Underemployment and Well-Being: Exploring the Dark Side of Meaningful Work

Abstract

The goals of this study were to examine the relations between underemployment, meaningful work, and well-being and evaluate whether the relation between underemployment and well-being was moderated by meaningful work. In a diverse sample of working adults, meaningful work significantly moderated the relation between underemployment and positive affect, negative affect, depression, and stress. However, contrary to expectations, having meaningful work did not protect against the negative relation between underemployment and well-being. Rather, having meaningful work was associated with a positive relation between underemployment and negative affect, stress, and depression. Additionally, people low in meaningful work had a positive relation between underemployment and positive affect. This unexpected finding may point to a potential “dark side” of meaningful work. Specifically, people with meaningful work who are unable to fully employ their skills and abilities may be at particular risk for poorer well-being.

Keywords

underemployment meaningful work subjective well-being overqualification depression

Due to a rapidly changing economy characterized by income inequality, globalization, and the exponential growth of technology, stable employment has become increasingly constrained (Brynjolfsson & McAfee, 2014; International Labor Organization [ILO], 2017). Moreover, organizations such as the ILO (2017) have argued that work in the global economy will remain precarious, with unemployment and underemployment continuing as problems. Given this economic environment, an area of particular concern is the psychological consequences of underemployment. Scholars have found that underemployment negatively relates to a host of well-being variables, such as optimism, psychological well-being, psychological health, life satisfaction, work well-being, and physical health (Bracke, Pattyn, & von dem Knesebeck, 2013; Cassidy & Wright, 2008; Friedland & Price, 2003; Lobene & Meade, 2013; Roh, Chang, Kim, & Nam, 2014). As a consequence, researchers have begun to identify moderators that might buffer the negative relation between underemployment and well-being (e.g., Lobene & Meade, 2013), and one promising moderator may be meaningful work (Allan, Douglass, Duffy, & McCarty, 2016). Following from latent deprivation theory (Jahoda, 1981), underemployment deprives people of meaningful purpose, which may decrease well-being. Therefore, people who maintain meaning in their work may be protected against the negative impact of underemployment. Therefore, using latent deprivation theory as a guide, the goals of this study were to (a) explore the relation between underemployment and well-being, (b) assess the relation between underemployment and meaningful work, and (c) examine meaningful work as a moderator of the relation between underemployment and well-being.

Underemployment

Underemployment is a multidimensional construct that scholars have operationalized and measured in various ways (Allan, Duffy, & Blustein, 2016; Allan, Tay, & Sterling, 2017; McKee-Ryan & Harvey, 2011), but the term generally refers to work that is lesser or inferior compared to some standard (Feldman, 1996). In his seminal article, Feldman (1996) described five types of underemployment: person possesses more education than a job requires; person is involuntary employed outside one’s field; person has higher skills and experience than is required; person is involuntarily engaged in part-time, temporary, or intermittent employment; or person is earning less wages than what is typical. Although this conceptualization of underemployment is comprehensive, it can be too broad and lack specificity. A more common and focused operationalization of underemployment is people’s perceptions that their work does not fully use their level of education, skill, and experience (Maynard, Joseph, & Maynard, 2006). For example, an architect may be working in fast-food service due to losing his or her regular job in architecture. This type of underemployment is also known as overqualification, which we will use as the definition of underemployment for the present study.

Theoretical Background

Broadly, Feldman (1996) argued that five antecedents (i.e., economic factors, job characteristics, career history, job search strategies, and demographic characteristics) lead to underemployment, which in turn affect five outcomes (i.e., job attitudes, psychological well-being, career attitudes, job behaviors, and social relationships). Many of Feldman’s hypotheses from this model have since been supported (McKee-Ryan & Harvey, 2011), but many questions remain unanswered. Chief among these is how or why underemployment leads to poorer well-being. One framework that address this question is latent deprivation theory (Jahoda, 1981). According to Jahoda (1981), employment affects well-being via its latent consequences. Work’s “manifest” consequence is income, which provides access to necessities such as food, shelter, and clothing. However, work has many benefits beyond income, which represent its “latent” consequences, including time structure, social contacts, collective purpose, personal status, and regular activity.

Scholars have generally found support for these latent consequences among diverse workers, even those working in unskilled manual labor, and latent consequences are greater in fully employed people (Creed & Machin, 2002; Paul & Batinic, 2010). Additionally, other perspectives like the psychology of working framework include similar benefits obtained from work such as survival, power, social connection, and self-determination (Blustein, 2008). Latent deprivation theory argues that unemployment deprives people of these latent benefits, leading to poorer well-being (Jahoda, 1981). Although initially applied only to unemployment, scholars have since applied latent deprivation to underemployment (Creed & Machin, 2002) and have argued that underemployment lies on a continuum between no employment and full employment (Dooley, 2003).

Supporting the predictions of latent deprivation, scholars have linked underemployment to poorer well-being using a host of variables. For example, in the work domain, underemployment has been associated with lower job satisfaction (Benedict, Gayatridevi, & Velayudhan, 2009; Maynard et al., 2006), organizational commitment (Lobene, Meade, & Pond, 2015), and affective commitment (Lobene & Meade, 2013) as well as to greater work stress (Maynard, Brondolo, Connelly, & Sauer, 2015). More broadly, scholars have found underemployment to positively predict psychological distress (G. J. Johnson & Johnson, 1996) and negative affect (Thompson, 2014) and negatively predict positive affect (Thompson, 2014) and subjective well-being (Wu, Luksyte, & Parker, 2015). In short, underemployment consistently links to poorer well-being inside and outside of the workplace.

Despite underemployment’s negative relation to well-being, scholars have begun to identify moderators of this relation. For example, job autonomy buffers the relation between underemployment and subjective well-being in individualistic cultures (Wu et al., 2015). More relevant to the present study, calling—a variable closely related to meaningful work—moderated underemployment such that the negative relation between underemployment and organizational commitment was weaker for those high in calling (Lobene & Meade, 2013). Other moderators of underemployment and well-being include empowerment (Erdogan & Bauer, 2009), negative affect (G. J. Johnson & Johnson, 2000), and emotional support (G. J. Johnson & Johnson, 1996). In summary, scholars have identified several moderators of the relation between underemployment and well-being.

Meaningful Work

Meaningful work is a work that is personally meaningful, aids personal growth, and contributes to the greater good (Steger, Dik, & Duffy, 2012). This definition maps well onto the latent consequence of “collective purpose,” which “links individuals to goals and purposes that transcend their own” (Jahoda, 1981, p. 188). Therefore, meaningful work is essentially a latent consequence of work that probably contributes to well-being. Supporting this claim, meaningful work correlates with a host of work and well-being constructs. For example, people who say their work is meaningful report greater life satisfaction and life meaning and lower levels of anxiety, hostility, stress, and depression (Allan, Dexter, Kinsey, & Parker, 2018; Allan, Duffy, & Douglass, 2015; Steger et al., 2012). Meaningful work also predicts positive work-related variables such as job performance and productivity (Allan, Duffy, & Collisson, 2017, Ariely, Kamenica, & Prelec, 2008), career and organizational commitment (Duffy, Dik, & Steger, 2011), and intrinsic work motivation (Steger et al., 2012). Finally, experiencing meaningful work correlates with lower withdrawal intentions and rates of absenteeism (Bunderson & Thompson, 2009; Steger et al., 2012) and greater job satisfaction and perceiving one’s job as a calling (Duffy, Bott, Allan, Torrey, & Dik, 2012). In short, meaningful work is an important correlate of both personal well-being and job-related outcomes.

The Present Study

The goals of this study were to (a) evaluate the relations between underemployment and well-being, (b) assess the relation between underemployment and meaningful work, and (c) examine meaningful work as a moderator of underemployment and well-being. We chose several well-being variables that captured the range of well-being: subjective well-being, depression, and stress. First, subjective well-being is a widely used operationalization of well-being that scholars have linked to work variables (e.g., Bowling, Eschelman, & Wang, 2010). Subjective well-being is typically composed of three dimensions: life satisfaction, positive affect, and negative affect. Although originally combined, these variables are better analyzed separately because they represent relatively independent constructs (Warner & Rasco, 2014). Second, we included depression and stress as well-being variables. Depression is a widely studied outcome of inadequate employment (Wanberg, 2012) and is a likely outcome of underemployment as well (Dooley, Prause, & Ham-Rowbottom, 2000). Stress is also a potential outcome of underemployment (Winefield, 2002) that is less studied than other outcomes. However, it represents a substantial part of well-being that risks physical health and other outcomes (e.g., Segerstrom & Miller, 2004).

Given the well-established relations between underemployment and well-being (see McKee-Ryan & Harvey, 2011), we hypothesized that underemployment would be associated with less life satisfaction and positive affect and greater negative affect, depression, and stress (Hypothesis 1). Second, scholars have not directly evaluated underemployment and meaningful work as correlates. However, theoretical predictions suggest that underemployment may deprive workers of meaningful work (Blustein, 2008; Jahoda, 1981). Therefore, we hypothesized that underemployment and meaningful work would be negatively related (Hypothesis 2). Finally, as suggested by latent deprivation theory, underemployment may reduce the experience of meaningful work, which subsequently could lead to poorer well-being (Steger et al., 2012). Although this is a mediator hypothesis, meaningful work may be better positioned as a moderator between underemployment and well-being. Several studies have found meaningful work to buffer the relations between risk factors and well-being (Allan, Douglass, Duffy, & McCarty, 2016; Lobene & Meade, 2013). Moreover, unlike people who are unemployed, many underemployed people may be able to maintain meaning in their work, which could function as a protective factor. Therefore, we predicted that meaningful work would buffer the relation between underemployment and each of the well-being variables, such that people high in meaningful work would have no relation between underemployment and well-being (Hypothesis 3).

Method

Participants

The sample consisted of 638 adults with ages ranging from 19 to 83 (M = 38.43, SD = 12.49). In terms of gender, participants self-identified as female (n = 346, 54.1%), male (n = 290, 45.3%), transgender (n = 1, 0.2%), androgynous (n = 1, 0.2%), and genderqueer (n = 1, 0.2%). In terms of ethnicity, participants mainly self-identified as European American (n = 495, 77.3%), with remaining participants identifying as African American (n = 60, 9.4%), Latinx American (n = 36, 5.6%), Asian American (n = 33?, 5.2%), Native American (n = 4, 0.6%), Arab American (n = 1, 0.2%), and multiracial (n = 10, 1.6%). In terms of education, 9.53% (n = 61) had a high school education or less, 5.8% (n = 37) had a trade or vocational school diploma, 25.6% (n = 164) had some college, 44.1% (n = 282) had a college degree, and 14.8% (n = 95) had a professional or graduate degree. Of the sample, 71.3% (n = 456) of participants were employed full time, 15.9% (n = 102) were employed part time, 5.9% (n = 38) were self-employed full time, 6.3% (n = 40) were self-employed part time, and 0.5% (n = 3) responding with other. The sample captured a wide range of occupations with 302 unique job titles represented. The most frequently reported job titles were salesperson (3.9%, n = 25), manager (2.4%, n = 15), teacher (2.4%, n = 15), cashier (1.7%, n = 11), administrative assistant (1.6%, n = 10), customer service representative (1.6%, n = 10), accountant (1.4%, n = 9), and office manager (1.4%, n = 9).

Instruments

Underemployment

Underemployment was measured with the 9-item Scale of Perceived Overqualification (Maynard et al., 2006). Participants responded on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items included “My job requires less education than I have” and “I have job skills that are not required for this job.” Item responses were summed to calculate total scores, with higher scores indicating greater perceived overqualification. Maynard, Joseph, and Maynard (2006) found an estimated internal consistency of α = .89. They also found scores on the measure to correlate in the expected direction with facets of work satisfaction and other measures of overqualification. The estimated internal consistency for the present study was α = .93.

Meaningful work

Meaningful work was measured using the 10-item Work as Meaning Inventory (WAMI; Steger et al., 2012). Sample items include “I have found a meaningful career” and “The work I do serves a greater purpose.” Participants answered items on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). Points from each item were summed to calculate a total score, with higher scores representing higher levels of meaningful work. Steger, Dik, and Duffy (2012) found the scale to correlate in the expected direction with variables such as career commitment, presence of life meaning, job satisfaction, and calling. Furthermore, Steger et al. (2012) found the WAMI to have an internal consistency reliability of α = .93. The estimated internal consistency for the present study was α = .94.

Positive affect and negative affect

Positive affect and negative affect were measured with the 20-item Positive and Negative Affect Schedule (Watson, Clark, & Tellegren, 1988). The scale presents participants with a number of positive and negative emotions such as “excited,” “enthusiastic,” “distressed,” and “upset.” Respondents then rate to what extent they had felt each emotion over the past week on a 5-point scale ranging from 1 (very slightly or not at all) to 5 (extremely). The 10 positive and 10 negative emotion items are summed to create respective subscales. For the Positive Affect subscale, Watson, Clark, and Tellegren (1988) reported internal consistency reliabilities ranging from α = .86 to α = .90 and test–retest reliabilities ranging from r = .47 to r = .68, depending on the time frame specified. For the Negative Affect subscale, Watson and colleagues (1988) reported internal consistency reliabilities ranging from α = .84 to α = .87 and test–retest reliabilities ranging from r = .39 to r = .71, depending on the time frame specified. Furthermore, Watson and colleagues (1988) found the subscales to correlate in the expected directions with general distress, depression, and other measures of affect and mood. The estimated internal consistencies for the Positive Affect and Negative Affect subscales in the present study were α = .92 and α = .93, respectively.

Life satisfaction

Life satisfaction was measured by the Satisfaction With Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985). The SWLS is composed of 5 items answered on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items include, “The conditions of my life are excellent” and “I am satisfied with my life.” Items were summed, with higher scores indicating greater perceived life satisfaction. Regarding validity, the SWLS has positively correlated with other measures of life satisfaction and positive affect (Diener et al., 1985). Regarding reliability, item responses on the SWLS have demonstrated high internal reliability, with an estimated internal consistency of α = .91 in a sample of U.S. working adults (Duffy, Allan, Autin, & Bott, 2013). The estimated internal consistency in the present study was .93.

Stress

Stress was measured with the 10-item Perceived Stress Scale (PSS-10; Cohen, Kamarck, & Mermelstein, 1983). Participants responded on a 5-point scale ranging from 0 (never) to 4 (very often). Sample items included, “In the last month, how often have you been upset because of something that happened unexpectedly?” and “In the last month, how often have you felt that you were on top of things?” Item responses were summed to calculate total scores, with higher scores indicated greater perceived stress. Cohen and Williamson (1988) found an estimated internal consistency of α = .78. PSS scores were correlated with reports of the amount of stress experienced during an average week and the amount of stress experienced now as opposed to a year ago. PSS scores are also moderately correlated with the number of stressful life experienced in the past year as well as how often people use health services (Cohen et al., 1983). The estimated internal consistency for the present study was α = .83.

Depression

Depression was measured by the 20-item Center for Epidemiological Studies—Depression Scale (CES-D; Radloff, 1977). For this measure, participants indicated how often they experienced a number of depression symptoms in the previous month on a scale ranging from “‘rarely or none of the time (less than 1 day)” to “all of the time (5–7 days).” Sample items include: “I was bothered by things that don’t usually bother me” and “I felt depressed.” Item responses summed to create a depression score, with higher scores indicated more symptoms. Radloff (1977) found an estimated internal consistency of α = .85 in the general population sample and α = .90 in the psychiatric patient sample. Radloff (1977) also reported test–retest reliabilities ranging from r = .32 (1 year) to r = .67 (1 month). CES-D scores also correlated in the expected direction with positive and negative affect, general psychopathology, and other measures of depression. The estimated internal consistency for the present study was α = .94.

Procedure

Participants joined the study through Mechanical Turk (MTurk). MTurk is an online participant source that allows people to complete surveys for monetary compensation, although most respondents report completing surveys for enjoyment (Buhrmester, Kwang, & Gosling, 2011). Recent reviews and studies examining MTurk have largely concluded that it produces valid data that are comparable to laboratory and other Internet recruitment methods (Buhrmester et al., 2011; Paolacci, Chandler, & Ipeirotis, 2010; Sprouse, 2011). Results across studies using a variety of sampling methods (e.g., laboratory-based, students samples, MTurk) often find comparable results and relations among variables, with some differences depending on the comparison sample (Casler, Bickel, & Hackett, 2013; D. R. Johnson & Borden, 2012; Sprouse, 2011). A link including an informed consent document and the survey itself was posted on MTurk, and participants self-selected themselves into the study. However, to join the study, participants had to (a) be over the age of 18, (b) reside within the United States, (c) not be a full-time student, and (d) be employed at least part time. Participants were given US$0.50 for completing the survey, which is consistent with typical amounts offered to participants of survey research on MTurk.

The initial sample consisted of 798 people. However, 31 only provided demographic information, 34 did not meet inclusion criteria (i.e., unemployed, full-time student, or unable to work), and 92 did not respond correctly to 3 embedded validity items (e.g., “Please select ‘strongly agree’ for this statement”). We removed these cases. Finally, as noted below, we removed two cases as outliers. This left a final sample size of 638. Of this sample, 634 (99.4%) participants had complete data and 4 (0.6%) were missing data on one study variable. Given the small amount of missing data, we removed cases pairwise for each analysis.

Results

Preliminary Analyses

No variables had scores greater than 3.5 standard deviations from the mean, except for negative affect, which had two extreme scores (z = 3.79 and z = 3.55). These cases were removed from the data set as outliers. No variables had absolute values of skewness or kurtosis greater than 1, except for negative affect that was slightly over 1 (skewness = 1.08). Therefore, we did not transform any study variables. Table 1 displays the correlations among the study variables. Underemployment was significantly related to all study variables, including meaningful work (−.45), life satisfaction (−.30), positive affect (−.19), negative affect (.16), depression (.25), and stress (.28).

Table 1.
Correlations and Descriptive Statistics for Study Variables.

Variable 1 2 3 4 5 6 7

1. Underemployment —

2. Meaningful work −.45* —

3. Life satisfaction −.30* .51* —

4. Positive affect −.19* .54* .52* —

5. Negative affect .16* −.28* −.38* −.27* —

6. Depression .25* −.40* −.56* −.46* .74* —

7. Stress .28* −.38* −.56* −.38* .66* .82* —

Mean 40.74 33.46 22.10 32.62 17.99 36.04 26.89

Standard deviation 13.28 9.84 8.09 9.02 8.26 12.73 7.35

Possible range 9–63 10–50 5–40 10–50 10–50 20–80 10–50

p < .01.

Moderation

To assess moderation, we used the PROCESS macro for SPSS Version 23.0 (Hayes, 2015). This macro allows users to probe the significance of interaction effects and also provides all relevant tests such as the omnibus F test, R ² increase due to interaction, and a Johnson–Neyman significance region. We z-scored underemployment and meaningful work, which reduced the multicollinearity of interaction terms by centering the variables. We then entered the predictors into a series of regressions predicting the five outcome variables. Table 2 displays the output for each outcome.

Table 2.
Meaningful Work as a Moderator of Underemployment and Well-Being.

Variable β SE B* R R ² R ²Δ F

Life satisfaction

Underemployment* −.09 .04

Meaningful work .46 .04

Interaction .03 .03 .52 .27 .001 74.60

Positive affect

Underemployment .10 .04

Meaningful work .61 .04

Interaction −.13 .03 .56 .31 .02 94.05

Negative affect

Underemployment .03 .04

Meaningful work −.28 .04

Interaction* .08 .04 .30 .09 .01* 20.33

Depression

Underemployment .08 .04

Meaningful work −.38 .04

Interaction* .09 .04 .42 .17 .01* 43.37

Stress

Underemployment .12 .04

Meaningful work −.35 .04

Interaction .10 .04 .41 .17 .01 42.13

*p < .05. **p < .01.

Life satisfaction

Meaningful work positively predicted life satisfaction (β = .46, p < .001) and underemployment negatively predicted life satisfaction (β = .09, p < .05). However, the interaction between meaningful work and underemployment did not predict life satisfaction (β = .03, p = .35). The variables in the regression explained 26.75% of the variance in life satisfaction.

Positive affect

Both meaningful work (β = .61, p < .001) and underemployment (β = .10, p < .01) positively predicted positive affect. However, these relations were qualified by a significant interaction between meaningful work and underemployment (β = −.13, p < .001). As seen in Figure 1, people low in meaningful work had a positive relation between underemployment and positive affect; however, people with moderate or high levels of meaningful work did not have a relation between underemployment and positive affect. The final regression explained 31.27% of the variance in life satisfaction.

Figure 1.
The moderation of underemployment and positive affect by meaningful work.

Negative affect

Meaningful work negatively predicted negative affect (β = .28, p < .001), but underemployment did not significantly predict negative affect (β = .03, p = .55). Regardless, these relations were qualified by a significant interaction between meaningful work and underemployment (β = .08, p < .01). As seen in Figure 2, people high in meaningful work had a positive relation between underemployment and negative affect; however, people with moderate or low levels of meaningful work did not have a relation between underemployment and negative affect. The final regression explained 8.93% of the variance in negative affect.

Figure 2.
The moderation of underemployment and negative affect by meaningful work.

Depression

Meaningful work negatively predicted depression (β = −.38, p < .001), but underemployment did not significantly predict depression (β = .08, p = .06). Regardless, these relations were qualified by a significant interaction between meaningful work and underemployment (β = .09, p < .05). As seen in Figure 3, people high in meaningful work had a steep positive relation between underemployment and depression; however, people with moderate and low levels of meaningful work had smaller positive relations between underemployment and negative affect. The final regression explained 17.32% of the variance in depression.

Figure 3.
The moderation of underemployment and depression by meaningful work.

Stress

Meaningful work negatively predicted stress (β = −.35, p < .001), and underemployment positively predicted stress (β = .12, p < .01). However, these relations were qualified by a significant interaction between meaningful work and underemployment (β = .10, p < .05). As seen in Figure 4, people with high and moderate levels of meaningful work had moderate positive relations between underemployment and stress; however, people with low levels of meaningful work had a small positive relation between underemployment and stress. The final regression explained 16.89% of the variance in stress.

Figure 4.
The moderation of underemployment and stress by meaningful work.

Discussion

The goals of this study were to evaluate the relations between underemployment and well-being, examine the relation between underemployment and meaningful work, and determine if meaningful work moderated the relation between underemployment and well-being. Based on latent deprivation theory, we predicted that meaningful work would serve as a buffer between underemployment and indicators of well-being, specifically, life satisfaction, positive affect, negative affect, depression, and stress. Results indicated that underemployment was linked to lower well-being and meaningful work and that meaningful work moderated the relation between underemployment and well-being. However, rather than serving as a buffer, meaningful work acted as a risk factor: People high in meaningful work had positive relations between underemployment and negative affect, depression, and stress. Additionally, meaningful work moderated the relation between underemployment and positive affect such that those low in meaningful work had a positive relation between underemployment and positive affect.

Supporting our first hypothesis, underemployment was associated with lower life satisfaction and positive affect and greater negative affect, stress, and depression. This is consistent with previous research linking underemployment to poorer well-being and greater psychological distress (Bracke & von dem Knesebeck, 2013; Cassidy & Wright, 2008; Friedland & Price, 2003; Lobene & Meade, 2013; Roh et al., 2014) and restates the potential negative effects of underemployment on par or on a continuum with unemployment (Dooley, 2003). Also supporting hypotheses, underemployment had a large, negative relation with meaningful work, which is consistent with theory (Jahoda, 1981). These findings justified our prediction that meaningful work would buffer the relation between underemployment and well-being.

Although meaningful work moderated the relations between underemployment and positive affect, negative affect, depression, and stress, they were in the opposite way that we predicted: People with meaningful work had a negative association between underemployment and well-being. There may be several explanations for these unexpected results, but the most likely may be that people who are underemployed in meaningful work become disillusioned. As discussed above, people who are underemployed are overqualified: They are not able to fully use their education, experience, and skills at work. People who have high meaningful work also deeply care about their work and are committed to doing a good job (e.g., Bunderson & Thompson, 2009). Therefore, people who find their work meaningful may be increasingly frustrated and disillusioned by underemployment, because they are not using their full abilities at a job they care about and view as important.

Although this is speculation about an unexpected finding, it points to a potential “dark side” of meaningful work, which is usually a protective factor positively related to well-being (e.g., Allan et al., 2016). This is similar to recent research finding, a downside to calling, which is a variable closely related to meaningful work (Duffy, Douglass, & Autin, 2015). In this case, having a calling but being unable to live it out is associated with poorer outcomes (Duffy et al., 2015). Similarly, the present study suggests that having meaningful work but being unable to fully employ one’s skills and abilities is associated with poorer outcomes. Broadly, these results suggest that variables typically positive for well-being can be harmful in certain circumstances. These situations may reflect cases where people have the presence of a positive state or situation but are frustrated in being able to live it out fully. Regardless, this opens up important avenues for future research.

Finally, meaningful work and underemployment significantly predicted life satisfaction but did not interact. This suggests that meaningful work and underemployment may have independent relations to life satisfaction. Again, this might be because the disillusionment caused by being underemployed in a meaningful job impact affective variables rather than life satisfaction in general. However, this is speculative, and caution should be used in interpreting this finding until more research is available.

Practical Implications

The results of this study have implications for career counselors working with people who are underemployed. As noted above, underemployment is pervasive and likely to remain a feature of the global economy well into the future (ILO, 2017). The ability of people to work full time in a satisfying job over many years has declined, particularly for marginalized clients with insufficient access to social, financial, and human capital (Das-Munshi, Leavey, Stansfeld, & Prince, 2012; ILO, 2017; Konrad, Moore, Ng, Doherty, & Breward, 2013; Slack & Jensen, 2011). Career counselors should be aware of this reality and understand underemployment’s likely impact on clients’ well-being. Moreover, counselors should know that helping a client find sufficient employment might mean they work multiple jobs or jobs that are not an ideal fit for their education, experiences, and skills. Empathizing with clients who face this reality is critical, but teaching clients effective employment strategies is also important to help them find employment opportunities when available.

The results of the current study also suggest that despite meaningful work’s relation to a host of well-being variables, it may exacerbate the negative impact of underemployment. Performing a simple assessment with all clients about their work experiences might help uncover these issues. Career counselors can ask their clients whether they believe they are overqualified for their current work and whether they find their work meaningful. Clients may have complex dynamics at play, perhaps believing that their work is central to their life meaning and makes a difference in the world but also feeling restricted in their ability to do more. Regardless, such an assessment may yield important information that can guide career counselors’ interventions to help clients cope or find better work. Of course, helping clients find meaningful work that uses their full abilities is an important goal, but validating clients’ experiences and acknowledging the constraints and barriers they face are also critical—not all clients will have the same access to opportunities or the same ability to choose their work (Blustein, 2008). Some clients may need more access to community resources, and others may need to find opportunities to use their skills or find meaning outside of work.

Limitations and Future Directions

The results and conclusions from this study need to be considered in light of a number of limitations. First, the study was cross-sectional, making it impossible to determine causality between variables. For example, having poorer well-being may lead to underemployment. Although this hypothesis is generally not supported in the literature (Dooley et al., 2000), well-being and underemployment may share some third variable predictor such as economic conditions (ILO, 2017). Regardless, longitudinal studies would help clarify the direction of the relations. Second, the sample was collected through an online data collection site, MTurk. There are several studies that demonstrate the validity of this method of data collection (Buhrmester et al., 2011, Sprouse, 2011), but every sample has bias. Specifically, the sample gathered for this study was highly educated, which likely reflects disparities in who has access to the Internet and the time to complete surveys for enjoyment (Rhodes, Bowie, & Hergenrather, 2003). Also, participants self-selected into the study, which further obscures generalizability. Therefore, this study should be complemented by other studies using a variety of samples. Finally, this study did not control for the possible influence of other variables described in latent deprivation theory, such as structure and status. It will be important that future research examines the relations between underemployment and well-being while accounting for all relevant latent consequences of work.

Variable	1	2	3	4	5	6	7
1. Underemployment	—
2. Meaningful work	−.45*	—
3. Life satisfaction	−.30*	.51*	—
4. Positive affect	−.19*	.54*	.52*	—
5. Negative affect	.16*	−.28*	−.38*	−.27*	—
6. Depression	.25*	−.40*	−.56*	−.46*	.74*	—
7. Stress	.28*	−.38*	−.56*	−.38*	.66*	.82*	—
Mean	40.74	33.46	22.10	32.62	17.99	36.04	26.89
Standard deviation	13.28	9.84	8.09	9.02	8.26	12.73	7.35
Possible range	9–63	10–50	5–40	10–50	10–50	20–80	10–50

Variable	β	SE B	R	R ²	R ²Δ	F
Life satisfaction
Underemployment*	−.09	.04
Meaningful work**	.46	.04
Interaction	.03	.03	.52	.27	.001	74.60**
Positive affect
Underemployment**	.10	.04
Meaningful work**	.61	.04
Interaction**	−.13	.03	.56	.31	.02**	94.05**
Negative affect
Underemployment	.03	.04
Meaningful work**	−.28	.04
Interaction*	.08	.04	.30	.09	.01*	20.33**
Depression
Underemployment	.08	.04
Meaningful work**	−.38	.04
Interaction*	.09	.04	.42	.17	.01*	43.37**
Stress
Underemployment**	.12	.04
Meaningful work**	−.35	.04
Interaction**	.10	.04	.41	.17	.01**	42.13**

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Allan

B. A.

Dexter

Kinsey

Parker

(2018). Meaningful work and mental health: Job satisfaction as a moderator. Journal of Mental Health, 27, 38–44. doi:10.1080/09638237.2016.1244718

Allan

B. A.

Douglass

R. P.

Duffy

R. D.

McCarty

(2016). Meaningful work as a moderator of the relation between work stress and meaning in life. Journal of Career Assessment, 24, 429–440. doi:10.1177/1069072715599357

Allan

B. A.

Duffy

R. D.

Blustein

D. L.

(2016). Under (and over) employment: Measurement and correlates of employment discrepancy. The Counseling Psychologist, 44, 815–840. doi:10.1177/0011000016654766

Allan

B. A.

Duffy

R. D.

Collisson

(2017). Task significance and performance: Meaningfulness as a mediator. Journal of Career Assessment. Advance online publication. doi:10.1177/1069072716680047

Allan

B. A.

Duffy

R. D.

Douglass

(2015). Meaning in life and work: A developmental perspective. The Journal of Positive Psychology, 10, 323–331. doi:10.1080/17439760.2014.950180

Allan

B. A.

Tay

Sterling

H. M.

(2017). Construction and validation of the Subjective Underemployment Scales (SUS). Journal of Vocational Behavior, 99, 93–106. doi:10.1016/j.jvb.2017.01.001

Ariely

Kamenica

Prelec

(2008). Man’s search for meaning: The case of Lego’s. Journal of Economic Behavior & Organization, 67, 671–677. doi:10.1016/j.jebo.2008.01.004

Benedict

J. N.

Gayatridevi

Velayudhan

(2009). Perceived overqualification, job satisfaction, somatization and job stress of MNC executives. Journal of the Indian Academy of Applied Psychology, 35, 283–289.

Blustein

D. L.

(2008). The role of work in psychological health and well-being: A conceptual, historical, and public policy perspective. American Psychologist, 63, 228. doi:10.1037/0003-066X.63.4.228

10.

Bowling

N. A.

Eschleman

K. J.

Wang

(2010). A meta-analytic examination of the relationship between job satisfaction and subjective well-being. Journal of Occupational and Organizational Psychology, 83, 915–934. doi:10.1348/096317909X478557

11.

Bracke

Pattyn

von dem Knesebeck

(2013). Overeducation and depressive symptoms: Diminishing mental health returns to education. Sociology of Health & Illness, 35, 1242–1259. doi:10.1111/1467-9566.12039

12.

Brynjolfsson

McAfee

(2014). The second machine age: Work, progress, and prosperity in a time of brilliant technologies. New York, NY: W. W. Norton.

13.

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6, 3–5. doi:10.1177/1745691610393980

14.

Bunderson

J. S.

Thompson

J. A.

(2009). The call of the wild: Zookeepers, callings, and the double-edged sword of deeply meaningful work. Administrative Science Quarterly, 54, 32–57. doi:10.2189/asqu.2009.54.1.32

15.

Casler

Bickel

Hackett

(2013). Separate but equal? A comparison of participants and data gathered via Amazon’s MTurk, social media, and face-to-face behavioral testing. Computers in Human Behavior, 29, 2156–2160. doi:10.1016/j.chb.2013.05.009

16.

Cassidy

Wright

(2008). Graduate employment status and health: A longitudinal analysis of the transition from student. Social Psychology of Education, 11, 181–191. doi:10.1007/s11218-007-9043-x

17.

Cohen

Kamarck

Mermelstein

(1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396.

18.

Cohen

Williamson

(1988). Perceived stress in a probability sample of the US. In Spacapam

Oskamp

(Eds.), The social psychology of health: Claremont symposium on applied social psychology (pp. 31–67). Thousand Oaks, CA: Sage.

19.

Creed

P. A.

Machin

M. A.

(2002). The relationship between mental health and access to the latent benefits of employment for unemployed and underemployed individuals. Psychological Reports, 90, 1208–1210. doi:10.2466/pr0.2002.90.3c.1208

20.

Das-Munshi

Leavey

Stansfeld

S. A.

Prince

M. J.

(2012). Migration, social mobility and common mental disorders: Critical review of the literature and meta-analysis. Ethnicity & Health, 17, 17–53. doi:10.1080/13557858.2011.632816

21.

Diener

E. D.

Emmons

R. A.

Larsen

R. J.

Griffin

(1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71–75. doi:10.1207/s15327752jpa4901_13

22.

Dooley

(2003). Unemployment, underemployment, and mental health: Conceptualizing employment status as a continuum. American Journal of Community Psychology, 32, 9–20. doi:10.1023/A:1025634504740

23.

Dooley

Prause

Ham-Rowbottom

K. A.

(2000). Underemployment and depression: Longitudinal relationships. Journal of Health and Social Behavior, 41, 421–436.

24.

Duffy

R. D.

Allan

B. A.

Autin

K. L.

Bott

E. M.

(2013). Calling and life satisfaction: It’s not about having it, it’s about living it. Journal of Counseling Psychology, 60, 42. doi:10.1037/a0030635

25.

Duffy

R. D.

Bott

E. M.

Allan

B. A.

Torrey

C. L.

Dik

B. J.

(2012). Perceiving a calling, living a calling, and job satisfaction: Testing a moderated, multiple mediator model. Journal of Counseling Psychology, 59, 50. doi:10.1037/a0026129

26.

Duffy

R. D.

Dik

B. J.

Steger

M. F.

(2011). Calling and work-related outcomes: Career commitment as a mediator. Journal of Vocational Behavior, 78, 210–218. doi:10.1016/j.jvb.2010.09.013

27.

Duffy

R. D.

Douglass

R. P.

Autin

K. L.

(2015). The dark side of callings. In Yaden

D.B.

McCall

T. D.

Ellens

J. H.

(Eds.) Being called: Secular, sacred, and scientific perspectives. (pp. 13–25). Santa Barbara, CA: Praeger.

28.

Erdogan

Bauer

T. N.

(2009). Perceived overqualification and its outcomes: The moderating role of empowerment. Journal of Applied Psychology, 94, 557–565. doi:10.1037/a0013528

29.

Feldman

D. C.

(1996). The nature, antecedents and consequences of underemployment. Journal of Management, 385–407. doi:10.1177/014920639602200302

30.

Friedland

D. S.

Price

R. H.

(2003). Underemployment: Consequences for the health and well-being of workers. American Journal of Community Psychology, 32, 33–45. doi:10.1023/A:1025638705649

31.

Hayes

A. F.

(2015). An index and test of linear moderated mediation. Multivariate Behavioral Research, 50, 1–22.

32.

International Labor Organization. (2017). World employment and social outlook: Trends 2017. Geneva, Switzerland: Author.

33.

Jahoda

(1981). Work, employment, and unemployment: Values, theories, and approaches in social research. American Psychologist, 36, 184. doi:10.1037/0003-066X.36.2.184

34.

Johnson

D. R.

Borden

L. A.

(2012). Participants at your fingertips using Amazon’s Mechanical Turk to increase student–faculty collaborative research. Teaching of Psychology, 39, 245–251. doi:10.1177/0098628312456615

35.

Johnson

G. J.

Johnson

W. R.

(1996). Perceived overqualification and psychological well-being. The Journal of Social Psychology, 136, 435–445. doi:10.1080/00224545.1996.9714025

36.

Johnson

G. J.

Johnson

W. R.

(2000). Perceived overqualification, positive and negative affectivity, and satisfaction with work. Journal of Social Behavior and Personality, 15, 167–184.

37.

Konrad

A. M.

Moore

M. E.

E. W.

Doherty

A. J.

Breward

(2013). Temporary work, underemployment and workplace accommodations: Relationship to well-being for workers with disabilities. British Journal of Management, 24, 367–382. doi:10.1111/j.1467 8551.2011.00809.x

38.

Lobene

E. V.

Meade

A. W.

(2013). The effects of career calling and perceived overqualification for primary and secondary school teachers. Journal of Career Development, 40, 508–530. doi:10.1177/0894845313495512

39.

Lobene

E. V.

Meade

A. W.

Pond

S. I.

(2015). Perceived overqualification: A multi-source investigation of psychological predisposition and contextual triggers. The Journal of Psychology: Interdisciplinary and Applied, 149, 684–710. doi:10.1080/00223980.2014.967654

40.

Maynard

D. C.

Brondolo

E. M.

Connelly

C. E.

Sauer

C. E.

(2015). I’m too good for this job: Narcissism’s role in the experience of overqualification. Applied Psychology: An International Review, 64, 208–232. doi:10.1111/apps.12031

41.

Maynard

D. C.

Joseph

T. A.

Maynard

A. M.

(2006). Underemployment, job attitudes, and turnover intentions. Journal of Organizational Behavior, 27, 509–536. doi:10.1002/job.389

42.

McKee-Ryan

F. M.

Harvey

(2011). “I have a job, but…”: A review of underemployment. Journal of Management, 37, 962–996. doi:10.1177/0149206311398134

43.

Paolacci

Chandler

Ipeirotis

P. G.

(2010). Running experiments on Amazon Mechanical Turk. Judgment and Decision making, 5, 411–419.

44.

Paul

K. I.

Batinic

(2010). The need for work: Jahoda’s latent functions of employment in a representative sample of the German population. Journal of Organizational Behavior, 31, 45–64. doi:10.1002/job.622

45.

Radloff

L. S.

(1977). The CES-D scale a self-report depression scale for research in the general population. Applied psychological measurement, 1, 385–401. doi:10.1177/014662167700100306

46.

Rhodes

S. D.

Bowie

D. A.

Hergenrather

K. C.

(2003). Collecting behavioural data using the World Wide Web: Considerations for researchers. Journal of Epidemiology and Community Health, 57, 68–73. doi:10.1136/jech.57.1.68

47.

Roh

Y. H.

Chang

J. Y.

Kim

M. U.

Nam

S. K.

(2014). The effects of income and skill utilization on the underemployed’s self-esteem, mental health, and life satisfaction. Journal of Employment Counseling, 51, 125–141. doi:10.1002/j.2161-1920.2014.00047.x

48.

Segerstrom

S. C.

Miller

G. E.

(2004). Psychological stress and the human immune system: A meta-analytic study of 30 years of inquiry. Psychological Bulletin, 130, 601–630. doi:10.1037/0033-2909.130.4.601

49.

Slack

Jensen

(2011). Underemployment among minorities and immigrants. In Maynard

D. C.

Feldman

D. C.

(Eds.), Underemployment: Psychological, economic, and social challenges (pp. 127–143). New York, NY: Springer Science + Business Media.

50.

Sprouse

(2011). A validation of Amazon Mechanical Turk for the collection of acceptability judgments in linguistic theory. Behavior Research Methods, 43, 155–167. doi:10.3758/s13428-010-0039-7

51.

Steger

M. F.

Dik

B. J.

Duffy

R. D.

(2012). Measuring meaningful work: The Work and Meaning Inventory (WAMI). Journal of Career Assessment, 20, 322–337. doi:10.1177/1069072711436160

52.

Thompson

K. W.

(2014). Antecedents and consequences of subjective underemployment in organizations: Measure development and model test (Doctoral dissertation). Retrieved from ProQuest. (3596594).

53.

Wanberg

C. R.

(2012). The individual experience of unemployment. Annual Review of Psychology, 63, 369–396. doi:10.1146/annurev-psych-120710-100500

54.

Warner

R. M.

Rasco

(2014). Structural equation models for prediction of subjective well-being: Modeling negative affect as a separate outcome. The Journal of Happiness & Well-Being, 2, 34–50.

55.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063. doi:10.1037/0022-3514.54.6.1063

56.

Winefield

A. H.

(2002). Unemployment, underemployment, occupational stress and psychological well-being. Australian Journal of Management, 27, 137–148. doi:10.1177/031289620202701S14

57.

Luksyte

Parker

S. K.

(2015). Overqualification and subjective well-being at work: The moderating role of job autonomy and culture. Social Indicators Research, 121, 917–937. doi:10.1007/s11205-014-0662-2