self-esteem (B 5 0.00, t 5 1.00, p 5 1.00).Thus, Hypothesis 2 was supported because approach crafting related positively to an increase in turnover intentions among employees with high (vs low) self-esteem.
Hypothesis 3 stated that turnover intentions relate positively to the likelihood of voluntary job changes.Table 4 shows that this hypothesis was supported and also reveals that there is no direct relationship between approach crafting and actual voluntary job change, after taking into account prior quit intentions (see M1 in Table 4).Lastly, we predicted that approach crafting relates indirectly to voluntary job changes because of the relationship with prior quit intentions conditional on (a) need for challenge or (b) self-esteem.To test the conditional indirect path, we modeled the relationships in a path model using the DWLS (Diagonally Weighted Least Squares) estimator (because voluntary turnover is a binary outcome).To simplify the complexity of the model we included manifest interaction terms, while we modeled turnover intentions as a latent variable to increase the precision of the mediated path.The model fit the data well (standard χ 2 5 12.34, p 5 0.780, scaled χ 2 5 24.65, p 5 0.100, df 5 17, RMSEA 5 0.03, CFI 5 0.99, TLI 5 1.00).
Supporting the Hypothesis 4, we found that at high (þ1SD) need for challenge there was a significant indirect effect of approach job crafting on voluntary job change at Time 1 via quit intentions at T2 (B 5 0.06, SE 5 0.04 and p 5 0.035), which was also the case at high (þ1 SD) self-esteem (B 5 0.08, SE 5 0.03, p 5 0.007).However, in the model (Figure 3) only the interaction with self-esteem was significant.This means that Hypothesis 4a received only partial support because the slope differences of approach job crafting did not significantly differ across high (vs low) need for challenge (interaction term), while the conditional indirect effect at high (þ1SD) need for challenge supported our predictions.In contrast, Hypothesis 4b received full support because the slope differences (interaction term) were significant and the conditional indirect effect was in line with our expectations.In sum, approach crafting may indirectly relate to an increased likelihood of a voluntary job change because of prior quit intentions, which is only present when either need for challenge or self-esteem are high (vs low).

Discussion
Adopting the wise proactivity framework (Parker and Liao, 2016;Parker et al., 2019), the present study examined for whom approach job crafting (i.e.seeking structural resources, social resources and challenges) may relate positively to quit intentions and actual voluntary job changes.Our findings showed that employees' need for challenge and self-esteem moderated the cross-lagged relationship between approach-oriented job crafting and quit intentions.The relationship between approach crafting and quit intentions became positive and significant for employees with a high (vs low) need for challenge and those with high (vs low) self-esteem.In addition, employees who engaged in approach-oriented job crafting and who had a high need for challenge or high self-esteem showed an increased likelihood of voluntarily changing their job because of their quit intentions.Collectively, these findings suggest that challenge-driven employees and those with high self-esteem may initiate approach crafting in preparation for future job changes (Bruning and Campion, 2018).These  (Akkermans and Tims, 2017) or to protect their positive self-image (Cai et al., 2015).Next, we discuss these findings from the employer and the employee perspective on job crafting (Parker et al., 2019;Tims and Parker, 2020).

Theoretical implications
By revealing the influence of employee characteristics on the relationship between approach crafting and turnover, this study contributes to clarifying the inconsistent findings in the existing literature (Rudolph et al., 2017).Recent studies yielded mixed results, with some indicating a negative association between approach crafting and turnover intention (Chu et al., 2022;Esteves and Lopes, 2017;Oprea et al., 2020;Zhang and Li, 2020), while others have found positive (e.g.Bruning and Campion, 2018) or non-significant correlations (e.g.Debus et al., 2020;Dominguez et al., 2019).Our findings demonstrate that the relationship between approach crafting and quit intentions becomes positive, only when employees possess a high Job crafting and job quitting need for career challenge or high self-esteem.This underscores the importance of considering employee characteristics, such as career motives and personality, to fully comprehend the impact of approach crafting on employee retention outcomes.Currently, the literature portrays a rather uniformly positive view on approach job crafting.There exists consensus that approach crafting is beneficial for employee well-being and performance outcomes (e.g.Oprea et al., 2019), while evidence regarding retention had been unambiguous and remained understudied (e.g.Rudolph et al., 2017).By showing that approach crafting can increase turnover intentions for certain people (i.e. when employees are high in need for challenge or self-esteem), our findings reveal potential negative outcomes of approach crafting from the employer perspective.Given that voluntary turnover can be costly, our findings encourage future research to investigate how crafting efforts could be supported by human resource (HR) or line managers to enhance the retention of job crafters.
By engaging in open conversations, the different parties may ensure positive outcomes from job crafting (Parker and Liao, 2016).Thereby, organizations may be able to effectively harness the benefits of approach crafting (e.g.enhanced well-being and performance), while mitigating adverse effects (e.g.increased voluntary turnover).
Lastly, this study contributes to sparking discussions about when job quitting may be wise for job crafters (Parker and Liao, 2016).To illustrate this phenomenon, we utilize the metaphor of seeing your job as a mountain.Job crafters experience considerable agency to tailor this "mountain of work" to their own needs and priorities.However, eventually it may be wise to switch to another mountain (i.e.job), especially when one notices difficulties in their crafting efforts and starts feeling "locked at their job" (Feenstra-Verschure et al., 2023).Ideally, employees guide their decisions on whether to continue crafting or to initiate quitting on a balanced consideration of their own interests, their social surrounding, as well as the larger organizational and economic context (i.e.wisdom; Sternberg, 2004).Cultivating such wisdom may help employee to become more effective job crafters (Fong et al., 2022;Tims and Parker, 2020), as well as to realize less disruptive job exits (Klotz et al., 2021).

Limitations and future research directions
This study made use of the wise proactivity framework (Parker et al., 2019) to understand the relationship between approach job crafting and job quitting.However, we did not empirically investigate the underlying mechanisms driving this relationship.Based on our findings, we encourage future research to explore the potential mediating roles of employability and threat to self-esteem in the link between approach crafting and quitting.By investigating these mediating mechanisms, researchers can shed light on the self-regulatory processes through which approach crafting influences turnover intentions and actual turnover behavior.In a related vein, it may be interesting to also examine when approach job crafting may prevent turnover intentions to translate to actual job quitting.In other words, job crafting may not only be an antecedent of job quitting but also a moderator of turnover intentionsvoluntary turnover relationship.We encourage future research to explore this possibility further.
In the present study, we went beyond existing research (e.g.Bruning and Campion, 2018), by examining actual voluntary job changes as an outcome of job crafting.However, we were not able to differentiate between external job changes (moving to another organization) and internal job changes (within the same organization).We note that the psychological processes underlying these two types of job changes may exhibit similarities, as both are likely preceded by turnover cognitions and preparatory actions such as job crafting (Klotz et al., 2021).Future research could investigate this assumption further by exploring how the processes for internal job changes differ from those associated with external job changes.
Moreover, it is important to consider the characteristics of our study sample when interpreting the findings.Although we included a diverse range of job positions in our sample, it is worth noting that all participants were employed within a large public organization and had relatively high job security.This context might have influenced the dynamics of job crafting, turnover intentions and voluntary job changes.To enhance the generalizability of the findings, future research could benefit from collecting data from a more heterogeneous group of employees representing various organizational and cultural contexts (Wong and Cheng, 2020).
Finally, we suggest that the broader economic context, including labor market conditions and social security, can play a significant role in shaping individuals' job change decisions.Labor market factors such as the availability of job opportunities, unemployment rates and the overall economic climate may either push job crafters to quit or pull them to stay in their current jobs (Mobley et al., 1979).Exploring the influence of these contextual factors on the dynamics of job crafting and job quitting can help to unravel the interplay between individual behaviors and external circumstances.

Practical implication
In the recent public debate on "the great resignation", job crafting is often mentioned as a strategy that may help to prevent voluntary turnover (Klotz, 2021).Although prior research offered ample evidence for the positive well-being effects of approach crafting (see Oprea Job crafting and job quitting et al., 2019), the literature was still unclear about the role this behavior may play in the voluntary job exit process.From a practical perspective, our findings suggest that when employees engage in approach job crafting this may signify the initiation of a voluntary job change process, especially among those who desire career challenges or have high selfesteem.Thus, rather than reducing the likelihood of voluntary turnover, job crafting may lead to an increase in turnover intentions for certain individuals.
Based on these findings, we advise HR professionals to explore how they can better support job crafting effortsparticularly among employees with high need for career challenges and those with high self-esteem.To support these individuals, managers can ensure that ample opportunities for development are within the job crafters' current positions to enhance their internal employability (Nelissen et al., 2017).Moreover, managers could implement practices that ensure that proactivity is accepted by co-workers (Junker et al., 2021).By having open conversations with employees about how job crafting aligns with their career needs, as well as the needs of others and the broader work context, managers can help employees to craft their jobs in a wise way (Parker and Liao, 2016).

Conclusion
Guided by the wise proactivity framework (Parker et al., 2019), this study identified the moderating role of the need for career challenges and self-esteem in the relationship between approach-oriented job crafting, quit intentions and actual voluntary job change.Specifically, employees who engaged in approach job crafting and had a high need for career challenge or high self-esteem were more likely to voluntarily change their job based on their prior turnover intentions.Overall, this study highlights the importance of considering the motivational factors behind job crafting and the influence of individual characteristics to understand when job crafting may turn into job quitting. 0 Figure 2. Relation between approach crafting and turnover intentions at high (vs low) selfesteem À4.06 (0.42)*** À4.00 (0.42)*** À4.06 (0.42)*** À3.99 (0.42)*** Standard errors in brackets, ***p < 0.001, **p < 0.01, *p < 0.05 and þ p < 0.10 Source(s): Authors' work