This is an open access article distributed under the terms of the Creative Commons
Attribution License (In recent years, research on adult attachment has provided a new and fruitful theoretical approach to investigate individual differences in psychological health and well-being not explained by traditional measures of personality or dispositional traits (e.g., Harms, 2011; Mikulincer & Florian, 2001). The present study extends this line of enquiry by examining the role of attachment style in predicting well-being at the workplace.
Nowadays, we witness an upsurge of interest in the predictors and correlates of the psychological well-being that we experience in relation to that part of life that we spend at work (Danna & Griffin, 1999; Hartel, Zerbe, & Ashkanasy, 2005; Harter, Schmidt, & Keyes, 2003). Research approaches on workers' well-being arise from the idea that positive feelings and positive perceptions result in behavioral, cognitive, and health benefits (Isen, 1987; Warr, 1999). The presence of positive emotional states and positive appraisals enhance workers' performance and the quality of their work and life (Harter, Schmidt, & Keyes, 2003). For example, workers experiencing poor health and well-being conditions in the workplace are more likely to be absent from work, be less productive, make bad decisions (Boyd, 1997), and, overall, make consistently declining contributions to the organization (Price & Hooijberg, 1992). Additionally, any physical, emotional, mental, or social experience that people live at work surely affects them while they are at the workplace. However, since workers spend about one-third of their daily time at work, and, when they leave the work site, they don’t necessarily leave their job behind (Conrad, 1988a), the impact of such work-setting experiences goes beyond the work domain.
Work-Related Well-Being, Job Satisfaction and Job Involvement
For many applied researchers, job satisfaction and psychological well-being are constructs that are closely linked together (Bowling, Eschleman, & Wang, 2010; Gechman & Wiener, 1975; Rahimnia & Sharifirad, 2014; Simon, Judge, & Halvorsen-Ganepola, 2010). Indeed the most common means of operationalizing the happiness or the psychological well-being components have been through the assessment of job satisfaction (Rahimnia & Sharifirad, 2014; Wright & Cropanzano, 2000). As much as about a fifth of the variation in adult life satisfaction - as an index of psychological well-being - can be accounted for by satisfaction with work (Campbell, Converse, & Rodgers, 1976). Job satisfaction can facilitate the experience of positive cognitions relevant to life satisfaction (Bowling et al., 2010; Simon et al., 2010). In the literature, satisfaction and felt emotions are typically construed as the end result of valenced events: when something pleasant happens, the person feels satisfaction, happiness and joy; when something unpleasant happens, the person experiences dissatisfaction, anger, depression or fear. Applied to the work context, this means that employees will feel happy when work conditions are good and unhappy when work conditions are bad (Lucas & Diener, 2003). The equation of 'happy' or 'psychologically healthy worker' with 'satisfied worker' is implicit or explicit in several conceptualizations of job satisfaction. One of the most widely used definitions of job satisfaction in organizational research is by Locke (1976), who defines it as "a pleasurable or positive emotional state resulting from the appraisal of one's job or job experiences" (p. 1304). A more recent definition is from Hulin and Judge (2003), who describe job satisfaction as an emotional feeling individuals have about their job, reflecting the degree of pleasure or happiness their job induces in general.
The concept of job involvement too is viewed as strictly associated to work-related well-being (Gechman & Wiener, 1975; Riipinen, 1997). Despite differences in its conceptualization (Kanungo, 1982; Lodahl & Kejner, 1965; Saleh & Hosek, 1976), job involvement is usually understood as referring to the psychological and emotional extent to which an employee is engaged in, and enthusiastic about, performing his/her work. It refers to the way a person looks at her job as a relationship with the working environment and the job itself. Brown and Leigh (1996) actually suggested an operational definition of psychologically healthy climate based on how employees perceive aspects of their organizational environment and interpret them in relation to their own well-being.
An interesting area of work-related research focuses on individual differences in employees' health and well-being, investigating emotional dispositions and work-related traits, such as workers’ personality traits (Burns & Machin, 2013; Judge, Heller, & Mount, 2002), workers’ attempt to regulate inappropriate felt and/or expressed emotions (i.e., emotional labor), i.e., when they conflict with internalized norms or job requirements (Pisaniello, Winefield, & Delfabbro, 2012; Zammuner & Galli, 2005a, 2005b), workers’ specific dispositions and ways of dealing and managing emotions, hypothesized to differentially influence well-being (Zammuner, 2011, 2012a, 2012b), or job-related variables and work orientation (Rahimian, Nouri, Oreyzi, Moulavi, & Foroughi Mobarakeh, 2006; Zammuner & Kafetsios, 2005; Zammuner, Lanciano, Casnici, Cappellato, & Prencipe, 2011).
In sum, according to the literature both job involvement and job satisfaction represent crucial indexes of work-related well-being.
Adult Attachment Styles, Workplace and Work-Related Well-Being
Although in recent years our understanding of the cognitive and emotional processes involved in psychological health and well-being has significantly been increased by adult attachment research (e.g., Mikulincer & Florian, 2001), few studies have focused on the role that attachment styles play as regards workplace situations and relationships (Littman-Ovadia, Oren, & Lavy, 2013). As a new research agenda emphasizes, working is intrinsically a relational act, performed within interpersonal contexts and relationships (Blustein, 2011; Bowen, Siehl, & Schneider, 1989). Each decision, experience, or interaction within the work setting is understood, inclined, and bent by such relationships. As a consequence, Blustein (2011) encouraged researchers to examine whether and how individual differences in relational functioning are predictive of work-related attitudes and behaviors (Bowen et al., 1989; Harms, 2011; Hazan & Shaver, 1990; Kark, 2011; Malach-Pines, 2005; Popper, 2004; Richards & Schat, 2011).
Attachment styles reflect internal working models of self, others, and relationships (e.g., Bartholomew & Horowitz, 1991) and express individuals’ motivations, abilities, and perceptions as regards relationships (Harms, 2011). Thus, conceptualizing work as a relational setting highlights how individual differences in relationship orientations are relevant for understanding individuals’ work attitudes and emotions. Based on these premises, a growing body of studies has been showing the influence of attachment style in shaping people’s work-related behaviors, motivations, attitudes, and emotional responses (e.g., Harms, 2011; Hazan & Shaver, 1990; Richards & Schat, 2011). Studies have shown that secure attachment facilitates work-related exploration activities, being positively associated with career exploration (Littman-Ovadia, 2008), ability to negotiate (Ketterson & Blustein, 1997), adaptive work-related adjustment (Blustein, Prezioso, & Schultheiss, 1995), progress in career decision making (Hazan & Shaver, 1990), and confidence in receiving good evaluations by colleagues (Blustein et al.,1995). Conversely, insecure attachment orientations (both avoidant and anxious attachment styles) were shown to correlate with lower levels of organizational commitment, of pro-social and productive behaviours, and of organizational commitment (Mikulincer & Shaver, 2007).
However, in the work setting avoidant individuals differ from anxious individuals in some important ways. Avoidant individuals evaluate themselves as lower in job performance and in how attractive their colleagues judge them to be (Hazan & Shaver, 1990). Avoidant individuals also exhibit more conflict with co-workers, more relational difficulties outside of work (Hardy & Barkham, 1994), stronger intentions to leave their job and their organization (Mikulincer & Shaver, 2007), lower levels of instrumental and emotional support-seeking, and greater use of surface acting as a means of regulating emotional displays at work (Richards & Schat, 2011). Anxious individuals, on the other hand, expect to be undervalued by coworkers (Hazan & Shaver, 1990) and are anxious about relationships at work and job performance (Hardy & Barkham, 1994). Attachment anxiety has a negative association with instrumental coworker-helping behaviors (Geller & Bamberger, 2009), and with higher levels of both instrumental and emotional support seeking (Richards & Schat, 2011). These findings may reflect anxious individuals’ insecurity and fear of rejection, which make it difficult for them to feel more emotionally committed and to provide unconditional assistance. Their insecurity and low self-worth seem to affect similarly their relationships at work and other relationships (Mikulincer & Shaver, 2007).
As regards work-related well-being, the literature provides empirical evidences about the direct or indirect role of adult attachment. To illustrate, in a broad survey of the workplace, Hazan and Shaver (1990) found that securely attached individuals reported significantly higher satisfaction with most aspects of their workplace (e.g., coworkers, job security, recognition, etc.). Similarly, in a sample consisting mostly of computer software workers, securely attached individuals reported higher levels of satisfaction for their job and various aspects of it (Krausz, Bizman, & Braslavsky, 2001). Likewise, in a large sample of university employees, securely attached individuals reported higher levels of job satisfaction than anxiously attached individuals (Sumer & Knight, 2001). Conversely, individuals higher on anxious and avoidant attachment were less likely to report being satisfied with various aspects of their jobs in a study by Hardy and Barkham (1994). Rahimnia and Sharifirad (2014) showed that attachment insecurity mediated the relationship between authentic leadership and employee well-being in terms of job satisfaction. Schirmer and Lopez (2001) found that the interaction of supervisor support and worker attachment orientation significantly predicted work stress intensity as well as job satisfaction. Likewise, Reizer (2014) showed that job satisfaction plays a mediator role in the association between adult attachment and workers’ well-being. Ronen and Mikulincer (2012) indicated that both managers' and subordinates' attachment insecurities predicted lower job satisfaction among workers. Finally, Rahimian et al. (2006), investigating the relationship between adult attachment styles and work orientation and work involvement in state hospital nurses of Isfahan, showed that individuals with avoidant and ambivalent attachment styles exhibited less work involvement than subjects with a secure attachment style, highlighting the power of attachment style to predict work-related attitudes and behaviors in organizational contexts.
Jointly considered, the above cited results encouraged us to investigate the hypothesis that adult attachment style has an important role in predicting work-related well-being assessed in terms of job satisfaction and job involvement.
Aims and Hypotheses
The present study adopted a descriptive correlational design in order to examine the role of attachment style in predicting work-related well-being, over and above dispositional trait measures traditionally associated with it (emotional traits and work-related traits; Pisaniello et al., 2012; Rahimian et al., 2006; Zammuner, 2012a, 2012b; Zammuner & Galli, 2005a, 2005b; Zammuner et al., 2011). On the basis of the aforementioned literature, we operationalized the work-related well-being construct in terms of Job involvement and Job satisfaction (Bowling et al., 2010; Gechman & Wiener, 1975; Rahimnia & Sharifirad, 2014; Riipinen, 1997; Simon et al., 2010). Based on findings of the above quoted literature, we expected that:
H1. Secure attachment style will be positively associated with positive emotional traits, positive work-related traits, and with work-related well-being (job involvement and job satisfaction), and negatively associated with negative emotional traits and negative work-related traits.
H2. Both anxious and avoidant attachment styles will be negatively associated with positive emotional traits, positive work-related traits, and work-related well-being (job involvement and job satisfaction), and positively associated with negative emotional traits and negative work-related traits.
H3. Dispositional trait measures, assessed in terms of emotional traits and work-related traits, will be associated with work-related well-being (job involvement and job satisfaction).
H4. Attachment styles will explain a statistically significant increment of variance in work-related well-being (job involvement and job satisfaction), over and above emotional and work-related traits.
Method
Participants
We surveyed 527 Italian volunteers (62% female; Mage = 28.13; SD = 8.72) employed in a wide variety of jobs, obtaining a heterogeneous sample that however represents a range of occupations and spans the adult age range. The sample was a convenience sample, recruited through advertisements, and among friends and acquaintances of the researchers and their students and collaborators.
Procedure
Participants were individually administered a paper-and-pencil test battery containing all instruments detailed below.
Measures
Attachment Style
The Relationship Questionnaire (RQ; Bartholomew & Horowitz, 1991) was used to assess attachment orientations. It comprises a brief-sentence prototypical-description of each of four adult attachment orientations, i.e., secure, fearful, preoccupied, dismissing; e.g., “I am comfortable without close emotional relationships. It is very important to me to feel independent and self-sufficient, and I prefer not to depend on others or have others depend on me”. The four orientations are rated on a seven-point scale. Following Griffin and Bartholomew (1994), the four adult attachment prototypes were converted into two insecure attachment dimensions representing anxious and avoidant attachment (Lanciano, Curci, Kafetsios, Elia, & Zammuner, 2012). The anxiety dimension was computed by subtracting the sum of secure and dismissing scores from the sum of fearful and preoccupied scores. The avoidance dimension resulted from subtracting the sum of secure and preoccupied scores from the sum of fearful and dismissing scores. To meet the study aims, the a) secure, b) anxious, and c) avoidant attachment styles were considered in the statistical analyses.
Emotional Traits
A twenty-three item scale, rated on 6-point false/true of myself, assessed optimism, emotional awareness and emotional expressivity as emotional traits. As the factorial structure of this scale and the reliability of its components was assessed in other studies (e.g., Zammuner, 2011, 2012a, 2012b) involving a large sample of young adults, both students and workers (N = 2079; henceforth LS), we thought it advisable to test it anew for this workers sample. We thus run an item-level Exploratory Factorial Analysis (EFA), using the principal axis extraction method. Examining the scree plot and retaining the factors corresponding to the first larger eigenvalues, until the slope of the graph changed from rapid to slow decline determined the number of factors. In the overall analysis oblique axes were determined using the Promax rotation method. The first eigenvalues were: 3.57, 2.30, 1.84, 1.41, and 1.19, suggesting a four-factor solution. The first factor, loading on the first dimension, named optimism (2 items; e.g., I always look on the bright side of things), accounted for almost 15.50% of the total variance; Cronbach’s α = .51 (LS α = .69; at re-test α = .74; Zammuner, 2012b). The second factor, named emotional awareness (5 items; e.g., I am aware of non-verbal signals that I send to others), accounted for 10% of the total variance; Cronbach’s α = .54. The third factor, named emotional unawareness (5 items; e.g., Sometimes I feel emotions which I do not understand thoroughly) accounted for 8.02% of the total variance; Cronbach’s α = .66. The fourth factor, named emotional expressivity (11 items; e.g., When I am angry, others can see it), accounted for 6.11% of the total variance; Cronbach’s α = .74. (As regards the latter three factors, note that for the young adult sample LS, the factorial structure resulted in a 10-items awareness-unawareness dimension, Cronbach’s α = .73; at re-test α = .77, and in a 10-item transparent-opaque expressivity Cronbach’s α = .79; at re-test α = .82; Zammuner, 2012a, 2012b).
Emotional Labor
To assess frequency of emotion-expression regulation at work, participants rated how they usually expressed their emotions at the workplace - on five 5-point items (1 = Never, 5 = Always). The scale measures emotional labor in terms of surface acting (or shallow regulation; Grandey, 1998; Zammuner & Galli, 2005a, 2005b; Cronbach’s α = .75). An item example is: “I fake a good mood”.
Orientation at the Workplace
Individuals’ orientation towards a variety of aspects of their work was assessed by eighteen 5-point scale items (0 = False, 5 = True). Subjects judged how much the orientation described by each sentence generally described their own behavior or their attitude (Zammuner & Kafetsios, 2005; Zammuner et al., 2011). The factorial structure of the scale was assessed through the EFA, using the principal axis extraction method and the Promax rotation method. The first eigenvalues were: 3,72, 2.60, 1.41, and 1.10, thus suggesting a two-factor solution. The first factor accounted for almost 20.67% of the total variance and loaded on the first dimension named work-management orientation (10 items; Cronbach’s α = .76; e.g., “After I have made a decision, usually I go back to make sure I was right”). The second factor loaded on the second dimension, accounting for 14.44% of the total variance, named relationship orientation (8 items; Cronbach’s α = .69; e.g., “I have a good relationship with my supervisor”).
Job Involvement
Workers’ involvement with their job was assessed by eight 5-point items (1 = Totally disagree, 5 = Totally agree (Galli & Zammuner, 2004; Kanungo, 1982; Zammuner & Galli, 2005a, 2005b; Cronbach’s α = .84). Item examples are: “Most of my interests are centered around my job”; “To me, my job is only a small part of who I am”.
Job Satisfaction
Workers’ job satisfaction was assessed by four 5-point items (1 = False of myself, 5 = True of myself; Brayfield & Rothe, 1951; Cronbach’s α = .72). Item examples are: “I find my job very enjoyable”; “Very often my job is exciting”.
Data Analyses
As the emotional traits and work-related trait measures are scales for which empirical evidence is limited to unpublished manuscripts in relation to the Italian context, factor analyses were performed on these scales to assess their dimensional structure; EFA and reliability results (including previous unpublished reliability values) are reported in the Measures section. Descriptive and reliability analyses were run on all measures. Cronbach’s α values obtained in our study for each scale are reported above in the Measures section. Zero-order correlation analyses assessed a) the associations between attachment styles and emotional traits, work-related traits, and work-related well-being; and b) the associations between emotional traits, work-related traits, and work-related well-being. Hierarchical multiple regression analyses tested a) the extent to which all measured variables predicted work-related well-being, and b) the role of attachment styles in explaining an incremental variance in work-related well-being, over and above dispositional traits measures.
Results
Descriptive and Correlation Analyses
Table 1 displays the descriptive analyses outcomes of all measures. Mean ratings showed that participants exhibited high secure attachment; as regards insecure attachment styles, participants exhibited higher avoidant than anxious tendencies (t = -4.00, p < .001). Mean ratings furthermore showed that individuals reported high levels of positive emotional traits (including optimism, emotional awareness, and emotional expressivity), medium level of emotional labor, and higher levels of relationship orientation than work-management orientation (t = 13.15, p < .001). Finally, participants overall reported medium to high levels of work-related well-being (job involvement and job satisfaction).
Table 1 also reports a) the results of the zero-order correlation analyses between attachment styles and emotional traits, work-related traits, and work-related well-being indices; and b) the results of the zero-order correlation analyses between emotional traits, work-related traits, and work-related well-being indices.
The correlation results showed that secure attachment appeared to be positively associated with emotional unawareness, emotional expressivity, work-management orientation, relationship orientation, and job involvement, and negatively associated with emotional labor. Instead, anxious attachment was negatively correlated with optimism, emotional expressivity, work-management orientation, and job satisfaction, and positively associated with emotional unawareness. Avoidant attachment exhibited negative associations with relationship orientation and job involvement.
Furthermore, relationship orientation appeared to be positively correlated with job involvement, whereas optimism and emotional unawareness showed respectively a positive and a negative association with job satisfaction.
Table 1
Descriptive Statistics and Correlations Among Measures
| Measures | M | SD | SE | AN | AV | OP | EA | EU | EE | EL | WMO | RO |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Attachment style | ||||||||||||
| SE secure | 2.77 | 1.47 | ||||||||||
| AN anxious | -.35 | 2.93 | ||||||||||
| AV avoidant | .51 | 2.98 | ||||||||||
| Emotional traits | ||||||||||||
| OP optimism | 3.06 | .93 | -.01 | -.12** | .06 | |||||||
| EA emotional awareness | 3.48 | .61 | .05 | -.01 | -.06 | |||||||
| EU emotional unawareness | 2.91 | .83 | .17** | .12** | .01 | |||||||
| EE emotional expressivity | 3.29 | .59 | .14** | -.14** | -.02 | |||||||
| Work-related traits | ||||||||||||
| EL emotional labor | 2.25 | .75 | -.12** | .10* | -.02 | |||||||
| WMO work-management orientation | 3.52 | .59 | .18** | -.11* | -.05 | |||||||
| RO relationship orientation | 3.10 | .61 | .14** | .04 | -.13** | |||||||
| Work-related well-being | ||||||||||||
| JI job involvement | 2.52 | .71 | .20** | .07 | -.11* | .04 | .06 | .06 | -.01 | .06 | .02 | .20** |
| JS job satisfaction | 3.45 | .78 | -.07 | -.12** | -.03 | .21** | .04 | -.30** | .04 | -.05 | .07 | -.07 |
*p < .05. **p < .001.
The Prediction of Work-Related Well-Being
To test a) the extent to which the measured variables - i.e., socio-demographic characteristics, emotional traits, work-related traits, and attachment styles - predicted work-related well-being, and b) the attachment style’s increment of variance in work-related well-being over and above the other measures, we ran several hierarchical multiple regression model (HMR) analyses. In each HMR, the dependent variable was the examined work-related well-being index (job involvement and job satisfaction), with gender (male = 1; female = 2) and age entered at Step 1, emotional traits at Step 2, work-related traits at Step 3, and attachment styles at Step 4.
For the job involvement index, as shown in Table 2, at Step 1 the model is significant (F = 3.19, p < .05) with gender as the only significant predictor. At Step 2, when emotional traits are introduced, the model and the incremental change are not significant (F = 1.99, p = .07; R2 change = .01, p = .24). At Step 3, when work-related traits are added, the model and the incremental change are significant (F = 3.30, p < .001; R2 change = .03, p < .001), with gender and work-management orientation as significant predictors. At Step 4, when attachment styles are introduced in the HRM, the model and the incremental change are significant (F = 4.60, p < .001; R2 change = .04, p < .001), with gender, work-management orientation, and secure and anxious attachment styles as significant predictors.
Table 2
Hierarchical Multiple Regression of Job Involvement
| Measures | Step 1
|
Step 2
|
Step 3
|
Step 4
|
||||
|---|---|---|---|---|---|---|---|---|
| β | p | β | p | β | p | β | p | |
| Age | -.11 | .01 | -.13 | .01 | -.10 | .04 | -.09 | .04 |
| Gender | -.01 | .78 | -.01 | .77 | -.02 | .65 | -.02 | .63 |
| Step 2 | ||||||||
| Optimism | .02 | .63 | .02 | .74 | .04 | .43 | ||
| Emotional awareness | .06 | .21 | .03 | .58 | .03 | .50 | ||
| Emotional unawareness | .08 | .07 | .07 | .14 | .01 | .79 | ||
| Emotional expressivity | -.00 | .97 | .02 | .75 | .01 | .91 | ||
| Step 3 | ||||||||
| Emotional labor | .04 | .40 | .05 | .27 | ||||
| Work-management orientation | -.03 | .61 | -.04 | .41 | ||||
| Relationship orientation | .18 | .00 | .15 | .00 | ||||
| Step 4 | ||||||||
| Secure attachment | .23 | .00 | ||||||
| Anxious attachment | .14 | .01 | ||||||
| Avoidant attachment | .03 | .55 | ||||||
| R2 | .01 | .04 | .01 | .06 | .04 | .00 | .08 | .00 |
| ΔR2 | .01 | .24 | .03 | .00 | .04 | .00 | ||
For the job satisfaction index, as shown in Table 3, at Step 1 the model is not significant (F = 2.54, p = .08). At Step 2, when emotional traits are introduced, the model and the incremental change are significant (F = 14.05, p < .001; R2 change = .13, p < .001), with gender, age, optimism, and emotional unawareness as significant predictor. At Step 3, when work-related traits are added, the incremental change is not significant (F = 9.82, p < .001; R2 change = .01, p = .27). At Step 4, when attachment styles are introduced in the HRM, the model and the incremental change are significant (F = 8.30, p < .001; R2 change = .02, p < .05), with gender, age, optimism, emotional unawareness, anxious and avoidant attachment styles as significant predictors.
Table 3
Hierarchical Multiple Regression of Job Satisfaction
| Measures | Step 1
|
Step 2
|
Step 3
|
Step 4
|
||||
|---|---|---|---|---|---|---|---|---|
| β | p | β | p | β | p | β | p | |
| Age | .03 | .50 | .11 | .01 | .09 | .04 | .10 | .02 |
| Gender | .10 | .03 | .01 | .02 | .10 | .02 | .09 | .03 |
| Step 2 | ||||||||
| Optimism | .19 | .00 | .19 | .00 | .19 | .00 | ||
| Emotional awareness | .02 | .70 | .02 | .66 | .02 | .69 | ||
| Emotional unawareness | -.30 | .00 | -.30 | .00 | -.26 | .00 | ||
| Emotional expressivity | -.02 | .60 | -.04 | .41 | -.05 | .30 | ||
| Step 3 | ||||||||
| Emotional labor | -.02 | .72 | -.02 | .71 | ||||
| Work-management orientation | .05 | .29 | .04 | .37 | ||||
| Relationship orientation | -.08 | .08 | -.07 | .10 | ||||
| Step 4 | ||||||||
| Secure attachment | -.10 | .06 | ||||||
| Anxious attachment | -.15 | .00 | ||||||
| Avoidant attachment | -.13 | .01 | ||||||
| R2 | .01 | .08 | .13 | .00 | .13 | .00 | .14 | .00 |
| ΔR2 | .13 | .00 | .01 | .27 | .02 | .02 | ||
Jointly considered, the regression analyses showed that emotional traits (especially optimism and emotional unawareness), work-related traits (especially work-management orientation) and attachment styles (secure, anxious, and avoidant attachment) play a role in defining the level of work-related well-being - as assessed by job involvement and job satisfaction. As hypothesized, attachment style moreover explained a statistically significant increment of variance in work-related well-being indices. Age overall induced in older participants, in comparison to younger ones, higher levels of job satisfaction. Concerning the role of gender, women exhibited higher levels of both job involvement and job satisfaction than men did.
Discussion
Attachment theory, although it is generally explored in personality research, has received little attention from researchers investigating the role of individual differences in the work setting. As a review by Harms (2011) points out, a few studies only have examined the role of attachment styles in relation to a variety of behaviors, attitudes, and experiences in the workplace setting, primarily addressing issues related to leader–follower dynamics, job attitudes and stress, and job performance outcomes. With the aim of integrating theories of attachment with workplace theories and findings, the present study investigated the role of adult attachment style in predicting work-related well-being, over and above dispositional trait measures traditionally associated with well-being in the literature (Pisaniello et al., 2012; Rahimian et al., 2006; Zammuner, 2012a, 2012b; Zammuner & Galli, 2005a, 2005b; Zammuner et al., 2011). Work-related well-being was measured in terms of job involvement and job satisfaction (Bowling et al., 2010; Gechman & Wiener, 1975; Rahimnia & Sharifirad, 2014; Riipinen, 1997; Simon et al., 2010).
Confirming Hypotheses 1 and 2, correlation findings showed that secure attachment was positively associated with positive emotional and work-related traits, whereas insecure attachment styles (especially anxious attachment) were negatively correlated with dispositional trait measures. More specifically, both attachment styles – secure and anxious - were positively associated with emotional unawareness. According to attachment theory (Bowlby, 1969/1982, 1973), a secure attachment leads to constructive affect regulation and emotional awareness (Berlin & Cassidy, 2003; Fonagy, Gergely, Jurist, & Target, 2002). As regards anxious attachment, people with strong dependence have less difficulty recognizing and expressing their own affective experience. The higher distress level expressed by anxiously attached individuals, as well as their fear of abandonment and their need to solicit relationships to avoid abandonment representations, are linked to a greater awareness of their own feelings and to their expression in order to attract attention from others (Fantini-Hauwel, Boudoukha, & Arciszewski, 2012). For anxiously attached people, emotional awareness is a key feature for obtaining support and care from others (Cassidy, 1994).
Concerning work-related well-being, in support of Hypotheses 2 and 3, job involvement was positively associated with secure attachment and relationship orientation, and negatively associated with avoidant attachment, whereas job satisfaction was positively associated with positive emotional traits, and negatively associated with anxious attachment.
Confirming Hypothesis 4, regression results showed that attachment styles explained a significant incremental variance in work-related well-being, over and above dispositional trait measures associated to well-being at the workplace. Secure and anxious attachment explained the job involvement level experienced by workers: Higher levels of security and lower levels of anxiety predicted workers’ greater involvement in their job. As any job requires interactions with others (colleagues, work team, clients, customers, employer, etc.), proximity and intimacy with others represent crucial and unavoidable aspects of a person's work life. It may thus be argued that individual (relational) differences in terms of attachment styles affect well-being at the workplace (e.g., Harms, 2011; Hazan & Shaver, 1990; Richards & Schat, 2011). On the other hand, the work-role attachment theory (Carter & Cook, 1995) suggests that the degree to which individuals are committed to their work-role influences their desire to remain a member of the workforce. According to this theory, job involvement refers to an individual’s affective attachment to a particular job (Carter & Cook, 1995) and reflects the degree to which individuals view their job as a central part of their life (Adams, Prescher, Beehr, & Lepisto, 2002). This work-role attachment theoretical perspective might be integrated by arguing that secure and not very anxious workers are likely to exhibit higher levels of involvement and commitment in their job. Anxiously attached individuals, on the other hand, have a negative view of the self, leading to an obsessive need for assurance from others, overdependence, hypervigilance to social and emotional cues from others, fear of rejection, and to general preoccupation and anxiety with and about relationships (Fraley, Niedenthal, Marks, Brumbaugh, & Vicary, 2006; Mikulincer & Shaver, 2005, 2007). Such features prevent them from being fully engaged and involved in their job.
Our study also showed that both anxious and avoidant attachment styles explained the degree of job satisfaction: lower levels of either anxiety or avoidance (or both) predicted higher levels of well-being in terms of job satisfaction. These findings might be partially interpreted recalling that individuals characterized by avoidant attachment view others as unavailable, unresponsive, or punishing (Bowlby, 1973; Mikulincer & Shaver, 2005), and thus distance themselves from others, attempt to avoid emotional closeness and intimacy, and deny their need for proximity (Mikulincer & Shaver, 2005). Avoidant individuals have furthermore a negative view of others, leading to obsessive self-reliance, independence, and difficulties in trust and dependence on others (Mikulincer & Shaver, 2007). However, that both insecure (anxious and avoidant) attachment styles negatively predicted job satisfaction seem to suggest that other variables may mediate these relationships, or are necessary to fully explain them. A supervisor’s emotional and/or instrumental support may be such a mediator candidate to explain how attachment style is related to this well-being index. For instance, Harms (2011) argued that high anxiously-attached workers reported levels of satisfaction similar to those with low anxious attachment when supervisor’s support was available; when support was lacking, they reported significantly lower job satisfaction. Interestingly, individuals high on avoidant attachment reported significantly higher job satisfaction when supervisor support was low. In sum, to fully account for such unexpected results it might be necessary to focus on the different behaviors and reactions that insecure individuals have in different work contexts.
Our study, in line with results obtained in previous studies (Borys & Perlman, 1985; Clark, Oswald, & Warr, 1996), showed that age and gender too contributed to some extent to participants’ evaluation of their work-related well-being. Our results showed that older workers experienced more job satisfaction compared to younger participants, and that women exhibited higher levels of work-related well-being than men did.
Finally, in accordance with our expectations, attachment explained a statistically significant increment of variance in work-related well-being in terms either of job involvement or of job satisfaction.
In conclusion, the present study shows that the manner in which people experience interpersonal relationships affects their well-being, highlighting the importance of individual attachment-related characteristics to better understand several aspects of people's behavior and attitudes in the workplace. If we accept the premises that a person's attachment style influences her ability to deal with intimate relationships, and that the workplace is essentially a relational context, then we can argue that a person's attachment style plays a crucial role in her workplace too, with workers characterized by high-secure and low-anxious attachment being generally more involved in their job, and workers characterized by low-anxious and low-avoidant attachment being more satisfied in their job.
Our study is not however without limitations and several options are open for further investigation. First, our findings are based on a mostly correlational study, so caution should be taken when inferring conclusions. Second, the adopted emotional-trait measures showed low reliability values for the sample in our study (especially Optimism). Third, some of our measures (emotional traits and work-related traits) were validated in the Italian context only, making it difficult to compare results across studies. The findings obtained in our study thus need to be replicated both with other employee samples, and in studies that include other well-established emotional trait measures - e.g., measures of emotional intelligence and competence, including emotion regulation abilities and empathy (Gresham & Gullone, 2012; Kafetsios, Athanasiadou, & Dimou, 2014; Lanciano et al., 2012; Wei, Liao, Ku, & Shaffer, 2011; Zammuner, Dionisio, Prandi, & Agnoli, 2013). Fourth, a more exhaustive assessment of well-being at the workplace might be obtained by employing other measures, such as measures of job importance, job withdrawal, work burnout, occupational stress (Littman-Ovadia et al., 2013). Fifth, a better prediction of well-being might be obtained by controlling or modelling other variables likely to be associated with it - e.g., marital status, socio-economic status, and preferred coping styles or affect regulation strategies (e.g., Lopez, Mauricio, Gormley, Simko, & Berger 2001; Wei, Vogel, Ku, & Zakalik, 2005), perhaps even including personal attachment-history variables, such as attachment in relation to parent- and peer-interaction in adolescence (e.g., Ma & Huebner, 2008). Sixth, when looking at the incremental validity of attachment styles at work, further interaction variables might be usefully controlled for, such as type of work (e.g., jobs requiring interactions vs. individual and independent jobs). Finally, since gender has been shown to influence psychological well-being, with women reporting more symptoms of lower well-being than men (although the magnitude was fairly small; Danna & Griffin, 1999; Harris, Heller, & Braddock, 1988; Schaubroeck & Jones, 2000), the relationship of attachment with well-being at work might be better explored by explicitly modeling gender differences too.
In sum, although attachment styles have been studied across a variety of workplace phenomena, it is increasingly obvious that a large amount of work still needs to be done, many specific issues remain to be investigated. The present study is a (not exhaustive) attempt to address the issue of individual differences in work-related well-being from an attachment dispositional point of view, and, despite its limitations, adds to our knowledge of salient variables and processes that underlie well-being in work settings.