English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Introduction
The agenda of Organizational and Work Psychology (OWP) over the last fifty years has included, as one of its main concerns, understanding the bonds that workers form with an organization, especially organizational commitment. There has been a longstanding interest in investigating the antecedents, correlates, and consequences of organizational commitment, such as performance, motivation, and job satisfaction (Bastos et al., 2013). More recently, the field has begun to invest in research that relates this bond to another phenomenon that has only lately attracted the attention of scientists and society in general: the well-being experienced by workers in their work context.
Despite the disparity in the amount of scientific production on these two phenomena, they have often been studied together since the 2000s. The accumulated evidence leaves no doubt that well-being at work and organizational commitment are strongly related (e.g., Meyer & Maltin, 2010). However, it is still unclear how the pattern of association between them occurs. The literature presents an ambiguous relationship between these two phenomena, with different perspectives considering them as equivalent (e.g., Fisher, 2010), correlated (e.g., Ryk, 2021), or even as antecedents or consequences of each other (e.g., Brunetto et al., 2012; Raina, 2013). This complexity reflects the historical ambiguity surrounding the relationship between commitment and job satisfaction (e.g., Çınar et al., 2022), a variable considered by some theoretical perspectives of well-being as one of its dimensions (e.g., Diener, 1984). Although this is not the definition adopted in the present study, the proximity between job satisfaction and workers’ well-being can, to some extent, contribute to the discussion about how organizational commitment and well-being at work are related.
The present study aims to verify the association between well-being at work and organizational commitment and to determine whether they behave as equivalent phenomena, correlated phenomena, or whether well-being is a precursor or an outcome of commitment. Given the challenge of using the dominant standard of cross-sectional research to test, in alternative theoretical models, the direction of influence between the two phenomena, we seek, in effect, to gather additional evidence for the debate on a classic question that remains unanswered. In addition to advancing the scientific field, this identification may have benefits for interventional practices aimed at fostering more positive relationships between workers, their work, and the organization.
Organizational commitment
Organizational commitment gained prominence as a research topic following the seminal book by Mowday et al. (1982), which consolidated a body of research developed in the 1970s. The field has advanced based on the Organizational Commitment Questionnaire, which was constructed from a definition by Porter and Smith (1970) and validated by Mowday et al. (1979). It is a one-dimensional perspective of measuring the construct that emphasizes the affective nature of commitment, defining it as a state in which the individual identifies with an organization and its objectives and wishes to remain a member of it.
Since 1991, however, the theoretical-conceptual model that has gained greater adherence in the literature is the three-dimensional model by Meyer and Allen (1991), which systematizes organizational commitment on three bases: 1) affective: identification with the organization, shared values, and personal involvement; 2) normative: internalization of norms and the reciprocity associated with the psychological contract established between the worker and the organization; 3) continuance: assessment of the costs associated with leaving the organization due to the investments made and the perceived lack of alternatives in the job market. In the last decade, the hegemony of this model has been questioned due to its conceptual inconsistencies - such as the possibility of overlap between affective and normative dimensions - and its empirical inconsistencies - such as the opposite behavior of the continuance basis in relation to the affective and normative bases when the construct was tested in relation to its antecedents and consequences (Bastos et al., 2013).
In Brazil, the efforts of Bastos and his research group (2006) to better delineate the theoretical and conceptual boundaries of the construct are highlighted. Consistent with these efforts, there is a movement to return to the original definition of organizational commitment (COMM) with respect to its affective basis (Bastos et al., 2011), and to separate the other bases for the composition and proposal of new constructs to represent the worker’s bonds with the organization. As a result of these processes, organizational consent emerged from the normative basis and organizational entrenchment from the continuance basis. The distinction between these three types of bonds reflects, respectively, the degree to which an individual is driven by desire, obligation, or necessity to remain in the organization (Bastos et al., 2013).
Comparing the three constructs, it can be said that COMM is the positive bond that the worker establishes with the organization, since it is characterized by shared values between the individual and the organization, and by the employee’s active commitment to achieving organizational objectives. This bond can then be described as a relatively stable psychosocial process that is the result of a relationship built over time (Bastos et al., 2018). It is strongly associated with various positive indicators that are important for both the worker and the organization (Bastos et al., 2013). One such indicator is work-related well-being.
Well-being at work
The different theoretical propositions surrounding well-being are mainly anchored in two philosophical-epistemological streams. On the one hand, based on the hedonic perspective, there are authors (e.g., Diener, 1984) who understand well-being as a synonym for pleasure, a state resulting from a cumulative balance of affective states, in which the amount of positive affective states (such as happiness, willingness) stands out in relation to the amount of negative affective states (such as sadness, anger). From this stream comes, for example, the concept of well-being at work used by Ferreira et al. (2007), who define it as a phenomenon resulting from a cognitive assessment related to job satisfaction and an affective state consistent with the presence of positive and negative emotions that workers direct toward their work. On the other hand, there are authors (e.g., Ryff & Keyes, 1995) who conceive of well-being based on the eudaimonic perspective, which understands well-being as a process of individuals moving toward personal achievement and growth. In line with this, Bartels et al. (2019, pp. 3-4) propose to evaluate eudaimonic well-being at work as a combination of intrapersonal (internal or personal aspects) and interpersonal (external or social aspects of work) that “shape an individual’s experience of fulfilling his/her potential and intrinsic goals”, while Paz (2004) describes personal well-being in organizations as the result of satisfying needs and fulfilling desires of individuals when performing their role in organizations.
Recently, a movement has emerged in the literature that aims to highlight the importance of assessing well-being by combining these theoretical bases (e.g., Seligman, 2003; Henderson et al., 2014; Paschoal & Tamayo, 2008; Simone, 2014; Warr, 2007). Peter Warr (1978), the pioneer in the study of well-being at work (WBW), understands this phenomenon as a combination of the hedonic and eudaimonic perspectives, defining it as a positive subjective experience related to work that depends on two central aspects: 1) the prevalence of positive affect over negative affect at work (PREV_PANA hedonic basis), and 2) having work experiences that provide development of individual potential and personal fulfillment (EF eudaimonic basis) (Warr, 2007).
Regarding the individual’s overall well-being, several studies compare different constructs derived from hedonic and eudaimonic approaches to determine whether there is indeed empirical evidence to support the notion that they are independent phenomena (e.g., Chen et al., 2012; Cooke et al., 2016; Gallagher et al., 2009; Joshanloo, 2016; Joshanloo et al., 2017). Although the results are not conclusive, it is clear that, in general, hedonic- and eudaimonic-based constructs of well-being tend to exhibit similar behavioral associations with most of the phenomena recognized as correlates of well-being, but there are some phenomena that still fall outside the rule (Cooke et al., 2016). Thus, efforts are needed to gather empirical evidence that allows for a discussion of these issues in relation to the constructs of well-being specific to the work context as well (Turban & Yan, 2016).
Understanding these theoretical distinctions is essential when attempting to verify the relationship between well-being and other phenomena, since the use of similar nomenclatures does not always clearly show that one is working with the same concept of well-being. In this regard, the following section presents the associations between well-being and organizational commitment, as identified in various studies. Special attention will be given to distinguishing the specific constructs of well-being used as a basis for a comprehensive analysis of the results (Carneiro & Bastos, 2020).
Associations between work-related well-being and organizational commitment
The theoretical confusion that still surrounds work-related well-being contributes to making its relationships with organizational commitment unclear. Specific research can be found that examines commitment as a representative of well-being when the latter is understood from a more generalist and inclusive perspective of any positive experience that the individual has with their work or the organization in which they work (e.g., Baptiste, 2008; Fisher, 2010; Van Horn et al., 2004). Siqueira and Padovam (2008), for instance, advocate this position of including organizational commitment as a representative of well-being, but they do so not from a generalist perspective, but rather from an operational one, describing well-being at work as “(...) a concept consisting of three components: job satisfaction, work involvement, and affective organizational commitment” (p. 206). Siqueira et al. (2014) describe these variables as bonds, with commitment being a bond toward the organization, and satisfaction and involvement referring to the relationship that the individual has with their work. This kind of operationalization can also be found in Mahima and Gurpreet’s (2019) study, which conceptualizes well-being as consisting of three dimensions: job satisfaction, affective organizational commitment, and emotional exhaustion.
This type of proposal shows no grounding in either of the two main theoretical-epistemological streams that have guided the study of well-being, since both understand this phenomenon as a positive subjective state and not as a bond or attitude, as COMM is classified (Bastos et al., 2013; Bastos & Aguiar, 2015). Although the definitions adopted here for WBW and COMM share the affective foundation, this is not enough to define them as interchangeable factors. In addition, it is important to note that COMM includes affect directed toward the organization, whereas WBW includes affect directed toward the work. Thus, the difference in the levels covered is also an important element to consider when addressing these phenomena. This scenario therefore demonstrates the importance of verifying whether COMM and WBW can be treated as equivalent phenomena or whether they are merely correlated.
Despite the reality found in the empirical research, it is believed that the use of COMM as a representative of WBW is not supported in the theoretical-epistemological bases traditionally used to conceptualize this phenomenon, and this use may entail a theoretical expansion of what is understood as WBW, in which a set of positive phenomena may be used indistinctly in an attempt to access a phenomenon that has more specific characteristics. As Lucas, Diener, and Suh (1996) point out, the broader the definition of a phenomenon, the greater the need to analyze the discriminant validity between it and other phenomena to which it is related. With this in mind, the first hypothesis of the study is presented:
H1: Organizational commitment (COMM) and well-being at work (WBW) are two positively related but independent phenomena.
To test H1, the strength of the level of correlation between COMM and WBW will be assessed, as well as how they behave in relation to other variables. To this end, their interactions with several sociodemographic variables will be analyzed, along with two phenomena in the field of OWP with which they are often associated: self-efficacy and organizational support.
As shown in some reviews (e.g., Bastos et al., 2013; Meyer & Maltin, 2010), in general, studies dedicated to verifying the relationship between well-being and commitment point mainly to positive correlations between the two constructs, with few studies attempting to identify antecedence/consequence relationships between these phenomena. However, those that do attempt to verify the direction of this relationship do not present consistent results. Some point to well-being as an outcome of affective organizational commitment (e.g., Andrade, 2008; Bastos et al., 2013; Panaccio & Vandenberghe, 2009; Raina, 2013), while others identify well-being as an antecedent (e.g., Aggarwal-Gupta et al., 2010; Albrecht, 2012; Brunetto et al., 2012; Garg & Rastogi, 2009; Kolakowski et al., 2020; Yalcin et al., 2021).
Research that considers commitment as a predictor of well-being bases its theoretical support primarily on the fact that organizational commitment reduces the impact of stressors in the work environment (Meyer & Maltin, 2010; Panaccio & Vandenberghe, 2009). Therefore, by helping to cope with negative factors related to work, affective organizational commitment would promote the worker’s pleasure or well-being. It should be noted that the well-being constructs used in these studies are predominantly based on the hedonic perspective (e.g., Panaccio & Vandenberghe, 2009) and/or consist of constructs that are not specific to the work context and aim to assess only the general well-being of the individual (e.g., Raina, 2013), whereas research pointing to well-being at work as an antecedent of organizational commitment tends to use a eudemonic perspective on the first phenomenon, understanding it as a process of development and expression of individual potential that can generate a positive attitude of the worker toward the organization (e.g., Aggarwal-Gupta et al., 2010; Albrecht, 2012; Brunetto et al., 2012; Garg & Rastogi, 2009; Kolakowski et al., 2020; Yalcin et al., 2021).
In short, it is recognized that different theoretical contributions have generated conflicting conclusions about the relationship between well-being at work and organizational commitment. Therefore, in view of the different directions indicated by the literature for understanding the relationship between the phenomena investigated, the intention here is to test these possible relationship patterns, based on the understanding that commitment is not a construct that constitutes well-being, but a phenomenon that acts in the well-being process, mediating its dimensions, based on the following hypotheses:
H2: Organizational commitment (COMM) presents different patterns of association with dimensions of well-being at work (WBW), representing different theoretical perspectives.
H2A: Well-being measured from the eudemonic perspective (WBW-EF) behaves as an antecedent of organizational commitment (COMM).
H2B: Well-being measured from the hedonic perspective (WBW-PANA) behaves as an outcome of organizational commitment (COMM).
Although no research has been found to date that investigates the relationship between these two phenomena using a unified concept of WBW that considers both the hedonic and eudaimonic bases, we reflect here that, theoretically, the subjective state of the individual is more likely to influence the construction of the quality and the type of relationship they will establish with the organization than the other way around. Therefore, our intention is to test the third and final hypothesis:
H3: The individual’s overall work-related well-being (WBW, assessed jointly in its hedonic and eudaimonic bases) positively predicts their level of organizational commitment (COMM).
Method
This study is quantitative, extensive, and cross-sectional.
Participants
The sample was non-probabilistic and consisted of 360 workers from different segments/occupations who were approached based on their interest and availability. The main eligibility criterion was to have at least a 3-month relationship with the organization, a period that is considered reasonable for a worker to get to know and adapt to their work activities and the organization.
The age ranged from 19 to 66 years (M=35.80; SD=10.64). Most of the participants were female (69.6%) and had completed some university education (51.7%), although there were also workers with a lower level of education, such as elementary and secondary school (9.1%). In terms of occupational data, most of the workers were engaged in activities related to the service sector (41.2%), mainly concentrated in private organizations (57.8%), but also in public organizations (32.7%) and in the third sector (9.5%). Only 20.5% of the participants were in a leadership position when they responded to the survey, and 58.6% of the sample were paid up to 4 times the minimum wage.
Data collection instrument
The instrument consisted of five parts:
(1) Sociodemographic and occupational information: questions that trace the trajectory of each worker, their social origin, family context, and entry into the world of work.
(2) Well-being at Work Scale, validated by Paschoal and Tamayo (2008), consisting of the factors “positive affect - PA” (9 items, α=0.93), “negative affect - NA” (12 items, α=0.91), and “expressiveness/fulfillment at work - EF” (9 items, α=0.88). Items had to be evaluated based on a 5-point response scale, with an intensity scale for the first two factors mentioned and agreement for the last factor.
(3) Organizational Commitment Scale, one-dimensional, validated by Bastos and Aguiar (2015) (12 items, α=0.86). On this scale, the participant had to choose between six response levels, the first three being disagreement and the last three being agreement.
(4) Perceived Organizational Support Scale, validated by Siqueira and Gomide Júnior (2008) (9 items, α=0.86). In this instrument, there are seven response options organized on an agreement scale.
(5) Self-efficacy General Scale, one-dimensional, proposed by Chen et al. (2001), adapted and validated in Brazil by Balsan et al. (2020) (6 items, α=0.83). The items had to be answered using a 5-point Likert scale of agreement.
Data collection procedures
Data collection took place through a “snowballing” mechanism, initiated by inviting people in the researchers’ social network to respond to the survey. Those who received the invitation were also asked to forward it to others who might be apt to respond. The questionnaire was available in both digital version and printed versions and was administered by self-report. Response time was approximately 30 minutes. The research followed ethical guidelines of the APA, whose basic principles were explained in the free and informed consent form, which was signed by each of the participants. The contact details of the researchers were made available in case of doubts or need for assistance.
Data analysis procedures
The data were tabulated in the SPSS (Statistical Package for Social Sciences) 17.0 software, through which we performed: descriptive data analyses; factor analyses to ensure that the scale used properly fit the study sample; Pearson’s correlation analyses to verify the association between the key variables of the study, as well as to analyze their patterns of association with other variables. The AMOS 18 software was also used to perform structural equation modeling (SEM) using the maximum likelihood estimation method, which allowed the comparison of alternative models of association between the study variables (Figure 1).
Results and discussion
Measuring the fit of the instruments to the sample
First, we tested the fit of the structure of the instruments used for the sample in question, as suggested by authors such as Borsa and Seize (2017). The structure was tested through both exploratory and confirmatory factor analyses, as testing the scale measurement model is a prerequisite for placing the measurement in a structural model (Byrne, 2010; Pilati & Laros, 2007).
In general, the scales were organized as expected and showed satisfactory reliability indicators; however, in the case of WBW, the analyses indicated the need to exclude three items because they cross-loaded into two factors: item 7 (“bored”) and item 14 (“frustrated”), originally from the NA factor, and item 21 (“quiet”), originally from the PA factor (see more in - author reference to be included after peer review). Subsequent analyses of WBW used its factors independently (PA, NA, and EF) as well as a general indicator resulting from a theoretical composition (WBW={PREV_PANA+EF}/2), in which the mean of the expressiveness/fulfillment dimension was considered with the prevalence of the PA over NA dimension (PREV_PANA={PA-NA+6}/2). It was not necessary to change the item composition of the COMM scale, which was assessed using a single indicator.
Confirmatory factor analysis (CFA) showed the need to re-specify the model by inserting modification indices. This insertion was done using as criteria: 1) insertion of parameters whose modification indices were greater than 30; 2) insertion of parameters only between the residual errors of the observable variables; 3) theoretical coherence between the variables for which the associations were suggested. In each of the scales, a total of three adjustment parameters were inserted, one at a time, with subsequent reassessment of the model. It was then considered that the insertion of the newly suggested parameters was not theoretically justified and would not significantly increase the fit indices of the model (Byrne, 2010; Perry et al., 2015). The final fit data for each of the scale models after CFA are shown in Table 1.
It is noteworthy that the RMSEA index for the COMM scale, although not at an ideal level, is within the maximum cut-off point (<0.10) established by Browne and Cudeck (1993). The same applies to the GFI index for the WBW scale, which can be classified as borderline but acceptable (Browne & Cudeck, 1993).
Organizational commitment and well-being at work: two related but independent phenomena
Following the descriptive analyses, we found that in the sample evaluated, the mean values of WBW (both in its general indicator and in its positive factors alone) and COMM are above the midpoint of their respective scales, which shows that, in general, there is a positive experience of well-being and commitment among the participants (Tables 2 and 3).
Regarding the correlations tested, the first important data to analyze relates to the level of overall association between the COMM and WBW phenomena. One way to confirm convergent validity between two phenomena is to find a strong association (r>0.800) between them (Joshanloo et al., 2017; Souza et al., 2017). As shown in Table 2, the association between COMM and WBW is positive but moderate (r=0.598; p<0.01), suggesting that although related, these phenomena can be considered independent.
The assessment of the pattern of associations of these phenomena with other variables can also be used as a sign of convergent validity, although this assessment is based more on qualitative parameters than on quantitative criteria. Table 2 shows the result of the relationships identified between the key phenomena of the study (COMM and WBW) and several variables with different characteristics: self-efficacy, which is a more subjective individual characteristic; age and education, which refer to the individual’s characteristics but have a more objective nature; organizational support, which refers to a characteristic that the worker perceives or does not perceive as being present in the organization; and compensation, which is an objective occupational characteristic. If we look at the correlations established by COMM and WBW with the variables mentioned, we can see clear differences (Table 2).
Table 2. Correlation matrix between well-being at work (WBW), organizational commitment (COMM), and other variables, with means and standard deviations.
** p<0.01; * p<0.05.
Note: Education and compensation were assessed as ordinal variables.
Self-efficacy, which refers to the beliefs a person develops about their ability to initiate, perform, and successfully complete tasks (Bandura, 1977), was more strongly associated with WBW (r=0.313; p<0.01) than with COMM (r=0.193; p<0.01). On the other hand, organizational support, which reflects workers’ perceptions of “(...) the extent to which the company that employs them is concerned with promoting their well-being” (Siqueira & Gomide-Júnior, 2008, p. 285), was more strongly associated with COMM (r=0.698; p<0.01) than with WBW (r=0.529; p<0.01). Both self-efficacy and organizational support can be classified as resources (Nielsen et al., 2017) that contribute to a more positive individual-work-organization relationship. Thus, it is not surprising that the greater the presence of these resources, the greater the experience of WBW and the stronger the establishment of COMM. However, it is interesting to note that self-efficacy, which is an individual resource, has a greater influence on WBW, while organizational support, which is a resource provided by the organization, has a stronger influence on COMM. Such data can be taken as an indication that WBW and COMM are phenomena with different levels of expression.
As for the more objective individual variables (age and education), a different pattern of association is also observed. Age showed a weak but significant association with COMM (r=0.119; p<0.05), whereas it was not significantly associated with WBW (r=0.095; p=0.074). While education was significantly and weakly associated with WBW (r=0.119; p<0.05), it was not significantly associated with COMM (r=0.103; p=0.052). Hence, it can be observed that the younger the worker, the less likely they are to be committed to the organization in which they work, although their age does not affect how they feel about the work they do. On the other hand, the higher the level of education, the greater the possibility that the worker will perform tasks that make them feel good, although this factor does not intervene in how much they will be committed or not to their organization.
Finally, compensation, which is a more objective occupational variable, showed a significant and positive correlation with both phenomena, but with a slightly higher strength with COMM (r=0.208; p<0.01), indicating that the financial return that the worker receives in exchange for their work is more associated with a stronger affective bond with the organization than with a more positive subjective experience in relation to their work (r=0.153; p<0.01).
Therefore, it is concluded that, taken together, the data presented support H1 of the study, allowing us to consider that WBW and COMM are different and independent phenomena, even if they are positively related. The evidence that we are dealing with two different phenomena allowed us to test hypotheses 2 and 3, in order to promote a more detailed understanding of the relationship between WBW and COMM.
Organizational commitment and well-being at work: what is the direction of the relationship?
Table 3 shows that all correlations between COMM and WBW proved to be significant (p<0.01), with positive and moderate associations with well-being in its overall measure (r=0.598), with the EF factor (r=0.576), and with the positive affect factor (r=0.567), and a negative and weak association with the negative affect factor (r=-0.274). The association of COMM with the second-order dimension PREV_PANA, formed from the prevalence of PA over NA, also proved to be significantly positive (r=0.495) and almost as strong as in the overall WBW measure. This result confirms previous studies that had already verified a positive association between COMM and the positive components of WBW and a negative association between COMM and the negative components of WBW, the latter phenomenon being assessed through different theoretical propositions (Andrade, 2008; Dávila De León & García, 2014; Souza et al., 2018; Traldi & Demo, 2012).
To verify the best-fit pattern of association between the variables, theoretical-empirical models 2 and 3 (see Figure 1) were compared using SEM, which allows the relationship between multiple dependent and antecedent variables to be tested simultaneously. No additional adjustment parameters were included beyond those already specified in the CFA for each of the scales.
Table 3 Correlation matrix between organizational commitment (COMM) and the dimensions of well-being at work (WBW), with means and standard deviations.
** p<0.01.
Well-being was represented as an overarching construct, considering the three dimensions (PA, NA, and EF) as first-order factors, with PREV_PANA as a second-order factor in model 2 and overall WBW as a second-order factor in model 3. This choice was guided by the fact that this type of model allows the phenomenon to be evaluated simultaneously from both a general and a specific perspective (Brunner et al., 2012).
Previous studies that had demonstrated the predictive power of eudaimonism with respect to COMM (e.g., Aggarwal-Gupta et al., 2010; Albrecht, 2012; Brunetto et al., 2012; Garg & Rastogi, 2009) and COMM with respect to hedonism (e.g., Panaccio & Vandenberghe, 2009) did so using well-being constructs anchored in only one of these theoretical bases in isolation. In the model tested here (model 2), in which both bases are tested simultaneously in their association with COMM, it was found that the eudaimonic basis of WBW did indeed have a moderate impact on COMM (β=0.634), just as COMM had a moderate impact on the hedonic basis of WBW (β=0.672) (Figure 2).
However, comparing the fit indices of this model with those of model 3 (Table 4), it is clear that model 2 is viable but less adequate, since the fit indices of model 3 are slightly better.
Table 4. Fit indices of the alternative models tested (2 and 3) on the relationship between organizational commitment (COMM) and well-being at work (WBW).
Looking at the indications for re-specification through the modification indices (MIs), it is noted that the most prominent recommendation of the analysis for model 2 is the insertion of a path linking the expressiveness/fulfillment factor (EF) to the hedonic basis of prevalence of PA over NA (M.I. 35.504; Par. Change 0.473), and the second recommendation is the insertion of a path between NA and PA (M.I. 35.074; Par. Change 0.479). Although these MIs were not inserted because they would not allow the model to be tested as proposed (with the hedonic and eudaimonic bases of well-being evaluated independently), the indications for re-specification suggest that the association between the three bases of WBW needs to be considered for a better representation of this phenomenon, which is represented by model 3.
Thus, although the model is statistically viable, the empirical support for the second hypothesis of the study (H2) can be considered only partial, since the relationship between COMM and WBW is better explained by an integrated assessment between the hedonic and eudaimonic bases of well-being, confirming the idea that these theoretical perspectives, although they may present some differences, refer to a common higher-level construct (Cooke et al., 2016; Paschoal & Tamayo, 2008; Warr, 2007), thus directing the focus of the analyses toward model 3 (Figure 3). Thus, confirmation of hypotheses H2A and H2B was not possible.
Assessing the fit indices of model 3 (Table 4), it can be said that H3 was confirmed, indicating that WBW can predict COMM. Compared to model 2, in which the eudaimonic basis of WBW appears as a predictor of COMM, while the hedonic basis appears as a consequence, model 3 shows greater robustness, with all fit indices slightly higher in terms of quality. On the other hand, considering that both model 2 and model 3 were statistically viable, it should be noted that the SEM was not sufficient to definitively clarify the directionality between these two variables.
In short, the comparison of the tested models shows that there is evidence to support the hypothesis that the association between WBW and COMM is cyclical in nature, based on feedback loops. This hypothesis provides a possible explanation for the contradictory results found in the literature regarding the pattern of association between these two phenomena. In this sense, the association between these two phenomena could be seen as a process, a flow, characterized by inherent complexity that poses challenges for cross-sectional research to fully capture. In addition to the empirical evidence obtained, such an ad hoc explanation is consistent with theoretical considerations accumulated in the socio-behavioral sciences about the nature of the phenomena under investigation. Given that most of our constructs are processes rather than fixed or rigid states or traits, it is important to recognize that linear relationships between antecedents and consequences may not adequately capture the complex bidirectional, cyclical relationship involved in variables that are in fact procedural.
In a meta-analysis conducted by Çinar et al. (2022) to examine the relationships between organizational commitment and job satisfaction among healthcare professionals, the findings pointed to a reciprocal cause-and-effect relationship between these two constructs. Given the close relationship between job satisfaction and worker well-being, this study strengthens the plausibility of the hypothesis proposed here, which warrants further investigation in future research using appropriate methodological designs to explore this relationship more comprehensively.
Conclusions
An in-depth analysis of the relationships between well-being at work (WBW) and organizational commitment (COMM), two crucial constructs in the OWP field, provides valuable insights for both theoretical and practical knowledge. In terms of theoretical advances, this study helped to clarify some issues, such as the fact that well-being and commitment are not the same thing. In addition, while this study did not yield definitive conclusions regarding the exact pattern of the relationship between these two phenomena, it did facilitate the development of theoretically consistent hypotheses that warrant further exploration in future research. In terms of practical knowledge, the findings suggest that people management practices aimed at cultivating positive states in individuals in relation to their work may have significant implications for fostering positive connections with their organization, and vice versa.
This study advances by providing evidence that WBW is a distinct construct from COMM. Despite this contribution, new evidence of discriminant validity needs to be sought. To this end, it is suggested that future studies use other methods to test the distinction between COMM and WBW to strengthen the understanding that there is independence between these phenomena. It is also recommended that discriminant validity analyses between the WBW measure and other positively correlated constructs, such as work engagement and job satisfaction, be included in future research agendas.
Moreover, the present study makes progress in explicitly comparing two contradictory patterns of association between these phenomena reported in the literature. Although the methodological design, which relies on cross-sectional data matrices for both models, is a limitation, the findings suggest that WBW may indeed function as both a precursor and an outcome of COMM.
According to Kline (2011), when a researcher is uncertain about which direction adequately defines the relationship between two variables, one of the options for analyzing this relationship through SEM is to specify, test, and compare alternative models that represent different causal directions. The lack of similar results in the existing literature justifies this first empirical effort to examine the direction of influence between the two phenomena by understanding how the theoretical bases (hedonic and eudaimonic) that make up WBW relate to COMM. However, considering that both models tested are statistically viable, the pattern of association between COMM and WBW can be inferred as cyclical: the more committed workers are to their organization, the better they tend to feel about their work; likewise, a higher level of work-related well-being also encourages workers to be more committed to their organization.
Nevertheless, these indications require further investigation in new studies that can include larger and more representative samples, as well as include in the models tested other variables that can help explain when COMM acts as an antecedent of WBW and when it acts as a consequence. Alternatively, even cross-sectional studies that include in their proposal the analysis of predictive relationships between WBW, COMM, and other variables may also help to understand, for example, whether the direction of the relationship between WBW and COMM can be modified in association with other variables. Certainly, longitudinal research designs that evaluate these phenomena over time, based on repeated or continuous measures, offer promising avenues for advancing the understanding of the complex processual relationships that can be established between WBW and COMM. This is because, as explained by Pilati and Laros (2007), SEM alone may not sufficiently control for the multiple factors that can intervene in the variation of the phenomena. Furthermore, as in other research traditions in organizational behavior, longitudinal studies of single cases, mapping changes in the two phenomena (commitment and well-being) over time in relation to personal and contextual factors, represent a methodological challenge to be addressed in future studies.
