Job crafting, meaningful work and employee engagement

In the current knowledge era, employees are increasingly responsible for their own performance, with employers expecting them to actively adapt to new work practices and tasks (Olafsen et al., 2024; Wu et al., 2023). Thus, this rapidly changing world poses a major challenge for maintaining work engagement (WE) (Kujanpää and Olafsen, 2024). Highlighting this changing nature of work, new workplace redesign practices are proposed through which employees can excel in their roles while better coping with changing tasks and work practices (Mukherjee and Dhar, 2023). In this regard job crafting (JC) represents a proactive, employee-initiated process in which individuals intentionally reshape aspects of their work such as the number, scope, or type of tasks they perform in order to influence the meaning and identity of their job (Holman et al., 2024; Mainka et al., 2024; Napier et al., 2024; Shin, 2024; Zhan et al., 2024).

A rapidly growing body of research has shown that JC is associated with a wide range of positive outcomes on employee well-being, engagement and performance (Guo and Hou, 2022; Olafsen et al., 2024; Urbanaviciute and Lazauskaite-Zabielske, 2022). For example, recent research has shown that JC positively influenced the WE and performance of healthcare workers, with a reduction in stress and a reduction in turnover intention (Iida et al., 2024). These positive impacts of JC contribute greatly to employee engagement (Jindal et al., 2023; Khan et al., 2022). However, much of the existing research has primarily focused on the direct effects of JC, while paying less attention to the underlying psychological processes through which these effects unfold. Scholars argue that a fundamental precursor of employee engagement is work meaningfulness (WM), defined as the perceived value of a work goal or purpose that aligns with an employee’s personal values (Beer et al., 2022; Zhang et al., 2024). Employees who experience their work as meaningful are more motivated to invest effort, perform effectively, and pursue both personal and organisational goals (Ni et al., 2024). Yet, despite strong theoretical arguments, the role of WM in explaining how JC enhances employee engagement remains underexplored, leaving an important gap in the current understanding.

Our study addresses this gap by empirically testing the mediating role of WM in the JC–work engagement relationship. By focusing on non-managerial employees in Central Europe, we provide novel evidence that both enriches self-determination theory (SDT) and extends the geographical scope of existing findings. Our findings underscore the need for a deeper examination of the issue particularly in light of the contradictory findings of Sabri et al. (2025), who argue that WM plays a less significant role in the relationship between JC and WE.

SDT can be used to further explore the relationship between these constructs. This theory states that humans have three basic psychological needs–autonomy, competence and relatedness–the satisfaction of which provides motivation, proactivity and employee engagement (Kujanpää and Olafsen, 2024; Putter et al., 2024). The current demands on employees can lead to stress and frustration, which impede the satisfaction of basic needs, reduce meaningfulness and weaken WE (Kujanpää and Olafsen, 2024; Olafsen et al., 2024; Putter et al., 2024).

We suggest that JC can directly satisfy the psychological needs defined by SDT, thereby helping to foster WE through WM as a mediator. Therefore, the aim is to examine the impact of JC on employee engagement assuming WM as a mediator in the relationship under study. Our research analysed questionnaire data from 451 employed individuals in Central Europe who do not hold managerial positions, using the partial least squares structural equation modelling (PLS-SEM) and bootstrapping method. This study contributes to literature about SDT by showing how JC satisfies basic psychological needs and enhances WM, which in turn fosters employee WE. It also provides practical guidance for managers on how to strengthen engagement through fostering autonomy, competence, relatedness, and meaningful work.

The following sections will provide an overview of the current literature that discusses the issues of JC, WM and employee WE. The methodology section presents a description of the baseline population and how the baseline sample of respondents was determined. It also describes the method of data collection, the characteristics of the background variables, and a description of the statistical methods used and their assumptions for use. In the third section, the main results of our research will be described in detail and discussed with the results of existing scientific articles. We will also provide implications of our results for science and for practice. Finally, we summarise the most relevant findings of our research and describe its strengths and weaknesses, as well as provide suggestions for future research.

Recent management research highlights that the contemporary world of work is increasingly shaped by rapid technological advancement, digitalisation, and intensified performance expectations, fundamentally altering employees’ work experiences (Ferrer-Serrano et al., 2024; Olafsen et al., 2024). In this context, employees are not only required to continuously update their skills and competencies, but also to adapt to ongoing changes in work processes, task structures, and forms of organisational control driven by new technologies such as artificial intelligence (Dong et al., 2024; Hossain et al., 2024; Rózsa et al., 2023; Stephany and Teutloff, 2024; Nguyen Van et al., 2025). While these developments may enhance flexibility and efficiency, they simultaneously introduce new sources of strain, uncertainty, and perceived loss of control, raising critical questions about how employees maintain motivation, meaningfulness, and engagement under conditions of persistent change.

Technological innovations inevitably bring technology-driven changes that transform employees’ tasks, restructure workflows and integrate information across organisational processes. While these changes can empower employees by offering greater flexibility and new opportunities for task autonomy, they may also intensify organisational control through increased monitoring and the creation of new interdependencies between work activities. Understanding how employees engage in JC to preserve their autonomy, competence, and relatedness is particularly challenging in such a complex and evolving environment (Chan et al., 2025). Therefore, a necessary step is to provide employees with continuous training in new digital technologies such as artificial intelligence (Canal et al., 2024; Rózsa et al., 2023; Toma and Hudea, 2024).

Effective training can strengthen employees’ confidence and lower barriers to adapting to digital technologies (Chan et al., 2025; Jiang et al., 2025). For senior employees in particular, training focused on AI-related knowledge and digital skills can mitigate concerns about employability. Organisations should also aim to enhance employees’ broader understanding of technological innovations and their practical application (Tang and Zhao, 2025). In addition, training programs should address social, soft, and leadership skills, including group management, as these competencies are critical for reducing job insecurity and alleviating fears of job loss (Li et al., 2025; Stek et al., 2025).

The introduction of new technologies into the work process constitutes, on the one hand, an increase in business performance, but, on the other hand, it triggers employees' feelings of stress and frustration from the overflow of a lot of information and changes, and, last but not least, the fear of losing their jobs (Moncada et al., 2024). In addition, employees are increasingly expected to demonstrate initiative, proactivity, and accountability in their work (Olafsen et al., 2024). Leadership thus plays a critical role in facilitating JC, both by modelling proactive behaviours and by continuously developing leadership competencies that support employees in navigating change (Chan et al., 2025; Jiang et al., 2025). At the same time, leaders must consider the effects of digitalisation on key job resources, including autonomy, feedback, communication, and interpersonal relationships with colleagues and supervisors. When digital transformation is managed in a way that strengthens these resources, it can foster higher employee satisfaction and engagement (Zeshan et al., 2025).

The rapidly changing world of work, higher demands and expectations, and especially the intensification of work, poses a challenge to employees that may impact their experienced WM and ultimately their engagement (Kujanpää and Olafsen, 2024). Changes brought about by digitalisation can significantly shape and, in some cases, constrain employees’ work experiences. When tasks become monotonous or overly standardised, boredom may arise, leading to feelings of alienation and a diminished sense of purpose (Chen and Choi, 2025; Zhao et al., 2025). Such outcomes reflect a misalignment between employees’ perceptions of their work environment and their internal needs, including personal values, aspirations, and well-being. JC, understood as self-directed changes aimed at adjusting tasks, relationships, or cognitive framing, can help restore this alignment by enhancing working conditions and imbuing work with a greater sense of meaning (Chan et al., 2025; Li et al., 2025; Zeshan et al., 2025).

Against this backdrop of intensified demands and technological changes, JC has emerged as a central construct in contemporary management research for understanding how employees proactively shape their work experiences. Originally conceptualised as self-initiated changes to task, relational, and cognitive boundaries of work (Wrzesniewski and Dutton, 2001), recent research conceptualises JC as a dynamic and adaptive self-regulation process through which employees align their work with personal needs, values, and capabilities in evolving work contexts (Demerouti et al., 2024; Jiang et al., 2025; Teng and Cheng, 2025).

This perspective positions JC not merely as discretionary behaviour, but as a critical mechanism for sustaining meaning and motivation when formal job design becomes insufficient or misaligned with individual preferences.

Employees customise their work by expanding or contracting task, cognitive, or social boundaries at work in order to gain a greater sense of meaning (Demerouti et al., 2024; Mukherjee and Dhar, 2023; Rózsa et al., 2023; Teng and Cheng, 2025; Wu et al., 2023). Boundary shifts are changes initiated by employees that eliminate the mismatch between their actual and preferred levels of job demands and resources (Demerouti et al., 2024; Zhan et al., 2024). By these resources, we mean physical, social and organisational characteristics of work that contribute to goal achievement, demand reduction and personal development (Ho et al., 2024a). Holman et al. (2024) adds that employees access these changes to secure personally beneficial outcomes such as improved well-being and engagement. The changes are implemented in three dimensions–task crafting (TC), cognitive crafting (CC) and social crafting (SC).

TC involves changing the number of tasks, where the employee pushes boundaries in order to incorporate one or more tasks in which he or she can use his or her strengths and competencies to the fullest (Clinton et al., 2024; Geldenhuys et al., 2021; X. Li et al., 2025; Shin, 2024). Such a boundary shift that is implemented to exploit the employee's strengths and competencies ensures greater meaningfulness from the work performed and thus higher job performance (Chen and Du, 2022). CC refers to a proactive strategy to achieve congruence with the work environment by changing one’s understanding of work meaning and work identity (Buonocore et al., 2020). Ikeda et al. (2024) argue that CC reduces emotional job exhaustion and attenuates the indirect effects of workload. Taneva and Peng (2024) add that CC increases the feeling of job security and thus contributes to higher employee well-being. (Demerouti et al., 2021; Geldenhuys et al., 2021). It involves reconstructing positive work beliefs, thereby changing perceptions of the meaning of work and self-identification. It involves changing the way employees view work, making it more meaningful. For example, employees are encouraged to consider how work gives meaning to life, contributes to the success of the organisation, and benefits the wider community. This approach can enhance reflective thinking, align individuals with their work, and positively impact algorithmic competence. It helps workers perceive algorithmic management as an opportunity, strengthening their self-efficacy (Wang and Liu, 2025; Zhou et al., 2025). SC reflects the need to create supportive and quality relationships in the workplace (Clinton et al., 2024). Employees can actively choose how and to what extent they interact with different colleagues and how much they engage in social activities (Cruz et al., 2022). More commonly referred to as “relational crafting”, this term describes changes in the frequency and nature of interactions in the workplace. This involves making an effort to get to know people at work, taking part in social events, and forming friendships with colleagues. This type of crafting is motivated by the human need to connect with others, thereby helping to build relationships and make work more meaningful (Li et al., 2025; Zhou et al., 2025).

Recent studies have further extended the JC literature by emphasising its dynamic, context-sensitive and motivational nature. Rather than viewing JC as a uniform or static behaviour, contemporary research emphasises that its effects depend on how crafting strategies evolve over time and how they align with employees’ underlying motives and psychological needs (Clinton et al., 2024; Demerouti et al., 2024). For example, longitudinal studies indicate that JC becomes self-sustaining, particularly when it contributes to goals that are personally meaningful and self-concordant, thereby reinforcing employees' sense of purpose and autonomy at work (Clinton et al., 2024).

Moreover, emerging research distinguishes between two types of JC: approach-oriented and avoidance-oriented. These types may have different effects on well-being and engagement (Lopper et al., 2025; Zhang et al., 2025). Approach-oriented JC, which involves expanding resources and challenges, has been consistently linked to higher levels of WE, vitality and performance. In contrast, avoidance-oriented JC primarily serves a protective function, reducing strain and preventing resource depletion in demanding situations (Ho et al., 2024b; Humayun et al., 2024). These findings suggest that not all forms of JC foster meaningful work experiences equally, highlighting the need to examine the psychological mechanisms through which different crafting strategies lead to sustained engagement.

Taken together, recent discoveries in the research on JC position it as a key self-regulatory process through which employees actively construct meaningful work experiences in increasingly complex, technology-driven environments. However, despite the growing body of evidence suggesting that JC enhances engagement, the specific role of WM as the central psychological pathway linking distinct JC behaviours to sustained WE remains unclear. Closing this knowledge gap is essential for advancing theory and informing managerial interventions in contemporary work contexts.

We base our research on SDT. SDT is a macro-theory of human motivation that includes incorporates all aspects of human life. The theory posits the existence of three basic psychological needs, which are understood as innate, basic and universal (Alonso and Kok, 2020; Li et al., 2025; Olafsen et al., 2024; Teng and Cheng, 2025). These needs are defined as aspects of a person's environment that when satisfied facilitate growth, well-being and optimal functioning. These are the needs for autonomy, competence and relatedness (Appolloni et al., 2023; Aydin et al., 2022; Olafsen et al., 2024; Tuan, 2022).

The need for autonomy refers to a way of performing work through one's own procedures and activities, provided that work goals are met. The proportionality of satisfying this need leads to a directly proportional increase in the demands on that need, thereby motivating the employee to come up with ways to continuously satisfy it (Liu et al., 2023). The basis of need satisfaction is the experience of personal choice and the willingness to accept responsibility for decisions in initiating actions (Baer et al., 2022; Putter et al., 2024; Vansteenkiste et al., 2020). The need for competence reflects individuals' desire to achieve a sense of mastery in the processes or tasks they perform and the opportunity to develop their abilities (Olafsen et al., 2024; Putter et al., 2024). This need is related to efficiency and growth, whereby an individual is motivated to use his or her strengths and perform tasks to the best of his or her abilities (Aydin et al., 2022; Nguyen et al., 2023). The last need is the need for relatedness, which consists of the desire of individuals to feel part of a community to have a sense of connection with other people (Olafsen et al., 2024). It also represents an individual's elementary need to be acknowledged and respected by other people in their environment (Baer et al., 2022; Lam et al., 2021; Putter et al., 2024; Vansteenkiste et al., 2020).

Based on the nature of JC, we hypothesise that by actively JC, employees can satisfy individual needs as defined by SDT and thereby influence WE directly and indirectly through WM.

WE represents an intrinsic commitment to the company that stems from the employee’s subjective proactivity and energy. It is an incentive for commitment to come to work, leading to a sense of loyalty to the employer (Badru et al., 2022). Toyama et al. (2022) further define engagement as a positive state of mind that is conditioned by energy, dedication, and absorption. It is a state of active energetic engagement with work in which employees have the opportunity to express their thoughts, feelings, and emotions (Costantini and Weintraub, 2022; Dong and Zhong, 2022; Jindal et al., 2023). However, despite extensive evidence documenting the benefits of engagement, considerably less attention has been devoted to understanding the proactive processes through which employees actively cultivate engagement in demanding and technologically intensive work environments.

Current approaches define engagement as a positive and fulfilling mental state in relation to work. It is characterised by vitality, dedication and absorption (Costantini et al., 2025; Ho et al., 2024a). Highly engaged employees demonstrate excellent work performance and greater overall well-being (Yue et al., 2024). It is also associated with higher self-efficacy and resilience (Toyama et al., 2024), and contributes to organisational sustainability (Zeshan et al., 2025). Furthermore, Costantini et al. (2025) argue that engagement reflects motivated and active participation in work and differs from workaholism, which is characterised by excessive and compulsive urges to work. Yue et al. (2024) also argue that high levels of energy and engagement are associated with high work performance and greater well-being in the workplace.

Existing research on engagement has dominantly focused on the issue of “being engaged” rather than on what engagement elicits (Linehan and O'Brien, 2024). JC is one of the main ways in which employees can boost their engagement and sense of meaning at work (Garrett, 2024; Jindal et al., 2023). For example, Jindal et al. (2023) found that JC directly stimulates employee engagement. Similar findings were also reported by Ho et al. (2024) and Chen et al. (2024), who observed that supportive leadership encourages JC, which in turn enhances engagement. Furthermore, Brennan et al. (2023) found that JC fosters optimism, thereby increasing WE. Ho et al. (2024) also showed that employees who actively engage in JC report higher engagement. Similarly, Guo and Hou (2022) argue that JC promotes employee engagement.

A crucial mechanism in this process is the concept of meaningful work, which is defined as the perceived value and purpose of an individual's job (Astakhova et al., 2024; Montani et al., 2020). Employees who perceive their work as meaningful believe that it is important and aligned with their personal values (Furstenberg et al., 2021), which motivates them to invest more effort, take responsibility, and proactively complete tasks (Ni et al., 2024; Zhang et al., 2024). Prior research shows that JC increases meaningfulness by allowing employees to align tasks with their strengths and interests, thereby enhancing the value and purposefulness of their work (Berg et al., 2023; Costantini, 2022; Geldenhuys et al., 2021). Liu et al. (2025) further argue that JC is a key mechanism through which employees proactively strive for engagement and meaningfulness. Taken together, these findings indicate that meaningfulness represents the most significant psychological pathway through which JC fosters engagement. Therefore, we propose the following hypotheses:

By WM we mean the value conveyed by the work goal or the value of the purpose itself (Montani et al., 2020). Employees who perceive meaningfulness believe that their work is important and aligns with their value system (Furstenberg et al., 2021). Perceiving meaning from the work performed can maximise employee performance and improve orientation towards personal and organisational goals (Han et al., 2020). In general, meaningfulness itself is a valuable aspect in every individual's life.

Employees who experience meaningfulness from the work performed are motivated to invest more effort provided personal goals are achieved (Zhang et al., 2024). A high level of perceived meaningfulness enables employees to take responsibility and gain a sense of usefulness, making them more proactive in completing tasks. Therefore, employees who perceive their work as meaningful may perform better (Ni et al., 2024). Astakhova et al. (2024) defines WM as the level of importance that a job has for an employee. The authors add that WM is the result of a number of factors at work that affect the employee. Berg et al. (2023) argues that employees who proactively engage in JC experience greater meaningfulness than employees who perform the job passively. Thus, JC increases the value and purposefulness of the work performed, which directly impacts overall job satisfaction and meaningfulness (Costantini, 2022; Geldenhuys et al., 2021). Guo and Hou (2022) and Wrzesniewski and Dutton (2001) suggest that JC is an essential factor for increasing WM. This argument is also supported by Liu et al. (2025) who state that JC represents a key mechanism through which employees proactively strive for WM. Beer et al. (2022) states that experienced WM is an essential factor for promoting employee engagement. This statement is supported by Mousa and Chaouali (2022) who link WM to many positive psychological manifestations such as job satisfaction or employee engagement. Li et al. (2023) argue that the various aspects of JC lead to a sense of WM via different processes. TC enhances WM primarily by aligning work tasks with employees’ skills and personal interests. In contrast, CC promotes WM by altering the interpretation and perception of work, while SC does so by reshaping the quality and scope of interpersonal relationships at work (Sabri et al., 2025; Taneva and Peng, 2024). Thus, we hypothesise that WM plays an important role in the relationship between JC and employee engagement. Therefore, we propose the following hypotheses:

The stated hypotheses can be expressed by the following scheme of relationships between variables (Figure 1).

The data used in this research was collected through MN Force and funded by Early Stage Grants project D05_2024 JC and sustainable employee performance. The project was supported by the European Union's NextGenerationEU and Recovery and Resilience Plan funds. The following subsections contain a description of the sample composition, a description of the background variables, and the data processing methodology.

The research sample consists of employees from Central European countries, specifically Austria, Czechia, Poland, and Slovakia. It was drawn from the overall dataset, which included 600 fully completed questionnaires. Only respondents who did not hold managerial positions were included in the analysis, in order to avoid bias caused by factors related to the higher level of responsibility and managerial duties typical of supervisory roles.

Our study was conducted in Central European countries. This region was deliberately selected because, unlike the United States or China, the topic of JC has been explored only to a very limited extent in the Central European context. Furthermore, the countries where most prior research has been carried out (primarily the United States and China) represent cultural extremes according to Hofstede’s cultural dimensions framework (Hofstede Insights, 2025). In contrast, Central European countries typically score in the mid-range across most of Hofstede’s dimensions, making them a theoretically valuable setting for investigating JC in a culturally balanced environment. The following Figure 2 presents a map of the regions in which studies on this topic were published between 2022 and 2025. It also illustrates the most frequent research collaborations recorded in the published articles. The map clearly shows that Central Europe remains significantly underrepresented in this field of research.

To determine the minimum base sample, we used the G*Power program. We used the statistical method PLS-SEM and we used the F-test Linear multiple regression: fixed model, R² deviation from zero. The test inputs were the effect size f² = 15, the probability of error α = 0.05, and the number of predictors = 3. We consider the independent variables as predictors. Starting from the input parameters, the G*Power program defined the following outputs:

• F tests–Linear multiple regression: fixed model, R² deviation from zero

• Analysis: A priori: Compute required sample size

• Input: Effect size f² = 0.15

  • $α err prob=0.05$

  • Power (1-β err prob) = 0.95

  • Number of predictors = 3

  • Output: Noncentrality parameter λ = 17.8500000

  • Critical F = 2.6834991

  • Numerator df = 3

  • Denominator df = 115

  • Total sample size = 119

  • Actual power = 0.9509602

The results of the F-test showed that a minimum of 119 measurements are required to achieve a 95% confidence level. This represents a minimum base sample of 119 respondents. In our research, we collected data using an agency. The agency returned a total of 451 fully completed questionnaires from employed people who do not hold a managerial position. The structure of respondents consisted of 109 (24%) respondents from Austria, 120 (27%) from the Czechia, 111 (25%) from Poland and 111 (25%) from Slovakia. Of the total, 226 (50%) were female and 225 (50%) were male. In terms of enterprise size, 60 (15%) respondents are in a micro enterprise, 87 (19%) in a small enterprise, 98 (22%) in a medium enterprise and 192 (43%) in a large enterprise. 14 (3%) respondents were unable to comment. In terms of age structure, 59 (13%) respondents are from Generation Z (1997–2012) cohort, 148 (33%) respondents are from Generation Y (1981–1996) cohort, 217 (48%) respondents are from Generation X (1965–1980) cohort and 27 (6%) respondents are from Baby Boomers (1946–1964).

The questionnaire and its individual items have been taken from validated sources that have been assessed and tested in numerous leading research studies by experts and academics who have directed their professional interest to the subject matter (Geldenhuys et al., 2021; Lips-Wiersma and Wright, 2012; Moreira et al., 2020; Navarro-Abal et al., 2023; Slemp and Vella-Brodrick, 2013; Wojcik-Karpacz, 2018). The questionnaire consisted of several basic parts. The first part of the questionnaire consisted of questions related to the issue of JC. The second part contained questions directed at job meaningfulness and employee engagement. The third part verified basic demographic indicators. As part of the response system, respondents were given the opportunity to comment on each question using a 5-point Likert scale, where 1 represented “almost never” and 5 represented “very often.” The 5-point Likert scale was primarily used because it enables the quantification of subjective attitudes and perceptions of specific research objects (Ebert et al., 2025; Virga et al., 2023). It provides a standardised, comparable method of measuring responses and supports quantitative analyses, including advanced methods such as structural equation modelling (Ebert et al., 2025). 5 – point Likert scales are also frequently used in the development and validation of measurement instruments for constructs such as JC, as they ensure reliability and validity across various contexts. Likert scales are also suitable for cognitive-behavioural constructs that capture how individuals perceive and experience different aspects of life (van Zyl et al., 2023).

The questionnaire was distributed through the research agency MN Force in the countries of Austria, Czechia and Poland. Respondents were approached by random sampling. Data collection within Slovakia was facilitated by Survio and promoted on social media.

For the purpose of our research, the following independent variables were used:

JC is measured using the JC Questionnaire (Geldenhuys et al., 2021; Slemp and Vella-Brodrick, 2013) and measures 3 factors:

1. Factor 1: TC (TC1 – TC5)

2. Factor 2: CC (CC1 – CC5)

3. Factor 3: SC (SC1 – SC5)

We assessed the internal consistency of the items selected for measurement using Cronbach's alpha. The results of measuring the internal consistency of the selected items reached an excellent level, which in the result represents the reliability of the measured variables. TC Cronbach α = 0.837. CC Cronbach α = 0.844. SC Cronbach α = 0.806. The specific questions of the questionnaire are set out in the  Appendix.

WM. The variable was measured using the Comprehensive Meaningful Work Scale (Lips-Wiersma and Wright, 2012) questionnaire. This questionnaire measures seven factors:

1. Factor 1: Unity with others (WM1 – WM6)

2. Factor 2: Serving others (WM7 – WM10)

3. Factor 3: Expressing full potential (WM11 – WM14)

4. Factor 4: Developing and becoming self (WM15 – WM17)

5. Factor 5: Reality (WM18 – WM20)

6. Factor 6: Inspiration (WM21 – WM24)

7. Factor 7: Balancing tension (WM25 – WM28)

WE. The variable was measured using the Utrecht Work Engagement Scale (UWES 17) questionnaire (Moreira et al., 2020; Navarro-Abal et al., 2023; Wojcik-Karpacz, 2018). The questionnaire measures the following 3 factors:

1. Factor 1: Vigour (WE1. WE4, WE7, WE10, WE13 and WE15)

2. Factor 2: Dedication (WE2, WE5, WE11, WE16 and WE17)

3. Factor 3: Absorption (WE3, WE6, WE8, WE9, WE12 and WE14)

We assessed the internal consistency of the items selected for measurement using Cronbach's alpha. The results of measuring the internal consistency of the selected items reached an excellent level, which in the result represents the reliability of the measured variables. WM Cronbach α = 0.948. WE Cronbach α = 0.955. The specific questions of the questionnaire are set out in the  Appendix.

The data obtained from the questionnaire survey were analysed using the partial least squares structural equation modelling (PLS-SEM) method. The chosen method is appropriate based on its property of estimating models in terms of understanding individual hypothesised relationships (Ringle et al., 2023). Further, the method was chosen because the theoretical model we proposed is complex, robust and proposes a higher order construct for which the method provides higher statistical power. Another justification for using PLS-SEM is that it combines more advanced methods and complex bootstrapping procedures compared to covariance-based structural equation modelling (Hair et al., 2023). The PLS-SEM algorithm relies on a series of linear regressions in conjunction with linear combinations (Ringle et al., 2023). The statistical software SmartPLS 4 was used to perform the method (Sarstedt et al., 2024). The process of performing PLS-SEM consists of two basic steps. The first is the measurement model, which represents the basic assumptions for the execution of the test. The second is the path model, which expresses the resulting relationships between the variables (Sarstedt et al., 2024).

The first step in the assumptions that are part of the measurement model is to validate the indicator loadings. Through factor analysis, we test the loadings of individual indicators, which must take a value > 0.708 to explain more than 50% of the variance of the indicators (Sarstedt et al., 2021). Furthermore, the measurement model was assessed for reliability through Cronbach alpha test and composite reliability tests (rho_a, rho_c). The above reliability tests were validated at the >0.700 level. Validity was tested in convergent and divergent (Hair et al., 2024). We verified convergent validity with the average variance extracted (AVE) test at the >0.500 level. We verified divergent validity by the Heterotrait-monotrait ratio test (HTMT) at a level <0.85 and the Fornell-Larcer criterion (FL) at a level < $AVE$ (Sarstedt et al., 2021). We further tested divergent validity by cross loadings, verifying that the tested indicators should have larger loadings within their variable than for the other variables.

The second step is to perform bootstrapping testing within the path model. For the testing of the 5 variables, we assumed the value of 1,000 bootstrapping samples (Magno et al., 2024). The PLS-SEM method itself is based on generating confidence intervals of p < 0.05 and t -statistics <1.64. We also examined the coefficient of determination R² and the predictive power of the model Q² predict. Q² predict values higher than 0 confirm the predictive power of the model. A comparison of prediction errors (root mean square error of approximation (RMSE), mean absolute error (MAE)) with a naive model (LM_RMSE, LM_MAE) shows that if most items have lower errors than linear model (LM), the prediction is strong; if half have lower errors than LM, the prediction is moderate; and if all items have higher errors than LM, the predictive power is weak (Hair et al., 2019). Positive path coefficient results close to 0 represent a weak to no relationship. Values of ± 1 indicate a weak relationship. Values of ± 3 represent a moderate relationship and values of ± 5 represent a strong relationship.

Our study focuses on identifying the impact of JC on employee engagement, with job meaningfulness acting as a mediator. The analysis was conducted on a sample of 451 non-managerial employees. Excluding managerial employees allowed us to eliminate the confounding influence of excessive responsibility for managing people and focus on the relationships between the variables under study. The respondents came from Central European countries: Austria, Czechia, Poland and Slovakia. The following Table 1 presents the descriptive statistics for respondents’ answers from individual countries regarding the variables under investigation.

Within the measurement model, we tested the factor loadings of the individual indicators that were used to measure the given variables. The first testing provided loading results, and we did not consider indicators that did not reach >0.708. Table 2 presents the results of the factor analysis after adjusting for the eliminated indicators.

We further tested the adjusted model for data that did not represent a sufficient burden with tests of reliability and convergent and divergent validity. Reliability was tested by Cronbach's alpha, rho_a and rho_c tests at the >0.700 level (Hair et al., 2024). The results are presented in Table 3.

Cronbach's Alpha test results represent reliable internal consistency of the scales, with the tested variables reaching levels >0.700. The composite reliability tests also produced excellent results for testing the internal consistency of the variables with the measured values reaching the >0.700 level. The results of testing the convergent validity of the individual variables reached values higher than the >0.500 level. Divergent validity was verified by HTMT and FL and through cross loadings tests.

Table 4 shows the results of divergent validity testing through the HTMT test. Individual correlations between variables are examined at the >0.85 level, and we do not identify any cases where this value is exceeded. The identified correlation between WM and WE reaches the critical level of 0.848, with this value still within the required level.

Table 5 reports the results of the Fornell–Larcker test of discriminant validity. The results indicate that, for most construct pairs, the inter-construct correlations were lower than the square root of the AVE (√AVE), as shown on the main diagonal. A slightly increased correlation was identified within the WM and WE relationship, which reached a value of 0.813. To verify the criticality of the measured value, we performed a cross loadings test, which compares individual indicators and their correlation among other variables. The results of the cross loadings are contained in Table 6.

Cross loading testing consists of verifying the correlation of indicators with other variables. The values that an indicator achieves within its associated variable should be higher than the values of the correlation of the indicator with other variables (Henseler et al., 2015). We do not identify any correlation at the indicator level when examining the relationship between WM and WE, and hence we consider the result of the FL test within the correlation of WM and WE to be insignificant.

Through bootstrapping based on n = 1,000 subsamples, we obtained the path coefficient results, which are contained in Table 7. Our primary focus was on the direct and specific indirect effects of TC, CC and SC on WE, assuming WM was the mediator.

Table 7 presents the results of testing the assumed relationships involving direct effects and specific indirect effects. Through the analysis of specific indirect effects, the mediating role of WM in the relationship between the dimensions of JC (TC, CC, and SC) and WE was examined. The results of testing the direct effect showed that the relationship between TC and WE shows a robust design when the value of the sample mean differs by only 0.002 units from the original sample. The standard deviation (STDEV) shows a deviation of 0.042 units from the path coefficient, marking its consistency across multiple bootstrapping samples. The T-value of the statistic yielded a result >1.64 and p = 0.001. These results indicate a statistically significant effect of TC on WE. The tested relationship between CC and WE shows robustness to the values of the original sample of 0.084 and the mean sample of 0.080, where both samples take almost the same value. The STDEV value explains a deviation of 0.048 units from the path coefficient, indicating that the path coefficient is consistent across multiple bootstrapping samples. The T statistic took a value > 1.64, with a p value = 0.039. These results represent a statistically significant effect of CC on WE. The relationship between SC and WE does not show statistical significance when the value of T statistic does not reach the level of 1.64 and p = 0.448. The relationship of TC to WM represents a robust construct when the value of the sample mean is close to that of the original sample. The deviation from the path coefficient is 0.056 units and thus the coefficient is consistent across multiple bootstrapping samples. The T statistic results in a value > 1.64, with p = 0.011. These results indicate a statistically significant effect of TC on WM. The relationship between CC and WM shows strong robustness when the values of the original and average samples show the same value of 0.282. The STDEV explains a deviation of 0.055 units from the path coefficient, representing its consistency across multiple bootstrapping samples. The T statistic took a value > 1.64 and the p value = 0.000. The obtained results indicate a statistically significant effect of CC on WM. The relationship between SC and WM shows a robust construct when the value of the sample mean of 0.282 is close to the value of the original sample of 0.284. The STDEV value represents a deviation of 0.057 units from the path coefficient, indicating its consistency across multiple bootstrapping samples. The T statistic took a value > 1.64 and statistical significance p = 0.000. The results denote a statistically significant effect of SC on WM. The most robust construct is represented by the relationship between WM and WE when the difference in the mean and original sample is 0.001 units. The STDEV marks a deviation of 0.035 units from the path coefficient, indicating its consistency across multiple bootstrapping samples. The t-value of the statistic is > 1.64, with a statistical significance of p = 0.000. These results show a strong statistically significant effect of WM on WE.

The results of specific indirect effects show that in the case of the mediating effect of WM in the relationship between TC and WE, this effect reaches a value of 0.090. The average value of the effect from the bootstrap samples was 0.093 and the STDEV was 0.039. The T-statistic value 2.285 together with the p-value = 0.011 confirm that this is a statistically significant indirect effect. In the case of CC, the indirect effect through WM on WE reached a value of 0.199, with the average effect value being identical and the STDEV being 0.041. The T-statistic value of 4.853 and p-value = 0.000 indicate the statistical significance of this relationship, which can be interpreted as a medium-intensity effect. A similar result was found for relationship between SC and WE, where the indirect effect reached a value of 0.200, the average effect value was 0.199, and the STDEV was 0.041. The T-statistic reached a level of 4.834 and p-value = 0.000, confirming the statistical significance of the medium-intensity effect. Overall, these results indicate that WM plays an important mediating role in the relationship between the dimensions of JC and WE, with the most significant mediated effect occurring in CC and SC, while the effect of TC is weaker but still statistically significant.

Figure 3 shows the path model of the hypothesised relationships. The individual relationships between the variables are expressed by the path coefficient and the significance level p. Specifically, TC exerts a weak positive influence on WM at a significance level of p = 0.011. The influence of CC on WM shows a moderate influence of 0.0282 at a significance level of p = 0.000. Similarly, SC exerts a moderate positive influence of 0.284 on WM at a significance level of 0.000. Within the relationship of JC on WE, TC shows a weak influence of 0.135 on WE at a significance level of 0.001. The influence of CC on WE presents a significantly weaker positive influence of 0.044 at a significance level of 0.039. A negative influence was identified within the relationship of SC on WE when the value of the path coefficient acquired −0.006. However, this relationship is not statistically significant as the p level = 0.448. WM mediated by JC exerts a strong positive influence of 0.704 on WE at a significance level of p = 0.000. JC explains 37.8% of the total variance in work meaningfulness (WM R² = 0.378). On the other hand, WM explains 68.7% of the total variance in work engagement (WE R² = 0.687).

Table 8 reports the results of the model’s predictive power. The Q² predict values are higher than 0 in all items, which confirms the predictive relevance of the model. When comparing prediction errors (RMSE and MAE) with a naive LM, we see that most items achieve lower or very similar values than LM, which indicates the model's moderate to strong predictive ability. Only in a few cases (e.g. WM_10, WM_11 and WM_14) does PLS-SEM have slightly higher values than LM, indicating locally weaker prediction.

Table 9 reports the results of hypothesis testing using standardised coefficients. These coefficients incorporate both the estimated path values and their statistical significance. The PLS-SEM analysis provides support for hypotheses H1, H2, H4, H5, and H6, whereas hypothesis H3 is not supported. Our results suggest that JC contributes to WM which encourages employees to be more engaged. Based on SDT, which assumes that employees satisfy their psychological needs through JC, our findings show that WM significantly mediates the relationship between JC and employee WE. In contrast, the direct effect of JC on employee engagement was weaker, and in the case of SC, not statistically significant.

Our study examined the relationship between task, cognitive and social JC and employee WE, with WM as the central mediator. The results revealed that WM is a strong predictor of engagement, and that the indirect effects of all three crafting dimensions on engagement via WM are significant. Although task and CC demonstrate weak yet significant direct effects on engagement, SC exhibits no direct impact; its contribution to engagement is entirely mediated by enhanced WM. Overall, the findings suggest that JC primarily promotes engagement by increasing employees' experience of meaningful work, rather than directly energising engagement.

Our results can be compared to existing research. For example, Jindal et al. (2023) examined the interactive effects of WE and work autonomy in enhancing job performance. The results showed that both WE and autonomy are high when the level of JC is highest. JC fully mediates the interactive effect of commitment and autonomy on WE. Neuber et al. (2022) further found that WE is positively related to future work performance. Research by Guo and Hou (2022) focused on the impact of JC on WE of managers. Their findings showed that JC positively enhances leaders’ person-job fit. They further found that JC has a positive impact on WM. Similarly, research by Liu and Zhang (2022) confirmed that JC promotes employee performance and hence the competitive advantage of the company. Napier et al. (2024) drew on SDT, focusing on how JC contributes to satisfying the need for autonomy in particular and thus contributed to WM. The results showed that the need for autonomy and the acquisition of meaningfulness are inherent in the JC process. The authors also added that the needs for competence and relatedness have received less attention in the research. A general factor of employee WE is well-being in both work and non-work life. Kujanpää and Olafsen (2024) investigated how proactive JC behaviours eliminate work and non-work stress and overall employee engagement. The results identified a positive effect of JC on the elimination of work and non-work stress. They also found that JC relieves frustration and increases well-being in the workplace and hence employee engagement. From another perspective on JC and WE, the research by Laguía et al. (2024) focused on examining how WE and the Big Five personality traits predict three dimensions of JC (towards strength, interests, and development). Results show that engagement (dedication and absorption) and traits such as extraversion, agreeableness, conscientiousness, and intellect/imagination are strong positive predictors of JC, while vigour and neuroticism are unrelated. These findings emphasise the importance of engagement and personality factors in encouraging proactive job redesign, and in supporting employee performance and well-being. Authors Petrou and de Vries (2025) investigate the effects of leisure crafting (proactive leisure activities aimed at goal-setting, human connection, and personal development) on personal and work-related outcomes. Study 1, a 4-week randomised controlled trial, showed that participants in the leisure crafting intervention reported greater improvements in meaning in life, self-efficacy, WE, and employee creativity compared to a control group. Study 2, a cross-sectional survey, replicated these findings and additionally demonstrated positive associations with subjective well-being and creativity as rated by others. The results highlight leisure crafting as a promising approach for fostering employee well-being, engagement, and creativity. A study by Rhee et al. (2024), grounded in the self-regulatory strength model, examines the daily interplay between recovery experiences, sleep quality, resilience, JC, and long-term job performance. The study found that overnight recovery and good sleep quality positively predicted next-day JC, mediated by morning resilience. Furthermore, daily resilience and JC jointly mediated the positive effects of recovery and sleep quality on job performance three months later. These findings highlight JC as a self-regulatory behaviour shaped by recovery processes and emphasise the importance of promoting employee recovery and sleep to support sustained WE and performance. Further research by Signore et al. (2024) investigates the strategic role of JC in mitigating the negative effects of job insecurity on WE, psychological well-being, and emotional exhaustion. The results demonstrate that JC enhances individual well-being, strengthens employability, and serves as a protective factor against stress-related mechanisms. These findings underscore the potential of JC interventions to foster both individual development and organisational growth.

Scientific implications. The findings of this study make an important contribution to the literature on JC and SDT. Within the JC literature (Demerouti et al., 2024; Geldenhuys et al., 2021, 2021; Wrzesniewski and Dutton, 2001), our results in our path model (Figure 3) emphasise the important role of WM in the relationship between JC and employee WE. These findings call for a re-evaluation of the prevailing focus on the direct impact of JC on WE, as our study revealed that its indirect effects, mediated through WM, were considerably stronger than its direct effects. The mediating role of WM emphasises its fundamental importance and reinforces the idea that the primary value of crafting behaviours lies in their ability to foster a sense of meaning in work.

Drawing on SDT (Olafsen et al., 2024; Putter et al., 2024), our model demonstrates that JC operates as a self-initiated mechanism through which employees satisfy their basic psychological needs for autonomy, competence, and relatedness (Napier et al., 2024), thereby enhancing their experience of WM and, in turn, WE. Consistent with SDT, the findings suggest that JC does not primarily influence WE through direct energising effects, but rather through employees’ subjective interpretation of their work as meaningful and personally valuable.

By engaging in task, cognitive, and SC, employees actively modify their roles to better align work demands with their strengths, preferences, and values, which enhances their sense of agency and control over their work. Task and CC, in particular, allow employees to shape tasks and perspectives in ways that foster a sense of mastery and effectiveness (Jindal et al., 2023), thereby satisfying the need for competence. SC supports the need for relatedness by enabling employees to build and maintain meaningful interpersonal connections, although our results indicate that such relational changes contribute to WE only insofar as they enhance WM.

Central to SDT is the assumption that activities experienced as meaningful are more likely to be internalised and autonomously regulated. This assumption is strongly reflected in our results, as evidenced by the robust relationship between WM and WE in the path model (Figure 3). Meaningful work appears to function as a proximal motivational state through which need-supportive crafting behaviours translate into sustained engagement, transforming work from a means of extrinsic reward attainment into a source of intrinsic fulfilment. The significant indirect effects of JC on WE therefore illustrate how self-determined actions foster intrinsic motivation through meaning-making processes. Taken together, these findings underscore the importance of WM as a central explanatory mechanism in the JC–work engagement relationship and help clarify why the motivational benefits of JC depend less on crafting behaviours per se and more on the extent to which such behaviours enable employees to experience their work as meaningful.

The findings of this study also have several implications for management practice, as they illustrate how the conditions of SDT can be translated into concrete managerial actions through JC. As the results demonstrate that JC primarily influences WE through WM rather than through direct effects, managerial interventions should not only encourage proactive behaviours, but also create conditions that help employees experience meaningfulness in their work.

Firstly, organisations should purposefully support the three motivational needs of autonomy, competence and relatedness, as these form the basis through which JC contributes to WM and WE. Autonomy can be fostered by giving employees discretion over how they structure and perform tasks, thereby enabling TC. Competence can be supported by providing opportunities for skill development, learning and reflection, helping employees recognise how their strengths and capabilities are utilised in their work. Relatedness can be encouraged by facilitating high-quality workplace relationships and psychologically safe interactions, rather than merely increasing collaboration frequency.

Secondly, organisations may benefit from training interventions that develop employees’ capacity for CC. Such interventions may include workshops focused on reframing tasks, aligning daily work activities with personal values and goals, and reflecting on the broader purpose of one’s role. These practices can help employees actively construct meaning in their work, thereby strengthening the motivational pathway from JC to WE identified in the present study.

Thirdly, at an organisational level, job design and leadership practices should explicitly emphasise the importance of roles within the organisation. Managers play a vital part in helping employees understand how their work contributes to organisational objectives and wider societal outcomes. Regular discussions about purpose, feedback emphasising impact rather than performance metrics alone, and communication connecting individual tasks to collective goals can enhance employees’ experience of meaningful work. In line with our findings, such practices are particularly important because WM is the key mechanism through which JC translates into sustained WE.

Taken together, these implications suggest that, to enhance employee WE, organisations should adopt a meaning-centred approach to JC that aligns managerial decisions with employees’ basic psychological needs, rather than providing generic encouragement of proactivity.

Our research conclusions confirm that JC contributes to employee WE, particularly through WM as a mediator. In this regard, TC, CC, and SC play a crucial role, as they positively shape employees’ perceptions of WM, which in turn fosters greater WE. The direct effect of JC on WE is relatively weak, highlighting the importance of perceptions of WM as a mediator of this relationship. These findings support the idea that satisfying the basic psychological needs of autonomy, competence, and relatedness through JC leads to higher employee motivation and engagement. Thus, our research suggests that efforts to foster the perception of work as meaningful may be an effective way to increase engagement and promote optimal employee functioning.

This study makes a valuable contribution to the literature on JC and SDT by emphasising the important role of WM in fostering employee WE. While previous research has primarily focused on the direct effects of JC on WE, our findings suggest that these effects are relatively weak and statistically inconclusive. In contrast, our model clearly confirms that WM is a significant mediator of the relationship between JC and WE. This emphasises that the real value of JC lies in its ability to enhance WM. This finding is an important theoretical contribution to the re-evaluation of existing JC models, suggesting that greater attention should be paid to the mechanisms influencing the subjective meaning of work rather than just the direct effect of JC behaviour itself. From the perspective of SDT, we demonstrate that JC not only satisfies autonomy, competence, and relatedness but also strengthens meaningfulness, which acts as a key mediator of engagement. By integrating these perspectives, the study highlights that JC behaviours matter primarily because they generate meaningful work, offering both theoretical enrichment and practical guidance.

The findings suggest that organisations should prioritise practices that increase autonomy, competence and relatedness, as these factors promote a sense of WM and WE. Managers can facilitate this by designing roles that enable employees to align their tasks with their personal strengths and values, and by offering opportunities for skill development. Training in cognitive techniques, such as reframing tasks or linking work to personal goals, can further increase perceived WM. At an organisational level, communicating the broader purpose of roles and linking individual contributions to company goals can strengthen meaning, thereby motivating employees and promoting sustained WE.

Our research also identifies some limitations of the approach used. The questionnaire we distributed contained more than 60 questions. This number of questions takes on average 10 min to complete, which may have an impact on respondents' interest, leading to biased results. Data processing through the PLS-SEM method carries several limitations. Mis-specification of the model can bias the results, especially if it contains more than four variables, which complicates interpretation. PLS-SEM also allows relationships outside the predicted model to be analysed, which can lead to a deviation from the main problem. The number of bootstrapping samples can also affect the results, where a large number increases statistical significance. In addition, PLS-SEM has limited capabilities in handling measurement errors that it does not take into account, which can lead to bias in the relationships between variables.

Future research could focus on refining the questionnaire to increase respondent engagement, and on exploring alternative data processing methods to mitigate potential biases in the results. Another important area for exploration is other theoretical frameworks, such as regulatory focus theory, to improve our understanding of the mechanisms through which JC influences WE. Additionally, examining the impact of JC on other aspects of working life would be beneficial, such as the subjective well-being of employees, their retention within the organisation, and the prevention of burnout syndrome. This would provide a more comprehensive understanding of the benefits of JC, contributing to the promotion of overall work well-being and long-term employee motivation.

• (TC1) I am introducing new approaches to improve my work.

• (TC2) I change the scope or type of work tasks I perform.

• (TC3) I introduce new work tasks that better suit my abilities or interests.

• (TC4) I participate in the decision to take on additional work tasks.

• (TC5) I prefer work tasks that match my abilities or interests.

• (CC1) I think about whether my work gives meaning to my life.

• (CC2) I remind myself of the importance of my work to the success of the organisation.

• (CC3) I remind myself of the importance of my work to the wider community.

• (CC4) I think about how my work positively impacts my life.

• (CC5) I reflect on the role my work plays in my overall well-being.

• (SC1) I try to get to know people well at work.

• (SC2) I organise or participate in work-related social events.

• (SC3) I participate in organising special events at work (e.g. a co-worker's birthday party).

• (SC4) I mentor new colleagues (officially or unofficially).

• (SC5) I make friends at work with people who have similar skills or interests.

• (WM1) I feel a sense of belonging at work.

• (WM2) When we make decisions at work, we can talk openly about our values.

• (WM3) At work, we talk about what is important to us.

• (WM4) At work, we support each other.

• (WM5) At work, we reassure each other.

• (WM6) We enjoy when we can work together.

• (WM7) I feel like I'm really helping our customers/clients.

• (WM8) My work contributes to creating products or services that improve people's well-being and/or the environment.

• (WM9) What we do at work is meaningful.

• (WM10) At work, we spend a lot of time on things that are really important.

• (WM11) At work, I create and apply new ideas or concepts.

• (WM12) At work, I make changes that make a difference to others.

• (WM13) At work, I experience a sense of accomplishment.

• (WM14) At work, I get excited about opportunities that are available to me.

• (WM15) At work, I lose my sense of what is right and wrong.

• (WM16) I don't like who I am becoming at work.

• (WM17) I feel disconnected from myself at work.

• (WM18) At work, I face reality.

• (WM19) At work we are tolerant, after all we are only human.

• (WM20) At work we realise that life is messy and that it's okay to be that way.

• (WM21) At work I feel inspired.

• (WM22) The work I do gives me hope for the future.

• (WM23) The vision we work on together at work inspires me.

• (WM24) I experience a sense of spiritual connection to my work.

• (WM25) I have time and space to think at work.

• (WM26) At work we strike a balance between focusing on getting things done and noticing how people are feeling.

• (WM27) At work, I create enough space for myself.

• (WM28) At work, I am able to maintain a balance between the needs of others and my own needs.

• (WE1) I feel full of energy at work.

• (WE2) I find the work I do meaningful.

• (WE3) When I work, time flies fast.

• (WE4) I feel strong and full of energy at work.

• (WE5) I am excited about my work.

• (WE6) When I work, I forget everything else around me.

• (WE7) When I get up in the morning, I feel like going to work.

• (WE8) I feel happy when I work hard.

• (WE9) I am absorbed in my work.

• (WE10) I can work continuously for a very long time.

• (WE11) My work challenges me.

• (WE12) I get carried away by my work.

• (WE13) I am mentally very resilient in my work.

• (WE14) It's hard to tear myself away from my work.

• (WE15) I always persevere at my work, even when things don't go well.

• (WE16) My work inspires me.

• (WE17) I am proud of the work I do.