Corps de l’article

1. Introduction

To cope with the scarcity of talent, employers are seeking to reduce employee turnover, which costs them from 90% to 200% of the departing employee's annual salary. The costs are financial and human: expense of employee recruitment and selection; loss of knowledge and productivity; decrease in customer satisfaction; increase in stress and workload for remaining employees; and so on (Bolt et al., 2022; Allen et al., 2010). Voluntary departures also make the remaining employees more inclined to leave (Laulié & Morgeson, 2021).

HR professionals have adopted strategies to retain their top talent and reduce turnover (Singh & Ramdeo, 2023), and many researchers have investigated the determinants of employee turnover. The intention to quit leads to job search behaviours (Bolt et al., 2022; Klotz et al., 2021; Hom et al., 2012), which have attracted the attention of many researchers (Hom et al., 2012; Laulié & Morgeson, 2021; McKee-Ryan & Harvey, 2011; Rubenstein et al., 2018; Tepper et al., 2009). Turnover intention, like turnover itself, is a multifaceted psychological process with individual, organizational and environmental determinants (Hom et al., 2012; Klotz et al., 2021; Singh & Ramdeo, 2023).

Despite the talent scarcity, workforce overqualification remains a frequent problem in many countries (Russell et al., 2016; van Dijk et al., 2019). Overqualification is "a situation where the individual has surplus skills, knowledge, abilities, education, experience, and other qualifications not required by or utilized on the job" (Erdogan et al., 2011, p. 217). Overqualification can be real or perceived (e.g., Erdogan et al., 2011; Maltarich et al., 2011). Real overqualification occurs when employees have more actual qualifications (education, skills, knowledge, abilities or experience) than their jobs require (Erdogan et al., 2011; Maynard et al., 2006). This study focuses on subjective overqualification, i.e., an employee’s perception of having more qualifications than what is required to do the job (Erdogan & Bauer, 2021; McKee-Ryan & Harvey, 2011; Zhang et al., 2016). By comparing their abilities and experiences with the job requirements, employees can determine the degree of alignment between their abilities and the skills needed to perform the job adequately (Maynard & Parfyonova, 2013). The match between the job’s requirements and the employee’s abilities, skills and experiences will strongly affect the employee’s attitudes, workplace behaviours, work relationships and job performance (Erdogan & Bauer, 2021). For example, perceived overqualification has adverse personal and organizational outcomes, such as poor job attitudes (e.g., dissatisfaction, lack of commitment), psychological health problems, counterproductive work behaviours, voluntary turnover, interpersonal problems and lower socioeconomic status (Arvan et al., 2019; Erdogan & Bauer, 2009; Harari et al., 2017; Maynard et al., 2006; McKee-Ryan & Harvey, 2011). According to a meta-analysis, mainly of studies where the employees were working in their native countries, an employee’s perceived overqualification is related to poorer job attitudes and well-being and higher turnover intention (Harari et al., 2017). According to Erdogan et al. (2018), there is a need to learn more about how overqualified employees can improve their situation or how the employer can put their skills to better use (e.g., Erdogan et al., 2018). This need is especially great in the case of immigrant employees, i.e., people working temporarily or permanently in a country that is not the one of their birth (McAuliffe & Kharida, 2020). They may particularly have trouble finding a permanent position that matches their skills and experience (Marino et al., 2022; Kalantaryan et al., 2021; Wassermann & Hoppe, 2019).

Through this study, we wished to understand how to retain overqualified employees and reduce their intention to quit voluntarily. Specifically, we explored the 3-way relationship between work-life balance, overqualification and turnover intention. We took a subjectivist approach (Kalliath & Brough, 2008) toward perceived work-life balance, which is an employee’s overall evaluation of the interactions between work and non-work domains that are negotiated and shared with role partners (Kelliher et al., 2019). Moreover, we studied these interactions in a specific geographic context, in Canada, and in two different groups of employees, Canadian-born and immigrant, the latter facing more challenges to getting a job that matches their qualifications. Such challenges have various reasons: language barriers; discrimination; undervaluation of skills, diplomas, qualifications or certifications obtained abroad; lack of work experience; and no professional network in the host country (Chen et al., 2010; Groutsis et al., 2019).

Our study will help advance the relevant literature in several ways. First, we have used conservation of resources (COR) theory (Hobfoll, 1989) to explore the processes underlying employee turnover (Lee et al., 1999; Mobley et al., 1979). Second, we have responded to a call for analysis of turnover intention in heterogeneous contexts and among different populations (Singh & Ramdeo, 2023). As the workplace becomes more heterogeneous, it is necessary to consider how cultural diversity impacts experiences and outcomes of work-life balance (Kelliher et al., 2019). Third, in terms of methodology, we have innovated by conducting two independent studies in Canada: one among Canadian-born employees and the other among immigrant employees, who are more likely to experience overqualification (Maynard & Parfyonova, 2013) and work-life conflict (Haar et al., 2018). Finally, we have obtained results with practical, personal, organizational and societal implications. They may help transform overqualified employees, either local or immigrant ones, into valued assets who, when effectively utilized, can contribute to the competitiveness of firms and to the well-being of society.

We structure this article as follows. First, we will present the theoretical framework and propose research hypotheses. Next, we will describe the methods, measures, statistical procedures and results of the two studies. Finally, we will discuss our results in terms of their theoretical and managerial implications, their relevance and their limitations, and we will present suggestions for future research.

2. Theoretical framework and research hypotheses

2.1. Conservation of Resources Theory and Turnover Intention

Conservation of resources (COR) theory (Hobfoll, 1989, 2002) posits that human motivation and behaviours will preserve or conserve valued factors, otherwise known as individual, social, tangible and symbolic resources, which may be "anything perceived by the individual to help attain his or her goals" (Halbesleben et al., 2014, p. 1338). Its central tenet is that people strive to acquire, maintain, protect and build resources to achieve their goals (Hobfoll, 1989). Resources are what people value, such as objects, conditions, personal characteristics, energies and social support. They can be tangible or intangible, self-generated or derived externally and existing to different degrees across work-life domains (Halbesleben et al., 2014). Resources provide individuals with instrumental value, such as owning a house or a car, being employed, having skills or experiences and possessing mental or physical energy.

An employee may suffer a real or perceived loss of resources in the work environment for various reasons: high workload; lack of autonomy; limited opportunities for advancement; monotony in work processes; insufficient support from the supervisor; and non-alignment between skills and tasks or responsibilities. If you are placed in such a situation, you may consider leaving to protect or conserve your resources by seeking alternative job opportunities that offer better resource availability, such as improved work conditions, higher pay or positions or jobs better suited to your skills or experience. Importantly, it is easier to develop and adapt to threats to your resources if you have more of them than if you have fewer. Having more resources means being better able to solve problems and less likely to have your well-being negatively affected by depletion of resources during stressful situations. This reality leads to certain principles and corollaries (Halbesleben et al., 2014; Hobfoll, 2002; Hobfoll et al., 2018).

2.2. The Relationship between Overqualification and Turnover Intention

According to COR theory, the more you believe you are overqualified, the more likely you will consider leaving your job (Hobfoll, 1989; 2002). The theory posits that people strive to acquire, maintain, protect and build resources. Resources might be anything you perceive as helping attain your goals (Halbesleben et al., 2014).

Consequently, overqualified employees will look for another employer with whom to maintain, protect and use their resources (qualifications) and reduce the negative impacts from the stress of not entirely using their skills. Many researchers have confirmed the negative relationship between perceived overqualification and turnover intention, albeit mostly among native-born employees (see the meta-analysis by Harari et al., 2017). This relationship may hold for all employees regardless of their specific conditions or contexts. Based on COR theory and the existing literature, we propose that Canadian-born and immigrant employees show a positive relationship between perceived overqualification and turnover intention.

  • Hypothesis 1.There is a positive relationship between overqualification and turnover intention among employees.

2.3. Moderating Impact of Work-Life Balance on the Relationship between Overqualification and Turnover Intention among Employees

Another key postulate of COR theory (Hobfoll, 2002, 2011), called resource scarcity, posits that, as employees have a limited supply of time, energy and attention, they may deplete their resources if they fail to perform their various roles. For instance, in the course of managing their work-life domains or inter-role conflicts, they will put pressure on their own resources, thus increasing their stress and their desire to leave their job. There is, however, another employee resource: work-life balance. In their review, Sirgy and Lee (2017) define work-life balance as "a high level of engagement in work life as well as non-work life with minimal conflict between social roles in work life and non-work life."

Consequently, an employee’s perceived work-life balance acts as a "coping resource," in line with another key postulate of COR theory: resources can generate new ones. Hobfoll (2002) speaks of "resource caravans," in which resources, once obtained, will accumulate. Specifically, when employees perceive they are overqualified, their perceived work-life balance (as a contextual or personal resource) may reduce their turnover intention.

Work-life balance has been studied mainly among native-born employees. In a meta-analysis, Rubenstein et al. (2018) confirmed that work-life balance is negatively related to turnover intention. Since resources are anything people perceive as helping them attain their goals, we can expect that overqualified Canadian-born employees know the resources (including social support) they might obtain in their community or their firm to reduce their work-life conflicts. In the context of the relationship between overqualification and turnover intention, we consider that work-life balance can determine the decision to stay or leave. A good work-life balance may deter departure despite strong intentions to leave. An employee may feel obliged to maintain a certain level of financial stability (Kracke & Klug, 2021; Wassermann & Hoppe, 2019). An employee may lack alternative opportunities that offer a better work-life balance (Renaud et al., 2021ab). Finally, an employee may fear not finding a similar job elsewhere that provides a better work-life balance, especially if the job market is competitive.

Nonetheless, even with a satisfactory work-life balance, an employee may feel dissatisfied because of monotonous tasks, lack of recognition or limited growth opportunities. These factors can still affect the commitment to a job and the intention to leave. A good work-life balance can only mask a lack of professional fulfillment. An employee may still feel that the job is lacking in challenge, intellectual stimulation or meaning. There may be career stagnation due to a lack of opportunities for promotion or advancement. Lack of prospects for professional growth can thus lead to an intention to leave, whatever the work-life balance.

Work-life balance has attracted scant attention from scholars in the case of immigrant employees. For immigrants, it is especially challenging because the move to another country affects not only the immigrant employee but also other family members. All of them have to adjust to a new institutional context with a different climate and language, education, daycare and healthcare systems and work-life conditions (Kracke & Klug, 2021; St-Onge et al., 2021 Santero & Naldini, 2020; Wassermann & Hoppe, 2019; Ballesteros-Leiva et al., 2017). Scholars emphasize the role of adjustments by spouses or children in the success, attitudes and behaviour of immigrants or expatriates at work (Ballesteros-Leiva et al., 2017; Harrison et al., 2018). Family members and other resources may help immigrants deal with the difficulties they and their close relatives experience personally and professionally. Based on our review of the literature, we might expect that perceived work-life balance, whether for immigrants or for Canadian-born employees, moderates the relationship between perceived overqualification and turnover intention. Figure 1 illustrates our research model.

  • Hypothesis 2. Work-life balance has a moderating effect on the relationship between overqualification and turnover intention.

INSERT FIGURE 1

3. Research method

We tested the above two hypotheses by conducting two independent studies: first, among Canadian-born employees; and, second, among immigrant employees in Canada. Our questionnaire was developed on Qualtrics software and made available in French and English, and the Likert scales were translated and back-translated by native English-French speakers (Brislin, 1970).

3.1. Variable Measurement

We utilized 7-point Likert scales, ranging from 1 (strongly disagree) to 7 (strongly agree), to evaluate all primary variables. We measured turnover intention on a scale from Tepper et al. (2009), which included an item such as "I plan on leaving this organization very soon." For this scale, Cronbach’s alpha is 0.97 among Canadian-born employees (Study 1) and 0.92 among immigrant employees (Study 2).

We measured perceived overqualification on a 9-item scale from Maynard et al. (2006). One item was: "My job requires less education than I have." For this scale, Cronbach's alpha was 0.92 among Canadian-born employees (Study 1) and 0.93 among immigrant employees (Study 2).

We measured work-life balance on a 6-item scale from Grzywacz and Carlson (2007). One item was: "I am able to negotiate and accomplish what is expected of me at work and in my family." For this scale, Cronbach's alpha was 0.92 among Canadian-born employees (Study 1) and 0.91 among immigrant employees (Study 2).

To control for factors that may influence turnover intention, we measured demographic variables such as age, gender, education and number of working hours per week. We also controlled for employability, given its link with turnover intention (Forrier et al., 2018; De Cuyper et al., 2011). When employees realize that they possess the skills and qualifications in demand on the job market, they may feel more confident about leaving their current position and looking for new opportunities elsewhere. Perceived employability was measured on the 4-item scale developed by De Witte (1992), which has good internal consistency. Cronbach's alpha was 0.89 among Canadian-born employees (Study 1) and 0.90 among immigrant employees (Study 2). An example is: "I am optimistic that I would find another job if I looked for one."

Finally, we measured the employee development climate within a company as a control variable, since it has been reported as an important determinant of turnover intention (Spell et al., 2014). A supportive climate for employee development offers opportunities for skills enhancement, internal promotion and professional development, which can strengthen job satisfaction and loyalty to the organization. When employees feel they have such a supportive climate, they may be more motivated and committed to their work (Bal et al., 2012), and their turnover intention will be correspondingly reduced. We measured this variable by using six items from Kooij (2010). One item was: "In our organization, workers are given training and encouraged to learn new things." The scale has good internal reliability. Cronbach's alpha was 0.92 among Canadian-born employees (Study 1) and 0.90 among immigrant employees (Study 2).

3.2. Data Analysis

We tested the first hypothesis by using structural equation modeling (SEM) with Mplus Version 8.3, which estimates and analyzes the relationships between variables in a complex system (Muthén & Muthén, 2012). This approach establishes the instruments’ factor structure among participant samples and avoids confounding associations among the variables due to measurement error. It is a multivariate statistical method that combines factor analysis and multiple regression analysis to examine the measurement and structural models within a single framework. We used robust maximum likelihood parameter estimation to make the statistical tests robust to non-normality. To test the interaction effects of work-family balance on the relationship between overqualification and turnover intention, we used the PROCESS macro (Stride et al., 2015, http://offbeat.group.shef.ac.uk/FIO/model1alatent.htm). To determine the research model fit, we looked at the chi-square value, the Root Mean Square Error of Approximation (RMSEA), the Standardized Root Mean Square Residual (SRMR), the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI). RMSEA and SRMR values between 0.05 and 0.08 indicate a good data fit (Hu & Bentler, 1998). CFI and TLI indices, whose values should be close to 0.90 (Bentler, 1983), were used to compare different models.

3.3. Study 1 among Canadian-Born Employees

3.3.1. Data Collection

We collected data through an online survey in 2018 among Quebec members of the order of Chartered Professionals in Human Resources (CPHR). In Quebec, HR experts can join this professional order by meeting the required competency requirements and devoting time to remain compliant with them. Only full-time human resources professionals were eligible for the survey, and we invited those who did not meet this criterion to forward the questionnaire to anyone they knew who did. We contacted potential participants via emails that included a link to the survey, and we sent reminder requests two weeks later. The sample consisted of 227 participants, 140 of whom were women (81%). The mean age was 33.2 years (SD = 9.2).

Regarding their positions, 55% had no supervisory responsibility, 27% were middle managers, 7% were senior managers or directors, 6% were first-line supervisors, 5% were technicians and 1% were administrative support staff. Almost all (98%) had a university degree (undergraduate or graduate). Nearly half (43%) were married, 27% lived in a common-law relationship and 28% were single or divorced.

3.3.2. Descriptive Statistics and Correlations among Variables

Table 1 shows that all internal consistencies were above 0.70. Among Canadian-born employees, perceived overqualification related positively to turnover intention. Perceived employability was related positively to turnover intention, perceived overqualification, work-life balance and perceived climate for employee development within the organization.

INSERT TABLE 1

3.3.3. Measurement Structure and Testing of Structural Relationship

We performed a confirmatory factor analysis to confirm the properties of the proposed model, including its three latent variables (Anderson & Gerbing, 1988)d. The proposed three-factor model provided a satisfactory overall fit (χ2 (242) = 422.1, p<0.01, CFI = 0.95, TLI = 0.94, RMSEA = 0.06, SRMR =0.05). All indicators loaded significantly on their latent variables. A one-factor model provided a poor fit (χ2 (252) = 2560,5, p<0.01, CFI = 0.34, TLI = 0.28, RMSEA =0.23, SRMR =0.20) with a significant deterioration in chi-square relative to the four-factor model (Δχ2 (10) = 2138.4, p<0.01). Hence, the proposed three-factor model appeared to be the best-fitting one.

3.3.4. Results

The results support Hypothesis 1 for Canadian-born employees (Table 2, Panel 1). Turnover intention was significantly increased by perceived overqualification (b= 0.32, t= 3.3 p < 0.00). On the other hand, it was significantly reduced by perceived climate for employee development (one of our control variables) (b= -0.41, t= -5.6, p < 0.00). Thus, among Canadian-born employees the more they were overqualified, the more they perceived a good climate for employee development, and the less likely they would consider leaving. The results do not support Hypothesis 2 for Canadian-born employees (Table 2, Panel 2). The perceived work-life balance did not significantly reduce the effect of perceived overqualification on turnover intention (interaction effect is 0.03 with 95% BC confidence intervals for the association being between -0.111 and 0.176, p < 0.708).

INSERT TABLE 2

3.4. Study 2 among Immigrant Employees

3.4.1. Data Collection

We recruited the immigrant participants in 2018 through the online social networks Facebook and LinkedIn, using snowball sampling (Goodman, 1961). We had four criteria: 1) being an immigrant in Canada; 2) being over 18 years of age; 3) working 30 hours or more per week; and 4) living in Canada. The sample included 237 immigrant employees, 126 (53%) of whom were female, and the mean age was 37.7 (SD= 7.9). On average, they had lived in Canada for 5.3 years (SD = 3.4 years), and 70% lived in Montreal, Quebec. In terms of work life, nearly 3.4% worked less than 25 hours per week, 74.9% between 25 and 40 hours a week and 21.7% more than 40 hours per week. In terms of family life, 62.1% had children, 21% lived with a spouse/partner without children and 17% were single. In terms of education, nearly 85% had an undergraduate or graduate degree. In terms of job position, 49% were professionals without supervisory responsibility, 22% middle managers, 8% administrative support staff, 7% technicians, 5% hourly-paid workers, 4% first-line supervisors and 4% senior managers or directors. On average, they had held their job position for 8.1 years with a standard deviation of 7 years. They came from more than 40 countries and were of different nationalities: French (24%), Colombian (11%), Brazilian (8%), Mexican (4%), Algerian and Belgian (4% each), Peruvian (3%), Venezuelan and Spanish (3%) and others (38%).

3.4.2. Descriptive Statistics and Correlations among Variables

In Table 3, all correlation coefficients were below 0.70, thus reducing concerns associated with multicollinearity (Tabachnick et al., 2007). Among immigrant employees, perceived overqualification was related positively to turnover intention. Their length of residence in Canada appeared to be related negatively to perceived overqualification and turnover intention.

INSERT TABLE 3

3.4.3. Measurement Structure and Testing of the Structural Relationship

We tested the measurement structure and the structural relationships by performing a confirmatory factor analysis to confirm the psychometric properties of the measurement model, including its four latent variables. The proposed four-factor model provided a satisfactory overall fit (χ2 (242) = 605.7, p<0.01, CFI = 0.93, TLI = 0.92, RMSEA = 0.08, SRMR =0.05). All indicators loaded significantly on their latent variables. A one-factor model provided a poor fit (χ2 (252) = 2987.3, p<0.01, CFI = 0.40, TLI = 0.35, RMSEA =0.21, SRMR =0.17) with a significant deterioration in chi-square relative to the four-factor model (Δχ2 (10) = 2381, p<0.01). The SEM provided a good fit to the data (χ2 = 639.5, df = 289, p = 0.000; CFI = 0.92; TLI = 0.91; RMSEA = 0.07; SRMR = 0.06).

3.4.4. Results

The results support Hypothesis 1 for immigrant employees (Table 4, Panel 1). Turnover intention was significantly increased by perceived overqualification (b= 0.21, t= 2.8 p < 0.00). Two of our control variables, climate for employee development and age, had a negative relationship with immigrant employees’turnover intention (b= -0.43 t= -6.2, p < 0.00; b= -0.24 t= -4.3, p < 0.00).

INSERT TABLE 4

The results also support Hypothesis 2 for immigrant employees (Table 4, Panel 2). Perceived work-life balance moderated the relationship between perceived overqualification and turnover intention (Estimate= -0.12, p < 0.05, BC intervals [-0.681 – -0.354]). Thus, if work-life balance was satisfactory, perceived overqualification had less impact on turnover intention (Figure 2).

INSERT FIGURE 2

4. Discussion and conclusion

Using COR theory and the existing literature, we studied two samples of employees in Canada: Canadian-born employees and immigrant employees, specifically the relationship between overqualification and turnover intention and the moderating effect of work-life balance. The results of both studies confirm a positive relationship between perceived overqualification and turnover intention (Hypothesis 1): the more the Canadian-born or immigrant employees perceived their skills as under-utilized, the more they intended to leave their organizations. This result is congruent with COR principles and existing research, which posit that the risk of losing resources may prompt overqualified employees to seek another employer to maintain, protect and use their resources (skills) and to reduce the negative impact of the stress of not fully using their skills (e.g., Erdogan & Bauer, 2009; Maynard & Parfyonova, 2013).

Hypothesis 2 posits that perceived work-life balance moderates the relationship between perceived overqualification and turnover intention: the better the perceived work-life balance, the weaker the relationship between perceived overqualification and turnover intention. This hypothesis is not supported by the study of Canadian-born employees but is supported by the study of immigrant employees. In other words, perceived work-life balance did not significantly reduce the effect of overqualification on turnover intention among Canadian-born employees. Regardless of perceived work-life balance (a personal factor), Canadian-born employees were more likely to express wanting to leave or look for another job that better uses their qualifications and experience (a job factor). Among immigrant employees, however, perceived work-life balance significantly moderated the relationship between perceived overqualification and turnover intention. Therefore, holding down a job that did not require all their skills may have had less effect on their turnover intention if they believed they had achieved a satisfactory work-life balance.

Why did perceived work-family balance reduce the effect of perceived overqualification on turnover intention among immigrants but not among Canadian-born employees? COR theory posits that "those with greater resources are less vulnerable to resource loss and more capable of orchestrating resource gain. Conversely, those with fewer resources are more vulnerable to resource loss and less capable of resource gain" (Hobfoll, 2011, p. 349). Mobley et al. (1979) also argues that an employee’s decision to leave will depend, among other things, on opportunities for alternative employment. Canadian-born employees are more likely than immigrants to have better resources, such as knowledge of the local job market, contacts, Canadian degrees and other advantages. A better work-life balance will be a weaker deterrent against quitting because they tend to have more opportunities for attractive employment in the job market and have higher expectations of fully using their skills elsewhere. By comparison, immigrant employees have fewer contacts and experience in their host country. Immigrant employees have a less extensive network of professional contacts and are, therefore, less in touch with and have less access to other employment opportunities on the market than do local employees (Marino et al., 2022; Porter et al., 2016). Immigrants might not rapidly find another place to work or might have fewer opportunities to get a better job elsewhere, even if they express a desire to quit. Immigrants weigh their potential losses in terms of supportive social connections (Kracke & Klug, 2021; Harrison et al., 2018) before they express an intention to quit if they are overqualified. Also, a better work-life balance will deter them from leaving because they feel they have fewer opportunities for attractive employment. Our results are consistent with existing findings that overqualified immigrant employees are less likely to say they plan to quit because their working visa is temporary or valid for a specific employer or because of other administrative rules related to their immigrant status (Chen et al., 2010; Wassermann et al., 2017). Immigrant employees might feel forced to please their employers to keep their jobs and remain in their host country (Groutsis et al., 2019; Harrison et al., 2018). They can better enjoy a job and an employer that allows them to balance work and family better. Finally, overqualification might not have the same meaning for immigrants as it does for local employees, since a large proportion of them have degrees, are highly overqualified, work in lowly positions and cannot use skills that are not recognized in Canada. Consequently, they are more likely to exchange a better professional life in their country of origin for a worse one of underemployment in a new country. With no real hope of having their qualifications recognized in another job in Canada, they are more inclined to consider their ability to balance their spheres of life, which is often more challenging for them as they tend to work in jobs with more difficult working conditions in terms of schedule and workload.

4.1. Practical Implications

Our results lead to several valuable practical implications. First, at a time of talent scarcity, it seems crucial to pay attention to how employees, whether Canadian-born or immigrant, perceive their job skills because that perception will significantly impact their turnover intention. Consequently, overqualified employees should have opportunities to engage in career development activities, such as job crafting, informal leadership and mentoring. They will make unique and valued job contributions when capitalizing on these opportunities and their additional resources or skills. In particular, they should have their skills put to best use through increased job responsibilities, through additional duties, which may or may not be associated with short- or medium-term rewards (e.g., bonuses, promotion), through diversified job involvement, through roles as coach or trainer with new employees, through participation in specific internal or external work committees, through reorganizing or restructuring of their work, and so on.

In addition, employers should take into consideration the specificities of overqualified immigrants. It is even more critical for employers to help immigrant employees achieve a satisfactory work-life balance. Overqualified immigrants should receive information, emotional support, tools and skill appraisal on an ongoing basis to help them manage their daily lives, balance work-life interactions and, ultimately, reduce their intent to quit. Examples could include policies and practices to support work-life balance through various activities (e.g., information on the healthcare system, medical referral information, useful websites, how to obtain a driver's licence), contact support, financial support, support to family members, including spouses (especially for job search or career counseling), children, elderly parents and even pets. For new immigrant employees, employers could consider posting immigrant stories and experiences, useful/practical information, ethical practices, and FAQs on the company website.

4.2. Limitations and Avenues for Future Research

Our two studies have limitations. Both were conducted with small and unequal samples of employees who differed in national origin, in job position, in personal profile and in organizational and environmental context. None of this could be controlled. This absence of control for both samples might not prevent certain factors (other than immigrant status, work-life balance and overqualification) from explaining some of our results. Another limitation is the lack of response rates for both samples, which may introduce uncertainties and biases that impact our interpretation of the findings. Since these limitations affect the generalizability of the findings, future researchers should replicate this study with other samples and use alternative research designs and measurements of the variables.

Another possible limitation is that we may have excluded certain immigrant profiles from other countries by collecting the data via social media. Social media use and presence may vary from culture to culture, thus significantly influencing the representation of certain immigrant groups in our sample. When using social media to recruit participants, future researchers should consider how the presence and patterns of social media use may differ across different cultures.

Our studies were cross-sectional, although only a longitudinal research design with data from multiple sources could help demonstrate the causal relationships between the variables. Qualitative studies using interviews with immigrant and Canadian-born employees could better reveal how immigrant employees differ from Canadian-born employees in terms of the determinants and impacts of overqualification. Future studies could combine subjective and objective measurements of overqualification to distinguish between actual voluntary departure and the desire to leave. Some employees might believe themselves overqualified despite factual evidence of being well-matched to their job (e.g., Arvan et al., 2019; Maltarich et al., 2011). We focused on turnover intention, the most important and frequently used determinant of voluntary turnover (McKee-Ryan & Harvey, 2011; Tepper et al., 2009). Meta-analyses, however, show that turnover intention explains only 15-20% of the variance in turnover behaviour (Griffeth et al., 2000). Finally, in light of COR theory, researchers ought to investigate how various contextual and personal resources can optimize the contributions of overqualified employees, be they immigrant or Canadian-born (e.g., Erdogan et al., 2011; Russell et al., 2016).

In conclusion, our results confirm the value of COR theory in understanding the relationship between overqualification and turnover intention among Canadian-born and immigrant employees and how work-life balance moderates this relationship among immigrants. Given the current talent shortage, employees, whether Canadian-born or immigrant, should have their skills better recognized and put to better use. In particular, immigrant employees can be retained by helping them better balance their work and personal lives.