Are University Students Ready to Work? The Role of Soft Skills and Psychological Capital in Building Sustainable Employability

Abstract

Soft skills are increasingly viewed as essential personal resources for sustainable employability, yet their combined role with Psychological Capital (PsyCap) and proactive career behaviors among university students remains insufficiently understood. Grounded in the Job Demands–Resources model, this study examines whether soft skills predict PsyCap, employability, job crafting (seeking challenges) and active job search behavior, and whether these relationships differ between STEM and non-STEM students. A sample of 501 Italian university students (mean age = 22.7) completed validated measures of soft skills, PsyCap (resilience and optimism), employability (employability, networking, social networks), seeking challenges and active job search. Structural equation modeling revealed that soft skills significantly predicted PsyCap (β = 0.57), employability (β = 0.45), seeking challenges (β = 0.61) and active job search (β = 0.25). Multi-group analyses showed configural invariance across STEM and non-STEM groups and generally comparable relationships, with slightly stronger effects of soft skills on PsyCap and employability for non-STEM students. These findings extend prior work by testing an integrated JD–R-informed employability model that links soft skills to both psychological resources and proactive career behaviors within the same SEM and across academic domains. Overall, findings highlight soft skills as foundational resources that enhance students’ psychological functioning and proactive career behaviors, ultimately supporting readiness for work and the development of adaptive, sustainable career paths.

1. Introduction

Soft skills are now widely recognized as essential transversal resources for building sustainable careers within a labor market characterized by accelerated transformations, globalization, and increasing organizational complexity. Although for a long time they were considered secondary to technical abilities, socio-behavioral competencies—such as communication, critical thinking, creativity, collaboration, and emotional regulation—are now understood as fundamental drivers of long-term employability (Heckman & Kautz, 2012; OECD, 2019). In a work environment shaped by digitalization and automation, where technical competences rapidly become obsolete, soft skills act as catalysts for continuous learning and adaptation, fostering individual resilience and re-skilling capacity (World Economic Forum, 2023; OECD, 2021).

A growing body of research highlights that specific soft skills—such as emotional intelligence, empathy, and stress management—contribute not only to performance but also to psychological well-being and engagement, which are essential components of long-term career sustainability (Feraco et al., 2024; Troshina et al., 2020). These competencies improve relational dynamics, enhance internal communication, and support participatory and responsible decision-making, which are critical to creating inclusive workplaces capable of promoting longer and healthier career trajectories (Deloitte, 2021; Carnevale et al., 2020). Furthermore, the value of soft skills extends to organizational and societal sustainability: abilities such as negotiation, collaborative leadership, and ethical responsibility are key to guiding innovation processes aligned with the common good and the reduction in inequalities (World Economic Forum, 2023). In this sense, soft skills act as a bridge between individual aspirations and collective goals, supporting the development of sustainable careers not only in terms of occupational continuity but also social impact. Promoting soft skills requires structured and integrated training initiatives in both educational and professional contexts. Mentoring programs, experiential learning, participatory work environments, and inclusive organizational cultures all facilitate the acquisition and long-term consolidation of these competencies (Battistelli & Odoardi, 2020; Robles, 2012).

1.1. The Relationship Between Soft Skills and Psychological Capital

The relationship between soft skills and Psychological Capital (PsyCap) is increasingly recognized as a key factor supporting both individual well-being and organizational performance. Soft skills—such as effective communication, emotional intelligence, problem-solving, and interpersonal competence—represent transversal abilities that facilitate adaptation to complex professional contexts and foster the development of the positive psychological resources that constitute PsyCap: self-efficacy, optimism, resilience, and hope (Luthans et al., 2007a, 2007b; Luthans & Youssef-Morgan, 2017).

A consolidated body of research shows that socio-emotional competencies are positively associated with workers’ ability to cope with organizational challenges while maintaining motivation and perseverance (Avey et al., 2011; Newman et al., 2014). Emotional intelligence and empathic communication support the regulation of emotions and the construction of trusting relationships, reinforcing core PsyCap components such as optimism and resilience (Salovey & Mayer, 1990; Cherniss, 2010). Recent contributions confirm the effectiveness of targeted training to enhance PsyCap, especially when linked to the simultaneous development of soft skills. Carter and Youssef-Morgan (2022) examined the effectiveness of face-to-face, online, and micro-learning interventions, finding that all modalities can successfully strengthen PsyCap. Furthermore, recent reviews (Reichard et al., 2024; Nguyen et al., 2024) confirm the growing centrality of PsyCap in learning and career processes.

Overall, the literature suggests a bidirectional relationship: developing soft skills contributes to enhanced PsyCap, while high levels of PsyCap, in turn, facilitate the further growth of socio-emotional competencies. These dynamic fosters constructive relationships, effective leadership, and proactive behaviors (Avey et al., 2011; Alessandri et al., 2018). Training modalities such as coaching, mentoring, and experiential learning have proven effective in strengthening these dimensions, promoting individual well-being and long-term career sustainability (Cameron & Spreitzer, 2012; Carter & Youssef-Morgan, 2022). From a conceptual standpoint, it is important to distinguish between soft skills and Psychological Capital, despite their empirical association. Soft skills refer to relatively stable behavioral and self-regulatory competencies (e.g., communication, adaptability, initiative) that guide how individuals act across situations, whereas PsyCap represents a set of positive psychological states—such as optimism and resilience—that reflect individuals’ momentary motivational and affective resources (Luthans et al., 2007a, 2007b; Luthans & Youssef-Morgan, 2017). Within this framework, soft skills can be theoretically positioned as antecedent resources that facilitate the development of PsyCap by enabling effective emotion regulation, proactive coping, and mastery experiences. In turn, PsyCap functions as a proximal psychological mechanism through which these competencies translate into adaptive attitudes and behaviors. This distinction is consistent with resource-based perspectives, which differentiate between capabilities (what individuals can do) and psychological states (how individuals feel and appraise their capacity to act), thereby supporting a directional pathway from soft skills to PsyCap rather than conceptual redundancy.

1.2. Soft Skills and Employability: A Multidimensional View

Employability is understood as a multidimensional capability that integrates competencies, psychological resources, and proactive career-management behaviors (Van der Heijde & Van der Heijden, 2006; Fugate et al., 2004; Cheng et al., 2022). In this perspective, soft skills represent a key element in supporting employability, as they enable individuals to translate technical knowledge into effective behaviors within dynamic and complex work environments. Far from being merely accessory competences, socio-emotional and self-regulatory abilities—such as communication, teamwork, flexibility, problem solving, leadership, and emotional management—enhance adaptability, continuous learning, and the development of professional networks, all of which are widely recognized as crucial for navigating career transitions successfully (Robles, 2012; Succi & Canovi, 2020).

Recent literature highlights that employability does not simply correspond to the possession of technical skills or the objective possibility of finding a job, but also includes psychological dimensions such as self-efficacy, optimism, professional identity, and the ability to manage opportunities proactively (Lo Presti & Pluviano, 2016; Jackson & Bridgstock, 2018). In this sense, soft skills act as “meta-competencies,” that is, transversal capacities that make it possible to exploit and operationalize technical expertise, while simultaneously supporting processes of self-regulation, resilience, and openness to change. Their transferable nature also makes them essential in a labor market characterized by digitalization, automation, and the constant redefinition of professional roles (Villegas, 2024; World Economic Forum, 2023; OECD, 2019). A significant contribution of soft skills to the construct of employability concerns their ability to facilitate the development and maintenance of social and professional capital. Effective relationships with peers, instructors, supervisors, and professionals increase opportunities for learning, visibility, and access to critical resources and information for labor-market entry (Artess et al., 2017). Communication skills, social sensitivity, and negotiation abilities support the activation of professional networks and participation in collaborative contexts—dimensions that the literature identifies as central to adapting to professional demands across different disciplinary domains (Yorke, 2006a, 2006b; Succi & Canovi, 2020).

From a psychosocial perspective, employability also encompasses attitudes and subjective perceptions such as confidence in one’s own resources and the belief in being able to successfully face the labor market. Soft skills help consolidate these aspects by facilitating uncertainty management, decision-making autonomy, and the ability to approach challenges positively—elements that converge with the role of psychological resources such as Psychological Capital (Luthans et al., 2007a, 2007b). Individuals with stronger socio-emotional competencies tend to perceive themselves as more effective, adaptable, and growth-oriented—characteristics strongly associated with higher perceived employability and more active job-search behaviors (Avey et al., 2011; De Vos, 2024). Overall, the literature converges in recognizing employability as the result of a dynamic interplay between behavioral competencies, psychological resources, and concrete career-oriented actions. Soft skills therefore constitute foundational resources that transform individual potential into effective behaviors, facilitating not only the transition from education to work but also the ability to sustain a flexible, sustainable career aligned with personal goals and values. Theoretical and empirical evidence indicates that employability emerges from an integrated system of behavioral competencies, psychological resources, and strategic career-management actions. In this perspective, soft skills function as higher-order enabling resources capable of transforming individual potential into adaptive and effective professional behaviors, supporting both the transition from education to work and the development of sustainable, flexible career trajectories aligned with one’s goals and values.

1.3. Soft Skills and Proactive Career Behaviors: Job Crafting and Seeking Challenges

The role of soft skills extends beyond shaping employability perceptions, influencing a wide range of proactive career behaviors that support long-term adaptability and career sustainability. Among these behaviors, job crafting—defined as the self-initiated process through which individuals modify their tasks, relationships, or perceptions of work—represents a central mechanism through which students and young professionals translate personal resources into developmental opportunities (Tims et al., 2012). Soft skills such as adaptability, initiative, emotional regulation, and interpersonal competence provide the cognitive and socio-emotional foundation that enables individuals to engage effectively in job crafting, particularly in the “seeking challenges” dimension. This facet reflects a proactive orientation toward exploring new tasks, assuming greater responsibility, and embracing learning opportunities, even in the presence of uncertainty or limited prior experience.

Research suggests that individuals with higher levels of soft skills are more likely to reinterpret demanding circumstances as opportunities for growth rather than as constraints, thereby activating motivational and behavioral pathways that foster professional development (Bakker & Demerouti, 2017). Competencies such as problem-solving, communication, and leadership enhance students’ readiness to take on complex or novel tasks, strengthening their ability to navigate transitions, expand their role boundaries, and cultivate a growth-oriented mindset. Furthermore, these proactive efforts contribute to a reinforcing cycle in which challenge-seeking behaviors build new competencies and psychological resources—such as resilience and self-efficacy—which in turn support further engagement in developmental activities (Avey et al., 2011; Luthans & Youssef-Morgan, 2017).

By integrating soft skills development with opportunities for job crafting—such as project-based learning, internships, and collaborative assignments—universities can foster students’ capacity to shape their own learning and career environments. Overall, soft skills and job crafting represent mutually reinforcing mechanisms: soft skills enable proactive career behaviors, while proactive engagement in challenging experiences catalyzes the development and consolidation of those same transversal competencies, ultimately contributing to adaptive and sustainable career trajectories.

1.4. Soft Skills and Active Job Search Behavior

Active job search behavior represents the point at which individual resources such as competencies, psychological capital, and perceptions of employability are translated into concrete actions aimed at entering the labor market. According to Blau’s (1994) classical definition, job search behavior comprises the intentional and observable activities individuals undertake to obtain employment, including both general search behaviors (e.g., exploring job postings, gathering information, consulting acquaintances) and focused behaviors (e.g., submitting targeted applications, contacting potential employers). In this sense, job search is a proactive process that demands time investment, planning, and sustained effort. Soft skills are key antecedents of active job search behavior, as they support its strategic (goal setting and task organization), relational (activation of professional networks), and self-regulatory (managing emotions and uncertainty) dimensions. Recent evidence shows that transversal competencies facilitate more effective and adaptive search strategies (Payne, 2018). Furthermore, several studies indicate that these personal resources are associated with greater intensity and persistence in job search activities and career Adaptability (Savickas, 2013) or career success (Zhang et al., 2022; De Vos et al., 2009), enhancing individuals’ capacity to cope with setbacks and maintain motivation over time (Fernández-Valera et al., 2020). Recent empirical evidence shows that individuals with stronger socio-emotional skills tend to use more diversified and proactive job search strategies such as activating social networks, leveraging informal connections, and engaging in exploratory behaviors (Jansen et al., 2020; Van Hooft et al., 2021; Pan et al., 2018).

In line with the JD-R model (Bakker & Demerouti, 2017), soft skills can therefore be conceptualized as personal resources that foster both internal assets (PsyCap, self-efficacy, motivation) and strategic and relational capabilities, increasing the likelihood that students engage in structured, persistent, and goal-directed job search behaviors.

1.5. The Moderating Role of Academic Domain (STEM vs. Non-STEM)

Although several studies have highlighted differences between STEM and non-STEM pathways, showing that students from different fields tend to attribute varying importance to transversal competencies (Succi & Canovi, 2020), more recent literature indicates that the impact of soft skills on career readiness and psychological well-being is generally stable and consistent across academic domains. Research conducted in heterogeneous university settings shows that competencies such as communication, adaptability, teamwork, and emotional regulation similarly predict higher perceived efficacy, elevated levels of Psychological Capital, and stronger action-oriented tendencies, regardless of students’ field of study (Knight & Yorke, 2003; Bridgstock, 2009). Likewise, multi-departmental studies demonstrate that soft skills are consistently associated with employability outcomes and proactive career behaviors, such as job crafting and job search engagement, without substantial differences between technical-scientific and social-humanistic disciplines (Tomlinson, 2017). In line with this perspective, several universities have progressively introduced structured programs and targeted interventions aimed at strengthening transversal competencies through workshops, training activities, and project-based initiatives accessible to students from different disciplinary backgrounds. Among the most relevant examples, various Italian universities have implemented initiatives designed to enhance soft skills and support students’ transition into the labor market (Ricchiardi & Emanuel, 2018; Emanuel et al., 2021; Chignoli et al., 2020; Ingusci et al., 2022). Although diverse in format and scope, these initiatives share a transversal and integrative view of soft skills development within higher education curricula.

To improve clarity, the discussion is organized according to the study hypotheses (H1–H5). We first discuss the direct effects of soft skills on Psychological Capital, Employability, Seeking Challenges, and Active Job Search Behavior (H1–H4), and then address the extent to which these relationships differ across academic domains (H5). Taken together, this body of evidence supports the hypothesis that the strength of the relationships between soft skills, Psychological Capital, employability, and proactive behaviors does not differ between STEM and non-STEM students. Overall, transversal competencies appear to operate as universal and widely transferable resources that contribute to the development of adaptive and sustainable career trajectories. Although each of the hypothesized links has been examined in prior research, the novelty of the present study lies in their theoretical integration and contextual application. Specifically, we (a) conceptualize soft skills as foundational personal resources within the JD–R framework and test their simultaneous associations with both psychological resources (Psychological Capital) and proactive employability-related behaviors (job crafting—seeking challenges; active job search) within a single SEM; (b) operationalize employability as a multidimensional higher-order construct (employability activities, networking, and social networks), thus capturing both internal adaptability and relational/digital capital; and (c) provide contextual evidence from Italian university students and examine whether the model holds across STEM vs. non-STEM domains, offering insight into the generalizability of this resource-based employability process in a higher-education setting.

1.6. Conceptual Model and Hypotheses Summary

The conceptual framework of this study is grounded in the Job Demands–Resources (JD-R) model (Bakker & Demerouti, 2017), which explains well-being, motivation, and performance as outcomes of the dynamic interplay between job demands—those aspects that require sustained effort and may lead to strain—and job resources, which promote growth, motivation, and resilience. Within this framework, personal resources such as optimism, self-efficacy, and adaptability play a crucial role: they help individuals cope with challenges, stimulate engagement, and initiate positive gain spirals that enhance both psychological functioning and performance (Hobfoll, 2011). Extending this logic to the educational context, university students can be seen as active agents who mobilize their resources to construct sustainable and meaningful career paths (Van der Heijde & Van der Heijden, 2006; De Vos et al., 2020). Higher education, in this perspective, is not merely a place for knowledge acquisition but a developmental arena where students learn to balance demands (e.g., academic workload, uncertainty about the future) and resources (e.g., social support, emotional competence) to prepare for transitions into the labor market. In the present study, Soft Skills are conceptualized as key personal resources that sustain this process. They include a broad range of socio-emotional and self-regulatory competencies—such as teamwork, communication, adaptability, initiative, empathy, and emotional control—that enable students to navigate complex environments and interact effectively with others (Robles, 2012; Succi & Canovi, 2020). According to the JD-R model, such resources enhance motivational processes by fostering individuals’ capacity to self-regulate, manage uncertainty, and engage proactively with their learning and career contexts (Bakker & Demerouti, 2017). One of the most relevant psychological mechanisms linking soft skills to positive outcomes is Psychological Capital (PsyCap)—a higher-order construct encompassing resilience, optimism, self-efficacy, and hope (Luthans et al., 2007a, 2007b). In line with this view, PsyCap is conceptualized in the present study as a theoretical mediating mechanism linking soft skills to employability-related outcomes. While the cross-sectional design does not allow for strict causal inference, positioning PsyCap as an intervening psychological process is theoretically justified by the JD–R model and by prior longitudinal and intervention studies showing that personal competencies foster PsyCap, which in turn predicts proactive and adaptive behaviors (Avey et al., 2011; Luthans & Youssef-Morgan, 2017).

PsyCap represents a personal resource that broadens individuals’ thought–action repertoires and increases their confidence in achieving goals despite obstacles. Theoretically, PsyCap can be enhanced through the development of soft skills that promote self-awareness, emotional regulation, and interpersonal effectiveness. In this sense, soft skills can be regarded as antecedents or enabling conditions for the development of PsyCap. At a behavioral level, the model integrates Employability as a multidimensional outcome that includes both competence-based and psychosocial components (Van der Heijde & Van der Heijden, 2006; Lo Presti & Pluviano, 2016). Employability reflects an individual’s ability to identify, obtain, and sustain work that fits personal and contextual needs. Within the JD-R perspective, employability can be seen as the outcome of effective resource mobilization: students who possess higher personal resources are more likely to transform them into proactive behaviors such as networking, adaptability, and continuous learning. In addition, Job Crafting, and particularly the dimension seeking challenges (Tims et al., 2012), is considered a behavioral manifestation of resource utilization. Job crafting reflects an individual’s proactive efforts to shape tasks and relationships to better align them with personal strengths and motivations. Students with well-developed soft skills may be more inclined to engage in such challenge-seeking behaviors, viewing demanding academic or professional situations as opportunities for growth rather than as threats. Finally, the model includes Active Job Search Behavior (Blau, 1994) as a concrete outcome of employability readiness. This construct captures intentional actions taken to explore and secure employment opportunities. In the JD-R framework, such proactive behaviors represent the translation of accumulated personal resources into external action: students who possess higher levels of adaptability, communication, and initiative are more likely to take strategic steps toward entering the labor market.

Overall, the model posits that Soft Skills act as foundational personal resources that enhance both internal psychological states (through PsyCap) and external employability-related behaviors (through job crafting, networking, and job search). Psychological Capital is expected to mediate these effects, functioning as a psychological mechanism that transforms competencies into sustainable motivation and engagement. Furthermore, the model assumes that these relationships may vary across academic domains (STEM vs. non-STEM), reflecting disciplinary differences in the relative importance of interpersonal versus technical competencies (Succi & Canovi, 2020; Artess et al., 2017). In doing so, this study moves beyond examining isolated associations by testing a resource-based employability process in which soft skills operate as a core personal resource linked to both psychological capital and observable proactive career behaviors within the same model. This integrated approach allows us to assess whether these pathways are stable across academic domains (STEM vs. non-STEM) in the Italian higher-education context. Considering the abovementioned literature, the hypotheses of this study are the following:

H1. 

Soft Skills positively predict Psychological Capital.

H2. 

Soft Skills positively predict Employability.

H3. 

Soft Skills positively predict Job Crafting—specifically, the dimension of Seeking Challenges.

H4. 

Soft Skills positively predict Active Job Search Behavior.

H5. 

The strength of these relationships did not differ between STEM and non-STEM students.

Hypotheses are graphically represented in Figure 1.

2. Methods

2.1. Participants and Procedure

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of University of Salento (protocol code 0037965—23 February 2022) for studies involving humans. All participants gave their informed consent prior to completing the questionnaire, in line with the ethical principles of the Declaration of Helsinki and its later amendments. The research protocol received approval from the Ethics Committee of the University of Salento (protocol no. 0037965—23 February 2022). Participants’ anonymity and data confidentiality were strictly ensured throughout all phases of the study, which involved 501 university students enrolled in Italian higher education institutions between 2022 and 2024. Participants’ ages ranged from 18 to 49 years, with an average age of 22.7 years (SD = 4.12), reflecting a heterogeneous sample composed primarily of young adults at different stages of their academic path. Overall, the majority were female (73.3%), while 26.7% identified as male. Regarding the level of education, most participants were pursuing a bachelor’s degree (76.2%), followed by master’s degree students (22.6%) and a small number enrolled in single-cycle programs (1.2%). The distribution across academic years showed that 22.6% of students were in their first undergraduate year, 40.1% in their third year, and 19.6% in the second year of a master’s degree program, indicating a prevalence of students nearing the end of their academic cycle. Only a small percentage (less than 2%) attended single-cycle programs such as medicine or law.

When distinguishing between disciplinary areas, 240 students (47.9%) were enrolled in STEM (Science, Technology, Engineering, and Mathematics) programs, while 261 students (52.1%) attended non-STEM programs, including social sciences, humanities, and education-related degrees. The average age was slightly higher among STEM students (M = 23.0, SD = 3.13) than among non-STEM students (M = 22.3, SD = 4.83), suggesting that technical and scientific programs may attract a slightly older population, possibly due to longer academic trajectories or delayed entry. In terms of gender composition, a strong imbalance emerged between the two groups. Among STEM students, 57.1% were female and 42.9% male, while in non-STEM fields, 88.1% were female and only 11.9% male, reflecting a pattern commonly observed in the Italian university system, where women tend to predominate in social and humanistic disciplines, whereas men are more represented in technical-scientific programs. Additionally, only 8.6% of respondents declared being outside the regular course duration, with a slightly higher proportion among STEM students (10.8%) than among non-STEM students (6.5%).

The data collection was conducted through an online questionnaire administered via Google Forms, distributed through institutional mailing lists, academic communication channels, and student associations. The survey was accessible for several weeks, allowing broad participation across departments and degree programs. Participation was entirely voluntary, anonymous, and uncompensated. Respondents were informed about the objectives and procedures of the research and provided their informed consent before beginning the survey. The research protocol was reviewed and approved by the Ethics Committee of the University of Salento, in accordance with the ethical standards of the Declaration of Helsinki (2013 revision). Participants were assured of confidentiality, the right to withdraw at any time, and that their data would be used solely for research purposes.

2.2. Measures

All constructs were measured through validated self-report questionnaires administered online via Google Forms between 2022 and 2024. Responses were collected on a 5-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Internal consistency and convergent validity were assessed using Cronbach’s alpha (α), McDonald’s omega (ω), and Average Variance Extracted (AVE). All scales met the conventional thresholds (α and ω ≥ 0.70; AVE ≥ 0.50), indicating satisfactory reliability and convergent validity. In particular:

• Soft skills were assessed through an 18-item scale developed to measure transversal behavioral competencies relevant to both academic and professional settings. The instrument captures areas such as initiative, flexibility, analytical and conceptual thinking, teamwork, leadership, sensitivity to others, and emotional self-control. The scale demonstrated excellent reliability (α = 0.94; ω = 0.95) and acceptable convergent validity (AVE = 0.51). This instrument was adapted from multidimensional frameworks of transversal competencies used in higher education contexts (Spencer & Spencer, 2008). As no validated Italian version of the original framework was available, the items were translated and adapted into Italian by the authors, ensuring conceptual equivalence with the original multidimensional competency model proposed by Spencer and Spencer (2008).
• Psychological Capital (PsyCap) was conceptualized as a higher-order construct composed of Resiliency and Optimism. Resiliency (4 items; α = 0.86; ω = 0.90; AVE = 0.64) reflects the ability to recover from setbacks and maintain focus under stress (e.g., “I am good at reacting when I encounter obstacles (e.g., when I fail an exam).”). Optimism (4 items; α = 0.88; ω = 0.90; AVE = 0.64) captures positive expectations about the future and confidence in one’s own capacities (e.g., “Even if unexpected events occur, I remain confident about the future”). The items were adapted from the Psychological Capital Questionnaire (Mazzetti et al., 2018), ensuring conceptual consistency with the original four-component model (hope, efficacy, resilience, and optimism).
• Employability was modeled as a second-order construct including three subdimensions: Employability (2 items; α = 0.82; AVE = 0.69), Networking (2 items; α = 0.80; AVE = 0.68), and Social Networks (2 items; α = 0.75; AVE = 0.61). The subdimensions respectively assess perceived career adaptability, the ability to maintain professional relationships, and the effective use of digital social connections to support professional growth. An example item is “Knowing new people is an opportunity that I miss rarely”. The measure was developed following the multidimensional conceptualization of employability by Fugate et al. (2004) and later operationalizations in higher education contexts (e.g., Lo Presti et al., 2019).
• Active job search behavior was assessed through the Job Search Behavior Scale developed by Blau (1994), which measures the frequency of proactive behaviors oriented toward employment seeking. Participants rated how often they engaged in job-seeking actions such as contacting potential employers or browsing job listings. An example item is “Filled out a job application”. The four-item scale showed excellent internal consistency (α = 0.87; ω = 0.90; AVE = 0.66), supporting its reliability in evaluating active career self-management behaviors. Since a validated Italian version of the Job Search Behavior Scale was not available, the scale was translated into Italian by the authors using an ad hoc translation procedure aimed at preserving the meaning and intent of the original items (Blau, 1994).
• The Seeking Challenges dimension of job crafting was measured using three items from the Italian validation of the Job Crafting Scale by Ingusci et al. (2018), which captures employees’ proactive efforts to engage in new learning opportunities or take on demanding tasks. An example item is “If there are new developments, I am one of the first to learn about them and try them out.” The subscale demonstrated acceptable internal consistency (α = 0.72; ω = 0.75) and good convergent validity (AVE = 0.51), confirming its adequacy in representing proactive, growth-oriented behavior.

3. Results

3.1. Discriminant Validity Assessment

Discriminant validity was first assessed through the Fornell and Larcker (1981) criterion and the Heterotrait–Monotrait (HTMT) ratio. As shown in

Table 1. Fornell-Larcker matrix values.

Construct	Soft Skills	Employability	Network	Challenges	Social Network	Resiliency	Optimism	Active Job Search
Soft Skills	0.67	0.41	0.39	0.61	0.34	0.49	0.52	0.25
Employability activities	0.41	0.83	0.79	0.48	0.69	0.26	0.28	0.12
Networking	0.39	0.79	0.82	0.46	0.66	0.25	0.26	0.12
Seeking challenges	0.61	0.48	0.46	0.70	0.40	0.38	0.40	0.27
Social Network	0.34	0.69	0.66	0.40	0.78	0.22	0.23	0.10
Resiliency	0.49	0.26	0.25	0.38	0.22	0.80	0.78	0.20
Optimism	0.52	0.28	0.26	0.40	0.23	0.78	0.80	0.21
Active Job Search	0.25	0.12	0.12	0.27	0.10	0.20	0.21	0.81

, the square roots of the Average Variance Extracted (AVEs; diagonal values) were higher than the corresponding inter-construct correlations, indicating satisfactory discriminant validity across all factors. The highest correlations emerged between Employability activities and Network (r = 0.69–0.78), reflecting their conceptual overlap within the employability domain, yet the square roots of AVE remained higher than these coefficients (√AVE ranging from 0.67 to 0.83). The Resiliency and Optimism factors of Psychological Capital also showed moderate inter-correlation (r = 0.78), consistent with theoretical expectations. The HTMT ratios (

Table 2. HTMT matrix.

Constructs	Active Job Search	Seeking Challenges	Employability Activities	Networking	Optimism	Resilience	Social Networks	Soft Skills
Active Job Search	—	0.32	0.17	0.08	0.23	0.20	0.14	0.27
Seeking challenges		—	0.51	0.40	0.49	0.46	0.38	0.63
Employability activities			—	0.80	0.29	0.31	0.70	0.43
Networking				—	0.27	0.23	0.72	0.35
Optimism					—	0.75	0.38	0.54
Resilience						—	0.31	0.52
Social Networks							—	0.28
Soft Skills								—

) were all below the conservative threshold of 0.85 (Henseler et al., 2015), further confirming that each construct captured a distinct yet related facet of students’ psychological and behavioral resources. Taken together, these results supported the assumption of discriminant validity for the measurement model.

3.2. Measurement Model

The overall SEM model was tested on the full sample (N = 501) using the robust maximum likelihood estimator (MLR). The measurement model demonstrated adequate fit to the data: χ²(682) = 1651.19, p < 0.001; CFI = 0.90; TLI = 0.90; RMSEA = 0.058 (90% CI [0.054, 0.062]); and SRMR = 0.060. These indices fall within the range considered acceptable for complex models (Hu & Bentler, 1999; Kline, 2016). All factor loadings were positive and statistically significant (p < 0.001). Within the Soft Skills latent construct, standardized loadings ranged from 0.52 (Persuasiveness) to 0.79 (Transferability), with most indicators exceeding 0.65, confirming a strong and coherent factor structure. Indicators reflecting teamwork (Teamwork, 0.75), leadership (0.64), adaptability (0.70), and trust (0.71) were among the most salient contributors, highlighting the multidimensional and integrative nature of soft skills in academic and work-related settings. For Psychological Capital, both subdimensions loaded strongly on the higher-order factor (Resiliency = 0.86; Optimism = 0.91), confirming the internal coherence of this resource-based construct. Within each subdimension, all items displayed substantial loadings (0.66–0.87 for Resiliency; 0.75–0.86 for Optimism). Regarding Employability, the second-order latent variable was well represented by its three components: Employability (0.91), Network (0.87), and Social Networks (0.76). The pattern of standardized loadings was consistent with prior conceptualizations (Fugate et al., 2004), suggesting that students’ perceived employability integrates both internal adaptability and external relational capital. For Job Crafting—Seeking Challenges, the three indicators loaded between 0.63 and 0.83 (p < 0.001), supporting the reliability of this proactive behavioral dimension. Finally, the Active Job Search Behavior scale also demonstrated strong internal structure, with standardized loadings ranging from 0.53 to 0.92, confirming the unidimensionality of the construct. Collectively, these results supported the convergent and discriminant validity of the measurement model, as indicated by high and statistically significant loadings, satisfactory fit indices, and acceptable AVE values for all latent variables (ranging from 0.51 to 0.69).

3.3. Structural Model Assessment

The hypothesized structural relationships were tested by examining the standardized path coefficients among the latent variables. As shown in

Table 3. Path coefficients of the overall model.

Path	β	SE	z	p-Value
Soft Skills → Psychological Capital	0.57	0.06	9.35	<0.001
Soft Skills → Employability	0.45	0.07	6.63	<0.001
Soft Skills → Seeking Challenges	0.61	0.05	11.29	<0.001
Soft Skills → Active Job Search	0.25	0.08	3.21	0.001

, all the structural paths from Soft Skills to the endogenous variables were positive and statistically significant (p < 0.001). A graphical representation of the model is presented in Figure 2. Specifically, Soft Skills exerted a strong positive effect on Psychological Capital (β = 0.57, p < 0.001), confirming that transversal competencies—such as adaptability, initiative, and emotional regulation—enhance students’ optimism and resilience. In turn, Soft Skills were also a significant predictor of Employability (β = 0.45, p < 0.001), suggesting that behavioral and interpersonal competences play a crucial role in shaping self-perceptions of career adaptability and professional networking. A moderate association was also found between Soft Skills and Active Job Search Behavior (β = 0.25, p < 0.001), indicating that individuals with higher levels of transversal competencies are more likely to engage in proactive behaviors to secure job opportunities. Moreover, Soft Skills were strongly related to Seeking Challenges (β = 0.61, p < 0.001), implying that students with higher soft skill levels tend to adopt a more exploratory and growth-oriented mindset in approaching academic and professional challenges. Overall, the structural model accounted for substantial portions of variance: 33% of Employability, 40% of Psychological Capital, and 25% of Seeking Challenges.

3.4. Multi-Group Analysis: STEM vs. Non-STEM Students

To test whether the model structure was invariant across academic domains, a multi-group SEM was performed by comparing students from STEM (n = 240) and non-STEM (n = 261) degree programs. Coefficients can be retrieved in

Table 4. Path coefficients in the multi-group analysis.

Path	STEM (β)	Non-STEM (β)	Δ
Soft Skills → Psychological Capital	0.51	0.62	+0.11
Soft Skills → Employability	0.44	0.46	+0.02
Soft Skills → Seeking Challenges	0.57	0.62	+0.05
Soft Skills → Active Job Search	0.25	0.22	−0.03

. The multigroup configurational model achieved acceptable fit indices: χ²(1364) = 2604.06, p < 0.001; CFI = 0.884; TLI = 0.874; RMSEA = 0.064; SRMR = 0.069. The results support configural invariance, indicating that the same latent structure applies across the two groups. Within the STEM group, all standardized loadings were significant (p < 0.001) and comparable to those of the overall sample, ranging from 0.57 to 0.86 for Soft Skills and from 0.73 to 0.94 for Employability components. The structural paths showed that Soft Skills significantly predicted Psychological Capital (β = 0.51, p < 0.001), Employability (β = 0.44, p < 0.001), and Seeking Challenges (β = 0.57, p < 0.001). The link with Active Job Search Behavior was smaller but still significant (β = 0.25, p = 0.005), suggesting that in STEM students, transversal competencies are associated with both psychological resources and proactive behaviors but are less directly related to job-seeking frequency. For the non-STEM group, a similar pattern emerged, but with slightly higher standardized effects on both Psychological Capital (β = 0.62, p < 0.001) and Employability (β = 0.46, p < 0.001). The path from Soft Skills to Seeking Challenges was also robust (β = 0.62, p < 0.001), while the effect on Active Job Search Behavior was moderate but significant (β = 0.22, p = 0.019). These results suggest that for non-STEM students—who often engage in less structured and more relational academic pathways—soft skills play an even more central role in enhancing both internal (psychological capital) and external (employability) resources. Comparing the variances explained across groups, the model accounted for 35% of Employability and 37% of Psychological Capital in STEM students, whereas the percentages were slightly higher for non-STEM students (42% and 39%, respectively). This pattern supports the interpretation that soft skills may represent a more salient driver of perceived employability and adaptive potential among non-STEM students, who rely more on relational and communicative competences to navigate career transitions.

4. Discussion

The present study examined the associations between soft skills, Psychological Capital, multidimensional employability, and proactive career behaviors within a JD–R–informed framework applied to university students. While the hypothesized relationships are individually well established in the literature, the main contribution of this study lies in the integrated modeling of these constructs within a single structural framework, allowing us to examine how transversal competencies relate simultaneously to internal psychological resources and to career self-management behaviors in higher education contexts. Moreover, by testing the model across STEM and non-STEM students, the study provides evidence on the generalizability of this pattern across academic domains, albeit in an exploratory manner. From a conceptual standpoint, potential overlap between soft skills, Psychological Capital, and employability was addressed both theoretically and empirically. Discriminant validity analyses supported the distinctiveness of the constructs, while their conceptual differentiation reflects different levels of analysis: soft skills as behavioral competencies, Psychological Capital as a malleable psychological resource state, and employability as a multidimensional outcome encompassing perceptions and behaviors. Treating these constructs as analytically distinct allows a clearer interpretation of how competencies are associated with psychological resources and with adaptive career-related actions. Consistent with H1, soft skills were positively associated with Psychological Capital, suggesting that students reporting higher levels of transversal competencies also tend to report higher optimism and resilience. This association aligns with positive organizational behavior perspectives (Luthans et al., 2007a, 2007b) and supports the idea that socio-behavioral competencies co-occur with more favorable psychological resource states. However, given the cross-sectional design, these findings should be interpreted as associative rather than causal. In line with H2, soft skills were positively associated with employability. This result reinforces competence-based models of employability (Van der Heijde & Van der Heijden, 2006), suggesting that students with stronger interpersonal and self-regulatory competencies also perceive themselves as more capable of navigating career transitions and mobilizing relational resources. Rather than reflecting the possession of technical skills alone, employability appears to be linked to students’ ability to translate competencies into adaptive career-relevant perceptions and behaviors. Regarding H3, soft skills showed a strong association with the job crafting dimension of Seeking Challenges. This finding indicates that students reporting higher transversal competencies also report a greater tendency to proactively pursue learning opportunities and demanding tasks. Within the JD–R framework, this pattern is consistent with the view that personal resources are associated with proactive behaviors aimed at growth and development (Bakker & Demerouti, 2017). Importantly, this result highlights the behavioral relevance of soft skills beyond employability perceptions. With respect to H4, soft skills were positively associated with Active Job Search Behavior, although the effect size was smaller compared to other outcomes. This suggests that transversal competencies may facilitate the translation of employability-related resources into concrete career actions, such as job search activities, particularly during early career transitions. Again, this association should be interpreted cautiously and not as evidence of directional or developmental effects. Finally, H5 was partially supported. Multi-group analyses indicated configural invariance across STEM and non-STEM students, suggesting that the overall pattern of relationships was comparable across academic domains. Although some coefficients appeared slightly stronger among non-STEM students, no formal tests of structural path differences were conducted; therefore, these between-group differences should be considered exploratory and descriptive, rather than confirmatory. This finding suggests that soft skills function as broadly relevant resources across disciplines, while leaving open the possibility of nuanced contextual differences. Taken together, these findings support a resource-based interpretation of sustainable employability in higher education, in which soft skills, Psychological Capital, and proactive behaviors are interrelated but distinct components associated with students’ readiness to manage career transitions. Rather than proposing new causal mechanisms, the present study contributes by empirically integrating these elements within a coherent framework and by clarifying their joint relevance in a university population.

5. Practical Implications

Practically, these findings support the growing consensus that universities play a pivotal role in developing employability not only as a set of competencies but also as a mindset (Knight & Yorke, 2003). By embedding soft-skills training within disciplinary programs and integrating reflective components that promote self-efficacy and optimism, institutions can help students internalize employability as part of their identity. Short-term interventions that explicitly target Psychological Capital—such as micro-learning modules, experiential laboratories, or coaching-based programs—may further amplify these effects, as meta-analytic evidence indicates that PsyCap can be effectively developed through structured experiences (Avey et al., 2011). Such initiatives are particularly relevant in a post-pandemic landscape marked by technological acceleration and occupational uncertainty, where adaptability, confidence, and social connectedness have become essential resources for sustainable transitions from education to work. From a pedagogical standpoint, the results suggest that employability training should move beyond isolated workshops or “add-on” activities. Instead, universities should embed transversal competencies as longitudinal learning outcomes integrated into curricula, assessment, and mentoring. Instructors can design project-based learning experiences, interdisciplinary collaborations, and community engagement activities that explicitly require teamwork, critical thinking, and reflective practice. Moreover, creating structured feedback systems—where students are encouraged to evaluate their soft skills alongside technical achievements—can reinforce awareness and agency in personal development. Embedding these elements within course design may also enhance Psychological Capital by providing mastery experiences and opportunities to experience progress and success. At the institutional level, the findings advocate for an ecosystemic approach to employability. Universities could strengthen their partnerships with employers, alumni, and local communities to co-design learning experiences that bridge academic and professional environments. Such collaborations may help students understand how transversal competences translate into career behaviors such as networking and challenge-seeking, while also allowing employers to clarify expectations and foster shared responsibility in skill development. Furthermore, introducing assessment tools that integrate self-evaluation and peer feedback could help universities monitor the evolution of students’ employability resources throughout their programs. In policy terms, the results contribute to the broader debate on how higher education can align with the United Nations Sustainable Development Goals, particularly Goal 8, which promotes decent work and economic growth. Investing in soft skills and Psychological Capital responds directly to the call for inclusive, lifelong learning and sustainable employment. Programs that foster resilience, adaptability, and social connectedness not only enhance students’ career readiness but also contribute to societal well-being by preparing graduates to navigate complex, changing labor markets ethically and collaboratively. Finally, these findings invite educators and policymakers to rethink employability as a shared responsibility across institutional levels. At the micro level, students should be empowered as active agents capable of shaping their learning and career paths. At the meso level, academic departments and faculty members should embed reflective, feedback-rich environments that stimulate growth-oriented mindsets. At the macro level, institutions and governments should promote structural initiatives—such as national soft-skills frameworks, certification systems, or inter-university alliances—that ensure continuity between education and employment ecosystems. Overall, the present results suggest that fostering soft skills within higher education is not merely a response to market demands but a strategic investment in sustainable human development. By linking behavioral competences to psychological resources and employability outcomes, universities can contribute to forming graduates who are not only work-ready but also future-ready—capable of maintaining employability, well-being, and purpose across transitions in an evolving world of work.

In conclusion, this study contributes to the understanding of how soft skills shape both the psychological and behavioral dimensions of employability. By integrating psychological capital and proactive behaviors within a single model, it offers empirical support for the view that employability is a sustainable, multidimensional capacity rooted in personal and contextual resources. The evidence that soft skills significantly enhance resilience, optimism, and career agency underscores their central role in preparing graduates for dynamic labor markets. These findings call for educational policies that prioritize the systematic development of transversal competencies alongside disciplinary expertise. Promoting soft skills within higher education is not merely a response to employers’ demands but a strategic investment in students’ long-term adaptability, well-being, and professional sustainability.

6. Limitations and Conclusions

Several limitations should be considered when interpreting the present findings. First, the study relied exclusively on self-report measures, which may introduce common-method bias and inflate associations among variables. Although this risk was partially mitigated through the use of distinct latent constructs and supported by discriminant validity analyses (Fornell–Larcker and HTMT), future research would benefit from multi-source designs, integrating peer, supervisor, or behavioral indicators of employability and career-related behaviors. Second, the cross-sectional design precludes causal inferences. Although Psychological Capital was theoretically framed as an intervening mechanism linking soft skills to employability-related outcomes, formal mediation analyses were not conducted. Accordingly, all interpretations referring to developmental or processual mechanisms should be understood as theoretically grounded rather than empirically tested causal claims. Longitudinal and experimental studies are needed to examine directionality and to test whether changes in soft skills are followed by increases in psychological resources and proactive career behaviors over time. Third, the sample was drawn from a single national context, which may limit generalizability. Cultural and institutional characteristics of higher education systems can shape how employability and psychological resources are developed and perceived. Cross-national replication is therefore required to assess the robustness of the proposed model across different educational and labor-market contexts. Fourth, although multi-group analyses supported configural invariance across STEM and non-STEM students, no formal tests of structural path differences were conducted. Consequently, observed between-group variations should be interpreted as exploratory rather than confirmatory. Future research should apply more fine-grained disciplinary classifications and formal invariance testing to examine contextual differences more rigorously. Finally, the study focused primarily on individual-level variables, without explicitly modeling contextual factors such as institutional support, mentoring quality, or labor-market conditions. Integrating these variables into multi-level or longitudinal models would provide a more comprehensive understanding of how personal and environmental resources jointly contribute to sustainable employability.

Despite these limitations, the present study contributes to the literature by offering an integrated, resource-based framework that links soft skills, Psychological Capital, employability, and proactive career behaviors within a higher-education context. Rather than proposing novel causal mechanisms, the study clarifies how these constructs are empirically distinct yet interrelated, and how transversal competencies are associated with both internal psychological resources and career self-management actions. These findings provide a useful foundation for future longitudinal research and for the development of evidence-informed educational interventions aimed at fostering sustainable employability among university students.