Job crafting, meaningful work and performance: a moderated mediation approach of presenteeism.

This study aimed to (1) analyze the relationship between the three job crafting dimensions (task, cognitive and relational crafting) and performance; (2) explore whether meaningful work fully mediates the links between the three job crafting dimensions and performance; (3) analyze if presenteeism moderates the indirect effect of meaningful work in the relationship between job crafting and performance. To achieve these goals, we collected data with 146 workers from a town hall. We analyzed the proposed moderated-mediation model using the PROCESS macro. The results showed that (1) the meaning in work mediated the relationship between the three job crafting dimensions and performance; (2) presenteeism moderated the mediated relationship between job crafting and performance via meaningful work, that is, lower levels of presenteeism were positively related to performance, in particular, when the levels of meaning in work were lower. These results show the importance of empowering employees to craft their work, as it appears to increase employees' performance, by enhancing the meaning attributed to the work. Promoting health conditions, at work, also seems to be relevant for performance once it may decrease levels of presenteeism. The cross-sectional design should be regarded as a limitation, and we assessed all the variables through self-reported measures. The present study contributes to the literature by analyzing the relationship of job crafting to performance via meaningful work, within the public administration context. Moreover, this study adds to the literature the conditional effect of presenteeism which has not been discussed in prior studies. This study aids in bridging this gap.

Introduction

Job crafting refers to the ability of reinventing and changing their work in a constructive, pleasant, and efficient way (Berg et al. 2013) and is related to the adaption of the function to oneself. Recently, scholars found favorable relationships between job crafting and several outcomes, such as performance (Weseler and Niessen 2016). Employees who assume their self as crafters adopt an active position in relation to their work and, consequently, to their performance (Tims and Bakker 2010). Employees can alter the boundaries of their jobs, by taking on more or fewer tasks (task crafting). They can also change their relationships at work, by changing the nature of their interactions with others (relational crafting). At last, workers can cognitively change their jobs by changing how they perceive their tasks (cognitive crafting). Thus, employees not only gain more control over their work, but also value and contribute to its maintenance, image improvement and ameliorates the relationship with coworkers and their supervisors (Wrzesniewski and Dutton 2001).

Recently, scholars (e.g., Lazazzara et al. 2020; Rudolph et al. 2017) found favorable relationships between job crafting and well-being outcomes, such as work engagement and job satisfaction. Besides, scholars reported a positive relationship between job crafting and work performance (e.g., Tims et al. 2015; Weseler and Niessen 2016). A meta-analysis, focused on job crafting and its outcomes, evidenced that it is related with work engagement, job satisfaction, and work performance (Rudolph et al. 2017). Similarly, Lazazzara et al. (2020), in a meta-synthesis of qualitative research regarding job crafting, demonstrated that individuals craft their work to align it with their preferences, abilities and motivations, improving work meaning and identity, which in turn enhances work-related well-being and performance. Individuals who craft their jobs appear to have higher levels of meaningful work (Sánchez-Cardona et al. 2020). The “meaning of work” is related to the degree of meaning that workers believe their work has, with their personal values, and with the relationships with colleagues and leaders (Rosso et al. 2010). Having meaningful work has also been related to higher levels of job satisfaction, motivation and work performance (Allan et al. 2019) leading to positive consequences both workers and organizations (Rosso et al. 2010).

According to Steger et al. (2012), having meaningful work encompasses three dimensions: (1) the psychological meaning of work which is related to the experience of work as significant and meaningful; (2) the making sense through work, which makes life, as a whole, more meaningful by assigning meaning to work, and (3) the good motivations that are related to the positive impact that work has on others. On the other hand, it is possible that job crafters, due to their active role in their tasks, present lower levels of presenteeism at work. Presenteeism explains the fact that people are present in the workplace, but due to physical or psychological problems, are unable to fully fulfill their functions (Giæver and Løvseth 2019). However, it does not seem to affect all employees in the same way.

From previous research, it is clear that job crafting is positively related to meaningful work, and that job crafting promotes performance (e.g., Tims et al. 2015). Despite this evidence, there are still no studies that demonstrate the relationship between job crafting, meaningful work, presenteeism and performance, in an integrated model, specifically in the public administration context. As such, a more integrated approach that combines all these concepts into a single model is needed. Thus, with the aim of contributing to expand knowledge on this model, we aimed to (1) analyze the relationship between each job crafting dimension and performance; (2) explore whether meaningful work mediates these mediated relationships; (3) analyze if presenteeism moderates the indirect effect of meaningful work in the relationship between job crafting and performance.

Theoretical background

Job crafting and performance

Job crafting has been introduced by Kulik et al. (1987), even tough, Wrzesnieski and Dutton (2001) have developed it further. The constant work changes, such as changes in work design due to technological advances (e.g., telework), have increased the need for strategies to promote performance, and work life quality. Job crafting emerged as a set of proactive actions that somehow adjust the function to the workers’ needs and characteristics, and at the same time enhances their job enjoyment, meaning and satisfaction.

By crafting their jobs, workers assume an active role in their function, by making physical or cognitive changes to the way the job is done. In addition, it is an informal process including work-related behaviors focused on changes to the tasks, or to the relational and cognitive job characteristics (Slemp and Vella-Brodrick 2014). These positive changes may be quantitative, that is, related to the change in the number of tasks to be performed, or qualitative, when the focus is on the change of the interpersonal interactions during the job, or in the way workers perceive their work as meaningful.

There are three ways trough which employees can craft their jobs (Wrzesniewski and Dutton 2001): (1) task crafting; (2) relational crafting, and (3) cognitive crafting. Task crafting occurs when there are changes in the formal tasks, for example, by adding or relieving tasks, or by changing the time and effort devoted to different tasks (Berg et al. 2013). Relational crafting arises from the change in the way employees interact with each other while at work (Berg et al. 2013) (e.g., being proactive about making friends or turning the interpersonal interactions to high-quality ones). At last, cognitive crafting involves altering how workers cognitively perceive and assess their job, with the view to making it more personally meaningful (e.g., making efforts to recognize the effect of work on the organization’ success) (Berg et al. 2013).

Job crafting by improving job resources and motivation may also contribute to positive work-related outcomes, such as performance (e.g., Weseler and Niessen 2016).

Performance theories differentiate between task, contextual and creative performance (Motowildo et al. 1997). Task performance (in-role) is related to the core tasks of the individual, in the organization, and therefore is related to the organizational goals (e.g., goal attainment, judgement and decision-making), and is part of someone’s job description. On the other hand, contextual performance (extra-role) is referred to all the activities and behaviors that contribute to work psychological climate and include, for example, helping colleagues engage in learning. Contextual performance is voluntary and intrinsically motivated (Williams and Anderson 1991). Finally, creative performance is “the production of novel and useful ideas, products, services, or organizational processes, as the basis of organizational innovation” (Gutnick et al. 2012, p. 2) promoting performance that triggers creative ideas (e.g., find new solutions to problems; Miron et al. 2004). In this study, we focused on in-role performance.

When individuals feel that their jobs are not offering challenges or motivations, they may increase their demands and resources as a means to use their skills (Gordon et al. 2015). Tims et al. (2012) showed that workers sought challenges when they were not using their abilities and skills (boredom), and that they would benefit by adjusting (i.e., job crafting) them. Increasing tasks, volunteering, and other challenges at work can lead to better performances and provide employees with increased energy toward goal attainment (LePine et al. 2005). In addition, increasing resources can promote performance, since resources help individuals accomplish their goals and minimize the negative consequences of excessive demands (Bakker and Demerouti 2007). Plus, the relationship between job crafting and performance may also occur when individuals reduce the number of tasks (task crafting) to decrease their demands to achieve a healthy level to attain their work goals without being overstretched (Tims et al. 2012).

Weseler and Niessen (2016) showed that job crafting was positively related to task performance. Similarly, Altinay et al. (2015), in a two-wave longitudinal study with police officers, showed that job crafting positively predicted task performance in a period of organizational changes. Gordon et al (2015) also presented evidence of the positive path between job crafting and task, and contextual performance. Tims et al. (2014) presented longitudinal evidence of the positive path between job crafting and performance. The positive relationship between job crafting and performance might be due to changes in the way workers perceive their work—the meaning of work.

The mediating role of meaningful work

When individuals can redesign and give a different dimension to the function, by crafting it, they tend to feel more connected to their work, feeling it more meaningful and, therefore, achieving higher levels of well-being and performance because they perceive more control over the environment (Sánchez-Cardona et al. 2020). Therefore, job crafting can promote performance, by improving meaning in work (Berg et al. 2008).

Depending on the individuals’ proactivity and motivation, these proactive actions can enhance the meaning attributed to work, once individuals, with the same functions, can perform different tasks (Rudolph et al. 2017). Thus, job crafting appears to be an antecedent of meaning in work (Nikolova and Cnossen 2020). By having the opportunity to modify the work, it leads the worker to create meaningful experiences.

Scholars have defined meaningful work in various ways. First, meaningful work has been investigated framed in the workplace spirituality literature (e.g., Giacalone and Jurkiewicz 2003). According to Pawar (2009), workplace spirituality promotes work attitudes, such as job involvement, and affective commitment, that appear to be relevant to improve performance, and reduce absenteeism and presenteeism (Rego and Pina e Cunha 2008). Another definition includes different terms, ‘meaningful work’, ‘meaningfulness’, or ‘meaning in work’ which refer to the significance or value of work (Lips-Wiersma et al. 2016).

One of the first organizational theories to integrate meaningful work was the Job Characteristics Theory (Hackman and Oldham 1976) which identifies conditions needed for people to be intrinsically motivated and have high performance at work. Hackman and Oldham defined meaningful work as a key psychological dimension that leads to higher job satisfaction, and work performance. For Hackman and Oldham (1976), meaning in work is related to workers’ perceptions that their work is worthwhile, important, or valuable.

Steger (2018) proposed that meaningful work is any work or occupational role that individuals fulfill and that is judged by them, to possess meaning, purpose, or significance. It is related to the meaning that workers believe their work has. Moreover, it is connected to the personal values and with the close relationships experienced by the individual (Rosso et al. 2010).

Steger et al. (2012) suggested that meaningful work includes three dimensions: (1) psychological meaning (workers’ perceptions of meaning or purpose in the job or career activities); (2) making meaning trough work (the capacity for work to be in harmony with and to help providing meaning in workers’ personal life), and; (3) good motivations (having the opportunity to positively impact or benefit the greater good of stakeholders in the worker’s community or society).

Several studies have demonstrated that meaningful work leads to higher levels satisfaction, motivation and performance (Sánchez-Cardona et al. 2020), which can have positive results for both workers and organizations (Steger 2018). Hulshof et al. (2020a, b), in their diary study, showed that job crafting was related to task performance via meaningful work, and work engagement. Similarly, Letona-Ibanez et al. (2021) demonstrated that job crafting was positively associated to work engagement, through meaning in work. By crafting their jobs, workers might, not only positively change their psychological meaning in the job, but also feel enhance meaning in their personal’s life and improve their good motivations to benefit the greater good of the others. Thus, we expect that:

H1a. Meaningful work will mediate the relationship between task crafting and performance.

H1b. Meaningful work will mediate the relationship between cognitive crafting and performance.

H1c. Meaningful work will mediate the relationship between relational crafting and performance.

The moderating role of presenteeism

Absenteeism and presenteeism are factors relevant for employees’ performance, and organizational productivity. Both absenteeism and presenteeism are frequent problems within the public administration context, even though with different features. Absenteeism occurs when employees miss work, due to illness. Presenteeism, on the contrary, was defined as the act of attending work while ill, or having any physical or psychological problem (Aronson and Gustafsson 2005). Turpin and colleagues (2004) also stated that presenteeism was related to productivity loss stemming from attending while ill.

Presenteeism appears to be predominant among the educational, welfare and health sectors (e.g., Ferreira et al. 2019; Ferreira and Martinez 2012; Martinez and Ferreira 2012). However, it has increased, in the public sector, which has resulted in serious problems, such as productivity loss. In the public sector, managers with their conservative policies, add pressure on workers to attend work, even when they have some limited health condition. An example of such practices may be the increased monitoring of short-term sick leave, or the reduction in seek pay. Some studies have shown that workplace pressures, such as management style or the fear of consequences, have been found to be causes to presenteeism (Baker-McClearn et al. 2010). Moreover, according to Johns (2010), work conditions, such as high job stress, high workload or understaffing contribute to the formation of presenteeism. Empirical evidence has demonstrated that presenteeism is also affected by strict absence standards, positive organizational factors (e.g., support and justice), critical organizational features (e.g., down- sizing), and HR practices (e.g., health-related practices) (Lohaus and Habermann 2019). Other reasons for presenteeism lie on personal motivations, for example, the belief that no one else can do the job, or the loyalty to one’s own professional image (Baker-McClearn et al. 2010). In addition, meta-analytic evidence demonstrated that job satisfaction, affective commitment, and work engagement were related to presenteeism (Miraglia and Johns 2016).

Despite being physically present, if the employee is not completely healthy, then it will have costs, not only for the individual, but also for the organization. The effect of presenteeism on ill-health and performance is well-documented and reflects the financial costs for organizations (Lohaus and Habermann 2019). Presenteeism impacts performance, because the presentee (the one who works through illness) cannot comply with the assignments satisfactorily. Other consequences of the presenteeism have been related to the team performance, poorer physical and mental health (Demerouti et al. 2009), decreased productivity and work ability (Skagen and Collins 2016).

Miraglia and Johns (2016) proposed a dual process model for presenteeism, suggesting that job demands and job and personal resources trigger presenteeism via both health impairment and motivational paths. That is, Miraglia and Johns (2016) suggested that some job demands (e.g., job insecurity, job control) job resources (e.g., support from supervisor) and personal resources (e.g., optimism) were closely related to job satisfaction, which in turn, improved presenteeism (motivational path), and that these job demands, and resources were related to health, decreasing presenteeism (health impairment path). Specifically, they demonstrated that presenteeism may be triggered from both a decline in health, following a negative strain path (excessive demands), and elevated motivation, deriving from positive job attitudes (job resources). In addition, Ma et al. (2019) discussed the motivation for presenteeism and differentiated between autonomous and controlled motivation. While autonomous motivation was like Miraglia and Johns (2016) attitudinal/motivational path, controlled motivation “refers to the state that one performs a behavior with the sense of being pushed, pressured or regulated by something other than one’s authentic will” (Ma et al. 2018, p. 111).

Working while sick has been consistently found to account for much more productivity loss than absenteeism (e.g., Collins et al. 2005). Building on the idea of an “employee’s felt obligation to attend” (Miraglia and Johns 2016), we expect that job crafting, as a way to promote job resources, may increase employees’ meaning in work, which in turn will enhance their performance, but only when they have autonomous motivation (when employees want to show presenteeism), and not in the presence of controlled motivation (when employees feel that they must show presenteeism), which might impair health and performance. Presentees will vary in their productivity due to a host of personal and contextual circumstances, such as the exact nature of their health problems. Nonetheless, they are unlikely to be fully engaged and fully productive while working. Thus, when workers are presentees, the relationship between job crafting and performance, via meaningful work, will become weaker. Presenteeism will be a condition through which the positive path of job crafting to performance via meaningful work will wilt, because individuals in worst health conditions will tend to devote more effort to the tasks required—those that they have to do—rather than crafting their job. As such, based on these arguments, we expect that (Fig. 1):

Fig. 1

Overview of the hypothesized moderated mediation model

H2a. The indirect effect of task crafting on performance through meaningful work depends on the level of presenteeism, such that the mediating effect of meaningful work in the positive relationship between task crafting and performance is expected to be weaker when presenteeism is higher (versus lower).

H2b. The indirect effect of cognitive crafting on performance through meaningful work depends on the level of presenteeism, such that the mediating effect of meaningful work in the positive relationship between cognitive crafting and performance is expected to be weaker when presenteeism is higher (versus lower).

H2c. The indirect effect of relational crafting on performance through meaningful work depends on the level of presenteeism, such that the mediating effect of meaningful work in the positive relationship between relational crafting and performance is expected to be weaker when presenteeism is higher (versus lower).

Methods

Participants and procedure

In this study participated 146 workers from a national town hall, from different occupational functions. Most part of the participants were women (74%) and were aged between 41 and 50 years old (70%). The majority worked in the town hall for at last three years (47%), between four and five years (38%), while 13% worked there in a period ranging from 6 to 10 years, and 2% worked there for more than 10 years. Of the overall sample, 86% had, at least, the high school complete, whereas 14% had a bachelor. Most participants were operational assistant (73%), followed by those who were technical assistants (24%), and at last 3% were superior techniques. The majority had missed at last one day in the last year, due to health injuries (85%).

We contacted the President of the town hall to ask the participation of employees in this study. Then, workers were contacted via an internal email sent by the head of the human resources department. This email explained the main goals of the study (i.e., to explore the effects of job crafting on performance), gave guarantees of anonymity and provided a hyperlink that redirected the participants to the online survey. Overall, we sent 150 emails to the participants, of which 146 returned valid responses (response rate = 97%).

Measures

Job Crafting. We used the 15-item Job Crafting Questionnaire (JCQ; Slemp and Vella-Brodick 2014) divided trough three dimensions: task (e.g., “Choose to take on additional tasks at work”, α = 0.72), cognitive (e.g., “Think about how your job gives your life purpose”, α = 0.83), and relational crafting (e.g., “Make an effort to get to know people well at work”, α = 0.83). Participants answered using a 5-point Likert scale (1—never to 5—always). The Cronbach’s alpha of the overall construct was 0.86.

Meaningful work. This was measured by the Work and Meaning Inventory (Steger et al. 2012), composed of 10 items that assessed greater good motivations (three items; α = 0.85, e.g., “The work I do serves a greater purpose.”), positive meaning (four items; α = 0.88 e.g., “I have found a meaningful career”) and meaning making through work (three items, α = 0.86, e.g. “My work helps me better understand myself.”). The answers were also given using the 5-point Likert scale (1—completely false to 5—completely true). The overall construct showed a Cronbach’s alpha of 0.86.

Performance. It was measured with six items from In-role performance questionnaire (Abramis 1994). This scale measures the perception of performance, over the last week (e.g., “I achieved my goals at work”). The responses were given using the 5-point Likert scale (1 “very bad” to 5 “very well”). The overall scale showed a Cronbach’s alpha of 0.87.

Presenteeism. This was measured by the Stanford Presenteeism Scale (SPS-6, Koopeman et al. 2002). It includes six items assessing the degree trough which individuals were “present” and focused at work (e.g., “My (health problem) distracted me from taking pleasure in my work”). The answers were also given using the 5-point Likert scale (1 “totally disagree” to 5 “totally agree”). The scale showed a Cronbach’s α of 0.91.

Statistical procedures

To test our hypotheses, we calculated the descriptive statistics and bivariate correlations among the variables under study. For data analysis, we used SPSS v. 27.0. To test the hypotheses, we conducted moderated-mediation analyses using bootstrapping (based on 5.000 bootstrapped samples using bias corrected and accelerated 95% confidence intervals (CI), recommended by Hayes (2018). To test hypothesis 1, we conducted a mediation through PROCESS (Hayes 2018), model 4. PROCESS tests the mediation by analyzing the indirect effect using the bootstrap method with 5000 corrected samples. Through the analysis of the confidence intervals, it is possible to avoid problems of the indirect effect power, resulting from the sample distribution, as is the case of asymmetric samples, and cross-sectional designs (MacKinnon et al. 2004). According to Preacher and Hayes (2004), mediation exists when the independent variable (X = job crafting) influences the dependent variable (Y = performance) through a mediating variable (M = meaningful work). The total effect of X on Y is defined as the total effect (c). The direct effect of X on Y after adding the mediator variable (M) is c '. The effect of X on M is effect a, and the effect of M on Y (controlling the effect of X) is effect b. The indirect effect between Y and X is defined as the ab effect. In most cases, the indirect effect (ab) represents the difference between c and c 'and, as such, the total effect (c) can be calculated as the sum of c and ab. As a rule, we are facing a partial mediation, when the value of the indirect effect (ab) is lower than the value of the total effect (c) with the same sign.

To test the second hypothesis, we split it in two phases. First, as suggested by Hayes (2018), we tested the isolated moderation effect of presenteeism on the relationship between meaningful work and performance, trough model 1 on PROCESS. Then, we tested the full moderated mediation model with PROCESS, model 14 (Hayes and Rockwood 2017).

Results

Preliminary analyses

Table 1 provides the descriptive statistics, correlations and reliabilities of the study variables.

Table 1

Descriptive statistics, bivariate correlations and reliabilities

Variables	M	SD	1	2	3	4	5	6	7	8	9
1. Task crafting	3.55	0.49	(0.72)
2.Cognitive crafting	3.82	0.58	0.50**	(0.83)
3.Relational crafting	3.64	0.70	0.39**	0.35**	(0.83)
4.Performance	4.01	0.53	0.16*	0.24**	0.11	(0.87)
5. Presenteeism	3.57	0.59	0.13	0.05	0.26**	0.21**	(0.91)
6. Meaningful work	3.73	0.36	0.39**	0.55**	0.36**	0.31**	0.12	(0.86)
7.Age	42.23	6.97	0.11	0.07	0.15*	0.04	− 0.00	0.16*	–
8.Gender	1.26	0.44	− 0.13	− 0.14	− 0.18*	− 0.03	− 0.02	− 0.17*	0.01	–
9. Absenteeism	1.48	3.61	0.00	− 0.07	− 0.20*	− 0.06	− 0.07	0.10	0.07	0.02	–

N = 146. **p < 0.01; * p < 0.05. Cronbach’s alfas are between brackets

Hypothesis testing

Hypothesis 1a assumed that meaningful work would mediate the link between task crafting and performance.

According to the results, the relationship between task crafting and meaningful work (a; B = 0.32, p = 0.00) and the relationship between meaningful work and performance (b; B = 0.38, p = 0.00) were significant. The indirect effect of task crafting on performance through meaningful work was 0.12 (p = 0.00), with 95% CI [0.05, 0.21], indicating a significant mediation effect. The total effect (c; B = 0.20 p = 0.00) was significant, but the effect (c'; B = 0.08, p > 0.05) between task crafting and performance after the entry of the meaningful work was not significant, revealing a full mediation. Hence, the hypothesis 1a received support.

Hypothesis 1b assumed that meaningful work would mediate the link between cognitive crafting and performance. The results showed that the relationship between cognitive crafting and meaningful work (a; B = 0.36, p = 0.00) and the relationship between meaningful work and performance (b; B = 0.43, p = 0.00) were significant. The indirect effect of cognitive crafting on performance through meaningful work was 0.15 (p = 0.00), with 95% CI [0.08, 0.25], indicating a significant mediation effect. The effect between cognitive crafting and performance after the entry of the meaningful work was not significant (c'; B = -0.08, p > 0.05), revealing a full mediation. Thus, the hypothesis 1b was also supported by the data.

At last, we tested the mediating model with relational crafting as the predictor. The indirect effect was 0.10 (p < 0.01) with a CI 95% [0.04, 0.42]. Moreover, the relationship between relational crafting and meaningful work (a; B = 0.32, p = 0.00) and the relationship between meaningful work and performance (b; B = 0.32, p = 0.00) were significant. The effect between relational crafting and performance after the entry of the meaningful work was not significant (c'; B = 0.13, p > 0.05), revealing a full mediation. The hypothesis 1c was supported by the data.

Hypothesis 2a assumed that the indirect relationship between job crafting dimensions and performance through meaningful work would be moderated by presenteeism, such that the indirect effect would be stronger for individuals with lower levels of presenteeism. The analysis of the simple moderation revealed a significant interaction effect between presenteeism and meaningful work (B = -0.36, β = 0.15, ΔR = 0.04, p = 0.00). The simple slopes of the interaction between presenteeism and meaningful work showed that, for individuals with low levels of presenteeism (1 SD below the mean), performance was higher for those who presented high scores of meaningful work (B = 0.51, β = 0.10, p = 0.00, CI 95% [0.30, 0.72]). However, this relation was not significant for those who scored high on presenteeism (B = 0.08, β = 0.14, p > 0.05, CI 95% [− 0.19, 0.35]).

Then, we tested the overall moderated mediation model for each job crafting dimension. First, we tested the moderated mediation model with task crafting. The findings showed a significant moderated mediation (− 0.12, CI 95% [− 0.26, − 0.01]), suggesting that the mediated effect (meaningful work) was conditional upon the levels of the moderator (presenteeism) (Table 2). The indirect effect was significant when the moderator presented lower levels (− 1 SD (B = 0.17, β = 0.06, p = 0.00, CI 95% [0.08,0.29], but not when the moderator presented higher levels (+ 1 SD (B = 0.03, β = 0.05, p > 0.05, CI 95% [− 0.06, 0.12] (Table 3). Specifically, the indirect effect of task crafting on performance via meaningful work was stronger for individuals with lower levels of presenteeism (Fig. 2). Thus, hypothesis 2a was supported.

Table 2

Moderated mediation results for task crafting

Meaningful work	B	(SE)	LL CI 95%	UL CI 95%
Task crafting	0.33**	0.06	0.20	0.45
Age	0.00	0.00	− 0.00	0.02
Gender	− 0.07	0.06	− 0.20	0.05
Absenteeism	0.01	0.00	− 0.00	0.03
R2	0.20**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000

LL lower limit, UL upper limit, CI confidence interval, SE standardized error

**p < 0.01

Table 3

Conditional indirect effects at the different levels of the moderator

Presenteeism	B	(SE)	LL CI 95%	UL CI 95%
Meaningful work
− 1 SD	0.17**	0.060	0.08	0.29
M	0.10**	0.03	0.04	0.17
+ 1 SD	0.03	0.05	− 0.06	0.12
R2	0.25**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000

LL lower limit, UL upper limit, CI confidence interval, SE standardized error. ** p < .01

Fig. 2

The moderation effect of presenteeism on the relationship between task crafting on performance, via meaningful work

After that, we tested the moderated mediation model with cognitive crafting. The findings showed a significant moderated mediation (− 0.14, CI 95% [− 0.27, − 0.02]). Thus, the indirect effect was significant when the moderator presented lower levels (− 1 SD (B = 0.21, β = 0.06, p = 0.00, CI 95% [0.10, 0.34], but not when the moderator presented higher levels (+ 1 SD (B = 0.05, β = 0.05, p > 0.05, CI 95% [− 0.04, 0.14] (Tables 4 and 5). Thus, the indirect effect of cognitive crafting on performance via meaningful work was stronger for individuals with lower levels of presenteeism (Fig. 3). Thus, hypothesis 2b was supported.

Table 4

Moderated mediation results for cognitive crafting

Meaningful work	B	(SE)	LL CI 95%	UL CI 95%
Cognitive crafting	0.36**	0.06	0.24	0.47
Age	0.01	0.00	− 0.01	0.03
Gender	− 0.08	0.06	− 0.21	0.04
Absenteeism	0.01	0.00	− 0.00	0.03
R2	0.27**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000

LL lower limit, UL upper limit, CI confidence interval, SE standardized error. **p < 0.01

Table 5

Conditional indirect effects at the different levels of the moderator

Presenteeism	B	(SE)	LL CI 95%	UL CI 95%
Meaningful work
− 1 SD	0.21	0.06**	0.10	0.34
M	0.13	0.04**	0.06	0.21
+ 1 SD	0.05	0.05	− 0.04	0.14
R2	0.21**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000

LL lower limit, UL upper limit, CI confidence interval, SE standardized error. ** p < 0.01

Fig. 3

The moderation effect of presenteeism on the relationship between cognitive crafting on performance, via meaningful work

At last, we tested the moderated mediation model with relational crafting. The results showed a significant moderated mediation (− 0.11, CI 95% [− 0.24, -0.01]). The indirect effect was significant when the moderator showed lower levels (− 1 SD (B = 0.16, β = 0.05, p = 0.00, CI 95% [0.06, 0.26], but not when the moderator showed higher levels (+ 1 SD (B = 0.03, β = 0.04, p > 0.05, CI 95% [− 0.06, 0.11] (Tables 6 and 7). Thus, the indirect effect of relational crafting on performance via meaningful work was stronger for individuals with lower levels of presenteeism (Fig. 4). Thus, hypothesis 2c was supported.

Table 6

Moderated mediation results for relational crafting

Meaningful work	B	(SE)	LL CI 95%	UL CI 95%
Relational crafting	0.32**	0.06	0.20	0.45
Age	0.00	0.00	− 0.01	0.02
Gender	− 0.05	0.06	− 0.19	0.08
Absenteeism	0.02*	0.01	0.00	0.03
R2	0.22**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000

LL lower limit, UL upper limit, CI confidence interval, SE Standardized error. **p < 0.01

Table 7

Conditional indirect effects at the different levels of the moderator

Presenteeism	B	(SE)	LL CI 95%	UL CI 95%
Meaningful work
− 1 SD	0.16**	0.05	0.06	0.26
M	0.09**	0.03	0.03	0.16
+ 1 SD	0.03	0.04	− 0.07	0.11
R2	0.21**

N = 146. Regression coefficients non-standardized. Bootstrapped samples = 5000.

LL lower limit, UL upper limit, CI confidence interval, SE standardized error. ** p < 0.01

Fig. 4

The moderation effect of presenteeism on the relationship between relational crafting on performance, via meaningful work

Discussion

The aim of this study was to analyze the mediating role of meaningful work in the relationship between job crafting and performance, and to explore the moderating role of presenteeism in the mediated relationship.

First, the relationship between job crafting and performance was mediated by the meaning in work. That is, when workers craft their jobs, they tend to attribute more meaning to their own work, which in turn, results in higher levels of performance. In this way, the more workers transform their work, adjusting it to themselves, it increases the meaning attributed to their work, and as a result performance. This is consistent with empirical studies focused on the effects of job crafting on performance (e.g., Rudolph et al. 2017). Accordingly, the meaning of work seems to be an indicator of the value held by employees about their own work and performance (Rosso et al. 2010). It is natural that workers, having the power to decide about the process of performing their tasks, may better manage their time, finding more meaning in their work which, in turn, may lead to better performances (e.g., Rosso et al. 2010). Likewise, the proactivity of workers, in shaping their work environment, can provide them more possibilities to better perform their work, because they tend to feel more control over the work environment and, as a result, attribute greater meaning to their work (Tims and Bakker 2010). Also, by taking the initiative to modify their work, people can find new opportunities to create meaningful experiences for themselves at work which can be a particularly important process to cultivate commitment for those who are increasingly dissatisfied with their jobs (Hulshof et al. 2020a, b) and performance.

Moreover, the mediated relationship between job crafting, meaningful work, and performance, is conditional on the levels of workers’ presenteeism, in such a way that the indirect effect is stronger when employees score low on presenteeism. That is, when presenteeism is high, the mediation does not occur. Thus, when the meaning attributed to work is low, the performance is significantly higher for those who experience less presenteeism. On the other hand, as the meaning in work increases, the levels of performance do not significantly differ, regardless of the level of presenteeism. Presenteeism is being physically in the workplace, even tough, in limited health conditions (physical and/or psychological). Thus, it seems that the meaning attributed to work is a significant predictor of the performance, as well as job crafting, however, presenteeism seem to influence these relations. According to Martinez et al. (2007), presenteeism is a way of being focused on the tasks at hand, even though there may be health restrictions. It is also about avoiding distractions, and getting the job done on a daily basis.

As such, these results lead to relevant insights to the public administration policies and to the organizations in general. First, job crafting is a positive predictor of job performance. Thus, the existing policies should support their workers and empower them to craft their tasks, in a cognitive or relational way. This may lead them to enhanced individual performances and organizational productivity (Berg et al. 2008). By extending the boundaries of their functions, through job crafting, workers can achieve higher levels of performance, through the meaning attributed to their job. However, supervisors have also to take into account the health conditions of their workers, once presenteeism may impair the quality of the job done. Despite this study is focused on the public administration, these results may be relevant for the employees and employers in general. First, because there are other empirical demonstrations of the positive path between job crafting and performance, in different work contexts (e.g., firefighters, health care professionals) (e.g., Gordon et al. 2015). Second, the relationship between job crafting and meaning in work has also been evidenced in many studies (e.g., Wrzesniewski et al. 2013). As such, employees in general may benefit from organizations that allow them to craft their jobs, because it is likely that they change the way of thinking about their work, by attributing more meaning to it, which might result in better performances. This evidences the utility of the model in different contexts and can aid organizations when developing trainings to positively influence job crafting, meaning in work and performance.

Limitations and future research directions

Despite the positive features of this study (being conducted in the public administration context), there are some limitations to bear in mind. First, the small sample size may limit the generalizability of these results. A bigger sample would lead to robust results. Moreover, our participants were all employees of a single town hall, that agreed to cooperate on this study, which limits the generalization of our results. Second, the study was conducted in a specific period. Data was gathered in the period of mandatory confinement, due to the imposed restrictions of the emergency state due to the pandemic situation promoted by COVID-19. As such, many of these workers were telecommuting. Several studies showed that presenteeism predicts worst performances as a reflection of declining health conditions (e.g., Martinez et al. 2007). Plus, presenteeism has been identified as one of the main reasons for intra-individual fluctuations in performance—quantity and quality of work—in particular for workers with a higher organizational tenure. Thus, and since many workers were in remote work, a special condition of work, this may have affected the way people feel and work, and thus biasing the results. Third, the use of self-reported measures also limits the generalizability of the findings. Although the anonymity and confidentiality of the data has been highlighted and guaranteed to the participants, the individual tends to respond according to social desirability. Bearing in mind that the questions were related to the work context, the answers may not be 100% genuine. Fourth, the use of a cross-sectional design, which may limit the justifications of the causal order among our tested variables, we suggest, for future studies, a longitudinal research design with at least three waves to confirm the causality of our hypotheses.

The findings open new avenues for research, which can be considered for future studies. Since job crafting is a process that occurs over time (Wrzesniewski and Dutton 2001), future studies should consider a longitudinal, or a daily, study analyzing its impact on job performance. On the other hand, a longitudinal study should cover a time period corresponding to a work cycle, between 6 and 9 months, because turnover and job adaptation tend to occur after this period of time, and also the worker’s ability to be a job crafter.

Practical implications

Given our findings, a job crafting training program aimed at increasing meaning in work would be a promising way to improve performance. The public sector needs workers who craft their jobs, in such a way that they can better cope with overdemanding jobs, characteristic of public management policies (Audenaert et al. 2019). For instance, public sector may benefit from practices such as empowering leadership, social support, knowledge sharing and HR flexibility (Tuan 2017). Additionally, job crafting workshops (group sessions) would be relevant to train job crafting skills, as workers can see their goals clearly defined, and thus know better how to craft their work (Hodson and Baker 2020). Plus, it should be useful to include an implementation intention technique that help to identify specific goal-directed behavior and coping strategies that strengths the association between relevant critical situations, and planned responses (Costantini et al. 2020). Moreover, individual coaching may also be a way to improve actions directed to craft the job (Junker et al. 2020) and to improve self-development regarding meaningful work. Job crafting behaviors may be enhanced by increasing personal awareness of its benefits, to improve positive attitudes, and others’ involvement in job crafting.

Conclusions

The purpose of this study was to explore factors explaining (meaning in work) and influencing (presenteeism) the relationship between job crafting and performance, in the public administration sector.

This study is innovative since it seeks to relate the structural changes made by the worker, in his function, and the relationship of such changes with performance. Moreover, the results contribute in an important way to the literature on public administration sector by providing evidence for the indirect effect of having a meaningful work in the relationship between job crafting and performance. In addition, by demonstrating that this indirect effect is conditional on presenteeism, it also gives insights about the relevance of being in good health conditions.