1. Introduction
The organizational environment and changes in the labor market pose challenges for researchers and managers, highlighting the need to broaden the understanding of human behavior and social interactions in the work context [
1]. This perspective arises from the idea that people represent essential competencies for organizational differentiation strategies [
2]. In this sense, the theoretical movement of positive organizational studies has gained prominence in the last decade, understood as an investigation track dedicated to improving organizations based on their internal strengths [
3].
Leadership significantly affects this context. In this regard, literature signalizes that the leader should serve as a model and inspiration for employees to join efforts towards achieving organizational goals [
4]. However, the theoretical and practical framework that relates the manager as the person who uses the set of “practices to mediate their relationship with the subordinate” is still underexplored, revealing a major gap in human resources management research [
5] (p. 54). Thus, the implementation of HRM practices mainly depends on the role of the leader [
6,
7].
In this scenario, organizational virtues (OV) positively affect an organization’s performance measures, gaining increasing attention from researchers in the context of positive psychology [
8]. To develop organizational practices with excellence, organizational virtuosity research began to stand out in the last two decades, emphasizing positive behaviors at work, considering that studying virtue as a source of positive resources for moral thinking can help to understand strategic management [
8,
9,
10]. In this sense, Areskoug Josefsson et al. [
11] suggest that investing in issues of organizational virtuosity, employee commitment, and joy at work can be useful in promoting a healthy workplace. In organizational terms, virtues are considered habits, desires, and actions that bring personal and social good [
12], and positively influence the organization’s ability to deal with adverse situations [
13]. In the organizational studies field, there is a tendency to study virtues based on two variables, justice and practical wisdom [
10,
14,
15,
16]. However, we observe that research still lacks theoretical and empirical work that directly observes organizational virtues [
17], confirming this is a fertile field for further.
In such a way, strategic human resources management arises to promote the rethinking of HRM strategies, policies, practices, and activities, taking into account contextual variables and the multiple actors involved [
18]. According to these scholars, this perspective enables an internal cohesion to the HRM function, along with a strategic integration with the organization’s objectives. Buren et al. [
18] indicate that such an approach facilitates the establishment of organizational plans, upgrades results, and increases resilience capacity. From the perspective of strategic HRM, practices can be strengthened to help achieve organizational goals, as tools are provided to communicate organizational goals in ways that can be more easily understood and executed by employees [
19].
Considering the relevance of HRM practices for organizations, it seems possible to state that a set of HRM practices, synchronized with the organizational strategy, promotes the capacity to reach a higher level of development. In the private sphere, the emphasis is on profitability and competitiveness. In the public context, in which society and citizens are the customers, the focus lies on optimizing resources, transparency, efficiency, and speed in service.
On that basis, researchers have addressed the importance of behaviors not prescribed by the formal system, such as organizational citizenship behaviors (OCB) [
20]. For instance, Organ [
21] states that organizational citizenship behaviors represent informal ways of cooperation and contributions provided by individuals based on job satisfaction and the perception of justice. In the study by Mostafa and Gould-Williams [
22], for example, the positive effect of the person–organization fit on the relationship between high-performance human resource practices, job satisfaction, and OCB was identified, culminating in the understanding that the adoption of high-performance human resource practices leads to improved attitudes and behaviors. From this angle, the literature reveals that organizational citizenship behaviors promote organizational effectiveness and long-term performance [
23].
Based on the above contextualization, additionally to the gap identified in the literature [
24], this paper intends to answer the following question: what is the relationship between leadership, organizational virtues, human resources management practices, and organizational citizenship behaviors?
Therefore, the main objective of this study is to identify the relationship between leadership, organizational virtues, human resources management practices, and organizational citizenship behaviors in a sample of employees from Brazilian companies. A total of two structural models were tested. In the first one, human resources management practices mediate the relationship between leadership and organizational citizenship behaviors. In the second one, human resources management practices mediate the relationship between organizational virtues and organizational citizenship behaviors. By testing these models, we also meet the agenda proposed by Demo et al. [
24] regarding the investigation of HRM practices as a mediating variable of organizational attitudes and behaviors.
Furthermore, we tested the measurement models of each of the research variables in the context of the general model, with the independent variables (leadership and virtues), the mediator (HRM practices), and the dependent (citizenship). Note that we explored the variables of this research at the individual level, considering the perception of employees. We understand perception as a meaningful and coherent image, being a possible, apprehensible, and measurable reality to study phenomena [
25].
Traditionally, HRM practices are studied as an antecedent or consequent variable in relational studies. Thus, by testing structural models in which HRM practices are mediating variables in unexplored relationships in the literature, such as between leadership, virtues, and citizenship, this study presents its main theoretical contribution. As practical implications, the findings compose a diagnosis for managers to better understand how such relationships happen, supporting decision-making to implement an increasingly effective, strategic, and humanized HRM.
2. Theoretical Framework and Research Hypotheses
Our main conceptual framework is that of Positive Organizational Studies. The theoretical movement of Positive Organizational Studies, coming from Positive Psychology, has gained prominence in the last decade as a line of investigation aimed at improving organizations, using their internal forces [
3]. It is based on health as opposed to disease, and on positive psychology; it comes from a leadership focused on people and relationships, and aims to comprehend human qualities, virtues, potentialities, and abilities; it seeks to support people for a happier, healthier, and more productive life in work organizations [
26]. In this context, it makes sense to study how leadership can inspire virtues in the work environment, studying the relationship between these variables, because organizational virtues are a variable of interest in positive organizational studies and they should be stimulated from leadership [
27].
Among the arguments that reinforce the potential of the theoretical framework of Positive Organizational Studies, we can highlight the focus on the well-being, health, and quality of life of employees to the detriment of negative aspects, such as experiences of suffering and illness, with the purpose of demonstrating that positive mental, social, and emotional aspects contribute to professional fulfillment and organizational development [
11,
28,
29]. It is interesting to emphasize in this perspective of positive psychology that positive behaviors at work have been gaining greater emphasis as they demonstrate that the organization is a place where virtues are necessary for the good development of management [
9].
We also build our theoretical background based on the idea that HRM practices must be established, shared, and implemented by an active leader capable of inspiring and directing the behavior of employees towards greater professional achievement and the achievement of better organizational results [
5,
30]. In this study, we follow this notion along with the conceptual framework of Barney’s Resource-Based View [
31]. In this regard, workers who are aware of the ethical behavior of their leaders tend to present organizational citizenship behaviors, because individuals learn social behavior through the repeated observation of leadership behaviors [
32].
Therefore, the Resource-Based View was another conceptual framework that inspired the construction of our research model. Within the proposal to understand employees as a core competence of organizations, the Resource-Based View (RBV) argues that people and their interactions are valuable resources in the production of knowledge and achievement of organizational goals, if articulated by a strategic leadership [
31]. Corroborating the RBV, Melo [
33] and Bianchi et al. [
5] explain that the leader’s performance is crucial for the achievement of organizational objectives in a context of a globalized economy, with leadership being a variable that significantly affects organizational results. Accordingly, it is also necessary to highlight that behaviors not prescribed by the formal system have also been gaining importance in obtaining organizational results, such as, Organizational Citizenship Behaviors (OCB), which have been analyzed in empirical studies as antecedents of organizational efficiency and, therefore, understood as essential to obtaining competitive advantage [
20,
34,
35].
Nevertheless, HRM practices must be formulated in line with the strategies and policies of organizations, thus laying the foundations for strategic human resources management [
36,
37]. In view of this understanding of the strategic role of people based on positive behaviors at work, added to the results of empirical evidence, such as some that will be demonstrated in the proposition of the hypotheses, it seems possible to affirm that a set of HRM practices, synchronized with the strategy organization, provides the organization, whether public or private, with the ability to reach a higher level of development [
38]. In order to understand the strategic role of HRM, we must cover the variables that affect HRM practices, particularly those related to the premises of positive organizational behavior [
5,
39].
Leadership is the exercise of influence so that individual and collective efforts achieve shared goals [
40]. There are different styles of leadership in the literature. Based on the evolution of studies on this topic, Melo [
33] consolidated three factors encompassing different perspectives. The first one is relationship, focusing on interpersonal relationships with subordinates, contemplating support, guidance, and facilitation to the work. The second is task, addressing the aspects of the definition of people’s roles on the team to achieve goals. Lastly, the situation indicates that the leader needs to understand the context and promote the necessary adaptations for decision making. In a more contemporary philosophical leadership approach, in addition to reaching organizational goals, the leader focuses on improving the human resources of organizations and the community, acting as an agent of change for the organization’s sustainable development and social responsibility actions to the community [
41].
According to an Aristotelian perspective, some researchers have pointed out that virtues are understood as the way people should be to generate a meaningful purpose in life. Rather than being understood by each thinker’s vision, such as Kant, for example, the most commonly chosen viewpoint in organizational studies is precisely the one that understands the virtues associated with the intellectual and moral spheres [
14]. The intellectual virtues of practical wisdom guide the achievement of organizational goals. Leaders must practice intellectual and moral virtues over time to improve managerial practices and decision making [
14]. There is an understanding that the most adequate way for managers to lead organizations is through example, presenting virtuous behaviors [
14,
42,
43]. In this way, Peterson and Park [
44] conceptualize organizational virtues as moral characteristics of the organization as a whole and as a permanent part of organizational culture. Therefore, Chun [
45] recommends that global companies focus on creating a differentiated image, based on social responsibility and the ethics of virtues, corroborating the studies of Paulraj, Chen, and Blome [
46] and Cugueró-Escofet and Rosanas [
14], who suggest that moral and virtuous reasons and people promote firms’ engagement with sustainability management practices.
In its turn, HRM practices are understood as articulated proposals of the organization regarding human relations with a view to obtaining desired results [
47].
Organizational citizenship behaviors are considered spontaneous actions, linked to the individual desire to cooperate, not prescribed in employment contracts, and independent of professional obligations [
48,
49,
50]. This condition can be explained from the Social Exchange Theory perspective [
51], which advocates the issue of reciprocity as a way to justify these behaviors. This means that employees adopt OCBs to reward fair treatment received by the organization and because they understand these behaviors as part of their work, demonstrating a sense of pride in being part of the organization [
52,
53,
54]. Thus, organizational citizenship behaviors are acts of social exchange offered voluntarily by workers to organizations [
55]. With this in mind, the possibilities of the relationship between the four variables will be analyzed, formalizing the hypotheses’ propositions that will be tested in two original mediation models.
2.1. Leadership and Human Resources Management Practices
Leadership serves as a bridge between a more effective HRM through strategies, policies, and practices, and the employees’ perceptions of such practices [
5]. Considering HRM practices as core elements of organizational culture and fundamental for organizational management [
56], the relationship between leaders and employees can be strengthened and generate greater commitment from the latter when HRM practices are aligned with the leader’s behavior [
57]. Thus, leaders must be committed to practices focused on the workers’ development while meeting the organizational performance goals [
47,
56]. Castro et al. [
58] reinforce the fundamental role of a committed leadership in the adoption and implementation of practices in the strategic HRM perspective.
Similarly, Aktar and Pangil [
59] and Nazarian et al. [
60] argue that HRM practices improve the level of employee involvement if they are introduced and properly managed by a leader who seeks to contribute to a greater effective organizational environment, in an increasingly challenging global environment. In this context, Karam et al. [
61] conclude that authentic leadership is indeed crucial to promote high-performance human resource practices, and it also helps to foster enhanced work engagement. Furthermore, according to the authors, by cultivating greater work engagement, individuals are motivated to bring their best, most authentic selves to the workplace and are more likely to achieve higher levels of both well-being and performance, in line with the assumptions of positive organizational studies.
On that basis, we propose the first research hypothesis:
Hypothesis 1 (H1). Leadership is positively associated with human resources management practices.
2.2. Human Resources Management Practices and Organizational Citizenship Behaviors
Lam, Chen, and Takeuchi [
62] highlighted that HRM practices related to training, development, involvement, and relationship had the greatest impact on organizational citizenship behaviors. Moreover, research suggests that strategic HRM practices presented a positive effect on organizational citizenship behaviors [
63], as also shown by Salas-Vallina, Pasamar, and Donate [
64].
Corroborating these findings, evidence indicates that socially responsible HRM exerts an indirect influence on organizational citizenship behaviors, especially in contexts of higher levels of ethics and corporate social responsibility [
65]. However, literature also signalizes evidence on the opposite. For example, Snape and Redman [
66] and Alfes et al. [
67] state that HRM practices influence OCBs, thus corroborating the existence of the relationship between the two variables. Therefore, we propose the second hypothesis of this study:
Hypothesis 2 (H2). Human resources management practices are positively associated with organizational citizenship behaviors.
2.3. Leadership and Organizational Citizenship Behaviors
Leadership is an important predictor of organizational citizenship behaviors, regardless of the small amount of research dedicated to this analysis [
68,
69,
70]. The study by Ghavifekr and Adewale [
71] showed the positive impact of transformational leadership on organizational citizenship behaviors. Additionally, scholars have shown that managers with a high level of leadership generate a positive influence in promoting organizational citizenship behaviors because there is a tendency for greater employee engagement when they identify with their leaders [
32,
72]. Furthermore, Freire and Gonçalves [
69] confirmed the relationship between leadership and organizational citizenship behaviors, with the variables of perception of social responsibility and organizational identification as mediators.
From these inferences, we present the third research hypothesis:
Hypothesis 3 (H3). Leadership is positively associated with organizational citizenship behaviors.
2.4. Organizational Virtues and Human Resources Management Practices
From the Positive Organizational Behavior perspective, individual behavior and organizational practices can operationalize virtuosity in organizations [
73]. This is because organizational virtues represent the moral and virtuous aspects of the work context, contributing to HRM practices as drivers of positive behaviors at work [
74]. Given this premise, literature indicates the possibility of positive associations between organizational virtues and HRM practices, envisioning a strategic human resources management based on programs that seek to promote ethics and virtues in organizations [
75].
Empirically, the results of the research conducted by Demo, Neiva et al. [
76] revealed the mediating role of HRM practices in the association between organizational virtues and well-being at work, whereas Coura, Demo, and Scussel [
77] showed the mediating role of organizational virtues in the relation between leadership and HRM practices. Both studies attested to the positive prediction of organizational virtues on HRM practices.
Thus, the fourth research hypothesis is proposed:
Hypothesis 4 (H4). Organizational virtues are positively associated with human resources management practices.
2.5. Organizational Virtues and Organizational Citizenship Behaviors
Authors such as Sun and Yoon [
78] indicate that an increase in the employees’ perception concerning organizational virtues can be a way to encourage organizational citizenship behaviors. In line with this proposal, the research by Rego, Ribeiro, and Cunha [
79] showed that the perception of employees about organizational virtues was a predictor of well-being and organizational citizenship behaviors. An explanation for the relationship between virtues and organizational citizenship behaviors is also pointed out by Pires and Nunes [
74], in which organizational virtues promote positive emotions and willingness to work in favor of the organization, which may represent an affective commitment of the worker, demonstrated through organizational citizenship behaviors. Positive perceptions of organizational virtues also lead to greater employee engagement, translating into organizational citizenship behaviors [
28]. Likewise, Mansur, Sobral, and Islam [
80] found a positive association between an ethical (virtue-based) leadership and organizational citizenship behaviors. From there, the fifth research hypothesis is based:
Hypothesis 5 (H5). Organizational virtues are positively associated with organizational citizenship behaviors.
2.6. Leadership, Human Resources Management Practices, and Organizational Citizenship Behaviors
The effectiveness of HRM practices in achieving organizational goals increases when practices are integrated and connected to each other [
39], being the role of the leadership to encourage and promote this integration [
6,
40]. In turn, several studies signalize a relationship between leadership and organizational citizenship behaviors, considering the effects of several mediating variables in such a relationship [
32,
72].
Moreover, HRM practices promote benefits for both workers and the organization, because organizational citizenship behaviors can emerge when the employee feels valued and recognized for such practices [
74]. Likewise, the study by Ababneh, Awwad, and Abu-Haija [
81] highlighted the influence of the interaction between leadership and HRM practices on employee engagement with regard to environmental initiatives. According to the authors, leadership has also been shown to be an important predictor of the adoption of the increasingly necessary green HRM practices. Additionally, the interaction between leadership and HRM practices also exerts a strong influence on the well-being of employees and on overall organizational performance as well [
82]. In turn, HRM practices also have a positive effect on organizational citizen behaviors, as presented in the results provided by Salas-Vallina et al. [
64].
Research indicates a gap regarding mediating models covering HRM practices and positive psychology variables, especially the elements of organizational culture [
24,
63]. Thus, the sixth hypothesis to be tested is:
Hypothesis 6 (H6). Human resources management practices mediate the relationship between leadership and organizational citizenship behaviors.
2.7. Organizational Virtues, Human Resources Management Practices, and Organizational Citizenship Behaviors
Assuming that elements of organizational culture, such as virtues, promote HRM practices that, in turn, raise the levels of organizational citizenship behaviors, it is reasonable that virtues lead to positive behaviors, which consequently lead to better organizational results [
52]. Furthermore, Snape and Redman [
66] observed that when workers perceive that the organization’s support extends beyond the performed work, HRM practices positively affect organizational citizenship behaviors. Therefore, similar studies propose investigations on the relationship between virtues and organizational citizenship behaviors, considering mediating variables such as job satisfaction [
17] and organizational support [
75].
Additionally, a significant number of studies indicate the role of HRM practices as predictors of organizational citizenship behaviors [
39,
62,
63,
83]. On the other hand, some studies relate the predictive role of organizational virtues on organizational citizenship behaviors, despite the need for further confirmation [
74,
75,
78].
By testing a multilevel model of mediation, Ruiz-Palomino, Linuesa-Langreo, and Elche [
84] found a mediating role of organizational citizenship behaviors in the relationship between leadership and team performance, ratifying the positive association between leadership and organizational citizenship. Moreover, Mansur et al. [
80] found a positive association between ethical (virtue-based) leadership and organizational citizenship behaviors. In the same perspective, Nemr and Liu [
32] concluded that ethical leadership has both direct and indirect effects on organizational citizenship behaviors, signaling the importance of studying variables that mediate the relationship between leadership and organizational citizenship behaviors, as we examined in the present study.
On that basis, corroborating the purpose of advancing and contributing to the recognition of the strategic role of HRM by investigating mediation relationships [
24,
39], as in the previous hypothesis, the bases are laid for the seventh and last hypothesis:
Hypothesis 7 (H7). Human resources management practices mediate the relationship between organizational virtues and organizational citizenship behaviors.
3. Method
This study reports the results of a survey with a quantitative nature and transversal time frame. The population of the study was employees of both public and private companies in the state of Minas Gerais, Brazil, from the service, commerce, and industry, in a total of 5 public companies and 6 private companies. We note that all the companies formally authorized the research in its workplace. The sample was non-probabilistic for convenience (adherence).
Data collection was conducted through a printed questionnaire administered in person by the researchers, in the second semester of 2019. Data analyses were performed in the second semester of 2021. The authors handed the respondents a sealed envelope containing the consent form and the questionnaire. To reach data reliability and to guarantee the anonymity of the respondents, we used two folders: one for the consent form and the other for the completed questionnaire deposit. Note that, according to the Sole Paragraph of Article 1 of Resolution No. 510/16 of the Brazilian National Health Council (CNS), consultative public opinion surveys that have their samples composed by unidentified subjects and the confidentiality of the data ensured, as is the case of this research, are exempt from ethical analysis by the Research Ethics Committees (CEP) and by the National Research Ethics Commission (CONEP) of Brazil.
According to Kline [
85], to use regression analysis through Structural Equation Modeling—SEM, the average sample must range between 100 and 200 subjects, criteria met by this study. For instance, to test measurement models, Kline [
85] suggests a minimum of 20 subjects per variable. Considering that the HRM practices model is the one with the highest number of variables (32), a minimum sample of 640 participants would be ideal. In this sense, counting eventual losses in the data processing process, we distributed 1.200 questionnaires—600 for employees of public companies and 600 for employees of private companies. We had a return of 362 questionnaires from public companies (46.9%) and 409 from private companies (53.1%), reaching a total of 771 responses. This number corresponds to 64.25% of the total of distributed questionnaires, which in the view of Baruch and Holtom [
86] represents a high response rate, considering that in organizational studies, it would be approximately 35.7%.
For data treatment, we performed a frequency distribution analysis (mean, standard deviation, variance, minimum, and maximum), the listwise procedure for missing values, the identification of outliers, and multicollinearity and singularity [
87]. We excluded 16 questionnaires due to missing values. Then, based on the Mahalanobis method, we removed 90 outliers. For the verification of tolerance values and variance inflation factor (VIF), we obtained numbers greater than 0.1 and less than 10.0, respectively. This means that there were no problems of singularity or multicollinearity for the sample. Regarding the assumptions for the use of multivariate analysis, we verified linearity, homoscedasticity, and normality of data distribution, using residual graphs and normal probability graphs in the AMOS software [
88]. No problems were detected. The final sample included 659 subjects, meeting the recommended minimum of 640 subjects.
The research instrument was a questionnaire composed of four scientific validated scales: the Scale of Evaluation of the Managerial Style—SEMS [
33], chosen based on the fact it was developed considering different leadership approaches; the Scale of Organizational Moral Virtues Perception Scale—SOMVP [
9]; the Human Resource Policy and Practice Scale—HRPPS [
89]; and the Organizational Citizenship Behaviors Scale—OCBS [
90]. We selected these scales because they are widely used in research, as well as for their very reliable psychometric indices, as
Table 1 shows.
Table 1.
Psychometric Indices of the Scales.
Table 1.
Psychometric Indices of the Scales.
Scale | Factors | Number of Items | Reliability Index (Cronbach’s Alpha) |
---|
SEMS | Task (T) | 6 | 0.72 |
Relationship (R) | 9 | 0.94 |
Situation (S) | 4 | 0.82 |
SOMVP | Organizational Goodwill (OG) | 17 | 0.95 |
Organizational Trustworthiness (OT) | 7 | 0.92 |
HRPPS | Recruitment and Selection (RS) | 6 | 0.81 |
Involvement (I) | 9 | 0.91 |
Training, Development, and Education (TDE) | 3 | 0.82 |
Work Conditions (CT) | 5 | 0.81 |
Performance Evaluation and Competencies (PEC) | 5 | 0.86 |
Remuneration and Rewards (RR) | 4 | 0.84 |
OCBS | Organizational citizenship behavior towards the individual (CBI) | 7 | 0.88 |
Organizational citizenship behavior towards the organization (CBO) | 4 | 0.75 |
For data analysis, we performed confirmatory factor analysis to assess the adjustment of the measurement models of the variables of leadership, organizational virtues, HRM practices, and organizational citizenship behaviors in the general model. Then, to specify and estimate the mediation models, we used path analysis through structural equation modeling, using the maximum likelihood test in the SPSS and AMOS programs. As for the mediation models, the analysis of the relationships between the variables were considered as independent variables leadership (L) in model 1 and organizational virtues (VO) in model 2. In both models, human resources management practices (HRMP) was the mediator, and organizational citizenship behaviors (OCB) was the dependent variable.
4. Findings
4.1. General Model and Measurement Models Tests
To analyze the fit of the model, we first analyzed the modification indices (MI). We identified a correlation between errors 11 and 12 of the Performance Evaluation and Competencies (PEC) and Remuneration and Rewards (RR) factors, respectively, adding this correlation in the model to improve the fit, based on the theoretical support from the scientific literature. In this regard, the performance evaluation and the competencies must be conducted systematically, periodically, and impartially, revealing aspects that can enhance both the development and remuneration of employees [
91]. Additionally, when performance evaluation has the character of subsidizing an effective plan for the development of competencies that affect the remuneration and reward system, instead of being merely punitive, there is a substantial improvement in the workers’ commitment, satisfaction, and productivity [
92].
According to Kline [
85], the analysis through SEM encompasses a measurement model (how the constructs are represented) and a structural model (how the constructs relate to each other). This analysis requires at least one incremental index and one absolute index, in addition to the chi-square value and the associated degrees of freedom to determine its acceptability. Hair et al. [
88] explain that a model that presents the normed χ2 value (CMIN/DF or NC, where CMIN is the χ2 statistic and DF represents the degrees of freedom of the model), CFI (Comparative Fit Index), RMSEA (Root Mean Square Error of Approximation), and SRMR (Standardized Root Mean Square Residual) has enough information for its evaluation.
For Kline [
85], the satisfactory fit values for a structural model are NC (CMIN/DF) of 2.0 or 3.0 and at most up to 5.0; CFI equal to or greater than 0.90; and an RMSEA and SRMR less than 0.06 or even 0.08.
Table 2 presents the values for the NC, CFI, RMSEA, and SRMR values for our model, all in accordance with the parameters recommended by the literature.
Next, to assess the internal validity of a scale or the quality of its items, we observe the factor loadings according to Comrey and Lee’s [
93] classification: loads below 0.32 are poor, between 0.32 and 0.54 are reasonable, between 0.55 and 0.62 are good, between 0.63 and 0.70 are very good, and greater than 0.70 are excellent. In the confirmatory factor analysis of the scales used in this study, the factor loadings of the factors (
Table 3) presented nine excellent items, two very good items, one good item, and one reasonable item, attesting to the quality of the items and, consequently, the internal validity of the scales. Additionally, all variables were significant, considering the
p-value <0.01 and the Critical Ratio (R.C) greater than |1.96|.
To analyze factor reliability, we used the Jöreskog’s Rho coefficient, a more accurate measure than Cronbach’s alpha for structural equation modeling, based on factor loadings and not on observed correlations between variables. Literature indicates ρ values above 0.6 as acceptable, above 0.7 as satisfactory, and above 0.8 as very satisfactory [
94,
95]. Leadership obtained Jöreskog’ Rho of ρ = 0.86; organizational virtues ρ = 0.87; human resources management practices ρ = 0.87; all considered very satisfactory. Organizational citizenship behaviors, on the other hand, obtained ρ = 0.66, being considered acceptable.
Figure 1 illustrates the test of the general research model, obtained in the confirmatory factor analysis, with the respective parameters.
Next, to attest to the validity of the constructs, we evaluated convergent, divergent, and nomological validity. In this step, we verify whether the dimensions (observable variables) effectively portray the theoretical constructs (latent variables) that intend to measure leadership, HRM practices, virtues, and organizational citizenship [
88]. As for the convergent validity, all factors had a factor loading greater than 0.50, Jöreskog’s rhos greater than 0.60, and extracted variances equal to or greater than 0.50 (
Table 3). Thus, we confirm the convergent validity of the measurement models [
73].
We also confirmed discriminant validity. According to the criterion proposed by Fornell-Larcker [
96], as shown in
Table 4, the estimated extracted variance of each variable was greater than the squared value of the correlation between them (values below the diagonal), proving that the four scales effectively measure different constructs.
Finally, we tested nomological validity, which checks the behavior of the scales when related to other constructs, in order to observe their conformity with the literature [
88]. The theoretical framework built for this study listed the possible theoretical and empirical relationships between the constructs, serving as the basis for the proposition of the hypotheses. The hypotheses tests, presented in the following section, confirm the nomological validity of the measures used because all the correlations between them were positive and significant. In summary, the findings indicated that the scales have reliability, internal and construct validity, and can be used in scientific research relational studies, as well as in managerial practice as a diagnostic tool for managers.
4.2. Hypothesis and Mediation Models Tests
In this step, we tested the predictions corresponding to hypotheses H1, H2, H3, H4, and H5. First, we verified the significances of the models. Next, we analyzed the regression coefficients (β), which indicate the magnitude and direction of the associations between the independent variables (IVs) and the dependent variable (DV). Finally, we checked the coefficient of determination (R
2), which indicates the percentage of DV variance explained by the IV, being a measure of adjustment of a linear statistical model: the higher the R
2, the more explanatory the proposed linear model [
87].
Table 5 presents these results.
The associations between the variables were all significant and positive. Regarding the regression coefficient R2, the prediction of organizational virtues on HRM practices (H4) showed results above 26%, indicating a great effect [
97]. The prediction of leadership in HRM practices (H1) had a medium effect (between 13 and 25%). For the other hypotheses, we verified a small effect (between 2 and 12%).
The next step was to construct the path models to test hypotheses H6 and H7 and verify whether HRM practices mediate the relationships between leadership and organizational citizenship behaviors, and between organizational virtues and organizational citizenship behaviors. In line with Baron and Kenny [
98], we tested four conditions simultaneously through structural equation modeling for each hypothesis. First, considering if the antecedent variable significantly predicts the mediating variable. Second, considering if the mediator significantly predicts the criterion variable. Third, considering if the antecedent variable significantly predicts the criterion variable. Fourth, considering if, when in the presence of the antecedent and the mediator variables, the relationship previously found to be significant between antecedent and criterion decreases (partial mediation) or disappears (total mediation).
The first mediation model tested (H6), concerning HRM practices mediating the relationship between leadership and organizational citizenship behaviors, can be seen in
Figure 2. All mediation assumptions were confirmed through the established hypotheses: H1 (β = 0.447; R
2 = 20%;
p-value < 0.01), H2 (β = 0.336; R
2 = 11.3%;
p-value < 0.01), and H3 (β = 0.236; R
2 = 5.6%;
p-value < 0.01), as shown in
Table 5. In the following, we calculated the indirect effect of leadership on organizational citizenship behaviors to test H6. We confirmed a partial mediation. The indirect effect was significant (
p-value < 0.01) and estimated at 0.128. The R
2 regression coefficient was 12.2%, that is, leadership and HRM practices explain 12.2% of the dependent variable organizational citizenship behaviors. According to Cohen [
97], this prediction reflects a small effect.
Table 6 summarizes the results, with all values being significant.
The results demonstrate that the direct relationship between leadership and organizational citizenship behaviors decreases in the presence of the mediator. With this, we confirm the last condition proposed by Baron and Kenny [
98] and the partial mediation of HRM practices in the relationship between leadership and organizational citizenship behaviors. On that basis, H6 has been confirmed.
The second mediation model (H7) tested, regarding if HRM practices mediate the relationship between organizational virtues and organizational citizenship behaviors, can be seen in
Figure 3. All mediation assumptions were confirmed through the established hypotheses: H4 (β = 0.629; R
2 = 39.5%;
p-value < 0.01), H2 (β = 0.336; R
2 = 11.3%;
p-value < 0.01), and H5 (β = 0.241; R
2 = 5.8%;
p-value < 0.01), as shown in
Table 5. Next, we calculated the indirect effect of organizational virtues on organizational citizenship behaviors to test H7. We confirmed a total mediation, as the direct effect of organizational virtues on organizational citizenship behaviors was not significant, attesting that the relationship between virtues and organizational citizenship behaviors only occurs through HRM practices.
Furthermore, the indirect effect was significant (
p-value < 0.01) and estimated at 0.191. The R
2 regression coefficient was 11.4%, that is, organizational virtues and HRM practices explain 11.4% of the dependent variable, organizational citizenship behaviors, with a small effect [
97].
Table 7 summarizes the results, with all values being significant. It is also worth mentioning that, according to Baron and Kenny [
98], the results highlight that the direct relationship between organizational virtues and organizational citizenship behaviors disappeared in the presence of the mediator, confirming total mediation, and thus confirming H7.
In short, the analyses of the findings confirmed the seven hypotheses of this study. Moreover, all relationships assumed in the mediation models tested were significant at the 0.01 level. Leadership indeed has a positive association with HRM practices, accounting for 20% of its explanation, corroborating the fundamental role of leadership in the perception of HRM practices [
56,
59,
60]. HRM practices were also positively related to organizational citizenship behaviors, influencing around 11% of their explanation. That is, the more HRM practices are perceived, the more organizational citizenship behaviors tend to be expressed [
63,
65].
Regardless of the small explanatory power (5.6%), leadership is also positively associated with organizational citizenship behaviors, revealing that strong and inspiring leadership stimulates extra-role behaviors, such as organizational citizenship [
56,
57,
58]. The greatest predictive effect found was in the relationship between organizational virtues and HRM practices (39.5%), possibly because both are elements of organizational culture. The more virtues employees perceive, the more they will perceive HRM practices [
73,
75]. Thus, managers must work on virtues and practices in an associated way. Similarly, virtues encourage citizenship behaviors [
28,
78], although in a more indirect or subtle way, because virtues only contribute to approximately 6% of the explanation of citizenship behaviors.
5. Discussion, Implications, Limitations, and Agenda
The tests of the mediation models are introduced as the main contribution of this paper, as they are unpublished. By confirming hypotheses 6 and 7, the important role of HRM practices in the relationship between leadership and organizational citizenship behaviors, as well as in the relationship between virtues and citizenship, was confirmed. In the relationship between leadership and citizenship, HRM practices were partial mediators. We understand that to foster organizational citizenship behaviors, leaders must resort to integrated and structured HRM practices, as organizational citizenship behaviors emerge as the employees feel valued and recognized for such practices [
74]. HRM practices acted as total mediators of the relationship between virtues and organizational citizenship behaviors. This means that citizenship behaviors will only be influenced by organizational virtues through effective HRM practices.
These findings support effective evidence-based management. Leaders must strive for good relationships through virtues and practices that encourage organizational citizenship behaviors, inspiring an increasingly humanized and strategic human resources management.
As theoretical contributions, our research advances to the areas of human resources management and organizational behavior by bringing an investigation of relationships still unexplored in the scientific literature, testing research hypotheses in more complex models, such as the mediation proposed here. We also emphasize the contribution to organizational studies dedicated to the antecedents and consequences of the variables in question. Furthermore, we foresee a methodological contribution regarding the use of diversified advanced statistical techniques, which brought evidence of validity and reliability to the proposed measurement and mediation models. In this context, students and researchers can rethink the ways in which the phenomenon has been studied and propose new epistemological, theoretical, and methodological approaches to its research.
As managerial implications, the study provides a diagnosis to managers of the organizations surveyed about how leadership, virtues, HRM practices, and citizenship relate to the work context. These findings support the search for continuous improvements in the management of these variables, as well as the analyses carried out herein, which can help managers in their strategies, policies, and practices aiming at a more effective human resources management, in order to promote more productive and healthier work environments.
This diagnosis contributes to implementing an increasingly strategic, consistent, and integrated human resources management area, in which virtues and citizenship behaviors are encouraged. In a complementary way, strong leadership tends to associate organizational virtues and HRM practices to inspire organizational citizenship behaviors to strengthen both the core of the organizational culture and the relationships of trust between members, increasing individual and organizational well-being, as well as promoting greater engagement at work, which will translate into superior results, both at the individual and at the organization level.
We also envisage a social contribution for this study because the promotion of more positive work environments that devise ennobling purposes can generate, in the context of public organizations, public employees more committed to effectively serving well, with diligence, speed, and transparency. This also applies to the private context, in which the service to the final customer in the relationship chain depends primarily on well-trained employees who are motivated to deliver superior quality products and services. In other words, investments in healthier work environments unveil a more humanized management, centered not only on results at the organizational level but also on individual and team levels. From this, we foresee fairer and more harmonious relationships within organizations, which will be reflected in more excellent service to customers, citizens, and society.
Regarding limitations, we highlight the quantitative nature of the study, precluding a deeper understanding of the phenomena beyond its measurement. Based on this, as the first agenda for further studies, we encourage research with multimethod design and triangulation strategies, bringing light to different nuances and perspectives to approach the studied constructs and their relationships. Additionally, the cross-section and the convenience sample are limiting, as the engendered results are restricted to the researched sample, preventing any possibility of generalization and causal inferences. Thus, longitudinal and time-series studies are welcome to shed light on how the relationships between variables occur over time. Furthermore, the cross-sectional nature, as well as self-report as a single source of data, can lead to common-method variance problems. However, although not portrayed in this article, the unifactorial structures of the measurement models did not present good fit indices, so we conclude that the common-method variance alone does not explain the results.
Another suggestion for future research lies in multilevel investigations that address perceptions not only of employees but also of managers and peers, which would provide a less biased view of the studied variables. Finally, we encourage further studies to improve the models tested so far, adding other variables of organizational behavior, such as well-being, commitment, identity, resilience, and justice in the workplace, seeking to investigate different relationships of prediction, mediation, and moderation.
6. Conclusions
Our study achieved the proposed general objective because the seven research hypotheses were confirmed. Our work represents an initial effort to investigate the still unexplored relationships between leadership, organizational virtues, human resources management practices, and organizational citizenship behaviors. We also proposed structural models of mediation between the variables, with the opportunity to inspire new studies that make progress in testing associations between different variables of positive psychology.
In organizational contexts in which we experience challenges and uncertainties, virtues need to be encouraged, and people need to be well cared for through practices to achieve their professional goals, collaborate to achieve organizational goals, and feel motivated to express citizenship behaviors. Given the importance of individual psychological well-being, especially in times of crisis and change, promoting healthy workplaces should be a priority for organizations.