**4. Discussion**

The main aim of this study was to contribute to the theoretical understanding of misaligned wellbeing patterns by considering the profiles emerging from the combination of different levels of job satisfaction and mental health. To accomplish this aim, we pursued two research objectives. The first was to identify four patterns of employee wellbeing based on a configurational variable that combines job satisfaction and mental health. Second, we examined some antecedents that can discriminate each of the four patterns. The antecedents we considered were role stress (role ambiguity, role conflict, and role overload), job importance, and overqualification.

Results showed that hypothesis 1 was partially confirmed. We confirmed that role ambiguity and job importance strongly differentiate between the unsatisfied-unhealthy and satisfied-healthy patterns and the rest of the patterns. On the one hand, our findings are aligned with previous studies on role ambiguity showing that the strength of relationships between role ambiguity, and job dissatisfaction and tension/anxiety are generally stronger than those for role conflict [20], whereas other studies have also shown that role ambiguity has adverse effects on employee job satisfaction and mental health [18,19,50]. On the other hand, with this hypothesis, we also confirmed that employees who perceive various facets of job importance, such as intrinsic (e.g., learning opportunity), extrinsic (e.g., job security), and social (e.g., societal contribution) facets, have optimal job satisfaction and mental health, compared to employees who have high role ambiguity. This result is consistent with previous studies [24].

Contrary to our expectation that role conflict would differentiate employees with the unsatisfied-unhealthy pattern from those with the satisfied-healthy pattern, our results instead showed that role conflict, along with role overload, characterized employees with the satisfied-unhealthy pattern. This result partly supports our hypothesis 2, confirming that role overload characterizes the satisfied-unhealthy pattern. Thus, role conflict and role overload had negative consequences on mental health, but less on job satisfaction. These results corroborate role expansion theory, which asserts that employees who engage in multiple roles at the same time receive incentives, status security, and position increments, which in turn have a positive effect on job satisfaction [21,22]. Furthermore, this study also confirms that focusing only on negative consequences of role stress is just one side of the issue, as asserted by McGowan et al. [50]. As the Job-Demand Control model indicates, highly demanding jobs can provide high decision latitude, control, and autonomy for employees, which, in turn, may decrease the negative effects of job demands on job satisfaction, although they can still produce negative effects on health [50].

Finally, we confirmed our hypothesis 3. Our argument was that overqualification would characterize employees with the unsatisfied-healthy wellbeing pattern. Results show that employees with higher levels of overqualification were characterized by job dissatisfaction and, at the same time, showed optimal levels of mental health. They may perceive that the salary, incentives, and other resources they receive from their work are not fair, given their qualifications. This argument substantiates equity theory. According to equity theory, employees compare the resources they put into the work (such as level of education, skills, knowledge, experience, etc.) to what they receive in return (such as payment, recognition, responsibility, etc.) in order to determine their sense of fairness [9]. If they perceive unfairness in what they receive, they may be dissatisfied with their job. Previous studies also have shown that overqualified employees have low job satisfaction [9,26] but higher satisfaction with their life and better mental health [6,10].

We also accomplished our first specific research objective, which was to identify these four wellbeing patterns involving job satisfaction and mental health. Surprisingly, the most populated cluster was the unsatisfied-unhealthy pattern, and more than half of our sample had a misaligned pattern i.e., either satisfied-unhealthy or unsatisfied-healthy. Traditionally, job satisfaction and mental health are believed to be harmoniously and positively correlated, with high (or low) job satisfaction positively correlated with high (or low) mental health. However, this might not always be the case. In this research, we focused on a new research paradigm by studying the combinations of different levels of job satisfaction and mental health. By combining different levels of job satisfaction and health, we identified four important wellbeing patterns and their antecedents. We especially focused on the anomalous or misaligned wellbeing patterns (satisfied-unhealthy or unsatisfied-healthy) as new typologies. Therefore, we believe that our research findings may motivate scholars to investigate the wellbeing patterns by using the current model as a framework. Furthermore, future studies could also combine the effects of role ambiguity, role conflict, role overload, job importance, and overqualification to move towards more generalizable empirical findings and theory development.

The results that job satisfaction and mental health together form four wellbeing patterns, indicate the need for theoretical precision; it is in fact important to integrate this complexity into the harmonious relationship between job satisfaction and mental health in order to study a broader taxonomy of relations and the conditions in which these patterns are elicited. For instance, our study questions the model of health and wellbeing in the workplace by Danna and Griffin [4], suggesting that it should be integrated with the four patterns of wellbeing and mental health described here. At the same time, our study questions whether wellbeing spillover always happens. Wellbeing spillover proposes that work and family domains of wellbeing have similar effects on one another [12]; therefore, we would expect low levels of job satisfaction to be related to low levels of mental health. Although our study suggests that this may be the case for the satisfied-healthy and unsatisfied-unhealthy patterns, and their antecedents in terms of role ambiguity and role importance, our results also suggest that this spillover may not always take place because spillover may not be present in the misaligned patterns. The results on their antecedents also suggest that work-related conditions and activities may affect work-domain (e.g., job satisfaction) and context-free (mental health) wellbeing at the same time, which challenges the idea that context-free wellbeing should be more responsive to health or family-domains [13]. We have already described why role conflict, role overload, and overqualification are separately related to the misaligned patterns, but we identified an alternative interaction explanation. For instance, in the introduction of this paper we argued that young employees who are overqualified may not have worse mental health because they may have jobs that do not stress them and that are viewed as stepping-stones to help them achieve higher goals (such as finishing college), all of which lead to the unsatisfied-healthy pattern. Aligned with this idea is that these employees also showed lower levels of job importance, thus confirming previous studies that showed that overqualified employees are more cynical about the meaningfulness of their job [51] and such reduce task importance, or significance, depends on how many other overqualified peers work in the same context [52]. We believe that low levels of importance to one's job may be, for young overqualified employees, a way to reduce the cognitive dissonance between their qualification and skill's usage and this may help in maintaining higher level of mental health. Therefore, future research should study the boundary conditions of the relationship between the antecedents here described (role conflict, role overload, and overqualification) and the misaligned wellbeing patterns.

Future studies might also investigate other potentially relevant antecedents, or moderators, of the mis/alignment between job satisfaction and mental health. In particular, following Danna and Griffin [4] and Nielsen et al. [53], it might be interesting to examine the sector of employment, in particular if it involves hazardous and stressful work settings (requiring for instance, night shifts or traveling), job resources as job autonomy, and also HR practices and social support. Literature has in fact showed that such factors may increase or decrease job satisfaction and wellbeing [4,53], but it should be examined if they differentially affect job satisfaction and mental health, also in relation to the family status of the employees (single, married, with children) and family history of mental health.

Finally, another major point concerns the sample being studied. The sample is composed of so-called millennials, thus, a very specific subgroup. These employees might differ in their general health (both mental and physical), wellbeing, satisfaction, etc., from other employees. As a group, millennials are in between twenties and late thirties, thus they do have a better physical health than older generations. Nevertheless, their lower tolerance to frustrations and their need to deal and face the economic crisis (initiated on 2008), which results in fewer career opportunities, may have as an effect a poorer level of mental health, especially in minor symptoms such as anxiety, life dissatisfaction, etc. Another specific situation of the millennials is that they face the transition from school to work in a situation that is in many cases not favorable. The support from their families, the resignation to have precarious/flexible jobs, to get incomes for subsistence or searching for jobs abroad, are some of the different ways of coping with the complex and difficult situation that millennials are facing during the actual economic crisis period.

#### **5. Limitations and Practical Implications**

One of the potential limitations of the current study is related to the sample, which is limited to young Spanish employees. To make better generalizations about the four wellbeing patterns, it is necessary to document their occurrence in other contexts. However, the sample was representative of all the regions of Spain, and the independent variables (role stress, job importance, and overqualification) that we tested in this research might be applicable to millennials in other contexts, which could make the generalization of these research findings more robust. Therefore, this limitation may be at least partially neutralized because the procedures and variables we used are applicable to millennials in other contexts. In addition, to test the external validity of the study, it would be useful to replicate it with millennials in other countries.

Another limitation is related to the measurement of job satisfaction and mental health. In job satisfaction measurement, cognitive/subjective biases may affect employees' evaluations of their satisfaction. Similarly, we assessed mental health by using the General Health Questionnaires (GHQ) in terms of a specific time period: '*during the past few weeks* ... '. However, events occurring "weeks ago" may be poorly recalled, and, therefore, induce some possible inaccuracy in mental processing [3]. However, the measurements of both job satisfaction and mental health are based on well validated and accepted instruments, and so we expect cognitive/personal biases and inaccuracy in mental processing to have little or no impact on the validity of the current research findings.

One of the main aims of organizational psychology is to improve employees' wellbeing. In this regard, our taxonomical approach provides relevant empirical evidence, facilitating the achievement of this endeavor. First, by combining job satisfaction and mental health, this study maps synergistic but also misaligned wellbeing patterns. Second, our study also provides valuable information of some personal and organizational variables related to them. In this way, our study informs organizational psychologists of when they may be improving at the same time job satisfaction and mental health, but also when this might not happen, creating misaligned wellbeing patterns instead. Thus, an important implication of our study is the provision of a useful wellbeing-pattern taxonomy from where to study and improve employees' wellbeing.

We also identify implications for other stakeholders. For instance, our results show that it would be worthwhile for organizations to find mechanisms to track and ensure the fulfillment of their commitments to millennials. Our results show, in fact, that only a small portion of employees are in the optimal job satisfaction and mental health category, whereas larger portions are in the unsatisfied-unhealthy and misaligned patterns. At the same time, organizations should carefully consider HR policies, such as staffing, to establish mechanisms to avoid phenomena such as role ambiguity, role conflict, role overload, and overqualification. These organizational and personal phenomena have been shown to have toxic effects on both job satisfaction and mental health.

Third, the results show that job importance is an important mechanism for a sustainable young workforce. In our study, young employees who reported having high job importance were characterized by being satisfied and healthy. Therefore, we argue that managers and employers should increase job importance by providing incentives related to job satisfaction and mental health. Often the jobs available for youngsters are "poor" overqualified and in some cases precarious. Thus, it is important that the companies enhance the meaning of work for youngsters offering jobs that are valuable and meaningful. This is the type of "incentives" that may make work more meaningful for youth and less toxic, dissatisfying and unhealthy.
