Start Submission

Reading: Who Gets Stuck in Their Workplaces? The Role of Matching Factors, between Individual and Job...


A- A+
Alt. Display

Original Article

Who Gets Stuck in Their Workplaces? The Role of Matching Factors, between Individual and Job, and Demographics in Predicting Being Locked In


Johanna Stengård ,

Department of Psychology, Stockholm University; Stress Research Institute, Stockholm University, SE
X close

Erik Berntson,

Department of Psychology, Stockholm University, SE
X close

Constanze Leineweber,

Stress Research Institute, Stockholm University, SE
X close

Claudia Bernhard-Oettel

Department of Psychology, Stockholm University, SE
X close


When a workplace/organization does not fulfill one’s needs and wishes anymore, many individuals change to other workplaces/organizations. However, for some individuals this is not feasible as they perceive a lack of alternatives; they feel stuck in a non-preferred workplace (being locked in), or they may be in the risk zone of becoming locked in. Few studies have investigated the reasons for becoming locked in, and it is the aim of this study to investigate whether matching factors between work and individual and/or demographic factors can predict locked-in positions. Multinomial logistic regression analyses were performed—cross-sectionally and longitudinally (N = 3633–6449)—and showed that mismatch in terms of over-qualification and lack of physical and mental work abilities increased the odds ratios for being in locked-in positions. In contrast, working in relatively higher socioeconomic categories of both manual and non-manual work, commonly demanding higher education (vocational or academic), protected against being locked in. This study contributes to the career research field by studying determinants of disadvantageous career positions, which have been neglected in past research.
How to Cite: Stengård, J., Berntson, E., Leineweber, C. and Bernhard-Oettel, C., 2019. Who Gets Stuck in Their Workplaces? The Role of Matching Factors, between Individual and Job, and Demographics in Predicting Being Locked In. Scandinavian Journal of Work and Organizational Psychology, 4(1), p.8. DOI:
  Published on 11 Oct 2019
 Accepted on 03 Jul 2019            Submitted on 25 Apr 2018


In the career research field, attention has typically been devoted to job transitions that either concern voluntary transitions between jobs or involuntary job losses (Fouad and Bynner, 2008; Greenhaus and Callanan, 2012). In terms of circumstances that are about remaining in a workplace/organization, it is often implicit that individuals themselves are responsible for such career decisions (Feldman and Ng, 2012) and hence, would leave a workplace when the job is no longer perceived as fulfilling (Fouad and Bynner, 2008). However, not everybody feels that they have control over where to work and can change workplaces whenever they want to. Such a situation, where individuals feel stuck in a workplace they no longer want to work in has been termed ‘being locked in’ (Aronsson, Dallner, and Gustafsson, 2000; Aronsson and Göransson, 1999). Such a position has been found to be associated with poor health (e.g., physical complaints, depressive symptoms) both cross-sectionally (Aronsson et al., 2000; Aronsson and Göransson, 1999; Fahlén et al., 2009; Furåker, 2010) and over time (Canivet et al., 2017; Stengård et al., 2016). However, there are very few studies about being locked in, and those that have been conducted differ in their ways of conceptualizing the phenomenon. Some of these studies build on non-preference toward one’s job, in terms of not wanting to remain in the current workplace in the future (Aronsson et al., 2000; Aronsson and Göransson, 1999; Muhonen, 2010). Other studies have used low employability perceptions (Furåker, 2010; Furåker, Nergaard, and Saloniemi, 2014), in terms of lacking opportunities to find another job in the labor market (Berntson, Näswall, and Sverke, 2008; De Cuyper et al., 2012). However, some recent studies have utilized a combination of those earlier conceptualizations (Fahlén et al., 2009; Stengård et al., 2016), and thus, being locked in is referred to as a combination of: 1) non-preference for the current workplace, and 2) low perceived employability. In the present study, this more recent conceptualization will be used.

An important question is whether there are some factors, related to the individual, that characterize those who are locked in, and whether such factors also can increase the odds for becoming locked in over time. There has been little research (for exceptions, see Bernhard-Oettel et al., 2018; Fahlén et al., 2009) on the potential role of work conditions and the fit of the work conditions to the individual, in relation to being locked in. Also, knowledge is scarce on whether there are demographic variables, such as gender, age, socioeconomic position, living area or family situation (married, having children) that put some individuals at greater risk of becoming locked in.

The aim of this study is twofold: first, to investigate if matching factors are associated with locked in status cross-sectionally, and with locked-in status changes over time. The second aim is to investigate whether relevant demographic variables is related to locked-in status.

Person–environment fit

Being locked in means that the individual is dissatisfied with and wants to leave the current workplace, which suggests that the phenomenon could be understood in light of the person–environment (PE) fit framework. Being locked in thus would imply an unfavorable match between the individual and his/her workplace/organization in a wider sense (as the workplace/organization is regarded as a non-preferred one). Such a misfit risks persisting because the individual, due to low perceived employability, perceives little chance to escape by leaving the workplace for another job.

According to the PE fit framework (Edwards, 1991; Kristof, 1996) the match between the person and the environment could (besides the needs-supplies dimension) be described along the dimension of demands–abilities fit (Edwards et al., 2006; Kristof, 1996). PE fit could as such be viewed on several levels: vocation (PV fit), organization (PO fit), job (PJ fit), and group (PG fit). In regard to PJ fit, ‘the match between the requirements of the job and the knowledge, skills, and abilities (KSAs) of the employee’ (Maynard, Joseph, and Maynard, 2006: 510) is targeted. In sum, what the job requires of the employee should correspond to what the individual feels he/she can offer the organization. If this mutual exchange is imbalanced, there is a misfit between the two, which according to two meta-studies has negative implications for job satisfaction and organizational commitment and also increases turnover intentions (Kristof-Brown, Zimmerman, and Johnson, 2005; Oh et al., 2014).

Since being locked in can be regarded as a prevailing general misfit between the individual and his/her workplace/organization (PE fit)—and all levels of fit (PV, PO, PJ, PG fit) have implications for the overall experienced PE fit (Jansen and Kristof-Brown, 2006)—it would be of value to examine whether some particular matching factors targeting different aspects of (mis)fit at the workplace are related to being locked in, and if those aspects can develop locked-in feelings over time.

Two earlier studies have conducted research in this area and found that mismatch between (quantitative) demand and control as well as effort/reward imbalance have associations with being locked in (Bernhard-Oettel et al., 2018; Fahlén et al., 2009). These studies focused on perceptions of work environment factors such as low control, (too) high demands or imbalance between effort and rewards, and showed that individuals with unfavorable reports on these work environment factors were at greater risk of being or becoming locked-in. Even though misfit between individual and job may explain such findings, neither of these two studies directly addressed the question of how perceived mismatches between individual and job, for example a misfit in regard to qualifications or physical/mental work abilities, relate to being locked in. Moreover, it is not known if such misfit at one point in time may have a scarring effect on the individual, meaning that a disproportion may leave a scar influencing the development of locked-in feelings later on even if this misfit no longer prevails.

Being locked in or in the risk zone for becoming locked in

In addition to being locked in, employees that could be considered as being in the risk zone for becoming locked in at their workplace are of interest (Stengård et al., 2016). Such employees are not yet locked in as they are satisfied with their current workplace at the moment, but they nevertheless risk becoming locked in: they want to change their workplace in the future, but perceive themselves as having low employability. There are different reasons for individuals accepting a job they do not want to remain in long-term. For instance, some individuals might take on a job opportunity as a stepping-stone (de Jong et al., 2009) in order to gain experience that could bring about future advantages and more attractive and secure job opportunities. However, regardless of the reason, if an individual perceives that his/her employability is low and does not improve while being in the job, there is a risk that the individual later on may experience him/herself as being locked in.

One study has indicated that individuals at risk of becoming locked in report poorer health (in terms of depressive symptoms and self-reported health) over time (Stengård et al., 2016), although not as poor as those who are actually locked in. This suggests that individuals in the risk zone—even though satisfied with their work arrangements at that moment—are affected by the knowledge of possible future PE misfit, and this seems to influence how they perceive their present situation (Shipp and Jansen, 2011). Thus, they might start to ruminate about their future careers as an effect of their low employability perceptions (Stengård et al., 2016).

Therefore, we distinguish three categories (locked-in status) in this study, namely being locked in, at risk of becoming locked in, and not being locked in.

Matching factors as predictors of locked-in status

Work environment studies have typically focused on the role of various job demands and resources (Bakker and Demerouti, 2007). However, with regard to the locked-in phenomenon, we argue that it is more important to understand how these conditions fit with a person’s own qualifications and abilities—as is consistent with the PE fit theory. In support of this, Jansen and Kristof-Brown (2006) have suggested that the fit between the employees and the working environment on different levels (e.g., vocation, organization, and job) should affect the overall experienced fit, related to the individuals’ preferences for their workplace/organization. One such aspect of fit (on the job level) is whether the person perceives him/herself as having the knowledge/skills needed to match his/her work tasks (Peiró, Sora, and Caballer, 2012; Wittekind, Raeder, and Grote, 2010). A mismatch in this aspect could either reflect the perception of having higher knowledge/skills than the job requires—over-qualification1—or may reflect the perception of lacking knowledge/skills for performing one’s work tasks—under-qualification. Furthermore, as it is well-established that having possibilities to utilize one’s skills at work is beneficial for job satisfaction (cf. skill variety of the job characteristics model [Hackman and Oldham, 1976]), it is also likely that over-qualification may be of importance with regard to locked-in positions. Supporting this notion, individuals who perceive themselves as being overqualified for their jobs have been found to often feel dissatisfied with their current work situation (Johnson and Johnson, 2000; Lobene, Meade, and Pond, 2015). Also, it has been indicated that they are less motivated to perform well (Bolino and Feldman, 2000; Maynard et al., 2006) and they have higher turnover intentions (Maynard et al., 2006). Moreover, when individuals are not utilizing their skills fully, they may, as a consequence, worry about losing their attractiveness on the labor market (Bolino and Feldman, 2000). In fact, it has been indicated that employers are often reluctant to hire individuals they feel are overqualified for the job (Erdogan et al., 2011). For instance, one field experiment showed that being overqualified (having one year of employment clearly below one’s skill and experience in the CV) had a similar scarring effect as one year of unemployment, when applying for a new job (of appropriate level) (Pedulla, 2016).

Furthermore, it has been indicated that overqualified employees to a lower degree develop career-enhancing strategies (Peiró et al., 2012), and consequently, the prospects for transferring to other jobs might decrease even further. Hence, a consequence may be that such individuals might experience/or develop locked-in perceptions. In addition, according to a German study among unemployed people re-entering the labor market, only a minority of those who took a job for which they were overeducated and thus overqualified for was transferred to appropriate employment within the next five years—hence they risk becoming trapped in a job below their qualifications (Voßemer and Schuck, 2016).

The other related scenario of misfit with regard to qualifications—under-qualification—may also be significant with respect to locked-in positions. In this scenario, an underqualified individual may not feel able to meet the work demands put on them, which thus may result in low preferences to remain in the job. Also, opportunities to change to a similar or better job position elsewhere may be limited, especially as lacking knowledge/skills for a position is often combined with the absence of the corresponding appropriate, formal education level (Thompson et al., 2015). On the other hand, lacking skills for performing one’s work tasks may feel inspiring for some individuals as it may be regarded as a chance for development. Empiric evidence is scarce with regard to under-qualification, but one study showed that under-qualification was related to lower perceived employability (Wittekind et al., 2010) and another study found that underqualified individuals received fewer hiring recommendations than overqualified individuals did (Thompson et al., 2015).

Taken together, being overqualified or underqualified might be related to locked-in status. Thus, our first hypotheses are:

Hypothesis 1: Over-qualification is positively related to locked-in positions a) cross-sectionally and b) over time.

Hypothesis 2: Under-qualification is positively related to locked-in positions a) cross-sectionally and b) over time.

As the level of mental and physical work abilities the worker possesses can be considered part of the individual’s KSAs (Ployhart and Bliese, 2006), we argue—in accordance with other researchers (see e.g., Oakman and Wells, 2016)—that also lack of mental or physical work ability could be viewed in light of the PE fit framework. Work ability fit would thus pertain to PJ fit between work demands and the person’s work abilities. Deterioration, especially in physical work ability, is a natural effect of aging (Ilmarinen, 2001), but work ability is also affected by things such as lifestyle factors, and (physical and mental) work demands (van den Berg et al., 2009). In the modern labor market, the mental pressure on individuals may increase due to, for instance, rapid technical advancements, implying that individuals constantly have to adapt to new requirements and learn new ways of working (Tuomi et al., 2001).

Shortage of physical or mental work ability implies that one’s resources/energy are depleted, which should result in strain (Edwards, 1996). A consequence might be a negative impact on job satisfaction as well as on the workplace preferences. There may also be a negative impact on employability if the individual perceives that due to the lack of work ability, other job alternatives he/she could attain and manage are scarce. Supporting this notion, there is some evidence indicating that advantageous work ability is associated with enjoyment of staying in one’s work (Tuomi et al., 2001) and with perceived employability (Nilsson and Ekberg, 2013). Thus, lack of work ability—physical or mental—might tie the person to the current workplace and/or might result in marginalization at the workplace, which over time may result in a locked-in position. Hence, our next hypotheses are:

Hypothesis 3: Physical work ability is positively related to locked-in positions a) cross-sectionally and b) over time.

Hypothesis 4: Mental work ability is positively related to locked-in positions a) cross-sectionally and b) over time.

Associations between demographics and locked-in status

A second aim of the present study is to investigate whether demographic variables (i.e., gender, age, socioeconomic position, place of residence, marital status, and parental status) are related to locked-in status. Demographic factors could be argued to reflect circumstances that might hinder or facilitate mobility in the labor market and the possibility to choose workplaces in accordance with preferences. A first important aspect to discuss is gender, and in Sweden, as in many other countries, the labor market is rather gender-segregated, both vertically and horizontally, which means that men and women on a structural level are exposed to different work environments and working conditions (OECD, 2014; Sverke et al., 2016). Hence, one could expect that there might be a difference in the prevalence of locked-in positions between men and women. However, the empirical evidence that possibly could shed some light onto this issue is mixed. For workplace non-preference, Aronsson et al., (2000) found no gender differences, while another study that investigated locked-in status indicated that men more often were locked in than women (Fahlén et al., 2009). However, the latter study targeted a specific, female-dominated workplace, and therefore it is uncertain whether this relationship would be reflected in the labor market overall. Regarding studies targeting gender and employability, some studies did not find any gender differences (Berntson, Sverke, and Marklund, 2006; Silla et al., 2009), while the bulk of studies indicate that men, compared to women, generally perceive themselves as having higher employability (see e.g., De Cuyper et al., 2008; Mäkikangas et al., 2013; Vanhercke et al., 2015). Therefore, we pose the following hypothesis:

Hypothesis 5: Gender is related to locked-in status, such that women are more often found in locked-in positions a) cross-sectionally and b) over time.

Also, over the course of an occupational career, aspirations and abilities are subject to change. Therefore, the risk for becoming locked in may be higher at some periods of one’s career than at others. At the beginning of the career, it is important for the individual to establish him/herself in the labor market. For many young adults it is difficult to get their most preferred position immediately, and as a result, they often start in a situation of PE misfit (Quintini, 2011a), which over time could turn into a locked-in position if things do not develop as expected. Toward the end of the career, older workers could be facing difficulties, for example due to age discrimination (Furåker et al., 2014; Ng and Feldman, 2012), or due to physical limitations (Ilmarinen, 2001). Earlier studies have indicated that older employees are less interested in changing jobs (Aronsson et al., 2000), and generally perceive lower employability (Berntson et al., 2006; Wittekind et al., 2010). The highest levels of employability appear to be among middle-aged employees (De Coen, 2012). However, no age differences could be found in relation to being locked-in when Furåker (2010) utilized a similar conceptualization of being locked-in as used in present article, although without disentangling the risk category from the locked-in category. Hence, the relationship between age and locked-in status is not clear. Based on the finding that employability is highest in middle age, and the arguments that labor market position may be less favorable for both younger and older workers, we pose the following hypothesis:

Hypothesis 6: Age is related to locked-in status, such as employees in their middle age are less often found in locked-in positions compared to younger and older employees.

The link between socioeconomic position (SEP)—traditionally measured by occupational position, education level, or income—and health and mortality is well established (Marmot and Wilkinson, 2006). Occupational position is suggested to be related to different levels of control and differences in overall working conditions, and this inevitably results in health differences (Geyer et al., 2006). This is in line with the ideas of a segmented labor market, for instance the ‘dual labor market’ (Doeringer and Piore, 1971) that suggests that workers could be allocated to either a privileged primary segment or to a secondary segment, reflecting a division in work conditions, wages, and future career prospects (Hanson, Martin, and Tuch, 1987). The primary segment typically consists of a larger number of non-manual workers and more highly educated individuals, which will facilitate job mobility, both externally and internally in the organization (Berntson et al., 2006; Leontaridi, 2002). Those in the second segment typically have a lower SEP and are often found in jobs with impaired work environment, low control, and high physical work stressors, which may limit their preferences to remain in these jobs for their entire working life. At the same time, their jobs require fairly little occupational training. Moreover, such jobs are about to disappear due to technology developments and outsourcing of such work to low-income economies (Baruch, 2015; van Eekelen, 2015). Following this line of thinking, lower SEP may be associated with locked-in positions.

The empirical evidence from the field is equivocal. For example, some studies have shown support for the notion that there is a tendency for non-manual workers to perceive higher employability compared to manual workers (Berntson et al., 2006; De Cuyper, Baillien, and De Witte, 2009; Furåker et al., 2014). Another study has found that non-manual workers were more often in non-preferred workplaces and in locked-in positions compared to manual workers (Furåker, 2010). This could indicate that it is too simplistic to divide employees into the broad categories of manual versus non-manual workers, because the two categories comprise both workers with valuable competence and those with little training doing routine work tasks, which translates into dissimilar job conditions and also differences in employability. In fact, Grusky and others (Grusky and Sorensen, 1998; Weeden and Grusky, 2005) have stated that this old division is outdated as society today has different views of different status groups. Consequently, the concept of social class needs to be refined to capture these changes and their effects on outcomes. For instance, skilled manual workers today can earn more money than highly educated employees such as physicians (Mayrhofer, Meyer, and Steyrer, 2007). Therefore, the present study discriminates between skilled and unskilled manual workers as well as between assistants and higher non-manual workers, and poses the following hypothesis:

Hypothesis 7: Socioeconomic position is related to locked-in status, such that unskilled manual workers and assistant non-manual workers (but not skilled manual workers) are more often found in locked-in positions compared to higher non-manual workers a) cross-sectionally and b) over time.

Place of residence (where one lives and works) might be an additional factor related to being locked-in. In less densely populated areas, unemployment rates are usually higher and the labor market less varied than in urban regions, and hence it may be more difficult to get a job (Berntson et al., 2006; Kirschenbaum and Mano-Negrin, 1999). Even though a move to an urban area could mean access to a more prosperous local labor market, in Sweden for example, few people move between municipalities. According to Statistics Sweden’s databases, over the past 10 years, on an annual basis, less than 5% of the population aged between 25 to 64 years moved (Statistics Sweden, 2018), indicating that the local labor market is crucial for perceived employability. Thus, one might assume that urban areas are less associated with locked-in positions compared to other regions. In line with this, earlier studies have shown that living in urban regions can be associated with higher perceived employability (Berntson et al., 2006; Furåker et al., 2014; McGuinness and Wooden, 2009), in times of recession as well as economic prosperity (Berntson et al., 2006). Thus, we pose the following hypothesis:

Hypothesis 8: Place of residence is related to locked-in status, such that individuals living in urban areas (compared to less populated areas) are less often found in locked-in positions a) cross-sectionally and b) over time.

When building family, one’s prioritizations in relation to working and private life are often impacted, for example, there could be restrictions on attainable working hours and commuting distances, as well as one’s possibilities to move (Bernhard-Oettel and Näswall, 2015; Furåker, 2010). For instance, in Sweden most families are dependent on double incomes and therefore a job opportunity that fits one partner could imply a deficient match to the work arrangements of the other partner, for instance when working hours become incompatible with family duties or if job changes increase commuting time. In a similar vein, it is possible that some parents will stay in or accept job conditions suitable for their private lives, but perhaps less in line with their own career visions (Bernhard-Oettel and Näswall, 2015). As a consequence, having family obligations might decrease opportunities to pick perfect job matches and may take a toll on employability, thus increasing the odds for becoming locked in. However, this has rarely been investigated. A previous study (Furåker et al., 2014) found no association between employability and either marital or parental status, but in this case only employability was tested (not locked-in status). We phrase our final hypothesis as follows:

Hypothesis 9: Family situation (in terms of having a partner or child responsibilities) is related to locked-in status, such that individuals having a partner or child responsibilities are more often found in locked-in positions compared to others without partner or child a) cross-sectionally, and b) over time.


Sample and procedure

Data were used from the 2014 and 2016 waves of the Swedish Longitudinal Occupational Survey of Health (SLOSH). SLOSH is a cohort study with a broad focus on work organization, work environment, and health (Magnusson Hanson et al., 2018). Data were collected via postal questionnaire every second year and all participants were informed of the study purpose and that participation was voluntary. All participants had previously responded to the Swedish Work Environment Surveys (SWES)—a representative data collection of Swedish workers conducted every second year by Statistics Sweden. In the present study, our sample was drawn from those who participated in SWES during either 2007, 2009, or 2011.

To investigate the potential role of work-related variables (first aim), participants comprised employees who answered the SLOSH questionnaire in 2014, and at that time had a permanent contract, were younger than 65, and provided full responses to questions about the work-related variables of interest (N = 6449). Of those meeting this criteria, 57.6% were women, 72.5% were non-manual workers, and the mean age was 49.1 (SD = 9.8) years. Longitudinal analyses were based on individuals who answered the questionnaire in 2014 and 2016, and had a permanent contract at both time points (N = 3898).

To investigate the potential role of demographic variables (second aim), a subset of the sample was used containing only those who also had sufficient answers on the demographic variables of interest (N = 5984 for cross-sectional analysis, and N = 3633 for longitudinal analysis).

Ethics approval was provided by the Regional Research Ethics Board in Stockholm.


Independent variables: Matching factors. Knowledge/skills fit was measured with a single item from Statistics Sweden (1997) reading ‘Comparing your skills and knowledge with the job you do, do you think you are…’ and which contained five answer alternatives (1 = very overqualified, 2 = overqualified in some respects, 3 = just qualified enough, 4 = would need some additional skills, 5 = would need many additional skills). The item was recoded into three categories: 1 = overqualified (1–2), 2 = just qualified enough (3), and 3 = underqualified (4–5), where the middle alternative served as the reference category. Work ability fit (physical and mental) was measured with two single items (Magnusson Hanson et al., 2018) modified from items from the work ability index (WAI) (see e.g., Ilmarinen, Tuomi, and Seitsamo, 2005) ‘How would you rate your work ability concerning physical (mental) demands?’ both scales having the answer alternatives (1 = very good to 5 = very poor). The answer alternatives were reduced to three, collapsing 1 with 2 and 4 with 5. Demographics. Information about gender (1 = male, 2 = female) and age were obtained from register data. Age was collapsed into the three nominal categories commonly used in career research, i.e., (1 = 34 years or younger, 2 = 35–49 years, 3 = 50 years or older), where the middle-aged group represents being in the ‘mid-career’ (De Vos, Forrier, Van der Heijden, and De Cuyper, 2017; van der Heijden, 2001), and serves as a reference category. Gets Stuck in Their Workplaces? The Role of Matching Factors, between Individual (Swedish socioeconomic classification) builds on the occupation (as reported in the questionnaire), in accordance with information about the length of education usually required (Statistics Sweden, 1984). In the current study employees were categorized into four categories; two levels of manual workers a) unskilled manual, and b) skilled manual workers, and two levels of non-manual workers c) assistant non-manual and d) intermediate- and higher non-manual workers. Place of residence. Register data for place of residence were placed into three categories (1 = urban areas, 2 = semi-urban areas, and 3 = sparsely populated areas), where urban areas included the three largest cities in Sweden (Stockholm, Gothenburg, and Malmö), and semi-urban areas included municipalities with more than 90,000 inhabitants. Marital status (0 = single, 1 = married/cohabitant) was measured with one item. Parental status was measured with the item ‘Do you have any children living at home? Include children living with you at least half of the time’, and had two answer alternatives (0 = no, 1 = yes).

Outcome variables: Workplace non-preference was measured with a single item ‘Is the company/workplace where you work today the place you wish to work at in the future?’ (modified version [Magnusson Hanson et al., 2018] of Aronsson et al. [2000], containing three answer alternatives [1 = yes, 2 = no, but I’m satisfied right now, 3 = no, I’m dissatisfied with the company/workplace]). Perceived employability was measured with a single item ‘How easy would it be for you to get another, similar job without having to change residence?’ (Statistics Sweden, 2004). The response alternatives ranged from 1 (very easy) to 4 (very hard) and 5 (I don’t know), where the last alternative (5) was excluded. Thereafter, the index was dichotomized, where 0 = (rather/very) high employability, 1 = (rather/very) low employability. By combining measures of workplace non-preference and perceived employability, employees were divided into three categories of locked-in status. The individuals who perceived themselves to have low employability and reported being in a non-preferred workplace were categorized as being locked in (LI). The individuals who perceived themselves to have low employability and reported being in a non-preferred job, but were satisfied right now were categorized as being at risk of becoming locked in (RLI). All other combinations were considered as not being locked in (NLI), see Table 1.

Table 1

Construction of the locked-in status variable (the new values in brackets).

Workplace non-preference Perceived Employability

Low High

1 = Preferred job NLI (1) NLI (1)
2 = Non-preferred workplace, but OK for now RLI (2) NLI (1)
3 = Non-preferred workplace LI (3) NLI (1)

LI: being locked in, RLI: at risk of becoming locked in, NLI: not locked in.

Data analysis

In order to test the study hypotheses, multinomial logistic regression analyses were conducted. This approach is suitable when the dependent variable (here: locked-in status) has more than two categories, and several independent variables (categorical and/or continuous) are being tested (Denham, 2010; Petrucci, 2009). Odds ratios and 95% confidence intervals (CI) are provided. Regarding cross-sectional associations between locked-in status and the independent variables, two separate independent multinomial logistic regression analyses were performed, one with matching factors and another with demographics2 as independent variables; both with NLI as the reference state. All variables, including locked-in status, were measured at Time 1. In the first analysis, matching factors were studied, namely knowledge/skills fit and work ability fit. In the second analysis demographics were studied, including gender, age, socioeconomic position, place of residence, marital status, and parental status.

Furthermore, to test our hypotheses with regard to longitudinal relations, two multinomial logistic regression analyses were performed separately for matching factors and demographics as independent variables. In the first analysis, the matching factors were measured at Time 1 while the outcome locked-in status was measured at Time 2. The baseline value of locked-in status (at Time 1) was controlled for as an independent variable. The same procedure was repeated for demographics as independent variables. The hypotheses are presented in Figure 1.

Figure 1 

The study hypotheses. For the cross-sectional hypotheses, both independent and dependent variables are measured at Time 1. For longitudinal hypotheses, the independent variables are measured at Time 1, whereas the dependent variable is measured at Time 2. Note that locked-in status baseline (Time 1) only is controlled for in the analyses of longitudinal relations (the dashed line).


In Tables 2 and 3 the distributions for all variables can be found.

Matching factors

The results from the multinomial logistic regression analyses including the matching factors (i.e., knowledge/skills fit, and work ability fit) as independent variables at Time 1 predicting locked-in status are shown in Table 2 (cross-sectional outcomes to the left, longitudinal outcomes to the right). All odds ratios for LI/RLI are in comparison to NLI.

Table 2

Multinomial logistic regressions. Matching factors between work and individuals. Odds ratios and 95% confidence intervals.

Locked-in status T1 (Cross-sectional) Locked-in status T2 (Longitudinal, including locked-in status baseline)


Odds ratios 95% CI Odds ratios 95% CI Odds ratios 95% CI Odds ratios 95% CI

Knowledge/skills fit
Overqualified 1.67*** [1.43, 1.93] 3.18*** [2.48, 4.07] 21.6 1.01       [0.80, 1.26] 1.80**   [1.25, 2.58] 21.1
Underqualified 1.22*     [1.03, 1.45] 1.26       [0.90, 1.75] 16.4 1.03       [0.80, 1.32] 1.22       [0.77, 1.91] 16.9
Well-matched (ref.) –       –       62.0 –       –       62.0
Work ability fit
Poor physical work ability1 1.33*** [1.18, 1.50] 1.10       [0.89, 1.37] 1.15       [0.96, 1.38] 1.38*     [1.04, 1.84]
Poor mental work ability1 1.50*** [1.30, 1.73] 2.72*** [2.24, 3.29] 1.39**   [1.13, 1.72] 1.40*     [1.01, 1.93]
Locked-in status baseline
LI 5.2 6.96*** [4.79, 10.13] 29.43*** [18.94, 45.74] 4.8
RLI 19.8 8.13*** [6.69, 9.90] 7.01*** [4.83, 10.18] 18.5
NLI (ref) 75.1 –       –       76.7
N 6449       3898      
Nagelkerke R2 .055       .245      

Reference profile: Non-locked-in at Time 2. Locked-in status distribution at Time 2 (LI: 4.7%, RLI: 17.2%, NLI: 78.1%).

* p < 0.05; ** p < 0.01; *** p < 0.001; † p < 0.10.

All independent variables at T1 (2014).

Cross-sectional results

Including all matching factors measured at Time 1 and the outcome variable locked-in status at Time 1 resulted in a Nagelkerke R2 of .06. The fit between knowledge/skills and one’s work tasks appeared to matter in accordance with LI/RLI prevalence, since being overqualified—as compared to being just qualified enough—was associated with higher odds ratios for being RLI (1.67, p < 0.001) and LI (3.18, p < 0.001) at Time 1. Thus, hypothesis 1a was supported. Regarding under-qualification, only a small increase in odds ratios for RLI at Time 1 was found (1.22, p < 0.05).

No relation between under-qualification and LI at Time 1 was found. Thus, hypothesis 2a was partly supported. Finally, poor physical work ability was associated with being RLI (1.33, p < 0.001) at Time 1. No relation to LI at Time 1 was found. Hence, hypothesis 3a was partly supported. Finally, poor mental work ability was associated with being RLI (1.50, p < 0.001) and LI (2.72, p < 0.001); thus supporting hypothesis 4a.

Longitudinal results

For the longitudinal model, measuring locked-in status at Time 2 and the matching factors at Time 1 and controlling for locked-in status at baseline (Time 1), the Nagelkerke R2 reached 0.25. Regarding fit between knowledge/skills and one’s work tasks, being overqualified at Time 1 was still associated with LI at Time 2 (1.80, p < 0.01). However, being overqualified at Time 1 was no longer related to RLI at Time 2. Thus, hypothesis 1b was only partly supported. Being underqualified at Time 1 was not associated with LI/RLI positions at Time 2, and hence, hypothesis 2b was not supported. Turning to work abilities; poor physical work ability increased the odds ratios for LI at Time 2 (1.38, p < 0.05) and poor mental work ability at Time 1 increased the odds ratios for RLI (1.39, p < .01) and LI (1.40, p < 0.05) at Time 2. Thus, hypothesis 3a was partly supported, whereas hypothesis 4a was fully supported.

Demographic variables

The results from the multinomial logistic regression analyses including demographic variables (i.e., gender, age, socioeconomic position, place of residence, marital status, and parental status) at Time 1 and locked-in status are shown in Table 3 (cross-sectional outcomes to the left, longitudinal outcomes to the right). All odds ratios for LI/RLI are in comparison to NLI.

Table 3

Multinomial logistic regressions. Demographic factors. Odds ratios and 95% confidence intervals.

Locked-in status T1 (Cross-sectional) Locked-in status T2 (Longitudinal, including locked-in status baseline)


Odds ratios 95% CI Odds ratios 95% CI Odds ratios 95% CI Odds ratios 95% CI

Gender (female) 0.94       [0.83, 1.08] 0.80†     [0.63, 1.02] 57.5 0.86       [0.71, 1.04] 0.62**   [0.44, 0.87] 57.7
Less than 35 years 0.95       [0.75, 1.21] 0.77       [0.49, 1.23] 9.6 0.92       [0.62, 1.35] 0.55       [0.26, 1.20] 7.8
More than 49 years 1.01       [0.87, 1.18] 0.99       [0.75, 1.31] 51.8 1.00       [0.80, 1.24] 0.93       [0.63, 1.36] 53.3
35–49 years (ref.) –       –       38.6 –       –       38.8
Socioeconomic position
Unskilled manual 1.61*** [1.32, 1.95] 1.61**   [1.14, 2.28] 11.9 1.27       [0.93, 1.73] 1.14       [0.64, 2.03] 9.8
Skilled manual 1.02       [0.84, 1.24] 0.87       [0.60, 1.27] 14.4 1.01       [0.75, 1.35] 0.92       [0.52, 1.61] 13.1
Assistant non-manual 1.56*** [1.29, 1.89] 1.90*** [1.38, 2.61] 12.6 1.50**   [1.12, 2.01] 2.77*** [1.78, 4.32] 11.0
Medium/higher non-manual (ref.) –       –       61.1 –       –       66.1
Place of Residence
Urban areas 0.78**   [0.66, 0.93] 1.10       [0.80, 1.51] 34.4 0.74*     [0.57, 0.95] 1.31       [0.82, 2.09] 34.7
Semi-urban areas 0.98       [0.83, 1.15] 1.17       [0.86, 1.58] 40.0 0.81†     [0.64, 1.02] 1.14       [0.72, 1.79] 40.2
Sparsely populated areas (ref.) –       –       25.6 –       –       25.1
Family situation
Marital status (married/cohabitant) 0.85*     [0.72, 1.00] 0.82       [0.62, 1.10] 79.7 0.94       [0.74, 1.20] 0.79       [0.52, 1.20] 80.1
Parental status (children at home) 1.16†     [1.00, 1.34] 0.96       [0.74, 1.26] 51.0 1.18       [0.95, 1.46] 1.01       [0.69, 1.49] 52.4
Locked-in status baseline
LI 5.1 7.63*** [5.19, 11.23] 38.84*** [24.63, 61.23] 4.8
RLI 19.5 8.33*** [6.79, 10.20] 7.82*** [5.28, 11.60] 18.3
NLI (ref) 75.4 –       –       76.9
N 5984       3633      
Nagelkerke R2 .018       .256      

Reference profile: Non-locked-in at Time 2. Locked-in status distribution at Time 2 (LI: 4.6%, RLI: 17.2%, NLI: 78.2%).

* p < 0.05; ** p < 0.01; *** p < 0.001; †p < 0.10.

All independent variables at T1 (2014).

Cross-sectional results

Including all demographics measured at Time 1 and the outcome variable locked-in status measured at Time 1 resulted in a Nagelkerke R2 of 0.02. Neither women nor men nor any age category showed increased odds for being LI/RLI at Time 1. Thus, hypotheses 5a and 6 were not supported. Socioeconomic position was associated with LI/RLI positions. Relative to intermediate/higher non-manual workers; unskilled manual workers had higher odds ratios for being RLI (1.61, p < 0.001) and LI (1.61, p < 0.01). Also, assistant non-manual workers had higher odds ratios for being RLI (1.56, p < 0.001) and LI (1.90, p < 0.001) compared to intermediate/higher non-manual workers. No significant difference in LI/RLI prevalence could be found comparing skilled manual workers to intermediate/higher non-manual workers. Taken together, hypothesis 7a was fully supported. Regarding place of residence, living in an urban area decreased the odds ratios for RLI at Time 1 (0.78, p < 0.05) compared to living in sparsely populated areas, whereas place of residence did not seem to matter for LI. Thus, hypothesis 8a was only partly supported. Turning to family situation, neither marital nor parental status showed any association with LI at Time 1. Parental status also showed no association with RLI. Contradicting hypothesis 9a, being married/cohabitant was indicated to slightly lower the odds ratios for RLI at Time 1 (0.85, p < 0.05). Hence, hypothesis 9a was not supported.

Longitudinal results

For the longitudinal model, measuring locked-in status at Time 2 and the demographics at Time 1, controlling for locked-in status at baseline, the Nagelkerke R2 reached 0.26. In the longitudinal analyses—contrary to being hypothesized—women had lower odds ratios than men of moving into the LI category at Time 2 (0.62, p < 0.01). No such gender difference was found for RLI. Thus, hypothesis 5b was not supported. Regarding the longitudinal associations between socioeconomic position and LI/RLI positions, assistant non-manual workers had higher odds ratios for becoming both RLI (1.50, p < 0.01) and LI (2.77, p < 0.001) compared to medium/higher non-manual workers, whereas no increased odds ratios were found for unskilled or skilled manual workers. Taken together, hypothesis 7b was partly supported. Living in an urban area compared to living in a sparsely populated area decreased the odds for becoming RLI (0.74, p < 0.05) (but not for LI) at Time 2. Hence, hypothesis 8b was only partly supported. Regarding family situation, neither being married/cohabitant nor parental status were risk factors for LI/RLI positions two years later. Thus, hypothesis 9b was not supported. Finally—when controlling for demographics—the odds ratios for an individual in RLI to become LI two years later was 7.82 (p < 0.001), and the odds ratios for an individual to stay in LI was 38.84 (p < 0.001) compared to a person in NLI.


The aim of this study was to identify determinants of being locked in at the workplace (LI: being in a non-preferred workplace combined with low perceived employability) or being at risk of becoming locked in at the workplace (RLI: being in non-preferred workplace—yet satisfied for now—combined with low perceived employability). The purpose was twofold; to investigate whether 1) matching factors between the individual and the work (over-qualification, under-qualification, and [physical and mental] work abilities), and 2) demographic variables (gender, age, socioeconomic position, place of residence, marital status, and parental status) were related to locked-in status cross-sectionally, and to change in locked-in status two years later (controlling for locked-in status at baseline).

The results indicate that the investigated matching factors were associated with both LI and RLI positions. In regard to the consequences of imbalance between individuals’ knowledge/skills and their work tasks, being overqualified was found to relate both to LI/RLI positions cross-sectionally and also to LI over time. This is in line with earlier studies, implying that over-qualification is associated with job dissatisfaction (Johnson and Johnson, 2000; Lobene et al., 2015), higher turnover intentions (Maynard et al., 2006), and lower chances of getting other jobs (Erdogan et al., 2011; Pedulla, 2016). Besides employers’ aversion to overqualified job seekers, the difficulty could be aggravated over time as a result of lost edge due to unutilized knowledge and skills as well as not developing career-enhancing strategies enough (Peiró et al., 2012). This may be why being in a job where one is overqualified also seems to increase the odds for becoming LI two years later. In contrast, being underqualified showed only small increased odds ratios for RLI cross-sectionally, and no significant relation to locked-in status longitudinally. Furthermore, no link to LI was found, either cross-sectionally or over time. Hence, the proposition that being underqualified would relate to locked-in status was largely not supported. This finding could indicate that receiving a promotion (or attaining a new job), despite lacking the appropriate knowledge/skills for that position, may not be so troublesome but rather be perceived as a chance for personal growth. Therefore, underqualified employees may choose to remain in the company and/or they may evaluate their employability as high since they previously succeeded in landing a better job than their qualification merited. To summarize, over-qualification seems to be more problematic than under-qualification since over-qualification is indicated to put individuals at danger of being or becoming stuck in their workplace.

Regarding work ability, associations were found with locked-in status for both poor physical and mental work abilities. The results showed that poor physical work ability increased the odds ratios for RLI cross-sectionally. Cross-sectionally, odds ratios were not increased for LI, but over time—when controlling for base level locked-in status—poor physical work ability increased the odds ratios for getting into an LI position. This indicates that poor physical work ability—despite not having an association with LI at Time 1—may act as a scarring factor that over time may get workers into LI. A possible explanation is that people with reduced work ability stay in a job that demands too much and this, over time, takes its toll. Another interpretation is that the individual over time becomes marginalized in the workplace as an effect of the limits in his/her physical work ability. For poor mental work ability, there were increased odds ratios for LI/RLI positions cross-sectionally as well as over time. Hence, poor mental work ability plays a significant role in relation to locked-in status. The results from the present study are in line with earlier research indicating a relationship between good work ability and high perceived employability (Nilsson and Ekberg, 2013) and a relationship between good work ability and enjoyment of one’s work (Tuomi et al., 2001). Furthermore, the study lends support to the suggestion that poor work ability matters to individuals’ feelings of being, or in the future becoming, stuck in their workplaces.

In summary, with regard to our first aim, our findings show that the match between demands and abilities appears to be of importance for avoiding locked-in positions. Skill mismatch—particularly over-qualification—as well as poor work abilities—physical as well as mental—were factors that contributed to locked-in positions and thus might make individuals feel stuck in their workplace with few perceived possibilities to escape to other employments or improve their current situation.

These findings add to earlier studies, which also showed that disproportions at work including high demands and low control, as well as effort/reward imbalance, increase the risk for being locked in (Bernhard-Oettel et al., 2018; Fahlén et al., 2009). In fact, somebody with too little work ability may find a job too (quantitatively) demanding. Therefore, to avoid locked-in positions it seems vital for the employees to continuously survey their match to their job with respect to several factors. More research is needed to understand in more depth how mismatches evolve and when a mismatch in qualification or work ability may turn into a locked-in situation.

Demographics and locked-in positions

With respect to the second aim of this study, the results indicated some associations between demographics and locked-in status. Contradicting our hypothesis, men had slightly higher odds ratios of becoming LI over time when the baseline level of locked-in status was controlled for, and cross-sectionally the same direction of an association approached significance. This finding partly mirrored the study by Fahlén et al., (2009) showing that men more often were locked in.

With regard to age, we found no associations with locked-in status. This is in line with an earlier study showing no age differences in regard to locked in (Furåker, 2010). A possible interpretation of this finding is that, despite possible age discrimination in the labor market (Furåker et al., 2014; Ng & Feldman, 2012), locked-in status may develop at any age.

In regard to socioeconomic position, assistant non-manual workers had higher odds ratios for both LI and RLI compared to medium/higher non-manual workers, cross-sectionally as well as over time. Cross-sectionally there were also increased odds ratios for unskilled manual workers to be in LI/RLI positions, whereas over time the odds ratios were not higher for this particular group to become LI/RLI compared to medium/higher non-manual workers. Interestingly, skilled manual workers showed no increased odds for being locked in at all. This implies, as suggested, that neither manual nor non-manual work as broad categories are associated with locked-in status. Instead, individuals in relatively lower-status jobs, no matter whether these involve manual or non-manual work, appear to be at greater risk for being locked in. The reason might be that lower-status positions (in both broad categories) in general may not be as satisfying and furthermore, due to lack of (higher) education those individuals might perceive more difficulties in attaining another job in today’s increasingly knowledge-driven and specialized labor market (Ployhart and Bliese, 2006; van Eekelen, 2015), which would require more highly educated non-manual workers as well as skilled manual workers, for example handicraftsmen.

Considering place of residence, living in an urban area decreased the odds ratios for RLI, both cross-sectionally and longitudinally, compared to living in a sparsely populated area. However, no significant association between place of residence and LI could be found. This was a little surprising since studies have shown that perceived employability is lower among employees living in sparsely populated areas (Berntson et al., 2006; Furåker et al., 2014; McGuinness and Wooden, 2009), since often the unemployment rates are higher and the labor market is less varied in such regions (Berntson et al., 2006; Kirschenbaum and Mano-Negrin, 1999). On the other hand, people who choose to live in the countryside may be more willing to stay in their workplaces and more satisfied with doing so (at least for the near future), as they know that they have few alternatives. This would explain why the link between sparsely populated areas and RLI was significant, whereas the link to LI was not.

Finally, with regard to family situation, neither marital nor parental status seemed to be of much importance in regard to locked-in status (this is in line with Furåker et al., 2014), as only a small cross-sectional association (in the opposite direction from what was hypothesized) was found between being single and RLI. This might indicate that family life is not what limits careers with regards to driving individuals into locked-in positions, at least not with regard to opportunities on the local labor market.

To sum up our findings on demographics and locked-in positions; it was found to matter in which segment the individual works, since employees in jobs with relatively lower status—in both manual and non-manual work—run a higher risk for being in locked-in positions. Furthermore, living in a sparsely populated area increased the odds for RLI, but interestingly not for LI, both cross-sectionally and over time. An explanation might be that even though individuals in the countryside acknowledge that there are scarce job opportunities for them on the local labor market, many feel content with or settle for what they have, while others perhaps feel that doing so may entail a risk in the long run.

Finally, our findings show, controlling for demographics, that being in locked-in positions increased the odds for being in locked-in positions two years later compared to a person who was not locked in. For the ‘risk category’ the odds ratios for remaining in an RLI position were similar to the odds ratios for them to becoming LI during this time frame. Thus, in accordance with the label placed on them, the risk group really is ‘at risk of becoming locked in’. For an individual in an LI position, besides high odds of remaining in that position, there were also increased odds for them to become RLI compared to those not being locked in. Hence, these findings suggest that locked-in status may change, and many of these changes will take place within locked-in (LI and RLI) positions.

Methodological considerations and future research

This study has several strengths, mainly because it uses a longitudinal design with a large cohort (generally representative for the Swedish working population) covering many different occupations. Also, the SLOSH study questionnaire covers many different aspects of working life and personal life, enabling us to study a variety of different variables that previously have been discussed as possible determinants of becoming or being locked in. However, as with all research, there are some methodological limitations to our study that the reader should bear in mind. First, locked-in status was constructed from two single items—perceived employability and non-preferred workplace. However, for some variables there is evidence that single items could work as well as multiple-item scales; this applies when a narrow content of a unidimensional construct is targeted (Gilbert and Kelloway, 2014; Wanous, Reichers, and Hudy, 1997). For instance, in several studies perceived employability has been measured with a single item (see e.g., Berntson et al., 2006; De Cuyper et al., 2010; Kirves et al., 2011; Silla, Gracia, and Peiró, 2005) and multiple-item scales of perceived employability often show extraordinary reliability (Berntson et al., 2008; De Cuyper and De Witte, 2011), as well as have been shown being of one-dimension (see, e.g., Berntson and Marklund, 2007), which may indicate the measurement of a fairly narrow content. Moreover, since the item used in our study is very similar to items in the multiple-item scales, we argue that a single item could be effectively used for perceived employability. Concerning the non-preference of the workplace variable, it is related to overall job satisfaction which in earlier studies has been demonstrated to work as a single item as well (Dolbier et al., 2005; Fisher, Matthews, and Gibbons, 2016; Wanous et al., 1997). Furthermore, this study contributes to the field by measuring the locked-in concept in more detail than was done in most earlier studies (see e.g., Aronsson et al., 2000; Furåker et al., 2014), since first, we covered the two dimensions of locked-in status (see e.g., Fahlén et al., 2009; Stengård et al., 2016), and second, we distinguished between being locked in and being at risk of becoming locked in (Bernhard-Oettel et al., 2018; Stengård et al., 2016). However, as this conceptualization does not directly ask whether an employee perceives that they are stuck in a non-preferred workplace, one direction for future research would be to test whether subjective perceptions and the classifications of locked-in situations as defined here are congruent. Yet another interesting avenue for future research would be to study the extent to which those who do not prefer their workplace anymore, but feel that there are few available alternatives, also remain with their current employer due to high levels of normative or continuance commitment (Meyer and Allen, 1991). This could be considered in combination with embeddedness that requires too high sacrifices to leave (e.g., due to benefits in the current organization, see Mitchell et al., 2001).

With regard to knowledge/skills fit, we also used a single item, but between over- and under-qualification, which is an improvement from many studies that mostly focused on misfit regardless of direction or only in terms of over-qualification (Erdogan et al., 2011; Thompson et al., 2015). Furthermore, the few studies about mismatch between tasks and skills in terms of under-qualification commonly measure a shortage of formal education in relation to position, as a proxy for skills, which may fail to capture lack of skills (Quintini, 2011a), which we aim to do. Finally, work abilities were also measured with two single items. Since the WAI is very long, several studies have compared it with just a single measure, which seems to work well as a proxy for the whole scale (Ahlstrom et al., 2010; Jaaskelainen et al., 2016; Mokarami et al., 2017). It should be noted that this may not be the case if work ability is measured to predict long-term sick leave/disability pension (Roelen et al., 2014; Schouten et al., 2016), which however, was not the focus of this study.

Finally, it should be noted that the effect sizes (prediction values) of the studied models and the independent variables turned out to be rather small, especially for demographic factors. This is not uncommon in research on demographics and psychological work-related outcomes, such as employability (Berntson et al., 2006) and job insecurity (Näswall and De Witte, 2003). Nevertheless, this indicates that there are probably a number of other individual and conditional factors that may increase the odds for being/becoming locked in, for example, the factors of needs–supplies fit, or social capital (such as support) should be targeted in future research. Also, personality traits may be of importance in relation to the locked-in phenomenon. One study for instance indicated that there was indeed a relationship between helplessness and being locked in (Stengård et al., 2017), but more studies should focus on trait-like variables, such as change resistance and proactivity, and their possible relationship with being locked-in.

Practical implications

Feeling stuck in one’s workplace is probably unsatisfying, and being in locked-in positions (being locked in or at risk of becoming locked in) has, in fact, been shown to be associated with poor (mental and self-reported) health over time, in earlier studies (Aronsson et al., 2000; Aronsson and Göransson, 1999; Canivet et al., 2017; Fahlén et al., 2009; Furåker, 2010; Stengård et al., 2016). Therefore, it is important to decrease the prevalence of locked in in the labor market. Are there any good ways of bringing this about? This study indicates that the match between demands and abilities is of importance, both in regard to levels of knowledge/skills and physical and mental work abilities. Hence, organizations should strive for balance for their employees by setting requirements that are appropriate based on each individual’s abilities. This is a continuous effort, since workers develop and age and hence, their abilities alter. Being overqualified or having deficient mental or physical work ability seems to be especially problematic vis à vis locked-in positions, thus it is important to provide employees with work tasks that utilize their knowledge/skills, but do not exceed their work abilities.

Furthermore, education seems to be of significance for avoiding locked-in positions, since working in upper socioeconomic positions in both manual and non-manual work, in this study, was found to be beneficial, and these positions generally require a university or vocational degree. Therefore, society should provide better opportunities for young as well as also older workers to acquire the ‘right’ education, which does not necessarily mean the highest possible education since that may result in over-qualification/over-education if there is no matching job on the labor market (Quintini, 2011b). One group that is of particular interest in this regard (but which this study could not single out due to having too few respondents) is immigrants, as many immigrants have to take on jobs which do not match their qualifications (Quintini, 2011b). Future studies should investigate whether immigrants are more often in locked-in positions.


By investigating mismatches—in the realm of the PE fit framework—between demands put on the employee and his/her abilities to meet these, the present study found that such misfits could be detrimental to one’s career, putting individuals at risk of feeling stuck in their current workplace, i.e., being locked in. The mismatches scrutinized in this paper cover knowledge/skills and work ability, where over-qualification as well as both mental and physical work abilities appeared to be most associated with both being locked in (being in a non-preferred workplace combined with low perceived employability) and at risk of becoming locked in (being in a non-preferred workplace—yet satisfied for now—combined with low perceived employability). These relations appeared to exist cross-sectionally and for developments over time.

Furthermore, we examined associations between demographics and locked-in status, and found that socioeconomic position was important as working in the lower segment of both manual and non-manual categories was related to being in locked-in positions. Living in a sparsely populated area was related to being at risk of becoming locked in.

An important contribution of this study is the differentiation between the risk category and being locked in, providing more insights into factors that elevate the risk of becoming locked in. Since being in the risk category, besides being detrimental to health in itself, was found to increase the risk for becoming locked in over time, even factors related to this risk category should be taken seriously to avoid potential development into a ‘real’ locked-in situation.

To conclude, these findings make a further contribution to the career research field, especially to knowledge of career strains, which is a research field that has been neglected in the past. Particularly, we explored the determinants of feeling stuck in one’s workplace. Nevertheless, more research is warranted, in particular in terms of scrutinizing how and why misfit develops and what organizations can do to intervene against such developments.


1Sometimes, overqualified equates with overeducated (see e.g., Quintini, 2011a), but often it reflects a combination of education and skills (which we comply to). 

2The analysis with the demographics was performed only with those individuals included in the analysis of matching factors; hence, N became slightly smaller. 


Funding: This research was supported by the Swedish Research Council for Health, Working Life and Welfare (FORTE grant no. 2004-2021, 2005-0734, 2009-1077, 2012-0979), by the Swedish Research Council (VR, grant no. 2009-6192, 2013-1645), and partial financial support came from the Stockholm Stress Center, a FORTE Center for Excellence (Ref. No. 2009-1758).

Competing Interests

Two of the co-authors are also editors of this journal, and were therefore removed from all editorial tasks for this paper. Another member of the editorial team has been assigned responsibility of overseeing peer review.


  1. Ahlstrom, L., Grimby-Ekman, A., Hagberg, M., & Dellve, L. (2010). The work ability index and single-item question: Associations with sick leave, symptoms, and health – A prospective study of women on long-term sick leave. Scandinavian Journal of Work Environment & Health, 36(5), 404–412. DOI: 

  2. Aronsson, G., Dallner, M., & Gustafsson, K. (2000). Yrkes- och arbetsplatsinlåsning: En empirisk studie av omfattning och hälsokonsekvenser [Locked-in, in the occupation and the working place. An empirical study of prevalence and health consequencies]. Arbete & Hälsa, 5, 1–25. 

  3. Aronsson, G., & Göransson, S. (1999). Permanent employment but not in a preferred occupation: Psychological and medical aspects, research implications. Journal of Occupational Health Psychology, 4(2), 152–163. DOI: 

  4. Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art. Journal of Managerial Psychology, 22(3), 309–328. DOI: 

  5. Baruch, Y. (2015). Organizational and labor markets as career ecosystem. In A. De Vos, & B. Van der Heijden (Eds.), Handbook of Research on Sustainable Careers (pp. 364–380). Cheltenham: Edward Elgar. DOI: 

  6. Bernhard-Oettel, C., & Näswall, K. (2015). Career continuance and transfer of competencies after job transitions: Insights from a Swedish study. In A. De Vos, & B. Van der Heijden (Eds.), Handbook of Research on Sustainable Careers (pp. 381–397). Cheltenham: Edward Elgar. DOI: 

  7. Bernhard-Oettel, C., Stengård, J., Leineweber, C., Westerlund, H., Peristera, P., & Östergren, P. O. (2018). Stuck at a workplace: What’s work control, demands and learning got to do with it? A longitudinal multilevel study on Swedish permanent employees in situations of ‘workplace locked-in’. The International Journal of Human Resource Management, Advanced online publication. DOI: 

  8. Berntson, E., & Marklund, S. (2007). The relationship between perceived employability and subsequent health. Work & Stress, 21(3), 279–292. DOI: 

  9. Berntson, E., Näswall, K., & Sverke, M. (2008). Investigating the relationship between employability and self-efficacy: A cross-lagged analysis. European Journal of Work and Organizational Psychology, 17(4), 413–425. DOI: 

  10. Berntson, E., Sverke, M., & Marklund, S. (2006). Predicting perceived employability: Human capital or labour market opportunities? Economic and Industrial Democracy, 27(2), 223–244. DOI: 

  11. Bolino, M. C., & Feldman, D. C. (2000). The antecedents and consequences of underemployment among expatriates. Journal of Organizational Behavior, 21(8), 889–911. DOI:<889::AID-JOB60>3.0.CO;2-G 

  12. Canivet, C., Aronsson, G., Bernhard-Oettel, C., Leineweber, C., Moghaddassi, M., Stengård, J., Östergren, P.-O., et al. (2017). The negative effects on mental health of being in a non-desired occupation in an increasingly precarious labour market. SSM – Population Health, 3, 516–524. DOI: 

  13. De Coen, A. (2012). Working & Aging: The Role of Age and Employability Regarding Job Search and Retirement Intention. (Doctoral dissertation). Leuven, Belgium: Catholic University of Leuven. 

  14. De Cuyper, N., Baillien, E., & De Witte, H. (2009). Job insecurity, perceived employability and targets’ and perpetrators’ experiences of workplace bullying. Work and Stress, 23(3), 206–224. DOI: 

  15. De Cuyper, N., Bernhard-Oettel, C., Berntson, E., De Witte, H., & Alarco, B. (2008). Employability and employees’ well-being: Mediation by job insecurity. Applied Psychology: An International Review, 57(3), 488–509. DOI: 

  16. De Cuyper, N., & De Witte, H. (2011). The management paradox: Self-rated employability and organizational commitment and performance. Personnel Review, 40(1–2), 152–172. DOI: 

  17. De Cuyper, N., De Witte, H., Kinnunen, U., & Nätti, J. (2010). The relationship between job insecurity and employability and well-being among Finnish temporary and permanent employees. International Studies of Management & Organization, 40(1), 57–73. DOI: 

  18. De Cuyper, N., Mäkikangas, A., Kinnunen, U., Mauno, S., & De Witte, H. (2012). Cross-lagged associations between perceived external employability, job insecurity, and exhaustion: Testing gain and loss spirals according to the conservation of resources theory. Journal of Organizational Behavior, 33(6), 770–788. DOI: 

  19. De Jong, J., De Cuyper, N., De Witte, H., Silla, I., & Bernhard-Oettel, C. (2009). Motives for accepting temporary employment: A typology. International Journal of Manpower, 30(3), 237–252. DOI: 

  20. Denham, B. E. (2010). Measurement of risk perceptions in social research: A comparative analysis of ordinary least squares, ordinal and multinomial logistic regression models. Journal of Risk Research, 13(5), 571–589. DOI: 

  21. De Vos, A., Forrier, A., Van der Heijden, B., & De Cuyper, N. (2017). Keep the expert! Occupational expertise, perceived employability and job search: A study across age groups. Career Development International, 22(3), 318–332. DOI: 

  22. Doeringer, P. B., & Piore, M. J. (1971). Internal Labour Markets and Manpower Analysis. Lexington, MA: Heath Lexington Books. 

  23. Dolbier, C. L., Webster, J. A., McCalister, K. T., Mallon, M. W., & Steinhardt, M. A. (2005). Reliability and validity of a single-item measure of job satisfaction. American Journal of Health Promotion, 19(3), 194–198. DOI: 

  24. Edwards, J. R. (1991). Person-job fit: A conceptual integration, literature review and methodological critique. In C. L. Cooper, & I. T. Robertson (Eds.), International Review of Industrial/Organizational Psychology (vol. 6, pp. 283–357). Oxford, London: John Wiley & Sons. 

  25. Edwards, J. R. (1996). An examination of competing versions of the person-environment fit approach to stress. Academy of Management Journal, 39(2), 292–339. DOI: 

  26. Edwards, J. R., Cable, D. M., Williamson, I. O., Lambert, L. S., & Shipp, A. J. (2006). The phenomenology of fit: Linking the person and environment to the subjective experience of person-environment fit. Journal of Applied Psychology, 91(4), 802–827. DOI: 

  27. Erdogan, B., Bauer, T. N., Peiró, J. M., & Truxillo, D. M. (2011). Overqualified employees: Making the best of a potentially bad situation for individuals and organizations. Industrial and Organizational Psychology-Perspectives on Science and Practice, 4(2), 215–232. DOI: 

  28. Fahlén, G., Goine, H., Edlund, C., Arrelov, B., Knutsson, A., & Peter, R. (2009). Effort-reward imbalance, “locked in” at work, and long-term sick leave. International Archives of Occupational and Environmental Health, 82(2), 191–197. DOI: 

  29. Feldman, D. C., & Ng, T. W. H. (2012). Theoretical approaches to the study of job transitions. In I. B. Weiner, W. C. Borman, D. R. Ilgen, & S. Highhouse (Eds.), Handbook of Psychology, Industrial and Organizational Psychology: Industrial and Organizational Psychology (pp. 292–307). John Wiley & Sons, Incorporated. 

  30. Fisher, G. G., Matthews, R. A., & Gibbons, A. M. (2016). Developing and investigating the use of single-item measures in organizational research. Journal of Occupational Health Psychology, 21(1), 3–23. DOI: 

  31. Fouad, N. A., & Bynner, J. (2008). Work transitions. American Psychologist, 63(4), 241–251. DOI: 

  32. Furåker, B. (2010). Inlåsning på den svenska arbetsmarknaden [Locked-in in the Swedish labour market]. Arbetsmarknad & Arbetsliv, 16(4), 55–70. 

  33. Furåker, B., Nergaard, K., & Saloniemi, A. (2014). Lock-in patterns among employees: A Nordic comparison. Journal of Comparative Labour Law and Industrial Relations, 30(4), 435–458. 

  34. Geyer, S., Hemstrom, O., Peter, R., & Vagero, D. (2006). Education, income, and occupational class cannot be used interchangeably in social epidemiology. Empirical evidence against a common practice. Journal of Epidemiology and Community Health, 60(9), 804–810. DOI: 

  35. Gilbert, S., & Kelloway, E. K. (2014). Using single items to measure job stressors. International Journal of Workplace Health Management, 7(3), 186–199. DOI: 

  36. Greenhaus, J. H., & Callanan, G. A. (2012). Career dynamics. In I. B. Weiner, W. C. Borman, D. R. Ilgen, & S. Highhouse (Eds.), Handbook of Psychology, Industrial and Organizational Psychology: Industrial and Organizational Psychology (2nd ed., pp. 593–614). John Wiley & Sons, Incorporated. 

  37. Grusky, D. B., & Sorensen, J. B. (1998). Can class analysis be salvaged? American Journal of Sociology, 103(5), 1187–1234. DOI: 

  38. Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior & Human Performance, 16(2), 250–279. DOI: 

  39. Hanson, S. L., Martin, J. K., & Tuch, S. A. (1987). Economic sector and job-satisfaction. Work and Occupations, 14(2), 286–305. DOI: 

  40. Ilmarinen, J. (2001). Aging workers. Occupational and Environmental Medicine, 58(8), 546–552. DOI: 

  41. Ilmarinen, J., Tuomi, K., & Seitsamo, J. (2005). New dimensions of work ability. In G. Costa, W. J. A. Goedhard, & J. Ilmarinen (Eds.), Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers (vol. 1280, pp. 3–7). Amsterdam: Elsevier Science Bv. DOI: 

  42. Jaaskelainen, A., Kausto, J., Seitsamo, J., Ojajarvi, A., Nygard, C. H., Arjas, E., & Leino-Arjas, P. (2016). Work ability index and perceived work ability as predictors of disability pension: A prospective study among Finnish municipal employees. Scandinavian Journal of Work Environment & Health, 42(6), 490–499. DOI: 

  43. Jansen, K. J., & Kristof-Brown, A. L. (2006). Toward a multidimensional theory of person-environment fit. Journal of Managerial Issues, 18(2), 193–212. 

  44. Johnson, G. J., & Johnson, W. R. (2000). Perceived overqualification and dimensions of job satisfaction: A longitudinal analysis. The Journal of Psychology, 134(5), 537–555. DOI: 

  45. Kirschenbaum, A., & Mano-Negrin, R. (1999). Underlying labor market dimensions of “opportunities”: The case of employee turnover. Human Relations, 52(10), 1233–1255. DOI: 

  46. Kirves, K., De Cuyper, N., Kinnunen, U., & Nätti, J. (2011). Perceived job insecurity and perceived employability in relation to temporary and permanent workers’ psychological symptoms: A two samples study. International Archives of Occupational and Environmental Health, 84(8), 899–909. DOI: 

  47. Kristof, A. L. (1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49(1), 1–49. DOI: 

  48. Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of individuals’ fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58(2), 281–342. DOI: 

  49. Leontaridi, R. M. (2002). Career, experience and returns to human capital: Is the dual labour market hypothesis relevant for the UK? Research in Economics, 56(4), 399–426. DOI: 

  50. Lobene, E. V., Meade, A. W., & Pond, S. B., 3rd. (2015). Perceived overqualification: A multi-source investigation of psychological predisposition and contextual triggers. The Journal of Psychology, 149(7), 684–710. DOI: 

  51. Magnusson Hanson, L. L., Leineweber, C., Persson, V., Hyde, M., Theorell, T., & Westerlund, H. (2018). Cohort profile: The Swedish longitudinal occupational survey of health (SLOSH). International Journal of Epidemiology, Advanced online publication. DOI: 

  52. Mäkikangas, A., De Cuyper, N., Mauno, S., & Kinnunen, U. (2013). A longitudinal person-centred view on perceived employability: The role of job insecurity. European Journal of Work and Organizational Psychology, 22(4), 490–503. DOI: 

  53. Marmot, M., & Wilkinson, R. G. (2006). Social Determinants of Health. Oxford: Oxford University Press. DOI: 

  54. Maynard, D. C., Joseph, T. A., & Maynard, A. M. (2006). Underemployment, job attitudes, and turnover intentions. Journal of Organizational Behavior, 27(4), 509–536. DOI: 

  55. Mayrhofer, W., Meyer, M., & Steyrer, J. (2007). Contextual issues in the study of careers. In H. Gunz, & M. Peiperl (Eds.), Handbook of Career Studies (pp. 215–240). Thousand Oaks: SAGE Publications, Inc. DOI: 

  56. McGuinness, S., & Wooden, M. (2009). Overskilling, job insecurity, and career mobility. Industrial Relations, 48(2), 265–286. DOI: 

  57. Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualization of organizational commitment. Human Resource Management Review, 1(1), 61–89. DOI: 

  58. Mitchell, T. R., Holtom, B. C., Lee, T. W., Sablynski, C. J., & Erez, M. (2001). Why people stay: Using job embeddedness to predict voluntary turnover. Academy of Management Journal, 44(6), 1102–1121. DOI: 

  59. Mokarami, H., Mortazavi, S. B., Asgari, A., & Choobineh, A. (2017). Work ability score (WAS) as a suitable instrument to assess work ability among Iranian workers. Health Scope, 6(1). DOI: 

  60. Muhonen, T. (2010). Feeling double locked-in at work: Implications for health and job satisfaction among municipal employees. Work, 37(2), 199–204. DOI: 

  61. Näswall, K., & De Witte, H. (2003). Who feels insecure in Europe? Predicting job insecurity from background variables. Economic and Industrial Democracy, 24(2), 189–215. DOI: 

  62. Ng, T. W. H., & Feldman, D. C. (2012). Evaluating six common stereotypes about older workers with meta-analytical data. Personnel Psychology, 65(4), 821–858. DOI: 

  63. Nilsson, S., & Ekberg, K. (2013). Employability and work ability: Returning to the labour market after long-term absence. Work, 44(4), 449–457. DOI: 

  64. Oakman, J., & Wells, Y. (2016). Working longer: What is the relationship between person-environment fit and retirement intentions? Asia Pacific Journal of Human Resources, 54(2), 207–229. DOI: 

  65. OECD. (2014). Achieving stronger growth by promoting a more-gender-balanced economy. Report prepared for the G20 Labour and Employment Ministerial meeting Melbourne, Australia. Retrieved from 

  66. Oh, I.-S., Guay, R. P., Kim, K., Harold, C. M., Lee, J.-H., Heo, C.-G., & Shin, K.-H. (2014). Fit happens globally: A meta-analytic comparison of the relationships of person-environment fit dimensions with work attitudes and performance across East Asia, Europe, and North America. Personnel Psychology, 67(1), 99–152. DOI: 

  67. Pedulla, D. S. (2016). Penalized or protected? Gender and the consequences of nonstandard and mismatched employment histories. American Sociological Review, 81(2), 262–289. DOI: 

  68. Peiró, J. M., Sora, B., & Caballer, A. (2012). Job insecurity in the younger Spanish workforce: Causes and consequences. Journal of Vocational Behavior, 80(2), 444–453. DOI: 

  69. Petrucci, C. J. (2009). A primer for social worker researchers on how to conduct a multinomial logistic regression. Journal of social service research, 35(2), 193–205. DOI: 

  70. Ployhart, R., & Bliese, P. (2006). Individual adaptability (I-ADAPT) theory: Conceptualizing the antecedents, consequences, and measurement of individual differences in adaptability. In C. S. Burke, L. G. Pierce, & E. Salas (Eds.), Understanding adaptability: A prerequisite for effective performance within complex environments (pp. 3–39). DOI: 

  71. Quintini, G. (2011a). Over-qualified or Under-skilled: A Review of Existing Literature (OECD Social, Employment and Migration Working Papers, No. 121). Paris: OECD Publishing. 

  72. Quintini, G. (2011b). Right for the Job: Over-qualified or Under-skilled? (OECD Social, Employment and Migration Working Papers, No. 121). Paris: OECD Publishing. 

  73. Roelen, C. A. M., van Rhenen, W., Groothoff, J. W., van der Klink, J. J. L., Twisk, J. W. R., & Heymans, M. W. (2014). Work ability as prognostic risk marker of disability pension: Single-item work ability score versus multi-item work ability index. Scandinavian Journal of Work Environment & Health, 40(4), 428–431. DOI: 

  74. Schouten, L. S., Bultmann, U., Heymans, M. W., Joling, C. I., Twisk, J. W. R., & Roelen, C. A. M. (2016). Shortened version of the work ability index to identify workers at risk of long-term sickness absence. European Journal of Public Health, 26(2), 301–305. DOI: 

  75. Shipp, A. J., & Jansen, K. J. (2011). Reinterpreting time in fit theory: Crafting and recrafting narratives of fit in medias res. Academy of Management Review, 36(1), 76–101. DOI: 

  76. Silla, I., De Cuyper, N., Gracia, F. J., Peiró, J. M., & De Witte, H. (2009). Job insecurity and well-being: Moderation by employability. Journal of Happiness Studies, 10(6), 739–751. DOI: 

  77. Silla, I., Gracia, F. J., & Peiró, J. M. (2005). Job insecurity and health-related outcomes among different types of temporary workers. Economic and Industrial Democracy, 26(1), 89–117. DOI: 

  78. Statistics Sweden. (1984). Socioeconomisk indelning (SEI) [Socio-economic classification] (MIS 1982:4). Örebro: Statistics Sweden. 

  79. Statistics Sweden. (1997). Undersökning om levnadsförhållanden 1997 [Survey of living conditions 1997]. Stockholm: Statistics Sweden. Retrieved from 

  80. Statistics Sweden. (2004). Arbetsmiljön 2003 [The work environment 2003] (Report No. AM68SM0401). Stockholm: Statistics Sweden. Retrieved from 

  81. Statistics Sweden. (2018). Statistical database: “Population by age and sex. Year 1860–2017” and “Domestic migration by type, age and sex. Year 2000–2017”. Retrieved from 

  82. Stengård, J., Bernhard-Oettel, C., Berntson, E., & Leineweber, C. (2017). Stuck in the job: Does helplessness precede being locked in at the workplace or vice versa? An analysis of cross-lagged effects. Journal of Vocational Behavior, 102, 15–27. DOI: 

  83. Stengård, J., Bernhard-Oettel, C., Berntson, E., Leineweber, C., & Aronsson, G. (2016). Stuck in a job: Being “locked in” or at risk of becoming locked in at the workplace and well-being over time. Work and Stress, 30(2), 152–172. DOI: 

  84. Sverke, M., Falkenberg, H., Kecklund, G., Magnusson Hanson, L. L., & Lindfors, P. (2016). Kvinnors och mäns arbetsvillkor: Betydelsen av organisatoriska faktorer och psykosocial arbetsmiljö för arbets- och hälsorelaterade utfall [Women’s and men’s working conditions: The importance of organizational factors and psychosocial work environment for work and health related outcomes]. Arbetsmiljöverket Kunskapssammanställning 2016:2. Stockholm: Elanders Sverige AB. 

  85. Thompson, K. W., Sikora, D. M., Perrewe, P. L., & Ferris, G. R. (2015). Employment qualifications, person-job fit, underemployment attributions, and hiring recommendations: A three-study investigation. International Journal of Selection and Assessment, 23(3), 247–262. DOI: 

  86. Tuomi, K., Huuhtanen, P., Nykyri, E., & Ilmarinen, J. (2001). Promotion of work ability, the quality of work and retirement. Occupational Medicine-Oxford, 51(5), 318–324. DOI: 

  87. van den Berg, T. I. J., Elders, L. A. M., de Zwart, B. C. H., & Burdorf, A. (2009). The effects of work-related and individual factors on the work ability index: A systematic review. Occupational and Environmental Medicine, 66(4), 211–220. DOI: 

  88. van der Heijden, B. (2001). Age and assessments of professional expertise: The relationship between higher level employees’ age and self-assessments or, supervisor ratings of professional expertise. International Journal of Selection and Assessment, 9(4), 309–324. DOI: 

  89. van Eekelen, B. (2015). Knowledge for the West, production for the rest? Journal of Cultural Economy, 8(4), 479–500. DOI: 

  90. Vanhercke, D., Kirves, K., De Cuyper, N., Verbruggen, M., Forrier, A., & De Witte, H. (2015). Perceived employability and psychological functioning framed by gain and loss cycles. Career Development International, 20(2), 179–198. DOI: 

  91. Voßemer, J., & Schuck, B. (2016). Better overeducated than unemployed? The short- and long-term effects of an overeducated labour market re-entry. European Sociological Review, 32(2), 251–265. DOI: 

  92. Wanous, J. P., Reichers, A. E., & Hudy, M. J. (1997). Overall job satisfaction: How good are single-item measures? Journal of Applied Psychology, 82(2), 247–252. DOI: 

  93. Weeden, K. A., & Grusky, D. B. (2005). The case for a new class map. American Journal of Sociology, 111(1), 141–212. DOI: 

  94. Wittekind, A., Raeder, S., & Grote, G. (2010). A longitudinal study of determinants of perceived employability. Journal of Organizational Behavior, 31(4), 566–586. DOI: 

comments powered by Disqus