Do Organizational Virtues Enhance Work Well-being? The Mediator Role of HRM Practices Theoretical-empirical Article

Context: over the last decades, positive psychology has brought up the need for research on positive emotions and attitudes, such as organizational virtues and well-being at work. Moreover, there is evidence that virtues may influence the perception about human resource management practices, constituting a driver of well-being in work context. However, literature signalizes the gap regarding research models covering the relation between these three variables. Objective: this study aimed to evaluate the relation between organizational virtues and well-being at work, mediated by the perception of HRM practices. Methodology: the research is classified as quantitative, carried out through a survey, with 286 participants, using structural equation modeling through the JASP software. Results: results confirmed the partial mediation played by HRM practices in the proposed model, bringing a theoretical contribution by joining the three variables into a single model, with psychometric evidence of the strong prediction of organizational virtues on HRM practices and well-being at work. Conclusion: as a managerial implication, our findings compose a diagnosis that shed light on the relevance for managers in organizations to prioritize organizational virtues and HRM practices in order to enhance healthier work environments.


INTRODUCTION INTRODUCTION
The organizational context and the changes in the labor market require the worker to be flexible and able to adapt to new demands and difficulties (Ribeiro, Mattos, Antonelli, Canêo, & Goulart, 2011). Consequently, challenges arise for researchers and managers, highlighting the need for broadening the understanding of human behavior and social interactions in the workplace context (Estivalete, Costa, & Andrade, 2014), since people have essential competences that provide strategic organizational differentiation (Horta, Demo, & Roure, 2012).
The Positive Organizational Studies theoretical movement has gained prominence in the last decade as a research stream focused in improving organizations, using their internal strengths and turning their attention to the study of variables such as virtues and well-being at work (Cunha, Rego, & Lopes, 2013). Studies on organizational behavior with positive guidance aimed to investigate and understand the potential of employees and enable the development of strengths and capabilities of each worker (Luthans, 2002).
The healthy and effective functioning of organizations and their workers cannot be achieved solely through corrective actions to prevent undesirable results (Marujo, Neto, Caetano, & Rivero, 2007). In this context, there is consensus on the importance of well-being for individuals and organizations. Among the variety of conceptual definitions, there has been a tendency in the scientific literature to approximate the terms well-being and happiness (Warr, 2007).
Researchers have developed theories about the relevance of organizational virtuosity, with reference to the organizational context where virtues are supported, practiced, disseminated, and preserved, both individually and collectively (Bright, Cameron, & Caza, 2006). From the individual perspective, virtue is a moral power or potency; a stable disposition to behave morally in a positive way (Oliveira & Tamayo, 2008).
In organizational terms, virtues strengthen organizations by providing a clear representation of what is desirable and honorable at individual and collective levels (Cameron, Bright, & Caza, 2004). Although it is still an underexplored variable, Moore and Beadle (2006) emphasize the importance of investigating virtues in the work context, since they can act as well-being boosters (Ryff & Singer, 1998). Likewise, Gomide, Vieira, and Oliveira (2016) argue that constructs from positive psychology, such as organizational virtues and well-being at work, should be studied in a more complex model, including organizational culture variables, as much as organizational practices.
As a central element of the organizational culture core (Hofstede, Neuijen, Ohayv, & Sanders, 1990), organizational practices, in which human resources management practices are included, have gained an important highlight on the agenda of organizational scholars (Yang et al., 2018). The reason for this is the patent influence on such practices on employees attitudes, behaviors, and emotions (Tamayo, Mendes, & Paz, 2000), as well as the impact on organizational results, like effectiveness and performance (Demo, Fogaça, & Costa, 2018;Guest & Conway, 2011;Nusrat, 2018).
The role of human resource management (HRM) in organizations nowadays is strategic, and its management should be in line with organizational planning and strategy (Legge, 2006). According to this author, strategic human resource management explores the way organizations manage work relationships in order to achieve goals, build healthier work environments and provide well-being to employees.
Therefore, the present study intends to evaluate the relation between organizational virtues (OV) and well-being at work (WBW), mediated by the perception of human resource management (HRM) practices. The relevance of our proposal lies in its contribution to the advance of organizational behavior research, from the positive psychology perspective, through the employees' potential in leading organizations to better results. We do that by investigating the unexplored relation between the three constructs simultaneously in a complex model, including mediation, highlighting the effects from organizational culture elements on workers' perceptions and emotions (Gomide, Vieira, & Oliveira, 2016).
Moreover, this research addresses other literature gaps. First, there is the identification of both antecedents and consequents of HRM practices in a same model, testing its role as a mediator (Armstrong & Taylor, 2017;Pires & Nunes, 2018;Singh, David, & Mikkilineni, 2018). Secondly, we address virtues in the work context as a predictor variable in organizational studies (Moore & Beadle, 2006).

Well-being at work
Considering people spend a good part of their lives in the workplace, or are working remotely, it is essential they feel good about it (Siqueira & Padovam, 2008). Deci and Ryan (2008) indicate that employees who feel good at In this context, scholars call for discussions and studies regarding well-being at work, since it is a broad area needing conceptual comprehension (Paschoal et al., 2013;Paschoal & Tamayo, 2008;Siqueira & Padovam, 2008). Luthans (2002) corroborates this perspective highlighting that most of the studies in academic literature focus on the understanding of negative psychological experiences of the worker and dysfunctional behaviors or results. Nevertheless, since the 1990s, a movement with the objective of fulfilling this gap of studies investigating positive phenomena related to individuals and groups started to consolidate and gain strength in the academic field, enabling a revaluation of human potentials and virtues (Paludo & Koller, 2007).
Under the organizational perspective, there seems to be a trend of incorporating intervention programs from positive psychology in stress management in the workplace (Tetrick & Winslow, 2015). To the authors, such programs can be successful for employees ' health promotion. Similarly, to Hirschle, Gondim, Alberton, and Ferreira (2019), stress level control in the work environment can help the workers preserve well-being at work by a better use of their regulatory processes. This was confirmed when Scheibe and Zacher (2013) highlighted that the ability of dealing with negative emotions in the workplace is an essential condition to reduce potentially adverse outcomes, like tensions, anxiety, and emotional exhaustion.
General well-being is divided into two main streams. First, subjective well-being addresses the subjective state of happiness (hedonic well-being) and represents a field of study that seeks to understand people's evaluation of their own lives (Albuquerque & Troccoli, 2004). Second, psychological wellbeing investigates human potential (eudemonic well-being) and is centered on personal expressiveness and self-realization experiences, known as central and defining elements of happiness (Deci & Ryan, 2008;Waterman, 1993).
In short, well-being configured in hedonic vision is an affective state, so positive affects predominate over negative ones. On the other side, well-being as personal fulfillment is characterized by the perception of development of potentials and progress in life purposes, representing the cognitive aspect of well-being (Waterman, 1993).
In this context, Paschoal (2008) defines well-being at work (WBW) as "the prevalence of positive emotions at work and the individual's perception that, in his work, he expresses and develops his potential and advances in the attainment of his life goals" (Paschoal, 2008, p. 23). This concept was validated in Brazil and brings together both cognitive and affective aspects of well-being. Since it embraces the hedonic and eudemonic perspectives, it is the concept adopted in the present study. Along with this definition, the scale validated by Demo and Paschoal (2016)in the United Statespreviously developed and validated in Brazil by Paschoal and Tamayo (2008), the Well-being at Work Scale (WBWS) -is the most recent scale in scientific literature and, for this, it is the instrument chosen for this research.
Regarding literature reviews on WBW, Paschoal et al. (2013) indicated the lack of studies on this topic, especially in Brazil, where the need for academic production about WBW in business and psychology journals is patent. Furthermore, scholars point out the relevance of the theme for contemporary organizations and the trend to encourage workers' positive experiences in the WBW study field (Warr, 2007). Nonetheless, theoretical and empirical advances are not resulting in effective changes in organizations (Bertoncello & Borges-Andrade, 2015). Based on this, Hirschle et al. (2019) reaffirm the need for acting toward the prevention and the offer of better physical and psychological conditions in the organizational environment.
Likewise, empirical developments indicated, over the years, the influence of several organizational factors on WBW, such as social climate, justice perception, and organizational support (Paschoal, 2008); organizational power (Dessen & Paz, 2010); turnover, absenteeism, and performance (Warr, 2007); and adaptive and functional strategies (Hirschle, Gondim, Alberton, & Ferreira, 2019). Besides these, other investigations must be underscored: Dose, Desrumaux, Sovet, and De Bosscher (2018) examined the relation between career success and WBW, mediated by psychological needs satisfaction, indicating they predict WBW. Mellor, Dufoix, Saunder, Albert, and Collange (2018) investigated factors that reinforce well-being at work, evaluating the mediator role of social support perception. As a result, they observed that positive situations at work, where social support is perceived, are associated with well-being increase. Braga, Andrade, Estivalete, Oliveira, and Costa (2017) studied the influence of organizational values on WBW perception, revealing a positive relation between variables, corroborating other papers in the same direction, like the works of Silva, Porto, and Paschoal (2010) and Souza (2012). Therefore, we understand that studies about WBW have composed the positive psychology agenda since its inauguration. Besides, we foresee a tendency for further investigation when the agenda is the promotion of healthier organizational environments. Regarding empirical studies on well-being, literature shows overall that organizational and labor variables are the main predictors of well-being at work (Paschoal et al., 2013), indicating the importance of studies including positive psychology constructs, such as organizational virtues (Magnier-Watanabe, Uchida, Orsini, & Benton, 2017).

Organizational virtues
In a context of deteriorating trust in organizations, resulting from corruption and lack of business ethics, among other factors (Bright et al., 2006;Cameron et al., 2004), it is crucial to rethink organizational performance (Peterson & Park, 2006). In this sense, the Positive Organizational Studies movement, which assumes health as opposed to disease (Scorsolini-Comin, Fontaine, Koller, & Santos, 2013), has proposed investigations at the organizational level with the objective of understanding the highest human potential, such as the virtues, promoting progress for people, communities, and institutions (Paludo & Koller, 2007).
Virtues are described as dignifying behaviors and results, the best of the human condition, the essence of humanity, and the highest human aspiration (Bright et al., 2006;Cameron, 2003). Virtues promote health, happiness, transcendence, meaning, and resilience (Ryff & Singer, 1998). However, even with the robustness of such conception, scientific literature still lacks a consensus regarding an exact number of virtues (Hamrahi, Najafbagy, Musakhani, Daneshfard, & Delavar, 2015).
Several studies have shown the role of virtues, at the individual level, in different aspects, for example, in the creation of love and empathy (Ribeiro & Rego, 2010); moral strength, determination, and tolerance toward difficulties (Baumeister & Exline, 1999); health and resilience to adversity (Ryff & Singer, 1998); and happiness and engagement at work (Singh et al., 2018). From this, we observe a scientific gap concerning studies investigating the importance of virtues in the organizational arena (Ahmed, Rehman, Ali, Ali, & Anwar, 2018;Liu, Wang, & Chen, 2019;McLeod, Moore, Payne, Sexton, & Evert, 2018).
In organizational terms, virtues strengthen organizations by providing a clear representation of what is desirable and honorable, at individual and collective levels . According to Zhang and Liu (2019), organizational virtue is an aggregate construct composed by a set of virtues. To these authors, the representation of a virtuous organization is not connected only to the virtues of its members, being linked to a unified organism with its own systems, structures, processes, and culture.
In addition, virtues in the work context are understood as qualities representing employees' belief in ethical and moral conducts (Ferrell, Fraedrich, & Ferrell, 2001). Virtues are also habits, desires, and actions that carry personal and social good (Cameron, Dutton, & Quinn, 2003) and positively influence the organization's ability to deal with adverse situations that could be harmful (Bright et al., 2006). According to Caza, Barker, and Cameron (2004), when organizations face uncertainty, virtues can play the role of guiding the organization to better performance through two attributes. The first one is the amplification of qualities that can favor and reinforce behaviors, attitudes, and positive contexts. The second is buffering qualities that protect the organization from negative consequences and create resilience, enabling the organization to recover faster from trauma -for example, in changing environments and market ambiguity. With reference to the collective level, Bright, Cameron and Caza (2006) present virtues by two ways: virtues in organizations and virtues through organizations. The first concerns individual behavior in the organizations and the second is related to the facilitators promoting and stimulating virtues in their members. Park and Peterson (2003) also divide organizational virtues in two types: (a) tonic virtues, a general condition, existing in any given time; and (b) phasic virtues, dependent on an external event, occurring only when a situation raises its need. In the same direction, Seligman and Csikszentmihalyi (2000) concluded that virtues play an important role in fighting psychological stress, addictions, and dysfunctional behaviors.
For Bright et al. (2006), in the organizational scenario, groups demonstrate virtues including actions that would be more difficult to be accomplished individually. Additionally, the effect of collective virtues may support a condition in which the drive to pursue excellence becomes part of the organization's culture . The authors add that, at the aggregate level, virtues have been associated with organizations, communities, and cultures, arguing that economic prosperity and the society longevity depend on them, since their existence implies the internalization of moral norms necessary to stability and social harmony.
With the purposed of increasing the number of studies on OVs, Gomide et al. (2016) developed, with evidence of validity, a measurement instrument for organizational virtues perception, with three factors: organizational good faith, organizational trust, and organizational generosity. This is the measure adopted in the present study, since it is the only scale validated in the Brazilian context, besides its good psychometric indices. Consequently, the concept for OV of this paper is "employee's set of beliefs that the employing organization has the ability to govern its relationships truthfully, to fulfill future commitments and to sacrifice its interests to reach the interests of its employees" (Gomide et al., 2016, p. 305).
Along these lines, Meyer, Stanley, Herscovitch, and Topolnytsky (2002) affirm that the individual's moral bond with the organization can be affected by the OVs, making it a pertinent and necessary subject for future studies on organizational behavior. Shahid and Muchiri (2019) proposed in a theoretical model the value of positive organizational behavior in the workplace, suggesting new roots to the development of studies relating OVs with leaders' authentic behaviors, psychological capital, prosperity, and performance. Notwithstanding the advances, OV researches are still incipient (Dawson, 2018;Hamrahi et al., 2015;Kooshki & Zeinabadi, 2015;Meyer, 2018), confirming it is a fertile field for forthcoming studies. In addition, Shahid and Muchiri (2019) highlight the need for more investigations regarding the antecedents and consequents of OVs.
In addition, recent studies related OVs to HRM and organizational structure (Hamrahi et al., 2015), organizational citizenship (Kooshki & Zeinabadi, 2015;Malik & Naeem, 2016), corporate social responsibility and organizational identity , ethical leadership and employee engagement for environmental protection (Zhang & Liu, 2019), performance and subjective wellbeing (Magnier-Watanabe et al., 2017), a specific interest for the present study. Hypothesis 1 derives from this discussion.
Hypothesis 1: Organizational virtues are positively related to well-being at work. Gomide et al. (2016) argue that constructs recently studied by positive psychology, such as organizational virtues and well-being at work, should be studied in a more complex model, including important variables of the organizational culture, as much as organizational practices.

Human resource management practices
Changes in the economic environment, the pursuit of people for better life quality at work, and the growing need for achieving results at individual and organizational levels have led companies to adapt to new contexts, seeking continuous innovation and advocating for healthier environments through investments in human capital (Demo et al., 2018). In this sense, HRM has evolved having people as an essential competency in achieving objectives and goals (Legge, 2006). In fact, before producing results, organizations should privilege the promotion of superior satisfactions in all their relationships (Demo et al., 2018).
According to Armstrong and Taylor (2017), HRM in the contemporary world should have as main characteristics: strategy with emphasis on integration; focus on entrepreneurship and business values; and the treatment of people as valuable resources for the organization (human capital), in order for them to be a competitive differential benefit. These arguments are in line with the resource-based view proposed by Barney (1991), the guiding theory of the present study. In this direction, HRM should follow coherent and consistent theories, aligned with organizational planning (Demo et al., 2018). Besides, HRM must play an active role in strategic choices, especially in the preparation of human talents for change (Bagnoli & Vedovato, 2014).
As stated by Legge (2006), there are differences between the classic personnel management and the current HRM, known as conservative (hard) and modern (soft), respectively. In the first one, the focus lies on integrating human resources practices with business strategy, emphasizing quantitative aspects, context in which people are considered an economic resource (Armstrong, 2014). The current perspective maintains the human resource (HR) activities and business strategies integration, proposed by the hard perspective (Legge, 2006). However, it considers employees as a valuable organizational inventory, as they propel production and knowledge management in organizations, translating it into better results: innovation and competitive advantage for private organizations; and productivity, speed, and transparency for public institutions (Legge, 2006).
Following this perspective, strategic human resource management (SHRM) is the rethinking of policies, practices, and activities of human resource management, taking into account contextual variables and multiple actors, assuming the role of HRM is internal cohesion (Buren, Greenwood, & Sheehan, 2011;Jackson, Schuler, & Jiang, 2014). These authors recognize that SHRM proposes an integration with organizational objectives in order to facilitate organizational plans establishment, to leverage results and their capacity for resilience.
In addition, Huselid (1995) and Pfeffer (1998) agree that the practices implemented by SHRM allow an increase in the knowledge, skills, and assertive attitudes of employees (KSAs). Hence, HRM should be seen as the ability to manage human resources through practices to gain and maintain competitive advantage (Mathis & Jackson, 2003). Oliveira and Oliveira (2011) define HRM as an integrated set of strategies, policies, and practices designed and planned to manage people in organizations.
To clarify the differences between the concepts of 'strategies, ' 'policies,' and 'practices,' Martín-Alcázar, Romero Fernández, and Sánchez-Gardey (2005) indicate that HRM strategies set the guidelines for work force management, while policies seek to coordinate practices so they are coherent and move in the same direction. Practices are at the end of the operation, assuming the lowest level among the three concepts, as they represent individual actions. Moreover, strategies are developed through policies, which, in turn, are implemented through practices, constituting the operational definition of policies (Demo, 2016). In such a way, practices assume the meaning of habit, routine, or policy component activities (Legge, 2006). Furthermore, HRM practices encompass the activities that collaborators actually deploy and experienced, enabling its objective verification and measurement by research instruments (Oliveira & Oliveira, 2011). The referred authors emphasize there must be an integration between HRM practices and organizational internal and external contingencies. Bringing together this perspective and the resource-based view, as well as the soft version of human resource management, HRM organizational practices are important when they are aligned with the goals of the organization, so they promote the conditions for employees to effectively contribute to results optimization, regardless of how they are measured (Legge, 2006). Thus, organizations should strive to focus on employee engagement through the implementation of successful HRM practices (Aktar & Pangil, 2018).
Nonetheless, there is a lack of research proposing validated scales that evaluate employees' perception of HRM practices in the workplace (Huselid, 1995). These scales, according to the referred author, need to be comprehensive, including the highest number of practices cited in the literature; and need to present good internal validity indices, as well as external validity and high reliability.
Therefore, the HRM practices model adopted in the present research was based on the proposal developed by Demo (2016), and in the study of Demo, Neiva, Nunes, and Rozzett (2014), who developed and validated the Human Resource Policy and Practice Scale (HRPPS), with crosscultural validation in Brazil and the United States, obtaining good internal validity and reliability indices. The HRPPS covers the six groups of practices most cited in the literature, namely: recruitment and selection; involvement; training, development, and education (TD&E); work conditions; performance and competency assessment; and rewards and remuneration.
With respect to the relationship between HRM practices and other organizational behavior variables, in order to identify antecedents and consequents, studies indicated positive relationships between HRM practices and constructs such as commitment, productivity, profitability, and quality (Guest & Conway, 2011); organizational learning (Li, Rees, & Branine, 2019); and organizational performance (Fuenzalida & Riccucci, 2019;Nusrat, 2018). Regarding recent empirical production, research has advanced in terms of testing HRM practices mediation models (Sánchez, Marín, & Morales, 2015), while other studies have filled the gap concerning HRM practices background investigation (Demo, Fernandes, & Fogaça, 2017). There are studies that have confirmed established relationships, such as HRM practices and organizational commitment (Nassar, 2018) and organizational performance (Nusrat, 2018;Otoo, 2019).
Thus, it is fundamental to acknowledge the variables that affect and are affected by HRM practices for the advancement of human resource management academic literature, as well as recognizing their strategic role in organizations (Demo et al., 2018), demonstrated by the patent growth of studies in the area (Yang et al., 2018).
Indeed, organizations can improve results at individual and organizational levels by offering specific HRM practices (Veth, Korzilius, Heijden, Emans, & Lange, 2019). In this sense, special attention has been paid by researchers to the possible influences of HRM practices on well-being at work (Demo & Paschoal, 2016;Horta et al., 2012), laying the groundwork for Hypothesis 2.
Hypothesis 2: Human resource management practices are positively related to well-being at work.
Furthermore, an important agenda is the need to focus on the antecedents of HRM practices, as the construct has traditionally been investigated as an independent variable (Demo et al., 2018). Hence, the authors point out that testing positive psychology variables as predictors of HRM practices would strengthen the line of studies of these antecedents, especially concerning organizational culture elements, such as values and virtues. Likewise, some works have signalized the possibility of positive associations between virtues and HRM practices (Hamrahi et al., 2015;Malik & Naeem, 2016;Pires & Nunes, 2018), inspiring the proposition of the third research hypothesis.
Hypothesis 3: Organizational virtues are positively related to HRM practices.
Finally, in addition to the proposed hypothesis regarding the relationship between each pair of variables, the main contribution of this study lies in presenting a mediation model to test relationships among the organizational behavior variables that might influence HRM, as requested by Armstrong and Taylor (2017), including variables from the positive psychology perspective (Singh et al., 2018). Besides, Moore and Beadle (2006) emphasize the importance of exploring virtues in the work context and Paschoal et al. (2013) suggest investigations on the prediction role of HRM practices on well-being at work in more sophisticated models, including mediation.
Additionally, studies have found that, when perceiving organizational variables and HRM practices, employees tend to present an increase in well-being and health, promoting a general improvement in the overall performance of organizations (Fuenzalida & Riccucci, 2019;Ryff & Singer, 1998). Therefore, we propose the fourth and last hypothesis to be tested.

Hypothesis 4:
Human resource management practices can mediate the relationship between organizational virtues and well-being at work.

Instruments and sample
Participants responded to an instrument/ questionnaire composed by three scientifically validated scales with psychometric indices that confirm these instruments' reliability: (a) Perception Scale of Organizational Virtues (Gomide et al., 2016); (b) Human Resources Management Policy and Practice Scale (Demo, Neiva, Nunes, & Rozzett, 2014); and (c) the Well-being at Work Scale -WBWS (Demo & Paschoal, 2016). We also included demographic variables to characterize the sample.
The Organizational Virtues Perception Scale (Gomide et al., 2016) was the measure chosen to evaluate the independent variable of this study, organizational virtues. The scale is composed by 30 items distributed in three factors (organizational good faith, organizational trust, and generosity), with composite reliability indices greater than 0.95 (Coura, Demo, & Scussel, 2020).
For the mediating variable, the Human Resources Policy and Practice Scale was chosen to measure participants' perception of HRM practices regarding the organization in which they work. The scale consists of 32 items divided into six factors: (a) recruitment and selection; (b) involvement; (c) training, development, and education; (d) working conditions; (e) competency-based performance appraisal; and (f ) compensation and rewards. The scale has evidence of validity obtained by samples in the US and Brazil and composite reliability indices greater than 0.70 (Costa, Demo, & Paschoal, 2019;Coura et al., 2020;Demo et al., 2018).
Lastly, Demo and Paschoal (2016) presented crosscultural validity for the Well-being at Work Scale, which has 29 items in its final version, with good psychometric indices and composite reliability values above 0.91. Reliability indices were measured by Joreskog's rho (Demo & Paschoal, 2016). Exploratory factor analyses were performed to assess whether the initial structure of the three instruments was reproduced in this sample. Data shows the initial structures with signs of validity follow the same pattern previously found.
The sampling process was characterized as nonprobabilistic and by convenience. The instrument was applied and answered in person, on a voluntary basis, guaranteeing participants' anonymity. A ballot box was available in the researched companies for people to drop answered questionnaires. Four hundred questionnaires were distributed and 286 were answered, a return of 71.5%. The questionnaire also covered questions related to demographic information for sample characterization purposes, such as age, sex, schooling, and time working in the company.
The sample was composed most by adults from 29 to 39 years old (39%), reaching an average of 36 years old. Twenty eight percent of the respondents fell on the range from 18 to 28 years old, 23% from 40 to 50 years old, and, finally, 10% from 51 to 66 years old. Regarding gender, the sample had exactly 50% of male and 50% of female participants, after data treatment. Concerning schooling level, a question was included in the questionnaire for the participants to indicate their completed schooling, ranging from elementary school to doctoral level. Results indicated the majority of participants has higher education (73.4%), followed by those with high school education (22.3%), master's degree (3.8%), and only one individual with PhD, representing 0.5% of the sample. Regarding time working for the company, 38% of participants stated they have been working in their respective organizations for more than 10 years, 33% between 1 and 5 years, 19% between 5 and 10 years, and, lastly, only 10% less than 1 year.

Procedures of collection and data analysis
Regarding purpose, nature, means, and time horizon, the present study is characterized as a descriptive and explicative, quantitative, field survey and transversal research, respectively. To analyze the relationship between variables, the research model adopted organizational virtues as independent variable, HRM practices as mediating variable, and well-being at work as the dependent variable. Based on previous research (Demo et al., 2014;Demo & Paschoal, 2016;Gomide et al., 2016), we calculated factor scores (for each scale) and these measures were used to perform structural equation modeling. The second order concepts were used as latent variables in the model tested (organizational virtues, HRM practices and well-being at work). Data were analyzed in the software JASP version 0.14. Structural equation modeling used ML and DWLS (robust options) estimator and lavaan syntax. Mediation analyses were based on maximum likelihood -ML estimator and DWLS estimator (robust options). Considering each scale separately, exploratory and confirmatory factor analysis were performed using Factor and the results showed similar structure used before. The model with all scales together has not presented a good adjustment, so we tried using factor scores and second-order constructs only. In this study, we are showing only reliability measures for this sample.
To perform data processing, the first step was frequency distribution analysis, which was done using the listwise procedure to identify missing values. However, since the application was made and guided by the researchers themselves, there were no blank items. In the next step, we investigated the presence of outliers (Tabachnick & Fidell, 2013), which did not identify any outlier. These analyses considered only factors calculated. We tested the assumptions for the use of multivariate analysis, including normality tests considering JASP procedures. Univariate normality tests accused significance to all variables (factors). Mardia's test (multivariate normality) using R also accused non-normality (significance to skewness and kurtosis). Despite the indices presented, the structural model test was performed and presented a suitable adjustment. To solve problems with non-normality, we performed robust options in JASP software. Thus, the final sample consisted of 286 individuals.
Structural equation modeling was performed in JASP considering the following indicators: χ2 / gl -chisquare in relation to degree of freedom or NC (Watkins, 1989); CFI -comparative fit index (Bentler, 1990), standardized root mean squared residual (SRMR) and RMSEA -root mean square error of approximation (Browne & Cudeck, 1993). The model will be considered suitable for NC ≤ 3; CFI ≥ .95; RMSEA ≤ .08 (Browne & Cudeck, 1993;Kline, 2011). Factor scores and these measures were used to perform structural equation modeling. Second order latent variables were estimated in structural equation model.
This study used only one data source and is therefore subject to the common-method variance problems. Hence, confirmatory factor analysis was applied using one-factor structural equations, as recommended by Podsakoff, MacKenzie, Lee, and Podsakoff (2003). According to the authors, if the one-factor model presents adjustment, there is common-method variance. Nevertheless, results indicated the adjustment of the one-factor model was not acceptable (NFI = 0.52; CFI = 0.49; NNFI = 0.49). Thus, we concluded that the common-method variance alone does not explain the results.

RESULTS RESULTS
Firstly, descriptive analyses (mean, standard deviation, and mode) and reliability analysis by factor of each research variable are summarized in Table 1.  2,4,6,8,9,10,11,12,13,16,17,18,20,23,28,29)  Negative affects (items 2, 5, 6, 7, 9, 10, 12, 14, 15, 16, 18, 20) 2.08 0.90 1.00 0.94 0.94 Secondly, the correlations among variables are showed in Table 2. The factors of well-being, organizational virtues, and human resource management policies and practices are correlated among them. There are correlations also between gender, education level, and factors of the study variables. These correlations do not influence the model results because tests were performed including these variables in the model, but the fit measures were poorly adjusted.  In order to test the model, Mardia's test was performed and we found normality problems. So, we used robust option for ML and DWLS models. All the results are quite the same. Here we are describing ML resultsrobust option. After that, we tested a structural model using second order factors (we calculated factorial scores of all measures). The tested model (Figure 1) was evaluated through structural equations and overall fit (χ 2 = 166.257; df = 51; p < 0.001) was adequate. The proportion between qui-square and degrees of freedom was 3.26, below 5.0, as recommended by Kline (2011). The other adjustment indices that considered the sample size were also adequate (CFI = 0.96; NFI = 0.94; TLI = 0.94), although the residue was considered high (RMR = 0.03; SRMR = 0.04; RMSEA = 0.08; [IC 90% = 0.07, 0.10]). Thus, model fit results were considered quite acceptable. In this study, we tested the direct and indirect effects also through JASP software. Figure 1 illustrated the design of the model tested. The coefficients and variables are shown with indirect effects confirmed using JASP calculations and by comparing the restricted and non-restricted models. The hypothesis test results are shown in Table 3.
The standardized regression coefficient beta (β) indicates the intensity and direction of the correlation between independent variables (IVs) and the dependent variable (DV) (Hair, Anderson, Tatham, Black, & Babin, 2009). Some coefficients were significant, indicating positive and strong associations between variables. In addition, according to the model results, we observed organizational virtues are predictors of human resource management practices (β = 0.70; p < 0.001). Organizational virtues were also presented as drivers of well-being at work (β = 0.34, p = 0.003). This result did not support the second hypothesis of the study. Likewise, human resource management practices proved to be predictors of well-being at work (β = 0.77, p < 0.001). These results corroborate three hypotheses of the study. Additionally, according to the coefficients of determination (R 2 ), the organizational virtues explain 60% of HRM practices (high effect) and the organizational virtues added to HRM practices explain 20% of well-being at work (medium effect) (Cohen, 1992). Table 3 shows the model coefficients of the hypothesis tests. According to Hayes (2018), none of these paths is an essential condition for mediational effects.
Finally, mediation analyses supported Hypothesis 4: human resource management practices act as partial mediator in the relationship between organizational virtues and wellbeing (indirect effects b = 0.74; 0.001). In other words, human resource management practices indeed mediate the relationship between organizational virtues and well-being at work, as illustrated in Table 4.

DISCUSSION DISCUSSION
The objective of this paper was to test a mediational model with organizational virtues perception as an antecedent of well-being at work, a relation mediated by the perception of human resources practices in work organizations. The results have supported almost all hypotheses proposed by the research model. Thus, the predictive power of organizational virtues perception on well-being at work is optimized when this relation is present in organizations where people perceive the presence of polices and, particularly, effective human resources practices.
As theoretical contributions, this study advances by filling a gap regarding the application of moral and ethical aspects in organizational actions. Although these relations have been studied in previous works (Haidt, 2000;Pires & Nunes, 2018), this paper contributes by proposing human resource practices as mediators in an unheard-of model that employs moral virtues as predictors of well-being at work. Morally and ethically well-evaluated organizational policies and practices seem to be good drivers of positive feelings related to work and the employer system itself. Moreover, this study is positioned in the interface of studies about organizational behavior and human resources management, suggesting that in environments perceived as virtuous, organizational management practices tend to be more effective in promoting well-being.
In addition, from a systemic approach in organizational investigations, the results from this study suggest that organizational practices, effectively perceived as more humanistic and with ethical and moral contents, tend to lead workers to perceive the organization as a promoter of virtues that enables positive emotions, guiding them to act proactively in relation to the employer, in conformity with his objectives. Cameron and Winn (2012) and Pires and Nunes (2018) reached similar results when affirmed well-developed policies tend to become part of employees' daily life, facilitating the comprehension of demanded behaviors and their way of thinking, acting, or feeling. It is noteworthy that although Pires and Nunes (2018) have tested HRM practices as predictors of organizational virtues, they claim HRM practices must reflect the moral nature and virtuous aspects of the work context -that is, virtues should precede such practices, in agreement with the logic of our research model. In addition to the empirical evidence we have found, we reaffirm our positioning since the elements constituting the core of organizational culture, such as values and virtues (Hofstede et al., 1990), influence the perception of organizational practices indeed, including the HRM ones, as also shown by Demo (2010).
Likewise, concerning managerial implications, we understand that managers must lead by example, developing an evidence-based management. Hence, our results work as a diagnosis to inspire managers to promote a more strategic human resource management through more effective leadership. In this context, the commitment of leaders with consistent behaviors comprising high levels of honesty, respect, transparency, and reliability would conduct the managerial board and collaborators to internalize and diffuse these behavioral models in daily organizational operations, including conflict resolutions and the search for better resilience.
Furthermore, the presence of a strong leadership tends to influence the organizational culture, specifically HRM practices. According to Pires and Nunes (2018), management strategies, policies, and practices whose contents are guided by altruistic reasons help extract a richer meaning from the functions performed by the employee. In addition, the results generated might inspire managers for an increasingly humanized human resource management, since healthier work environments tend to generate better inter-organizational relationships as a practical implication and consequently better provision of service to customers and society in general, advancing to a social implication.
However, there are limitations in this study. First, we cite its cross section, which inhibits a comprehension of the phenomenon in its temporal complexity. Besides, the quantitative nature of this investigation did not enable the capture of aspects of the model that would be essential to understand the phenomenon in question, such as subjects' conception about the organizational virtues composing the model. Therefore, we recommend future research focus on longitudinal data and multimethod analysis, aiming at a systemic and continuous comprehension of the phenomenon, beyond its measurement. Yet, it would be interesting to compare results obtained in the different sectors researched in this paper, namely, public, private, and third sector.
As an agenda, we propose further research adapt the model from this study, seeking to investigate both the predictive role of organizational virtues and the mediating role of human resources management practices with other criteria-variables like positive leadership, organizational citizenship, and work resilience, suggested by Gomide et al. (2016). Such action could put light into the buffering or amplifying role of virtues, as previously suggested by Caza et al. (2004).

CONCLUSION CONCLUSION
Finally, despite the limitations, the present study has given new perspectives to different possibilities of relating virtues in organizational context with other variables of organizational behavior, as well as the important mediating role played by HRM practices. This study represented a seminal step in the investigation of the joint relationship between these variables. We aim to inspire new studies for the investigation of the phenomenon in the context of positive psychology, fostering the promotion of healthier organizational environments based on virtues, good organizational practices, and proficient work relationships, which should be translated into better results both at individual and organizational levels.