How socially responsible human resource management fosters work engagement: the role of perceived organizational support and affective organizational commitment

Purpose – In recent years, efforts to reinforce the links between corporate social responsibility and human resource management have highlighted employees’ role as crucial organizational stakeholders. This study aims to investigate whether workers’ perception of socially responsible human resource management(SR-HRM) based onemployee-focused practices is related toworkengagement (WE).This research also explored whether perceived organizational support (POS) and affective commitment (AC) can contribute to explaining this relationship. Social exchange theory and job demands-resources model wereused totheoretically frame the research. Design/methodology/approach – Data were collected from a sample of 222 employees working in diverse organizations, using individual online surveys. Several analyses were conducted to assure data robustness tocommonmethod bias. Findings – The results confirm that SR-HRM fosters WE and that this effect is subject to sequential mediation by POS and AC. Accordingly, SR-HRM practices contribute to higher level of POS, which then foster stronger affective bonds with employers and, in turn, higher levels of vigor, absorption and dedicationamong workers. Originality/value – The findings contribute to the expansion of the SR-HRM literature by providing a deeper understanding of how this management strategy affects employees’ job-related attitudes, particularly WEa much-overlookedvariableinthis realm.


Introduction
Workplaces can present employees with multiple challenges arising from globalization, international competitiveness and organizational change. These trends have wide-ranging impacts on workers, which they experience as inadequate regulation, insecure work relationships, unstable careers and increased job demands (Voegtlin and Greenwood, 2016). For decades, human resource management (HRM) has leaned toward prioritizing economic rationality over employees' interests and needs, especially after the advent of strategic HRM (de Gama et al., 2012;Jack et al., 2012). This emphasis on financial goals is a deviation from HRM's tradition of social values orientation and advocacy on behalf of workers' welfare. More recently, the responsibility for working conditions has increasingly been shifted onto organizations, so their perception of HRM as a guardian of good workplace environments has been re-emerging (Voegtlin and Greenwood, 2016). This recent trend is most likely associated with the growing importance of corporate social (Information about the authors can be found at the end of this article.) responsibility (CSR) and the push to gain legitimacy (Del-Castillo-Feito et al., 2022) and a good reputation (Chaudhary, 2017). However, the shift toward socially responsible HRM (SR-HRM) is also a reaction to the growing pressure exerted by stakeholders and society at large on organizations (Helmig et al., 2016). SR-HRM refers to HRM policies and practices implemented to guarantee employee-oriented CSR. This perspective thus implies that these strategies should encourage not only individuals' best work performance but also the fulfilment of employees' overall expectations, thereby contributing to their well-being and satisfaction (Barrena-Martínez et al., 2017;Sancho et al., 2018).
The increased popularity of SR-HRM ( Barrena-Martínez et al., 2019;Jamali et al., 2015;Sancho et al., 2018;Shen and Zhu, 2011) indicates that CSR fosters more responsible HRM that embeds social responsibility into HRM routines and focuses on employees as organizations' main stakeholders. Companies need dedicated employees who are deeply engaged in their work despite constantly changing business landscapes (Christian et al., 2011;Schaufeli, 2014). CSR practices are positively associated with engagement at work (Ferreira and Real de Oliveira, 2014; Glavas and Piderit, 2009;Gü rlek and Tuna, 2019), so they can strengthen HRM strategies that employees perceive as significant and be interpreted as demonstrating their organization's willingness to support and invest in them (Alfes et al., 2013).
The present study sought to respond to previous calls for more research on internal stakeholders concerning SR-HRM insofar as extant literature is still relatively limited (Diaz-Carrion et al., 2020;Jamali et al., 2015;Jer onimo et al., 2020;Newman et al., 2016;Omidi and Dal Zotto, 2022). To understand better how SR-HRM practices are related to workers' job attitudes, analyses need to be conducted to clarify underlying psychological mechanisms. One possible explanatory mechanism is perceived organizational support (POS) (Kurtessis et al., 2017), which is positively related to human resource practices and work engagement (WE) and reflects employees' perception of their organization's commitment to its workers (Eisenberger et al., 1986). Another possible mechanism is affective commitment (AC), which refers to employees' emotional connection to their organization (Meyer and Allen, 1991). As AC is positively associated with individuals' fulfilling work experiences and various other outcomes (Meyer et al., 2002;Riketta, 2002), it might help to explain SR-HRM and WE's relationship. Hence, the current investigation examined the evidence for SR-HRM's positive relationship with employees' WE via the sequential mediation of POS and AC. The present study's results contribute to the literature on SR-HRM and WE in different ways. First, the study is a pioneer in highlighting SR-HRM's significant role in stimulating employees' energy, absorption and dedication at work, thus expanding the literature on the antecedents of WE. Second, based on job demandsresources model (JD-R; Bakker and Demerouti, 2007) and social exchange theory (SET; Blau, 1964), the study unravels POS and AC as two psychosocial mechanisms underlying that relationship. Despite POS's important role in predicting employees' attitudes and behaviors (Eisenberger et al., 2020;Kurtessis et al., 2017), it has never been studied as a mediator of the link between SR-HRM and WE. The same applies to AC, despite previous studies revealing its mediating role regarding other individual outcomes. The current research thus sought to analyze the direct relationship between employees' perceptions of SR-HRM and WE, as well as the POS and AC's mediating role, providing added value to the existing knowledge. This study's findings also contribute to the discussion on the temporal process of SR-HRM effects on job attitudes (Omidi and Dal Zotto, 2022), taking WE as the focus, by addressing POS and AC's sequential intervention. A careful review of the relevant literature failed to uncover any research dealing with the sequential relationships between these four variables. The results provide information relevant to practitioners by clarifying the factors which foster WE, an important booster of the effectiveness of WE interventions (Knight et al., 2017) by managers and human resource (HR) professionals adopting a socially responsible approach to people management.

Work engagement
WE is seen as a crucial psychological capability in organizations (Knight et al., 2017;Lesener et al., 2020;Schaufeli, 2014). It was initially defined as the degree to which individuals are personally involved in their work, including employees' physical, cognitive and emotional engagement during work performance (Kahn, 1990). In more recent research, WE was characterized as an affective-cognitive state composed of employees' vigor at, dedication to and absorption in work, which is not limited to specific targets but is persistent and widespread (Schaufeli and Bakker, 2004). Vigor comprises high levels of energy, mental resilience and willingness to do work well, which generates persistence in the face of difficulties. Dedication is described as a sense of meaning, pride, enthusiasm and inspiration gained through meeting workplace challenges, which results in employees' strong identification with their work (Gonz alez-Rom a et al., 2006). Finally, absorption refers to a state of deep involvement in and concentration at work, which is experienced by workers as the rapid passage of time and difficulty in distancing themselves from their job (Schaufeli and Bakker, 2004).
Meta-analyses (Christian et al., 2011;Halbesleben, 2010;Rich et al., 2010) have found that WE can be advantageous for organizations because it is not only positively related to work performance and extra-role behaviors but also negatively related to turnover intention. Researchers have identified numerous antecedent factors related to personal characteristics, e.g. conscientiousness, core self-evaluations, positive affect, value congruence and optimism. Determining factors can also include work characteristics and organizational resources, such as training, autonomy at work, feedback, tasks' variety and meaning and perceived social support (Christian et al., 2011;Halbesleben, 2010;Lesener et al., 2020;Rich et al., 2010). Based on SET (Blau, 1964) and JD-R model (Bakker and Demerouti, 2007), this paper proposes that SR-HRM practices can be considered as resources that help workers to address work (e.g. training) and nonwork goals (e.g. work-family balance practices), as well as a form of discretionary treatment that signals that employees are valued by their organization. Employees might reciprocate such perceived investment and support with increased commitment and involvement at work.

Socially responsible human resource management and work engagement
HRM has changed and evolved in recent decades, expanding beyond its initial function of ensuring enough employees with the necessary skills and motivation to meet organizations' needs and achieve their established goals (Lengnick-Hall et al., 2009). HRM's focus shifted from carrying out administrative functions to becoming a fundamental factor in organizational effectiveness (Boxall and Purcell, 2000;Lengnick-Hall et al., 2009).
The growing focus on CSR has given emphasis to the entire organization's social responsibilities, encompassing activities that go beyond philanthropy (McWilliams and Siegel, 2001). Organizations depend on networks of different members of society, so they are accountable for not only achieving their shareholders' financial objectives but also managing a broad range of social factors (Boxall and Purcell, 2000;Carroll, 2016), including organizations' environmental footprint, contribution to society's welfare and provision of decent work conditions. The strongest reason for the growing interest in the CSR-HRM intersection is a widespread focus on employees as stakeholders given both CSR and HRM's crucial role in organizations' activities (Sarvaiya and Eweje, 2014). SR-HRM's initial main research goal was to identify functional areas of HRM that should be considered to achieve constructive economic and social results through CSR (Orlitzky and Swanson, 2006), including, among others, recruitment and selection, performance evaluation, compensation, training and development, and other more specific aspects (e.g. appraisal of employees' social and economic performance). Since then, other SR-HRM components have been proposed (Shen, 2011;Shen and Zhu, 2011), such as employee-oriented HRM (e.g. flexible working hours and employee participation in decision-making), HRM to facilitate general CSR (e.g. promoting and rewarding employees' social performance) and HRM's role in legal compliance (e.g. equality, health and safety at work). More recently, Sancho et al. (2018) identified five dimensions of SR-HRM in the Spanish context: training and continuous development, work-life balance, attention to diversity, communication and professional career.
SR-HRM practices seem to have the potential to enhance employees' positive behaviors and attitudes (Jamali et al., 2015;Omidi and Dal Zotto, 2022). Previous research has indicated that these practices are positively associated with individual performance and organizational citizenship behavior (Lee et al., 2022;Newman et al., 2016;Shao et al., 2019;Shen and Benson, 2016), as well as organizations' intellectual capital and its three components: human, social and organizational (Barrena-Martínez et al., 2019). SR-HRM also contributes to increased satisfaction, well-being, motivation, confidence and commitment and to decreased turnover intention (Kundu and Gahlawat, 2015;Newman et al., 2016;Zhang et al., 2022;Zhao et al., 2022). Furthermore, these practices were found to help inhibit organizational unethical behaviors, notably by reducing employees' unethical pro-organizational behavior (Luu, 2023) and increasing their moral voice (Liao et al., 2022;Zhao et al., 2023).
Despite the abovementioned findings, few researchers have explored SR-HRM's effect on employees' WE (Jer onimo et al., 2020) despite evidence that both HRM (Alfes et al., 2013) and internal CSR practices (Bayode and Duarte, 2022;Glavas and Piderit, 2009;Gü rlek and Tuna, 2019) are positively related to WE. This relationship can be explained by SET (Blau, 1964). Accordingly, employees' perception of these practices as a demonstration of employers' willingness to support their workers and respond to their needs (Alfes et al., 2013;Chaudhary, 2017;Ferreira and Real de Oliveira, 2014) can activate the norm of reciprocity, leading workers to reciprocate the perceived organizational investment with increased engagement at work (Kundu and Gahlawat, 2015;Saks, 2006;Shen and Zhu, 2011). An alternative explanation can be provided by the JD-R model (Bakker and Demerouti, 2007). SR-HRM practices can be considered as providing resources (e.g. training, information, improved work conditions) that help workers to deal with job and nonjob responsibilities. The availability of these resources would constitute a motivational trigger and subsequently lead to increased positive outcomes, such as higher levels of energy, absorption and dedication at work. The present study's first research hypothesis was thus formulated as follows:

H1. A positive relationship exists between employees' perception of SR-HRM practices and their WE.
Mediating role of perceived organizational support POS consists of employees' beliefs about the extent to which their organization values their contributions and shows concern about their well-being (Eisenberger et al., 1986). This perception fosters workers' expectation that their efforts will be equitably rewarded by their organization through future support in difficult situations (e.g. illness), payment of fair wages and interesting, relevant work (Eisenberger et al., 1986). According to organizational support theory (Eisenberger et al., 1986), POS results from workers' interpretation of their employer's reasons and intentions underlying favorable or unfavorable treatment of employees in terms of policies, norms, procedures and actions affecting the staff (Eisenberger et al., 2001;Kurtessis et al., 2017). This interpretation is influenced by factors that shape reciprocity perceptions (Blau, 1964), such as the frequency, intensity and sincerity of their organization's praise and approval, as well as other work-related rewards.
Therefore, POS triggers a social exchange process that causes employees to feel an obligation to reciprocate their organization, bearing in mind organizational goals and prosperity to compensate their employer for the favorable treatment received (Eisenberger et al., 2001;Kurtessis et al., 2017). POS also induces employees to incorporate organizational belonging and workplace status into their social identity, which affects how well their socio-emotional needs are fulfilled. In addition, POS reinforces these individuals' belief that putting more effort into achieving organizational goals will bring better rewards (Eisenberger et al., 1986;Rhoades and Eisenberger, 2002).
HR practices and working conditions are included in POS's main antecedents (Eisenberger et al., 2020;Kurtessis et al., 2017). Moreover, HR practices and working conditions can be seen as ways to enrich employees' experience (e.g. personal development, work safety, autonomy, rewards, participation in decision-making and family support practices). These HRM strategies are positively related to POS (Kurtessis et al., 2017;Rhoades and Eisenberger, 2002). Employees with strong POS show fewer symptoms of fatigue and burnout and report a better work-family balance (Eisenberger et al., 2020;Riggle et al., 2009). By perceiving a greater amount of organizational support, workers are less likely to intend to leave their organization or do so and are less often late or absent (Kurtessis et al., 2017;Rich et al., 2010).
The inference can be made that SR-HRM can also increase POS, through regular policies and practices designed to ensure the quality of employees' working and living conditions. In turn, workers with strong POS have positive expectations and a sense of security due to their organization's favorable treatment, thereby showing higher levels of WE to reciprocate this internal organizational investment (Rich et al., 2010;Saks, 2006). In addition, the provision of resources by organizations can also serve as a motivational boost and lead to increased WE. Based on these findings, a second hypothesis was proposed: H2. POS mediates the relationship between employees' perception of SR-HRM practices and their WE.

Mediating role of affective commitment
Organizational commitment refers to a psychological state that leads employees to desire, need and/or feel obliged to maintain membership in their organization (Meyer and Allen, 1991). This commitment can be conceptualized as a "force that binds an individual to a course of action of relevance to one or more targets" (Meyer and Herscovitch, 2001, p. 301), which functions as intrinsic motivation. The most prominent model of this construct has three components (Meyer and Allen, 1991): affective, continuance and normative commitment.
More specifically, AC comprises employees' positive emotional connection to their employer, which contributes to more enthusiasm for their work and a greater commitment to organizational success (Meyer and Allen, 1991). Continuance commitment derives from the perception of costs associated with leaving an organization and the absence of more favorable alternatives. Finally, normative commitment reflects employees' feeling of obligation to maintain their bonds to their organization, which arises from a sense of moral duty to their employer and is reflected in the performance of work devoid of enthusiasm and commitment (Meyer and Allen, 1991).
The current research focused on AC because of the growing interest in understanding commitment's emotional dimension and AC's stronger association with work-related attitudinal and performance variables as compared to continuance and normative commitment (Meyer et al., 2012;Riketta, 2002). Meta-analyses on AC's consequences indicate negative associations with employee turnover and/or perceptions related to departure and absenteeism (Guchait and Cho, 2010;Mercurio, 2015). Work performance is positively correlated with AC (Meyer et al., 2002;Sotiropoulou et al., 2022). The same occurs with innovative work behaviors (Muñ oz et al., 2022). Additionally, organizational citizenship behaviors have a strong relationship with AC, and both stress and work-family conflict are negatively correlated with AC (Mercurio, 2015;Meyer et al., 2002). Work experiences are further reflected on POS, authentic leadership and organizational justice, and are positively correlated with AC, in contrast to role ambiguity and role conflict's negative correlation with AC (Lee and Wei, 2017;Duarte et al., 2021).
HRM practices that are a part of employees' daily lives, such as employee reward and recognition processes, have been identified as AC enhancers (Guchait and Cho, 2010;Jawaad et al., 2019;Morrow, 2011). Prior findings indicate that SR-HRM relates to stronger AC (L opez-Fern andez et al., 2018; Shen and Zhu, 2011), by increasing trust in the company (Iqbal and Deng, 2020). AC is also positively associated with WE (Asif et al., 2019;Bizri et al., 2021;Meyer et al., 2002;Sotiropoulou et al., 2022;van Gelderen and Bik, 2016) due to the meaning employees attribute to their work and their enthusiasm toward it (Poon, 2013). When workers develop an emotional bond with their organization this influences their willingness to be engaged (Gupta et al., 2016). Following SET framework (Blau, 1964), emotionally connected employees feel the need to reciprocate the resources received from the organization with increased involvement and dedication at work. Given the available evidence, the following hypothesis was developed: H3. AC mediates the relationship between employees' perception of SR-HRM practices and their WE.
Perceived organizational support and affective commitment's sequential mediation POS is positively related to organizational commitment, with a particularly strong connection with AC (Lee and Peccei, 2007;Gupta et al., 2016;Rhoades et al., 2001;To and Billy, 2023); that is, employees reciprocate favorable treatment with greater commitment. Moreover, AC is a potential mediator of the relationship between POS and WE. Using the SET framework, Gupta et al. (2016) explored employees' access to specific resources inherently associated with POS and revealed that these conditions may lead to the development of positive affective states, which, in turn, contribute to positive attitudes toward (and behaviors at) work, including stronger WE. Empirical evidence has been found for this AC mediation, suggesting that commitment can be an explanatory mechanism for POS and WE's relationship (Gupta et al., 2016;Poon, 2013). Considering that SR-HRM can be interpreted by workers as a sign of the organization's concern and care for their members, it is proposed to motivate a chain of positive effects leading to increased WE by means of stronger POS and AC. This chain of effects is based not only on the need workers feel to reciprocate organizations' investment in their teams with outcomes inherently valued by the firms (i.e. stronger AC and WE; SET), but also on the effective use of resources provided by SR-HRM to face the challenges of their daily lives (JD-R model). Based on the rationale inherent in the aforementioned findings, the last hypothesis was written as follows: H4. POS and AC are joint sequential mediators of the relationship between employees' perception of SR-HRM practices and WE.

Sample and procedures
A cross-sectional survey was conducted in Portugal to collect data for an empirical examination of the research model. The survey, which was made available online, included an informed consent section in which the research goals were explained. Data anonymity and confidentiality were also guaranteed as recommended by the Declaration of Helsinki. Besides items collecting sociodemographic data, the questionnaire included questions designed to measure all the selected variables. The survey was disseminated using researchers' professional and social networks to reach as many participations as possible.
Data collection occurred in February 2020. To take part in the present research, participants needed to be employed at their current organization for at least six months. This assured that the participants were sufficiently familiar with their workplace's HRM practices.
A total of 222 completed questionnaires were received, which comprised a non-probability convenience sample. The respondents had a mean age of 42.95 years (standard deviation [SD] = 11.26; minimum = 21 years; maximum = 68 years), and nearly two-thirds were female (61.7%). The majority of the participants had a higher education degree (74.8%), and a mean tenure of 13.79 years in their current organization (SD = 11.08; minimum = 0.5; maximum = 40), with 79.7% having a permanent employment contract. Around one-quarter held a managerial position (27.0%). Most respondents worked in a private organization (75.6%) with a for-profit orientation (82.9%); 70.3 had a job in a large organization (>250 employees), 10.8% worked for a medium-sized organization (50-250 employees) and 18.9% for a micro to small organization (1-50 employees). Most employers were from the tertiary sector (84.6%).
The respondents indicated their level of agreement with each item using a Likert scale ranging from 1 ("Totally disagree") to 5 ("Totally agree"). As Sancho et al. (2018) did in their research, a composite score was calculated for each participant in the present study based on the five dimensions' averages (a = 0.90).
POS (mediator variable one) was measured by eight items taken from Eisenberger et al.'s (1986) scale (e.g. "My organization really cares about my well-being."; a = 0.91). The participants responded to the items using a Likert scale ranging from 1 ("Totally disagree") to 7 ("Totally agree").
AC (mediator variable two) was assessed using six items from Meyer and Allen's (1997) scale (e.g. "I really feel as if this organization's problems are my own"; a = 0.89). The respondents also reacted to the items using a Likert scale ranging from 1 ("Totally disagree") to 5 ("Totally agree").
Finally, WE (the criterion variable) was measured with the utrecht work engagement scale-9 scale , composed of nine items (e.g. "Time flies when I am working."; a = 0.94). The participants answered the items using a Likert scale ranging from 1 ("Never") to 7 ("Always").

Discriminant and convergent validity of measures
This study's cross-sectional design meant that common method variance (CMV) could potentially weaken the results' validity (Bozionelos and Simmering, 2022). Several techniques were applied to check whether the survey items captured distinct constructs as opposed to demonstrating CMV. Harman's single-factor test (Podsakoff et al., 2003) revealed that the first factor accounts for less than 50% of the total variance, that is, only 16.64% out of total variance of 69.29% (Kaiser-Meyer-Olkin = 0.94; p < 0.001).
Confirmatory factor analyses were then conducted using AMOS version 26 software. The results reveal that the four-factor model fits the data well (e.g. root mean square error of approximation [RMSEA] = 0.06; Tucker-Lewis index [TLI] = 0.90; comparative fit index [CFI] = 0.91). In contrast, the single-factor model and three other alternative models produced unacceptable fit statistics (Hu and Bentler, 1999), indicating that the variables in question capture different constructs (Table 1).
A model based on latent methods was also tested by adding an unmeasured latent method factor to the four-factor model. All items were allowed to load on their theoretical constructs and on this latent method factor, with results indicating a good fit to the data (RMSEA = 0.06; TLI = 0.91; CFI = 0.93) ( Table 1). The goodness-of-fit of the two models were compared using the CFI difference test. The CFI changed by 0.02 between the models, which is below the recommended minimum of 0.05 (Bagozzi and Yi, 1990). This result indicates that including the latent method factor in the model does not significantly improve its overall fit, which reduces the likelihood that CMV is present.
To ensure discriminant validity, average variance extracted (AVE) values were also estimated and compared to the squared correlations between all pairs of variables. As recommended by Fornell and Larcker (1981), the AVE values are greater than the shared variance between variables (Table 2). Overall, the different techniques' results indicate that the four constructs show discriminant validity and that no serious CMV is present. Convergent validity was also confirmed as both the composite reliability (CR) values and the AVE values are above the recommended cut-off points of 0.70 and 0.50, respectively (Fornell and Larcker, 1981) (Table 2).

Descriptive statistics and correlations
Descriptive statistics were computed using IBM SPSS version 26 software, which revealed high or medium average levels for all the variables (Table 2). In addition, Spearman's coefficients show that variables are all significantly intercorrelated and that they have moderate correlation values (all p < 0.05). Some socio-professional characteristics, namely, Notes: df = degrees of freedom; RMSEA = root mean square error of approximation; TLI = Tucker-Lewis index; CFI = comparative fit index; SR-HRM = socially responsible human resource management; POS = perceived organizational support; AC = affective commitment; WE = work engagement Source: Created by authors the participants' age (r s = 0.14; p < 0.05) and holding a managerial position (r s = 0.38; p < 0.01), are significantly correlated with their reported level of WE. The same is true of organization size (r s = 0.17; p < 0.05). These variables were thus included as covariates in subsequent analyses.

Hypotheses testing
PROCESS macro for SPSS version 26 software (Hayes, 2013) was used to test the hypotheses. Table 3 presents the sequential mediation analysis results (Model 6).
The first hypothesis predicted that employees' perceptions of SR-HRM practices would increase their WE. The results confirm that SR-HRM's total effect on WE is statistically significant (unstandardized coefficient [B] = 0.63; p < 0.001), indicating that the respondents' perception of their organization's practices increases their engagement with their work. H1 was thus supported.
The second hypothesis stated that POS mediates the link between SR-HRM practices and employees' WE. The results verify that SR-HRM practices significantly predict workers' perceptions of organizational support (B = 1.17; p < 0.001), but the latter has no statistically significant effect on levels of WE (B = 0.07; p > 0.05). In addition, POS's indirect effect is not statistically significant (B = 0.09; 95% confidence interval [CI] = À0.04; 0.21). Therefore, H2 was rejected.
The third hypothesis stated that AC mediates the link between SR-HRM practices and WE.
Although the results indicate that AC helps explain employees' WE (B = 0.43; p < 0.001), their level of perceived SR-HRM does not have a significant impact on their affective bond to their organizations (B = 0.14; p > 0.05). The indirect effect is thus not statistically significant (B = 0.06; 95% CI = À0.02; 0.14), so no noteworthy mediation effect exists. H3 was thus also not supported.
Finally, H4 posited that POS and AC are sequential mediators of the relationship between workers' perception of SR-HRM practices and these individuals' WE. SR-HRM's indirect effect on WE through POS and AC is statistically significant (B = 0.19; 95% CI = 0.11; 0.30).
The results thus confirm that employees' perception of organizations' SR-HRM practices is associated with these workers' stronger POS (B = 1.17; p < 0.001). This POS then fosters stronger affective bonds to their organization (B = 0.39; p < 0.001), which subsequently contributes to higher levels of WE (B = 0.43; p < 0.001). These findings support H4.
As SR-HRM's direct effect on WE (B = 0.29; p < 0.001) is smaller than the total effect but still significant, the results indicate that the mediation is partial. Table 3 shows the covariates' effects on WE, indicating that participants who held a managerial position reported higher WE than those without such responsibility. Participants' age and Notes: Spearman's correlations below the diagonal; Cronbach's alpha between parentheses in italic; squared correlations above the diagonal; Ã p < 0.05; ÃÃ p < 0.01; a managerial position: 0 = no, 1 = yes; b organization size: ranges from 1 (less than 10 employees) to 4 (more than 250 employees); SR-HRM: socially responsible human resource management Source: Created by authors Table 3 Hypothesis testing results  p < 0.001; a managerial position: 0 = no, 1 = yes; b organization size: ranges from 1 (less than 10 employees) to 4 (more than 250 employees); POS = perceived organizational support; AC = affective commitment; WE = work engagement; B = unstandardized coefficient; SE = standard error; SR-HRM = socially responsible human resource management; ES = effect size; BootLLCI = bootstrap lower limit confidence interval; BootULCI = bootstrap lower limit confidence interval Source: Created by authors organization size did not significantly relate to WE. The full model explains half of the unique variance of WE (R 2 = 0.50; F [6, 215] = 36.32; p < 0.001). Figure 1 summarizes the main results.

Discussion and conclusions
This study explored the relationship between employees' perceptions of their organization's SR-HRM practices and their WE and analyzed POS and AC's possible sequential mediation of this relationship. The existing literature reports that a more comprehensive understanding of how SR-HRM contributes to employees' positive job attitudes and behaviors is needed (Diaz-Carrion et al., 2020;Jamali et al., 2015;Jer onimo et al., 2020;Newman et al., 2016;Omidi and Dal Zotto, 2022). Using SET and JD-R frameworks, this study contributes to this endeavor by providing evidence that POS and AC act as psychosocial mechanisms that link workers' perception of SR-HRM practices to improved WE.
First, the present study's results highlight the positive association between organizations' adoption of SR-HRM practices and their employees' WE. Previous research has explored CSR (Bayode and Duarte, 2022;Glavas, 2016;Glavas and Piderit, 2009) and HRM's effect (Alfes et al., 2013) on WE, but fewer studies have examined SR-HRM's effect (Jer onimo et al., 2020). Similar to prior research, the current findings reveal that workers benefit from socially responsible management. This strategy includes practices promoting training and continuous development opportunities, work-life balance, diversity, communication and professional growth. SR-HRM fosters higher levels of energy, absorption in and dedication to work. According to Jamali et al. (2015), SR-HRM appears to encourage positive responses from workers specifically in terms of their level of WE as a way to reciprocate in return for the resources their organization makes available.
The present results also show that SR-HRM practices contribute to increased POS among workers (Eisenberger et al., 2020). Unexpectedly, the latter factor has no significant effect on employees' levels of WE. Increased levels of perceived support have been positively related to higher levels of energy, dedication and absorption at work, but our results seem not to support this relation (cf. Rich et al., 2010;Saks, 2006). The correlation between this pair of variables is positive and of moderate intensity but the regression analysis produces no significant effect between them. So, no support was found for H2. H3 concentrated on Notes: B = unstandardized coefficient;***p < 0.001 Source: Created by authors AC's mediating effect and also received no support. In this case, AC had a positive effect on employees' WE (Asif et al., 2019;Bizri et al., 2021;Poon, 2013;van Gelderen and Bik, 2016), even though SR-HRM failed to alter significantly employees' affective bond to their organization (cf. L opez-Fern andez et al., 2018;Shen and Zhu, 2011). The positive and moderate correlation coefficient between SR-HRM and AC was not sufficiently strong to produce a significant effect when considering other variables (e.g. POS) in the regression equation. As previous research suggests that POS and AC might individually mediate the principal relationship under analysis, further studies are needed to clarify the present results. Why increased POS resulting from higher perceived investment in SR-HRM did not increase involvement at work? Why the same investment in SR-HRM practices did not stimulate a higher affective bond to the company? One possible explanation for this research's failure to find support for H2 and H3 could be related to the intervention of contextual factors not considered in this research, such as workers' attributions regarding companies' motivation to invest in such practices. Zhang et al. (2022) have recently shown how substantive attributions (i.e. considering that the organization genuinely wants to take care of their members) have a positive conditional effect on how SRHRM affects employee well-being via perspective-taking (i.e. by influencing them to take the perspective and values of others when thinking about a situation), while symbolic attributions (i.e. considering it is an impression-management strategy to benefit organizations' reputation and profit) negatively moderate this indirect effect. A similar moderating effect might have occurred regarding the chain of effects here analyzed. Hence, future research might dedicate attention to potential boundary conditions for SR-HRM effects on workers' job attitudes and behaviors. Another potential explanation might relate to the analytical procedures followed as PROCESS macro software estimates direct effects for all the model variables at once. This approach may have reduced the significance of the paths between POS and WE and between SR-HRM and AC. Future studies should check the current findings' stability across different samples and with diverse analytical approaches.
Despite H2 and H3's lack of support, the results provide evidence of POS and AC's role as psychosocial sequential mediators that help explain the relationship between SR-HRM practices and WE. Consistent with SET and JD-R frameworks, SR-HRM practices contribute to a higher level of POS, which then fosters stronger affective bonds with employers and, in turn, higher levels of vigor, absorption and dedication among workers.

Theoretical and practical contributions
Regarding this study's theoretical contributions, the findings expand the literature on SR-HRM's consequences by strengthening the evidence for its positive effect on WE, a muchoverlooked variable (Omidi and Dal Zotto, 2022). The results also indicate that this influence occurs via increased POS and AC, which provides a deeper understanding of the complex psychosocial mechanisms underlying the relationship between workers' perception of SR-HRM and their WE. By exploring a sequential mediation model, the study contributes to the discussion on the temporal process of SR-HRM effects on job attitudes (Omidi and Dal Zotto, 2022) and helps to shed light on the black box of SR-HRM effects. SET and JD-R frameworks provide theoretical support for proposed links.
These findings additionally have managerial implications because they suggest that managers and HR professionals should value and implement SR-HRM practices, thereby supporting employees and increasing their perception of being valued by their organization. POS is a key variable in the literature on organizational behavior and HRM because POS is related to various desirable employee outcomes (Eisenberger et al., 2020;Kurtessis et al., 2017;Rich et al., 2010;Riggle et al., 2009), such as AC. The present results confirm this relationship and indicate that managers and HR professionals need to provide support to their workers. Workers benefit from their organization's emotional support (e.g. providing employment stability and diminishing job insecurity), resources (e.g. informational support), better communication (e.g. stimulating symmetrical internal communication) and continuous learning and development opportunities (e.g. upskilling and/or reskilling opportunities), among other forms of support. Employers can provide more positive work experiences, informing and involving employees in work processes and outputs, considering the continuous improvement of their individual well-being and personal development. This will foster employees' motivation and commitment to their organization's values and goals. Workers' increased AC will then contribute to more WE, made visible as increased energy, dedication and absorption in fulfilling work responsibilities.
SR-HRM practices go beyond traditional HRM functions as the former address more current issues such as work-family-life balance, diversity and inclusion (Barrena-Martínez et al., 2017;Sancho et al., 2018). Implementing SR-HRM practices can be challenging for organizations, but the benefits gained from employees' responses can be considerable (Jamali et al., 2015). SR-HRM can also have a positive effect on other stakeholders as fair, responsible treatment of employees is a sustainable development goal defined by supranational institutions and multiple agents. As others suggest (Dyllick and Muff, 2016), these practices may be considered part of a broader scope of sustainable HRM, a "Common Good HRM" (Aust et al., 2020), which should serve ecological and human rights purposes beyond the strictly economic and social spheres, as pointed in the United Nations' Sustainable Development Goals. Therefore, managers and HR professionals need to be more aware of SR-HRM positive outcomes and assume new organizational responsibilities in addressing global sustainability challenges.

Limitations and future directions for research
Despite the theoretical and practical contributions, the above findings should be interpreted and generalized with caution given this study's limitations. First, the use of convenience sampling to facilitate data collection restricts the results' usefulness in other settings. Future research could benefit from generating more representative samples of organizations' members to ensure greater generalizability. In a different direction, future research could provide an industry-specific analysis (e.g. labor-intensive industries, such as tourism and hospitality, versus others), examining eventual differences in the chain of effects here explored. Despite a socially responsible approach to people management being of value for all workers, employees in labor-intensive sectors might benefit even more from such an approach than employees in less labor-intensive contexts. Future research can also explore if different SR-HRM dimensions have dissimilar effects on workers' responses to their organizations' approach to people management.
Second, this study's design prevented any clear conclusions to be drawn about the causal nexus between the variables in question and increased the risk of CMV. Causality in the present conceptual model was established based on the relevant literature, but models subjected to empirical analyses tend to produce recursive statistics. The analyses carried out indicated that CMV is not a serious concern in this research (Bozionelos and Simmering, 2022;Podsakoff et al., 2003), but further studies on this topic should conduct time-lagged data collection to reduce the potential for CMV more effectively.
The proposed model explains half of the variance in the participants' WE, indicating that the variables selected are important predictors of workers' levels of energy, dedication to and absorption in their work. However, other variables also need to be incorporated into this model so that researchers can more fully understand how organizations' SR-HRM enhances their employees' WE. Plausible mediating variables other than POS and AC include, among others, passion at work because SR-HRM can contribute to increased job-related passion, which then stimulates more WE (Tr epanier et al., 2014). As perceived socio-emotional support can lead to higher WE (Soares and Mosquera, 2019), psychological contract fulfillment might also serve as an intervening variable in the relationship between SR-HRM and WE. Exploring boundary conditions to the relationships here analyzed can constitute another avenue for future research. National culture, for example, can serve as a relevant moderator because cultural characteristics seem to be associated with companies' socially responsible performance (Koprowski et al., 2021). In the same vein, SR-HRM practices can also be influenced by the national culture where the company operates. Despite these limitations, the present findings contribute significantly to the field of SR-HRM by providing evidence that these practices foster organizational members' WE.