Team boosting behaviors and collaborative job crafting: a multilevel examination of empowering leadership, organizational support, and psychological safety

Project teams increasingly operate under changing tasks, deadlines, and client demands. Their ability to adapt depends on collective effort, shared energy, and mutual adjustment, especially when work is knowledge-intensive and interdependent (Junker et al., 2025; Magni et al., 2026). Research has clarified many features of collaboration, but the mechanisms through which teams develop a collective capacity for proactive work redesign remain underdeveloped (Barker Scott and Manning, 2024; Edelmann et al., 2026; Urien et al., 2026). This study examines whether empowering leadership and perceived organizational support (POS) are associated with collaborative job crafting through team boosting behavior (Kim and Yoon, 2025; Park and Park, 2023).

This question involves a multilevel coordination problem. Members experience resources and opportunities individually, while project delivery requires those experiences to become coordinated action and shared expectations. Project teams combine heterogeneous expertise and frequently revise roles and relationships. Interaction, mutual adjustment, and convergence can turn dispersed individual experiences into collective patterns that support adaptation (Junker et al., 2025; Kozlowski and Klein, 2000).

Job crafting was originally conceptualized as an individual-level behavior through which employees proactively modify aspects of their work to improve fit with their needs, strengths, and preferences (Wrzesniewski and Dutton, 2001). Subsequent research extended this perspective to the team level (see Leana et al., 2009), emphasizing that teams may collectively craft tasks, relationships, and work processes to better align their work with shared goals and changing demands (Lazazzara et al., 2020; Tims et al., 2022). However, the antecedents and mechanisms that enable collaborative job crafting remain insufficiently understood (Demerouti, 2026; Wang et al., 2025). In particular, more research is needed on how contextual inputs are translated into shared proactive behaviors within teams (McClelland, 2026).

We focus on two contextual inputs. Empowering leadership is a proximal, leader-based resource that promotes autonomy, participation, and ownership over work decisions (Arnold et al., 2000; Quinn and Spreitzer, 1997). POS is an organization-based resource reflecting employees' perceptions that the organization values their contributions and cares about their well-being (Eisenberger et al., 1986, 2014). These inputs capture the immediate leadership context and the wider organizational context associated with proactive team behavior.

We propose that these inputs are associated with collaborative job crafting through team boosting behavior: mood-enhancing, energizing, and uniting actions directed toward the team (Fortuin et al., 2021, 2023). Team boosting mobilizes social energy and mutual support. Collaborative job crafting concerns the proactive modification of tasks, relationships, and work processes (Leana et al., 2009). This conceptual distinction provides the basis for the proposed mediating relationship.

The Input-Mediator-Outcome (IMO) model provides the process architecture linking contextual inputs, interpersonal processes, and outcomes (Hoch and Dulebohn, 2017). The Job Demands-Resources (JD-R) perspective explains why job and social resources are associated with energetic and proactive behavior (Bakker and Demerouti, 2017). Empowering leadership and POS are the inputs, team boosting behavior is the mediator, and collaborative job crafting is the outcome. The multilevel analysis separates employees' perceptions from shared team properties and tests the pattern within and between teams.

Psychological safety is examined as a boundary condition because collaborative job crafting involves interpersonal risk (Edmondson, 1999). Team boosting should be more strongly associated with work redesign when members feel safe to speak up, experiment, and challenge routines. Empowering leadership may also develop psychological safety through autonomy, participation, and expressed confidence (Frazier et al., 2017; Joo et al., 2023). The moderation model addresses a conditional relationship and does not assume that psychological safety is exogenous. The cross-sectional design cannot resolve temporal ordering.

By developing and testing this multilevel moderated mediation model, the study makes three main contributions. First, it clarifies the distinct relevance of empowering leadership and POS as complementary inputs for proactive team behavior, distinguishing between leader-based and organization-based sources of support. Second, it identifies team boosting behavior as a mediating mechanism that explains how these inputs are associated with collaborative job crafting. Third, it advances a multilevel understanding of psychological safety by examining whether it strengthens the relationship between team boosting behavior and collaborative job crafting at both the individual and team levels. In doing so, the study responds to calls for a more precise understanding of how leadership, organizational support, and emergent team processes jointly shape proactive team outcomes (de Jong et al., 2025; Gopalan et al., 2026; Tajeddini et al., 2026).

The IMO model explains how contextual inputs are associated with outcomes through processes and emergent states (Hoch and Dulebohn, 2017). JD-R theory supplies the motivational logic: resources support goal attainment, help employees manage demands, and sustain energetic and proactive engagement (Bakker and Demerouti, 2017; Bakker et al., 2023). Because JD-R theory originated mainly at the individual level, its application to teams requires an explicit account of how shared resources and motivation emerge (Junker et al., 2025).

At the individual level, empowering leadership and POS can be experienced as leader-based and organization-based resources. Team boosting may mobilize them through affective energy, focused effort, encouragement, information, and peer assistance. Mood-enhancing actions can replenish energy, energizing actions can focus effort, and uniting actions can connect complementary knowledge. These interpersonal actions make resources socially accessible and may facilitate collaborative job crafting.

At the team level, shared properties require more than score aggregation. Common exposure to leadership and organizational practices, followed by repeated exchanges of cues, emotions, support, and expectations, may produce a shared resource climate and a collective boosting pattern (Chan, 1998; Kozlowski and Klein, 2000; Urien et al., 2026). IMO specifies the input-process-outcome sequence, and the multilevel extension of JD-R explains the motivational role of resources within that sequence.

This integrated perspective helps clarify the role of team boosting behavior within the model. Team boosting behavior is conceptualized as an active interpersonal process involving mood-enhancing, energizing, and uniting actions directed toward other team members (Fortuin et al., 2021, 2023). These behaviors generate relational energy, strengthen collective motivation, and facilitate adaptive coordination within teams (Gillet et al., 2025). In this sense, team boosting behavior operates as a dynamic mechanism connecting contextual resources to collaborative job crafting.

The process-outcome distinction is central to the model. Team boosting behavior consists of actions that energize the collective, strengthen positive affect, and foster cohesion. Collaborative job crafting consists of joint efforts to modify tasks, roles, interactions, and work processes in response to team resources and demands (Tims et al., 2022). Team boosting captures socio-relational activation; collaborative job crafting captures coordinated work redesign. The constructs capture distinct but related aspects of collective adaptation.

This perspective is also consistent with foundational work on collaborative job crafting (Leana et al., 2009), which emphasizes that collective forms of crafting emerge through social interaction, mutual adjustment, and coordinated action among team members. In project-based teams, where expertise is distributed across members, collaborative job crafting involves adaptive coordination processes through which team members collectively adjust workflows, redistribute responsibilities when needed, share knowledge, and modify patterns of collaboration to better respond to changing demands (Liu et al., 2025; Zhang et al., 2026). Team boosting behaviors may facilitate these adaptive processes by strengthening trust, openness, and collective motivation within the team.

Accordingly, the present study applies the integrated IMO and JD-R perspectives to address three key gaps in the literature: (1) the limited understanding of team boosting behavior as a dynamic interpersonal mechanism linking contextual resources to collaborative job crafting; (2) the limited conceptual differentiation between team boosting behavior and collaborative job crafting; and (3) the insufficient understanding of how psychological safety shapes the effectiveness of these proactive team processes.

Figure 1 presents the proposed multilevel moderated mediation model.

Empowering leadership promotes autonomy, participation in decision making, self-leadership, and shared responsibility (Arnold et al., 2000; Quinn and Spreitzer, 1997). Its focus on delegated authority, ownership, initiative, and self-management distinguishes it from leadership approaches centered mainly on inspiration, participation alone, or follower development (Tekleab et al., 2008; Wang et al., 2022).

This distinction is particularly relevant in the context of proactive team behaviors and collaborative job crafting. Empowering leadership encourages employees to actively shape their work environment and participate in collective adaptation and proactive work redesign processes (Coun et al., 2022). From a JD-R perspective, empowering leadership may be understood as a social resource that enhances intrinsic motivation, autonomy, and proactive engagement (Bakker and Demerouti, 2017). In the IMO framework, it also functions as a contextual input shaping team interaction processes. Specifically, empowering leadership creates conditions in which team members feel encouraged to contribute ideas, support one another, and actively participate in collective problem solving (Audenaert and Decramer, 2018).

Empowering leadership may also contribute to psychological safety by legitimizing participation, reducing status barriers, and signaling confidence in members' competence (Frazier et al., 2017; Joo et al., 2023). Empowering leadership describes leader behavior; psychological safety describes beliefs about interpersonal risk. Their distinct meanings permit moderation analysis, while their possible temporal relationship requires cautious interpretation.

Recent studies suggest that team boosting behavior represents a particularly important interpersonal mechanism through which empowering leadership influences collective functioning (Fortuin et al., 2021). Team boosting behaviors include mood-enhancing, energizing, and uniting actions directed toward other team members that reinforce collective motivation and relational coordination (Fortuin et al., 2023). These behaviors are especially important in demanding and interdependent work contexts because they sustain collective energy and facilitate adaptive collaboration.

Accordingly, empowering leadership is likely to foster team boosting behaviors by encouraging initiative, mutual support, and cooperative interaction patterns within teams.

POS reflects employees' beliefs that the organization values their contributions and cares about their well-being (Eisenberger et al., 1986). It captures organization-wide support and socio-emotional investment, complementing the proximal leadership resource represented by empowering leadership (Tsachouridi and Nikandrou, 2019).

From a JD-R perspective, POS constitutes an important organizational resource that enhances engagement, commitment, and proactive behavior by strengthening employees' sense of reciprocity and psychological security (Bakker et al., 2023). Employees who perceive strong organizational support are more likely to develop trust in the organization and to reciprocate through cooperative and discretionary behaviors that benefit collective functioning (Nazir et al., 2019).

Although POS has been widely associated with positive work attitudes and innovative behaviors, its role in shaping proactive interpersonal team processes remains less explored (Azfar et al., 2026). We argue that POS may foster team boosting behavior because supportive organizational environments encourage employees to actively contribute to team morale, collaboration, and collective functioning (Tremblay, 2019).

Importantly, team boosting behavior reflects concrete interpersonal actions through which employees energize and support one another within the team context. In supportive organizational environments, employees may be more willing to engage in such actions because they perceive lower interpersonal risk and stronger collective reciprocity norms.

Accordingly, POS is expected to promote team boosting behavior by reinforcing cooperative interaction patterns and motivating employees to contribute positively to team functioning.

Collaborative job crafting is the coordinated modification of tasks, relationships, interactions, and work processes to align team resources with demands (Leana et al., 2009; Tims et al., 2022). It emerges through mutual adjustment among members. Team boosting behavior captures the energizing and uniting interactions that may support that redesign.

The constructs represent different elements of team adaptation. Team boosting concerns interpersonal activation through energy, cohesion, and encouragement. Collaborative job crafting concerns changes to work structures, responsibilities, and workflows. The first describes a social interaction process; the second describes the collective redesign of work.

This distinction is particularly important in project-based teams characterized by distributed expertise and interdependence. In such contexts, collaborative job crafting involves adaptive coordination processes through which teams redistribute activities, share knowledge, adjust collaboration patterns, and collectively respond to evolving demands (Haffer et al., 2021).

Within the integrated IMO and JD-R perspectives, team boosting behavior represents the motivational and relational mechanism that facilitates these adaptive processes. By fostering openness, collective energy, and cooperative interaction, team boosting behaviors create favorable conditions for collaborative experimentation and work redesign.

Accordingly, teams characterized by stronger team boosting behaviors are more likely to engage in collaborative job crafting.

Building on IMO and JD-R, team boosting behavior is proposed as the interpersonal mechanism linking empowering leadership and POS to collaborative job crafting. Contextual resources may support initiative, psychological security, and engagement. Mood-enhancing, energizing, and uniting actions circulate these resources as collective motivation, relational coordination, and adaptive collaboration, creating a pathway to proactive work redesign.

From a JD-R perspective, team boosting behavior may therefore be understood as part of the motivational pathway through which social and organizational resources may foster proactive adaptation and collaborative job crafting. Teams characterized by higher levels of interpersonal energizing and mutual support are more likely to collaboratively adjust workflows, responsibilities, and work practices to better align with changing demands.

Accordingly, we propose:

Psychological safety refers to the shared belief that the team environment is safe for interpersonal risk-taking (Edmondson, 1999). Psychological safety plays a critical role in enabling open communication, experimentation, and collaborative adaptation within teams (Frazier et al., 2017).

Psychological safety may shape voice and experimentation, develop through supportive leadership and positive interaction, or evolve reciprocally with those behaviors (Edmondson and Bransby, 2023; Joo et al., 2023). We specify it as a moderator because the focal question concerns when energizing and uniting interactions are more strongly associated with collaborative work redesign. This analytical role does not imply exogeneity.

At higher levels of psychological safety, members can use relational energy to question routines, negotiate responsibilities, and test alternatives with lower anticipated interpersonal cost. At lower levels of psychological safety, positive interaction may remain socially supportive without extending to changes in tasks or roles. The cross-sectional interaction is interpreted as a conditional association; temporal precedence remains untested.

The study was conducted within one of the Big Four management consulting firms in Italy, a context characterized by intensive use of project-based teamwork, high professional standards, and strong interdependence among team members. Employees regularly work in temporary and cross-functional project teams spanning areas such as audit, tax, consulting, and advisory services. In these environments, teamwork is essential for delivering complex and high-quality client solutions (Beau and Jerman, 2024).

Project-based work enables the integration of specialized expertise and supports innovation and adaptive problem solving. At the same time, it creates significant coordination demands because teams are continuously formed, reconfigured, and required to collaboratively respond to changing client expectations and operational constraints (Bailey et al., 2024). These characteristics make Big Four consulting firms particularly suitable contexts for examining how leadership, organizational support, and interpersonal team processes contribute to collaborative job crafting and proactive adaptation within teams.

Data collection began with two exploratory focus groups involving 12 participants each, drawn from different project teams within the organization. The focus groups lasted approximately two hours and were conducted to explore employees' experiences related to leadership, organizational support, team interactions, and psychological safety. The qualitative insights emerging from these discussions informed the refinement of the conceptual framework and survey instrument. In particular, participants described informal empowering leadership practices, as well as team-boosting actions such as emotional support, humor, informal mentoring, and energizing interactions among colleagues. These recurring themes reinforced the relevance of team boosting behavior as a central interpersonal mechanism within the proposed model.

Based on these insights, the questionnaire was pilot-tested with a small group of employees who were not included in the final sample to ensure clarity and appropriateness of the items. Minor revisions were introduced following participant feedback.

The final survey was distributed electronically to approximately 400 employees working within the consulting firm. To ensure consistency in the team-level analyses, organizational management identified employees primarily assigned to a single project team. Participants were instructed to answer the questionnaire with reference to their main team, defined as the team in which they spent most of their working time.

A total of 341 responses were initially collected (response rate ≈ 85%). Following data screening procedures, 320 responses were retained for the final analyses after excluding respondents with less than one year of experience within their focal team. This threshold was introduced to ensure that participants had sufficient exposure to the team environment and interaction patterns to provide reliable evaluations of leadership, team boosting behavior, psychological safety, and collaborative job crafting. Given the relational and emergent nature of the focal constructs, shorter periods of team membership may not allow employees to develop stable perceptions of team dynamics and collective processes.

The final sample comprised 320 employees in 60 project teams (M = 5.33 respondents per team). The mean formal team size was 7.85; the difference reflects survey completion and inclusion criteria. Team-size implications depend on task and coordination demands, although smaller project teams often experience fewer coordination losses (Hoegl, 2005; Stewart, 2006). These figures describe the study context and do not establish representativeness. Aggregation statistics were used to evaluate convergence among participating members.

Participants had a mean age of 28.4 years (SD = 3.5); 52% were male and 48% female. Mean organizational tenure was 4.2 years (SD = 1.8). Without population benchmarks, these statistics are not presented as representative beyond the focal firm.

All measures were based on established scales, with team-referent wording adaptations for collaborative job crafting. For scales not previously validated in Italian, the back-translation procedure was employed to ensure linguistic and conceptual equivalence.

Empowering leadership was measured using the Empowering Leadership Questionnaire (ELQ) developed by Arnold et al. (2000). The scale assesses leader behaviors promoting autonomy, participation, and self-leadership among team members. The ELQ includes 38 items rated on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” A sample item is “Encourages work group members to express ideas/suggestions.” Cronbach's alpha for the scale was 0.82.

POS was measured using the eight-item Italian version of the Survey of Perceived Organizational Support (SPOS; Eisenberger et al., 1986), validated by Battistelli and Mariani (2011). The scale consists of 8 items rated on a 5-point Likert scale. A sample item is “The organization values my contribution to its well-being.” Cronbach's alpha was 0.77.

Team boosting behavior was measured using the scale developed by Fortuin et al. (2021). The instrument includes 18 items covering mood-enhancing, energizing, and uniting behaviors directed toward the team. A sample item is “I make sure that there is laughter in our team.” Following Fortuin et al. (2023), the three dimensions were combined into a single composite index representing overall team boosting behavior. Cronbach's alpha for the scale was 0.83.

Collaborative job crafting was assessed using an adapted version of the Job Crafting Scale (JCS) developed by Tims et al. (2012). Following prior team-level job crafting research (e.g. Leana et al., 2009; Mäkikangas et al., 2017; Tims et al., 2013), the original individual-level wording was adapted to capture collective and collaborative modifications of work tasks, resources, and interactions within teams. For example, “I try to develop my capabilities” was adapted as “In our team, we try to develop our capabilities.” The adapted scale includes 21 items rated on a 5-point Likert scale ranging from “never” to “very often.” Cronbach's alpha was 0.86.

Psychological safety was measured using Edmondson's (1999) Psychological Safety Scale. The scale includes 7 items rated on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” A sample item is “It is safe to take a risk on this team.” Cronbach's alpha was 0.87.

Controls were selected a priori on theoretical grounds (Bernerth and Aguinis, 2016). Age, gender, organizational tenure, and work experience were included at the individual level because career experience may relate to job crafting and team perceptions (Rudolph et al., 2017). Team size and mean team tenure were included at the team level because composition and size may affect communication and coordination (Hoegl, 2005; Stewart, 2006). The substantive pattern was unchanged with controls.

Multilevel structural equation modeling (MSEM) in Mplus separated within-team and between-team variation and estimated indirect and conditional associations at both levels (Preacher et al., 2010). Continuous predictors in interaction terms were centered, and confidence intervals were used for indirect effects.

The cross-sectional field design provided an initial multilevel test in an access-constrained project setting. It allowed us to assess within- and between-team covariance, aggregation, and the proposed pattern of direct, indirect, and conditional associations in one field sample. It cannot establish temporal precedence, causal mediation, or direction. Empirical estimates are therefore described as associations and statistical indirect effects.

Before hypothesis testing, we assessed data adequacy for MSEM and centered continuous predictors used in interactions. To further assess multicollinearity, variance inflation factors (VIFs) were examined. All VIF values remained well below the recommended threshold of 10, suggesting that multicollinearity did not represent a significant concern in the analyses.

All hypothesized relationships were tested while controlling for age, gender, organizational tenure, and work experience at the individual level, and for team size and average team tenure at the team level. The inclusion of these control variables did not substantially alter the significance or overall pattern of the hypothesized relationships.

Table 1 presents the means, standard deviations, and correlations for the study variables.

Table 2 summarizes the confirmatory factor model comparisons.

Before testing the structural model, a confirmatory factor analysis (CFA) was conducted to assess the adequacy of the measurement model and the discriminant validity of the focal constructs. The hypothesized five-factor model – including empowering leadership, POS, psychological safety, team boosting behavior, and collaborative job crafting as distinct constructs – demonstrated a satisfactory fit to the data: χ² = 734.62, df = 395, p < 0.001; RMSEA = 0.052; CFI = 0.938; TLI = 0.931; SRMR = 0.044. Collectively, these indices indicated an acceptable measurement model and supported the discriminant validity of the focal constructs.

Table 2 compares the five-factor model with four models that combine related constructs: EL with POS (M2), TBB with CJC (M3), PS with TBB (M4), and EL with PS (M5). For M5, the results (χ² = 1,018.06, df = 399, p < 0.001, RMSEA = 0.070, CFI = 0.887, TLI = 0.876, SRMR = 0.073, ΔCFI = −0.051) indicated poorer fit than M1 and supported the distinction between EL and PS. Their observed correlations were r = 0.42 at the individual level and r = 0.26 at the team level (Table 1).

The CFA used the overall individual-response covariance matrix; separate item-level within- and between-team CFAs were not estimated. With 60 teams and a large item set, the between-team model would be vulnerable to unstable estimation. We therefore limit the claim to overall discriminant validity. Correlations below 0.50 at both levels and the aggregation tests provide complementary evidence, not a multilevel measurement test.

Procedural safeguards for common method bias included anonymity, confidentiality, and psychological separation of construct blocks (Podsakoff et al., 2003, 2012). The first unrotated factor in Harman's test accounted for 31.4% of variance. This indicates that one factor did not dominate, but the test cannot rule out method variance (Fuller et al., 2016). The five-factor CFA and safeguards reduce concern about a single common factor; common method bias remains possible.

To assess the appropriateness of data aggregation, we conducted a series of statistical tests to evaluate within-group agreement and between-group variance. We first calculated the intraclass correlation coefficients (ICC) for each focal construct: empowering leadership, POS, psychological safety, team boosting behavior, and collaborative job crafting. The ICC(1) values were as follows: empowering leadership (ICC(1) = 0.11), POS (ICC(1) = 0.21), psychological safety (ICC(1) = 0.18), team boosting behavior (ICC(1) = 0.15), and collaborative job crafting (ICC(1) = 0.24). Although ICC(1) values above 0.12 are generally considered supportive of aggregation, values slightly below this threshold may still be considered acceptable when supported by additional aggregation statistics.

To further assess aggregation adequacy, rwg(j) and ICC(2) values were also examined. The rwg(j) values for all focal constructs exceeded the commonly accepted threshold of 0.70, indicating strong within-group agreement. In addition, ICC(2) values for POS, psychological safety, team boosting behavior, and collaborative job crafting ranged between 0.60 and 0.72, suggesting acceptable reliability of the aggregated team-level means.

Finally, one-way ANOVAs revealed statistically significant between-team differences for all focal constructs (p < 0.01). Collectively, these results provided empirical support for aggregating individual-level perceptions to the team level and for treating the focal groups as meaningful team-level units.

The hypothesized relationships were tested using MSEM in Mplus. The MSEM simultaneously estimated relationships at both the individual and team levels and demonstrated an acceptable overall model fit: χ² = 287.45, df = 205, p < 0.001; RMSEA = 0.036; CFI = 0.943; TLI = 0.932; SRMRwithin = 0.038; SRMRbetween = 0.043.

At the individual level, empowering leadership was positively associated with team boosting behavior (β = 0.21, p = 0.01), supporting Hypothesis 1. POS was also positively associated with team boosting behavior (β = 0.19, p = 0.02), supporting Hypothesis 2.

Team boosting behavior was positively associated with collaborative job crafting (β = 0.39, p < 0.001), supporting Hypothesis 3. Significant indirect associations emerged for empowering leadership (β = 0.08, 95% CI [0.02, 0.15], p = 0.014) and POS (β = 0.07, 95% CI [0.01, 0.12], p = 0.022), consistent with Hypotheses 4 and 5. Simultaneous measurement precludes temporal mediation claims.

Psychological safety also moderated the relationship between team boosting behavior and collaborative job crafting, as indicated by a significant interaction effect (β = 0.26, p = 0.005), supporting Hypothesis 6. Specifically, the positive relationship between team boosting behavior and collaborative job crafting became stronger under conditions of high psychological safety.

At the group level, empowering leadership was positively associated with team boosting behavior (β = 0.25, p = 0.014), and POS was also positively associated with team boosting behavior (β = 0.22, p = 0.028), supporting Hypotheses 1 and 2 at the team level. Team boosting behavior was positively associated with collaborative job crafting (β = 0.41, p < 0.001), supporting Hypothesis 3 at the team level.

At the group level, significant indirect associations emerged for empowering leadership (β = 0.10, 95% CI [0.03, 0.17], p = 0.015) and POS (β = 0.08, 95% CI [0.02, 0.14], p = 0.022), consistent with Hypotheses 4 and 5 within the limits of the cross-sectional design.

Psychological safety moderated the team-level association between team boosting behavior and collaborative job crafting (β = 0.28, p = 0.003), supporting Hypothesis 6. Figure 2 shows the conditional associations at both levels. The interaction was positive at both levels.

The detailed results of the multilevel SEM analyses are presented in Table 3.

This study examined whether empowering leadership and POS are associated with collaborative job crafting through team boosting behavior and whether psychological safety conditions these associations at individual and team levels. The integrated IMO and JD-R model organizes contextual resources, interpersonal mobilization, and proactive collective adaptation.

Empowering leadership and POS were positively associated with team boosting behavior, consistent with leader-based and organization-based resources accompanying energizing and uniting interactions. This pattern extends evidence linking empowering leadership with proactive behavior (Kim and Beehr, 2023) and POS with cooperative and discretionary behavior (Nazir et al., 2019). The cross-sectional design does not rule out reciprocal relationships between supportive contexts and positive team interaction.

Team boosting behavior was positively associated with collaborative job crafting, consistent with work describing collective crafting as a socially coordinated process (Leana et al., 2009; Mäkikangas et al., 2017; Tims et al., 2013). Mood-enhancing, energizing, and uniting interactions may accompany coordinated work redesign in project teams. Collaborative crafting may also create opportunities for team members to energize and unite one another.

The indirect effects were consistent with a resource-mobilization account. Empowering leadership and POS were related to collaborative job crafting through their covariance with team boosting behavior. Interpersonal energy, support, and coordination specify the proposed motivational pathway. Longitudinal evidence is needed to establish transmission over time.

Psychological safety strengthened the association between team boosting behavior and collaborative job crafting, extending research on safety, job crafting, and proactive adaptation (de Jong et al., 2025; Lee, 2022). Lower anticipated interpersonal risk may help teams use relational energy to revise routines, responsibilities, and activities. Empowering leadership may also contribute to safety, and reciprocal development remains possible.

The similar individual- and team-level patterns have different meanings in project work. Individual actions become collective adaptation when team members notice, negotiate, and incorporate them into shared workflows (Kozlowski and Klein, 2000). Firm-wide practices and professional culture may create common interpretive frames; repeated project-team interaction may generate distinct subclimates (Schneider et al., 2013). These top-down and bottom-up processes can explain similar coefficients across levels. The team-level interaction is consistent with the relevance of shared psychological safety.

This study contributes to the literature in several interrelated ways by advancing current understanding of proactive team behaviors, collaborative job crafting, and the interpersonal mechanisms through which contextual resources shape collective adaptation processes.

First, the study integrates IMO and JD-R perspectives in a multilevel account of proactive team functioning. IMO specifies the process architecture, JD-R explains the motivational relevance of resources, and team boosting behavior describes their interpersonal mobilization as energy, encouragement, information, and peer support. At the team level, this pathway requires shared exposure and sufficient convergence before individual experiences can be treated as a collective resource climate (Chan, 1998; Junker et al., 2025; Kozlowski and Klein, 2000).

Second, the findings contribute to the literature on collaborative job crafting by reinforcing its collective, relational, and coordination-based nature. Earlier job crafting research primarily conceptualized crafting as an individual-level process through which employees proactively modify their work (Tims et al., 2012; Wrzesniewski and Dutton, 2001). More recent studies have emphasized that job crafting may also emerge collectively within teams through mutual adjustment and coordinated action (Leana et al., 2009; Mäkikangas et al., 2017; Tims et al., 2013, 2022). The present findings extend this line of research by linking team boosting behavior to collaborative job crafting in project-based environments. In doing so, the study responds to recent calls for greater attention to collective forms of proactive work redesign in highly interdependent and knowledge-intensive environments (Demerouti, 2026; Park and Park, 2023; Wang et al., 2025).

Third, the study contributes to the literature on team boosting behavior by clarifying its conceptual role within proactive team processes. Previous studies introduced team boosting behavior as a construct capturing mood-enhancing, energizing, and uniting actions within teams (Fortuin et al., 2021, 2023), yet its position within broader models of team adaptation remained underdeveloped. The present findings suggest that team boosting behavior represents a dynamic interpersonal mechanism connecting contextual resources to collaborative job crafting. More specifically, the study conceptually and empirically differentiates team boosting behavior from collaborative job crafting itself. Whereas collaborative job crafting concerns the collective redesign of work structures, tasks, and coordination processes (Leana et al., 2009; Tims et al., 2022), team boosting behavior concerns the interpersonal energizing interactions that help sustain collective motivation and adaptive coordination. By distinguishing these constructs while empirically linking them, the study contributes to greater conceptual precision regarding the relational mechanisms underlying collective adaptation processes.

Fourth, the findings contribute to the literature on empowering leadership by clarifying its distinct relevance for proactive collective adaptation. Prior research has associated empowering leadership with autonomy, intrinsic motivation, engagement, and proactive behavior (Arnold et al., 2000; Kim and Beehr, 2023; Xiang et al., 2021), while also highlighting conceptual overlaps with other positive leadership approaches. The present study reinforces the specific relevance of empowering leadership in contexts requiring collaborative adaptation because this leadership style explicitly promotes participation, shared responsibility, initiative, and self-management within teams. In line with recent discussions regarding leadership resources within JD-R processes (Bakker et al., 2023; Junker et al., 2022), the findings suggest that empowering leadership may facilitate proactive collective adaptation by fostering interpersonal interaction patterns characterized by participation, mutual support, and collective engagement.

Fifth, psychological safety is identified as a conditional team state. The observed interaction suggests that safety may strengthen the association between team boosting and collaborative job crafting. Empowering leadership may help develop safety, and repeated boosting and crafting may reinforce it (Frazier et al., 2017; Joo et al., 2023). The focal moderation is therefore one plausible role within a broader temporal process that should be tested directly.

Finally, the analysis separates individual perceptions from shared team properties. Project-level adaptation requires expertise integration, role negotiation, and coordination under changing demands. Organizational culture may create cross-level similarity, and project interaction may create team subclimates (Schneider et al., 2013). The team-level moderation estimate motivates longitudinal research on climate emergence.

The findings suggest practices for evaluation in project-based and knowledge-intensive organizations. The cross-sectional design does not demonstrate intervention effects.

First, leadership development may emphasize participation, autonomy, shared responsibility, and expressions of confidence in team members' capabilities. These behaviors were associated with team boosting and may also contribute to psychological safety. Their effects should be assessed over time.

Second, recognition, transparent communication, mentoring, and developmental support can signal that employees' contributions are valued. Longitudinal evaluation should test whether these practices increase team boosting or collaborative crafting.

Third, teams may benefit from routines for raising concerns, discussing mistakes, proposing changes, and testing alternative work arrangements. Monitoring these routines can clarify how psychological safety, leadership, and collaboration develop together.

The findings also offer practical implications directly at the employee and team levels. Teams may benefit from intentionally cultivating everyday interaction patterns that reinforce collective energy and adaptive coordination. For example, informal peer support, constructive feedback, collaborative problem solving, and active knowledge sharing may strengthen the relational conditions that facilitate collaborative job crafting. In project-based teams, where employees often operate with high levels of autonomy and self-responsibility, these interaction processes may help teams collectively adjust workflows and coordination practices without relying exclusively on formal managerial intervention (see Buonocore et al., 2026).

Finally, organizations may consider incorporating collaborative reflection practices into project routines. Structured debriefings, peer learning sessions, and team-based discussions regarding work organization and coordination challenges may help teams proactively identify opportunities for collective adaptation and work redesign.

This study presents several limitations that should be acknowledged.

First, simultaneous measurement limits causal and temporal conclusions. The indirect effects do not establish mediation over time, and the interaction does not establish the temporal position of psychological safety. Collaborative crafting may generate team boosting, empowering leadership may shape safety, and successful collaboration may reinforce both. The findings are theoretically grounded associations (Antonakis et al., 2010).

Second, all focal constructs were self-reported. Anonymity, questionnaire separation, the five-factor CFA, and Harman's test address some concerns, but common method bias cannot be excluded. Harman's test has low sensitivity, and a first factor explaining less than 50% of the variance does not establish the absence of common method bias (Fuller et al., 2016; Podsakoff et al., 2003, 2012). Future studies should add peer, leader, and archival indicators.

Third, the study was conducted within a single Big Four consulting firm in Italy. Teams operating in consulting and project-based environments possess characteristics that may differ substantially from teams in more traditional organizational settings. These teams often operate with relatively high levels of autonomy, temporary membership structures, fluid coordination patterns, expertise-based influence dynamics, and strong performance pressures. Such contextual characteristics may systematically shape leadership processes, interpersonal dynamics, psychological safety, and collaborative job crafting behaviors. Consequently, the transferability of the findings to non-project-based or less knowledge-intensive organizational contexts should be approached cautiously.

Fourth, collaborative job crafting was modeled as an overall composite. The aggregate may conceal different patterns for structural resources, social resources, challenging demands, and hindering demands. Reducing hindering demands may require clearer leader guidance and may have different consequences from approach-oriented crafting (Fong et al., 2022; Lopper et al., 2024; Rudolph et al., 2017; Tims et al., 2013).

Fifth, the CFA supported the overall five-construct model, including the EL-PS distinction, but separate within- and between-team item-level CFAs were not estimated. Sixty teams are limited for a fully latent multilevel model. Level-specific correlations and aggregation evidence are informative, but larger samples are needed for multilevel measurement invariance and discriminant-validity tests.

Relatedly, leadership dynamics within project-based consulting teams may differ from those characterizing traditional line-management structures. In project environments, leadership is often more fluid, temporary, expertise-driven, and interaction-based than in formally hierarchical teams. This contextual specificity may influence both employees' perceptions of empowering leadership and the functioning of collective adaptation processes.

Future research should use temporally separated, multi-source designs. A feasible three-wave study could measure empowering leadership, POS, and controls at Time 1; psychological safety and peer-reported team boosting at Time 2; and collaborative job crafting from members, leaders, or project records at Time 3. Waves could follow project milestones. Cross-lagged MSEM or random-intercept cross-lagged models could compare forward, reverse, and reciprocal paths, with diary bursts capturing shorter dynamics.

Future studies should track when individual perceptions converge into shared safety climates and boosting patterns. Larger team samples would permit separate within- and between-team CFAs, measurement invariance tests, and comparisons of psychological safety as moderator, mediator, and reciprocal outcome.

In addition, future studies should examine these relationships across different organizational and cultural contexts. Comparing project-based consulting teams with more stable or hierarchical teams may provide valuable insights into the contextual conditions under which collaborative job crafting and team boosting behaviors emerge.

Dimensional analyses should test whether resource-increasing, challenge-increasing, and hindrance-reducing forms of collaborative crafting have different associations with empowering leadership, team boosting, and psychological safety. Hindrance reduction deserves particular attention because it can protect the team, shift burdens, or conflict with project requirements.

Digital and hybrid teamwork may alter the focal mechanisms. Reduced co-presence can make humor, affective cues, and informal energizing less visible and can slow safety-climate emergence. Persistent digital channels can widen participation, preserve knowledge, and make work redesign transparent. Future research should test communication synchronicity, media richness, and virtual interaction patterns as boundary conditions (Bailey et al., 2024; Gupta et al., 2024).

This study was conducted in accordance with ethical standards for research involving human participants. Participation was voluntary and anonymous, and informed consent was obtained from all participants. According to institutional guidelines, formal ethical approval was not required for this type of research.