Depressive Symptomatology, Attachment Style, Job Insecurity and Burnout of Civil Servants in the Greek Economic Crisis

Maria Grigoriadou*a, George Kleftarasa


The purpose of the present study was to investigate the depressive symptomatology of civil servants in contemporary Greek society undergoing severe turmoil because of the profound economic crisis. More specifically, the study aimed at evaluating the severity of depression in this population in relation to attachment style, job insecurity and the three dimensions of burnout. The sample of the present study consisted of 207 employees working in Municipalities or Civil Services. Self-report questionnaires on depressive symptomatology, attachment style, job insecurity and burnout were distributed. According to the results, 38.20% of the sample indicated mild or severe depressive symptomatology. The hypothesis of significant correlations among depressive symptomatology, job insecurity and the dimensions of burnout were confirmed. Results also indicated significant correlations between depressive symptomatology and attachment styles involving a negative self model. Implications for counseling are discussed.

Keywords: depression, attachment, job insecurity, burnout, civil servants

The European Journal of Counselling Psychology, 2017, Vol. 6(1), doi:10.5964/ejcop.v6i1.126

Received: 2016-05-10. Accepted: 2017-07-18. Published (VoR): 2017-08-31.

*Corresponding author at: Department of Special Education, University of Thessaly, Volos 382 21, Greece. E-mail:

This is an open access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Since 2009, after a series of years of economic and social growth, Greek economy has been facing serious turmoil. In 2010, the Greek Parliament sanctioned a Memorandum of Economic and Financial Policies between the Greek government, the E.U. and the International Monetary Fund. According to the Annual Report of the Bank of Greece (2011), the Greek Gross National Product declined 4.5%, while unemployment reached 14.2% in the last quarter of the same year. Various studies have pointed out a link between economic variables and mental health. Subjective measures of financial well-being and unemployment seem to be related to the onset of depression and deterioration of physical and mental health (Bridges & Disney, 2010). Similar results were also found by Pickett and Wilkinson (2010) who claim that there is a definite link between mental health and income inequalities and that low social status may lead to psychological distress. In a study exploring changes in the Greek population between 2008 and 2011, the relationship between financial hardship and the onset of depression was further confirmed when the possibility of the sample suffering from major depression was found to be 2.6 times bigger in 2011 (Economou, Madianos, Peppou, Patelakis, & Stefanis, 2013).

Although the link between economic crisis and depression has been considerably discussed in literature, to the best of our knowledge, it has not been investigated in the population of civil servants. Moreover, the present study takes into consideration the factors of job insecurity, burnout and attachment style when examining the depressive symptoms of this specific population in contemporary Greece. The choice of these particular factors was based on the following facts: Several studies have shown that job insecurity is associated with both economic uncertainty (Adkins, Werbel, & Farh, 2001; Linz & Semykina, 2008) and depression (Burgard, Brand, & House, 2009; Witte, 1999). In the years of crisis, civil servants are exposed to stressful factors and dramatic changes affecting every aspect of their work life: working conditions, status and salary. Burnout is associated with exposure to stress in the working place (Maslach & Jackson, 1981) as well as depression (Ahola et al., 2005; Bellani et al., 1996; McKnight & Glass, 1995). Greek public sector in the years of economic turmoil, offered a new, interesting area to study these relationships. Finally, certain types of attachment seem to have an effect on depression (Abdul Kadir & Bifulco, 2013; Conradi & de Jonge, 2009; Cooley, Van Buren, & Cole, 2010). From a counseling point of view, there is no control over the actual economic circumstances that cause feelings of job insecurity and burnout. On the other hand, counseling can empower employees in order to face these feelings, change what can be changed and tolerate the rest. Good working and personal relationships could be a powerful means of empowerment. Understanding the way people relate to others could be useful in the process of improving their relationships. Therefore, counselors working with employees could benefit from examining whether these employees display a negative model of self, and the impact of this view on their working behaviour. In such cases, counseling could become the remedying experience that would help clients change the way they relate to others (Kafetsios, 2005) and thus become more resilient against the negative psychological consequences of economic hardship.

Attachment was defined by Bowlby (1982) as the behavior an infant exhibits in order to establish and preserve a bond of love and care with someone he/she considers important for his/her survival. Infants’ instinctive reactions such as smiling, crying and gripping are attempts to bond with the mother or the caregiver. The quality of this first bond in a child’s life leads to the formulation of mental representations or internal working models of the self and significant individuals and affects all future relationships (Bowlby, 1973, 1980). Ainsworth and her colleagues (Ainsworth, Bell, & Stayton, 1971) studied different kinds of infants’ attachment with the caregiver and identified different forms of communication, emotion regulation, and different responses to threatening situations. Based on the three styles of attachments in infancy, Hazan and Shaver (1987) studied attachment styles in adult life. A few years later, Bartholomew (1990) combined the internal working model of the self (worthy of love vs. not worthy) and the other (available vs. unreliable and rejecting) in order to describe four attachment patterns: The secure (positive model of self and other), the preoccupied (negative model of self, positive model of other), the fearful (negative model of self and other) and the dismissing (positive model of self, negative model of other). Several studies have shown that attachment style has an effect on depression. Insecure attachment styles and in particular attachment styles involving a negative model of self, have been identified as a vulnerability factor for depression (Abdul Kadir & Bifulco, 2013; Conradi & de Jonge, 2009; Cooley, Van Buren, & Cole, 2010).

Job insecurity, another psychological factor investigated in this study, describes the uncertainty about the continuity of a person’s current job, the probability of involuntary loss of desirable aspects of the job and the way people perceive this lack of assurance (Green, 2011). Research has shown that job insecurity is associated with an individuals’ employability (Green, 2011; Peiró, Sora, & Caballer, 2012), tenure status (Adkins, Werbel, & Farh, 2001) and personality characteristics (Näswall, Sverke, & Hellgren, 2005; Sverke & Hellgren, 2002). Moreover, job insecurity is consistent with the general conditions in the labor market and is increased in years of major economic uncertainty (Adkins, Werbel, & Farh, 2001; Linz & Semykina, 2008). As far as the link between job insecurity and depression is concerned, surveys indicate that long term job insecurity can significantly predict poorer self-rated health and depressive symptoms (Burgard, Brand, & House, 2009; Witte, 1999). Similarly, Ferrie and his colleagues (Ferrie, Shipley, Stansfeld, & Marmot, 2002) confirmed the negative link between job insecurity and mental health, while Hellgren and Sverke (2003) confirmed the hypothesis that job insecurity is a predictive factor of depression.

Burnout, the last construct investigated in this study, is a response to prolonged exposure to stressful conditions in the working place and consists of three factors: emotional exhaustion, cynicism or depersonalization, and inefficacy (Maslach & Jackson, 1981). Exhaustion is characterized by a lack of energy and feelings of emotional depletion. Exhausted employees feel that they are unable to offer, at a psychological level, to people relevant to their work; they assume that their expectations have not been met, and they are unable to perform their job as reliably as they did. Depersonalization refers to a distant and cynic attitude expressed with a lack of politeness or unwillingness to deliver suitable services. Employees adopt a negative and callous attitude towards their clients and colleagues and appear indifferent to their feelings. Finally, inefficacy is a person’s tendency to feel that he/she does not perform tasks adequately and is incompetent at work. Therefore, employees evaluate themselves negatively and as a result, their self-esteem decreases. Apart from the negative consequences in the workplace, burnout has also been found to have a detrimental effect on relationships and leads to a pessimistic attitude towards life in general (Iacovides, Fountoulakis, Kaprinis, & Kaprinis, 2003). From this point of view burnout has a lot in common with depression. Ahola and his colleagues (2005) studied this overlap between burnout and depressive disorders and came to the conclusion that higher levels of burnout indicate a higher probability of having a depressive disorder.

The purpose of this paper is to investigate depressive symptomatology in relation to attachment style, job insecurity and burnout in a sample of white collar employees working in the Greek public sector. A major limitation in previous research is that these factors have mainly been studied in samples of employees in health and education. Moreover, to our knowledge, attachment style has not been studied in relation to work environment. Hence, the need for further examination of the relationship between depressive symptomatology, attachment style, job insecurity and burnout in various populations is great and becomes even more important in the current socioeconomic conditions in Greece. More specifically, it is expected that (a) the subjects of the present study would manifest mild or high depressive symptomatology, job insecurity and burnout, (b) there are statistically significant relationships among depressive symptomatology, job insecurity and the three dimensions of burnout, (c) there are statistically significant differences among the four attachment types, with regards to the abovementioned variables, and finally, (d) job insecurity, the three dimensions of burnout and the type of attachment are predictive factors of depressive symptomatology. When employees face problems and feel unhappy, their performance does not meet their potential (Wentland, 2009). In these cases, counseling services could offer them a safe place where they would find new ways to deal with their problems in a non-judgmental, empathic environment. The counseling process has proven to be a valuable tool in helping not only employees to have better control over personal difficulties affecting their work, but also organizations to increase their productivity (Collins et al., 2012; Mcleod, 2008; Sharma, 2012). Since no similar study around civil servants has been undertaken, the present study enters a novel area of research.

Method [TOP]

Subjects and Procedure [TOP]

The population of this research was drawn from six mainland cities in Greece and was comprised of both executive and administrative employees in the public sector. A sample of 207 civil servants, 62 men (29.95%) and 145 women (70.05%) were studied. More than half of them (n = 114) were permanently appointed employees, whereas the rest of the participants had indefinite term contracts (n = 44), short term contracts (n = 42) or worked as interns (n = 7). The sample’s socio-demographic characteristics are presented in Table 1.

Table 1

Socio-Demographic Characteristics of the Sample (Ν = 207)

Characteristic n %
Up to 35 54 26.10
36 to 50 121 58.50
Over age 51 32 15.50
Marital Status
Single 50 24.20
Married 148 71.50
Single-Parent Family 9 4.30
Yes 141 68.10
No 66 31.90
Educational Level
Primary School/Junior High School 3 1.40
Senior High school 52 25.10
Technological Educational Institute 60 29.00
University 68 32.90
Postgraduate Studies 23 11.10
Other 1 0.50
Employment Contract
Permanent Employees 114 55.10
Indefinite Term Contracts 44 21.30
Short Term Contracts 42 20.30
Internships 7 3.40
Job Position
Employee 169 81.60
Head of Department 32 15.50
Director 6 2.90
Working Hours
Up to 40 hours per week 180 87.00
More than 40 hours per week 27 13.00
Professional Experience
Up to 15 years 138 66.70
16 to 25 years 40 19.30
26 years or more 29 14.00

Four hundred questionnaires were distributed in March 2013. They were either given personally or posted accompanied by a letter informing the participants on the purpose of the study and ensuring anonymity and confidentiality. Within a period of two months, 213 questionnaires were returned, but six of them were not included in the results, as they were not completed sufficiently.

Measures [TOP]

Except for the short demographic survey which was conducted to provide information on the characteristics of the subjects, as well as on their working conditions, participants completed the following four instruments.

Questionnaire of Self Evaluated Depressive Symptomatology (QD2) [TOP]

The QD2 was designed to assess depressive symptomatology in adults and derived from the content analysis of four inventories assessing depression: (a) the Beck Depression Inventory (BDI), (b) the Depression Scale of the Minnesota Multiphasic Personality Inventory (MMPI-D), (c) the Zung Self-Rating Depression Scale (SDS) and (d) the Hopkins Symptoms Check List (Pichot et al., 1984). It was constructed in France and consists of 52 items with good concurrent validity, good homogeneity and good reliability ranging from .92 to .95 (Kleftaras, 1991, 1997, 2000; Pichot et al., 1984). It was adapted and translated into Greek through back translations (Kleftaras & Tzonichaki, 2012) and the Greek version has good reliability (Kleftaras, 2006; Kleftaras & Katsogianni, 2012; Kleftaras & Psarra, 2012). It also shows satisfactory factorial structure, confirming the results of the content analysis in France, which revealed three basic dimensions: feelings of loss of general drive, depressive-pessimistic mood and anxiety (Kleftaras, 2004; Kleftaras & Tzonichaki, 2012; Pichot et al., 1984)

Relationship Questionnaire [TOP]

This questionnaire aims to assess adult attachment within Bartholomew's (1990) four-category framework (Bartholomew & Horowitz, 1991). It is made up of four short paragraphs, each describing a prototypical adult attachment pattern. Its adaptation into Greek was based on a sample of 232 people between the ages of 18 to 67 (Kafetsios, 2012). Correlation between RQ and (a) relevant psychological constructs such as loneliness and social support and (b) other tests of attachment styles, such as Experiences in Close Relationships Scale, demonstrate good validity (Kafetsios, 2005). This questionnaire has been widely used in the study of adult attachment (e.g. Whiffen, Kallos-Lilly, & MacDonald, 2001).

Maslach Burnout Inventory (MBI) [TOP]

MBI was designed by Maslach and Jackson (1981) to assess the three dimensions of the burnout syndrome. It consists of 22 items; nine of them assess the factor of emotional exhaustion, eight assess efficacy and the remaining five evaluate the factor of depersonalization. Response options range on a 7-point Likert scale (Maslach & Jackson, 1986). High scores in emotional exhaustion and depersonalization indicate higher burnout, while high scores in efficacy indicate lower burnout and vice versa. MBI has been used all over the world and therefore allows comparisons across not only occupational groups but also nations (Schutte et al., 2000). The Greek version of MBI shows good construct validity and internal consistency reliability with .84 for exhaustion, .55 for depersonalization and .71 for efficacy (Papadatou, Anagnostopoulos, & Monos, 2012).

Minnesota Satisfaction Questionnaire – Job Security Subscale [TOP]

Minnesota Satisfaction Questionnaire was developed to measure a worker’s satisfaction in 20 specific aspects of work and work environment (Weiss, Dawis, England, & Lofquist, 1967). Job security was measured using a subscale that consists of five items rated in a 5-point Likert-type scale (from 1 = very dissatisfied to 5 = very satisfied) and has good reliability (.84). Data reflecting the postulated relationship with other scales (i.e. Minnesota Importance Questionnaire) support the construct validity of the MSQ (Weiss, Dawis, England, & Lofquist, 1967). The questionnaire’s reliability was also tested in the Greek version (Amarantidou, Mantis, & Koustelios, 2009; Koustelios, Kouli, & Theodorakis, 2003).

Results [TOP]

Since the validity of the QD2 has not been studied in a Greek population, it was deemed necessary to study its factorial structure based on our sample. Principle-components factor analysis using Varimax rotations was conducted and showed two of the three primary factors (feelings of loss of general drive and depressive-pessimistic mood) that were found in the study of Pichot et al. (1984), explaining 34.25% of the variance. The third factor of anxiety that was proposed by Pichot, did not emerge in the present analysis, possibly because of the small sample studied, or the particularities of the participants. In the present study, QD2 had reliability coefficients of .95 (Cronbach’s alpha) and .92 (split-half reliability)

Levels of Depressive Symptomatology [TOP]

In order to confirm the first hypothesis of our study, that there are significant levels of depressive symptomatology in civil servants, the sample was divided into three groups according to their depressive symptoms. QD2 constructors do not give clear cut-off depression scores. By using samples of normal and clinically depressed individuals, they propose a rough estimation of scores for low and high depression groups and indicate their approximate mean depression scores. However, this did not pose problems, because the purpose of our research was not to investigate the variables of the present study in a clinically diagnosed depressive population, but to study depressive symptomatology and its fluctuation in a normal sample of civil servants. Consequently, given that QD2 scores represent the number of an individual’s depressive symptoms, we divided the sample into three groups based on the number of symptoms (0-15: lower depressive symptoms group; 16-30: mild depressive symptoms group; 31+: higher depressive symptoms group). These groups do not consist of clinically depressed or not individuals, but of persons with more or fewer depressive symptoms. Therefore, based on the purpose of the present study, we consider that the above sample division was suitable for testing our hypotheses. The differences among these three groups were statistically significant, F(2, 204) = 529.01, p < .001. The first group consisted of 128 individuals (61.80%) of the lower depressive symptomatology, with a score ranging from 0-15, Μ = 6.19, SD = 4.73. The second group consisted of 55 individuals (26.60%) of mild depressive symptomatology, with a score between 16 and 30, M = 21.78, SD = 4.76, and the third of 24 individuals (11.60%) of the higher depressive symptomatology, with a score of 31+, M = 36.88, SD = 4.70. Consequently, the biggest part of the sample (128 out of 207 individuals) manifests low depressive symptomatology. Post hoc comparisons using the Scheffé post hoc criterion for significance indicated that depressive symptomatology levels in the first group were significantly lower than the second and the third group.

Levels of Burnout [TOP]

In order to confirm the hypothesis that the greatest part of the sample demonstrates mild or high burnout, the sample was divided into three groups according to their score in each of the three dimensions of burnout using the cut off points described by Maslach and Jackson in MBI’s manual. As far as emotional exhaustion is concerned, the first group consisted of individuals whose score in exhaustion varied between 0 and 16. The second group consisted of those with a score ranging from 17 to 26, while the third was composed of those who scored 27 or higher, demonstrating high emotional exhaustion, M = 35.24, SD = 6.93. The analysis of variance showed that the differences among the three groups were statistically significant, F(2, 204) = 454.31, p < .001.

The group demonstrating lower emotional exhaustion, Μ = 10.33, SD = 4.05, included 76 participants (36.70%), the group demonstrating mild emotional exhaustion, M = 21.61, SD = 3.00, included 64 participants (30.90%), and the group demonstrating higher emotional exhaustion, M = 35.24, SD = 6.93, contained 67 participants (32.40%). According to this analysis only 36.70% of the participants manifests low emotional exhaustion, while the rest 63.30% manifests mild or high emotional exhaustion, thus confirming our hypothesis. According to Scheffé post hoc criterion emotional exhaustion levels in the first group were significantly lower than the second and the third group.

As far as depersonalization is concerned, the first group consisted of 108 individuals (52.20%) indicating low depersonalization with scores ranging from 0 to 6, Μ = 2.77, SD = 2.10. The second group consisted of 46 individuals (22.20%) exhibiting mild depersonalization with scores ranging from 7 to 12, M = 9.65, SD = 1.69, while the third group consisted of 53 individuals (25.60%) showing high depersonalization with scores 13 or higher, M = 17.23, SD = 3.32. Once more the differences among the three groups were statistically significant, F(2, 204) = 659.18, p < .001. Therefore, in the case of depersonalization the results do not confirm the hypothesis, since 52.20% of the sample demonstrated low scores. Post hoc comparisons using the Scheffé post hoc criterion for significance indicated that depersonalization levels in the first group were significantly lower than the second and the third group.

Finally, analysis of variance showed statistically significant differences among the three efficacy groups, F(2, 204) = 614.32, p < .001. The group that demonstrated low burnout (scoring 39 or higher), accounted for 38.60% of the sample and included 80 individuals, M = 43.46, SD = 3.03, while the group that demonstrated mild burnout (scores 32 to 38) accounted for 30.00% of the sample and included 62 individuals, M = 35.11, SD = 2.11. The third group demonstrated high burnout (scores 0-31) accounted for 31.40% of the sample and included 65 individuals, M = 24.75, SD = 4.12. Hence, the results confirm the hypothesis, as 61.40% of the sample demonstrates mild or high burnout. Post hoc comparisons using the Scheffé post hoc criterion showed that burnout levels in the first group were significantly lower than the second and the third group.

Levels of Job Security [TOP]

The hypothesis concerning job security has also been confirmed since the larger part of the sample (74.90%) demonstrated low or mild job security. The sample was divided into three groups according to their job security. Scores were ranked in percentiles in order to determine cut off points. Low and high levels were decided upon the 25th and the 75th percentile respectively. The analysis of variance showed statistically significant differences among the three groups, F(2, 204) = 489.37, p < .001. The first group of lower job security, Μ = 1.85, SD = 0.40, consisted of 44 participants, the second group of mild job security, M = 3.07, SD = 0.35, consisted of 111 participants (53.60%) and the third group of higher job security, M = 3.97, SD = 0.20, consisted of 52 participants (25.10%). According to Scheffé post hoc criterion job security levels in the first group were significantly lower than the second and the third group.

Relations of Depressive Symptomatology With Job Security and Burnout [TOP]

Statistically significant correlations were found between job security and depressive symptomatology. Results showed that there is a negative correlation between the two variables, r = -.31, p < .001). It seems that as job security decreases, depressive symptomatology increases and vice versa (Table 2).

Table 2

Pearson Correlations Between Depressive Symptomatology, and a) Job Security b) Burnout

Variable Depressive Symptomatology
Job Security -.31****
  Emotional Exhaustion .53****
  Depersonalization .43****
  Efficacy -.25****

****p < .0001.

As for burnout, its three subscales (emotional exhaustion, depersonalization and efficacy) were correlated to depressive symptomatology. Particularly, there is a statistically significant negative correlation between efficacy and depressive symptomatology, r = -.25, p < .001. This indicates that, the higher a person’s efficacy, the lower his depressive symptomatology (Table 2). On the other hand, positive correlations emerged between depressive symptomatology and 1. emotional exhaustion, r = .53, p < .001, and 2. Depersonalization, r = .43, p < .001. It seems that as emotional exhaustion and depersonalization increase, depressive symptomatology increases as well and vice versa (Table 2).

Differences Among the Four Attachment Types [TOP]

According to the present study, only 35.70% of the participants rated their attachment type as secure despite the fact that in the general Greek and English population this percentage reaches 50% or more (Kafetsios, 2005). The preoccupied attachment type was the least dominant type at 15.50%, a fact that is in agreement with other studies, while the fearful and the dismissive types were at 16.40% and 32.40% respectively.

In addition, the differences on depressive symptomatology among the four attachment types were explored. The analysis of variance (One-Way ANOVA) demonstrated statistically significant differences, F(3, 203) = 8.45, p < .001. Particularly, post hoc comparisons using the Scheffé criterion showed that fearful individuals had significantly higher depressive symptomatology, M = 21.50, SD = 10.75, compared to the secure, M = 11.96, SD = 11.22, and the dismissive individuals, M = 10.81, SD = 10.365. There were no significant differences among the other groups (Table 3). In other words, there was a statistically significant reduction in depressive symptoms in the groups having a positive model of self, in comparison with the group having both a negative model of self and a negative model of others.

Table 3

Comparison (One-Way ANOVAs) of Attachment Types

Variable M (SD)
Secure (n = 74) Fearful (n = 34) Preoccupied (n = 32) Dismissive (n = 67) df F
Depressive Symptomatology 11.96 (11.22) 21.50 (10.75) 16.72 (12.76) 10.81 (10.37) 3, 203 8.45****
Job Security 2.94 (0.77) 2.89 (0.93) 2.93 (0.76) 3.27 (0.74) 3, 203 2.88*
Job Burnout
Emotional Exhaustion 20.49 (11.66) 26.62 (10.85) 24.53 (11.86) 19.75 (10.64) 3, 203 3.79*
Depersonalization 6.74 (5.77) 10.21 (7.09) 10.22 (7.42) 7.21 (6.15) 3, 203 3.93**
Efficacy 36.35 (8.24) 33.24 (7.56) 32.53 (8.70) 35.85 (8.67) 3, 203 2.31

*p < .05. **p < .01. ****p < .0001.

Another set of comparisons was performed in order to explore the differences among the four attachment types with regards to job security. One-Way ANOVA comparisons indicated statistically significant differences, F(3, 203) = 2.88, p = .037. Particularly, Duncan’s multiple range test indicated that fearful individuals differ from dismissive ones, whereas, there were no statistically significant differences between the other groups. To sum up, results showed that the group having a positive model of self and a negative model of others had significantly increased job security in comparison with the group having both a negative model of self and a negative model of others (Table 3).

Statistically significant differences were also found among the four attachment types regarding two of the three burnout dimensions. As far as emotional exhaustion is concerned, comparisons of means and standard deviations indicated statistically significant differences among the attachment groups, F(3, 203) = 3.79, p = .011. In particular, according to Scheffé criterion, there was a significant difference between the fearful and the dismissive type, p = .04. In other words, fearful individuals exhibiting negative model of self and others had higher means in emotional exhaustion than the dismissive ones (negative model of others but positive model of self) (Table 3).

Concerning depersonalization, F(3, 203) = 3.93, p = .009, Tukey HSD criterion indicated a statistically significant difference between the secure and the fearful type, p = .047. More specifically, individuals having a positive model of both self and others, experienced less depersonalization than individuals having a negative model of self as well as of others. However, there were no statistically significant differences among the four attachment types concerning efficacy, F(3, 203) = 2.31, p = .078 (Table 3).

Predictive Factors of Depressive Symptomatology [TOP]

Finally, stepwise multiple regression analysis was performed in order to test the predictive value of job security, the three dimensions of burnout and the four attachment types, with regards to depressive symptomatology. Results indicated that emotional exhaustion, fearful type of attachment, depersonalization and job security are the four independent variables able to predict depressive symptomatology. There is a positive correlation between depressive symptomatology and (a) emotional exhaustion, β = .39, t = 5.61, p = .0001, (b) fearful type, β = .18, t = 3.22, p = .001, and (c) depersonalization, β = .17, t = 2.57, p = .011. On the contrary, there is a negative correlation between depressive symptomatology and job security, β = -.12, t = -2.00, p = .047: the higher an individual’s job security, the lower his/her depressive symptoms. More specifically, the first independent variable (emotional exhaustion) was responsible for 30.1% of the variance in explaining depression, F(1, 205) = 88.26, p < .001, the second variable (fearful type) for another 3.70%, F(1, 204) = 11.26, p = .001, the third variable (depersonalization) for an additional 2.30%, F(1, 203) = 7.20, p = .008, and the fourth one (job security) for another 1.20%, F(1, 202) = 3.98, p = .047. In other words, 37.30% of depressive symptomatology can be explained by these four variables (Table 4).

Table 4

Stepwise Multiple Regression Analysis for the Prediction of Depressive Symptomatology

Independent Variables B SE B β t
Emotional Exhaustion .39 .070 .39 5.61***
Fearful Attachment Type 5.77 1.79 .18 3.22***
Depersonalization .30 .12 .17 2.57***
Job Security -1.76 .88 -.12 -2.00*

Note. R2 = .37, F(4, 202) = 29.99, p < .001.

*p < .05. ***p < .001.

Discussion [TOP]

In brief, the hypothesis of the significant correlation between depressive symptomatology and (a) job insecurity, (b) burnout and (c) attachment styles involving a negative self model was confirmed. Moreover, the hypothesis that the sample shows mild or high depressive symptomatology, job insecurity and burnout was partly confirmed (with the exception of depersonalization). Finally, analysis of the results showed that emotional exhaustion, fearful type of attachment; depersonalization and job security are the four predictive factors of depressive symptomatology. These findings add to our existing knowledge of the relationship between the above mentioned variables, by verifying that these relationships apply in the specific population facing serious turmoil.

In the five years of the Greek economic crisis, unemployment rose, Gross National Product, as well as per capita income, declined and labor itself devaluated (Bank of Greece, 2011; Bank of Greece, 2013). The public sector, which used to be a safe working environment prior to the crisis, underwent great and violent changes too. Previous studies (Bridges & Disney, 2010; Butterworth, Rodgers, & Windsor, 2009; Economou et al., 2013), have indicated that depression is related to socio-economic factors. This fact is in line with our findings since an important percentage of the sample reported mild or high depressive symptomatology.

Various studies have also related the increase of job insecurity to periods of economic crisis, and the increase of job security to periods of economic stability respectively (Adkins, Werbel, & Farh, 2001; Linz & Semykina, 2008). Due to the current crisis, our assumption was that civil servants in Greece do not experience the security they used to feel any more. The sample’s mean job security confirmed this assumption.

Burnout has previously been associated with bad working conditions (Pines, 1986) and lack of fairness in the working environment (Maslach & Leiter, 2008). Therefore, bureaucracy, the inadequate number of employees, the growing workload, the complicated legislation and the fact that a person’s efforts are not easily recognized and rewarded, led to the hypothesis that the largest part of the sample would demonstrate mild or high burnout. This hypothesis was confirmed except for the case of depersonalization. This could be explained by the fact that cynicism is the last phase in the course of burnout (Cherniss, 1980; Edelwich & Brodsky, 1980, as cited in Theofilou, 2009).

Our hypothesis on the relationship between depressive symptomatology and job security was confirmed. In particular, our results showed that the higher a person scored in job security, the less depressive symptoms he would have. These results are consistent with the previous studies of Ferrie et al. (2002) and Hellgren and Sverke (2003). The effect of job security on psychological well-being can be explained on the assumption that loss of employment doesn’t only lead to loss of income, but also to withdrawal of social recourses, stability and previous time structure (Miana, González-Morales, Caballer, & Peiró, 2011). In other words it is associated with a series of negative effects, and therefore becomes a source of stress (Witte, 1999). Job insecurity presents a threat that an employee can neither predict, nor control. Thus, an individual is unable to take any actions, and this lack of control makes him experience a sense of entrapment and helplessness.

The findings of the present study suggest a correlation between the dimensions of burnout and depression which is also in line with a number of previous studies (Ahola et al., 2005; Bellani et al., 1996; McKnight & Glass, 1995). Emotional exhaustion was most strongly correlated with depressive symptomatology. There was also a strong positive correlation between depressive symptomatology and depersonalization. On the contrary, there was a negative correlation between efficacy and depressive symptomatology. In other words, depressed individuals are more likely to experience emotional exhaustion and depersonalization, while one’s efficacy and belief in his/her ability and capacity to accomplish a task and produce desired effects are related to lower depressive symptomatology. Taking all the above into consideration, it can be said that the increase of burnout on the whole, is related to an increase in depressive symptomatology and vice versa.

The time sequence between burnout and depressive symptomatology has not been established yet, but it can be assumed that depressive symptoms follow burnout and not vice versa (Iacovides et al., 2003). This assumption derives from the findings that burnout can lead to lack of sleep, social withdrawal, disappointment (Maslach, 1978, as cited in Galanakis et al., 2009) alcohol, coffee and tobacco abuse, as well as family and interpersonal problems (Bakker et al., 2000) thus affecting all aspects of a person’s functioning. Moreover, in modern western societies, job is a source of meaning and identity for many people, and problems at work can lead to an existential crisis (Lambie, 2006).

Another interesting finding of the comparisons between the four types of attachment (secure, fearful, preoccupied and dismissive), and the factor of depressive symptomatology was that fearful individuals had the highest levels of depressive symptoms compared to the secure and the dismissive ones, while the preoccupied type had the second highest score. According to literature (Akbag & Imamoglu, 2010; Bartholomew & Horowitz, 1991; Holmes, 2009) fearful and preoccupied types experience low self-esteem and social insecurity and face difficulties in close relationships; which the former avoids, whereas the latter craves but finds stressful and discomforting. On the contrary, the secure and the dismissive type experience high self-esteem and are not afraid of intimacy. What fearful and preoccupied individuals have in common is a low internalized sense of self-worth. On the other hand, what secure and dismissive individuals have in common is a positive model of self (Bartholomew, 1990). It is therefore confirmed that high depressive symptomatology is related to a negative internal working model of self. The primary prevention, thus, might include the use of methods that enhance self-esteem and the person’s belief that is worthy of love.

Attachment type is a relatively stable but not invariable trait of personality. It is gradually constructed through the experiences of infancy, childhood and adolescence and although it is resistant to change, it can still be altered (Bowlby, 1973). Stability of attachment depends on the extent to which one’s experiences confirm the notion formed in the early years of life regarding whether others can be trusted or not. Confirming experiences promote stability, while disconfirming experiences produce change. On the other hand, mental representations of the self and others seem to determine the kind of relationships one has and the kind of responses he/she evokes from his/her environment. Secure individuals are more likely to seek support, and therefore more likely to get it and confirm their internal models, while insecure individuals are more likely not to ask for social support and therefore not to receive any (Gerlsma & Luteijn, 2000). Moreover, avoidance of intimacy characterizing fearful and avoidant individuals, reduces the possibility of having close relationships that could help them change their working models of others (Bartholomew & Horowitz, 1991). Counseling and the therapeutic counselor-client relationship may, in these cases, offer the intimate experience that can change the way a client engages with others.

Finally, we tested the predictive value of the three dimensions of burnout, job security and the types of attachment, with regards to depressive symptomatology. Results indicated that emotional exhaustion is the first predictive factor (positive correlation), the fearful type is the second (positive correlation), depersonalization, is the third (positive correlation) and finally job security is the fourth independent variable that can predict depressive symptomatology (negative correlation). These four variables can explain 37.30% of depressive symptomatology.

Any counseling intervention should place the emphasis on dealing with the employee’s feelings of emotional depletion, as well as on creating a sense of personal value and trust at the workplace, and on balancing a job’s demands and an individual’s expectations. Emotional exhaustion, fearful attachment and depersonalization seem to predict depressive symptomatology. Therefore, every time we see public servants that are unable to perform their job as reliably as they used to and express a distant and cynic attitude, lack of politeness and unwillingness to offer suitable services, we should take it into consideration, explore the grave possibility of underlying burnout, and do our best to support them. This support should also aim to allow secure intimacy and reinforce a positive view of self and others.

This research comes with a few caveats for its results. First, a relatively small sample was studied which does not allow for generalization of the results to the general population of Greek civil servants. Another limitation concerns the fact that the measures used in the present study, despite their adaptation into the Greek language, their good reliability and good factorial structure, are not standardized in Greece. Furthermore, they are all self-reports and as a result the case of individuals answering in a socially acceptable manner (social desirability), cannot be ruled out. Moreover, claims based on these analyses should be made with caution, given their correlational nature that does not allow “cause and effect” conclusions. Lastly, the results of the present study are also limited by the fact that, to the best of our knowledge, there are no previous studies on this specific population. As a result, the scores of this sample were compared to the scores of different populations used in previous studies, whereas it would be far more interesting to compare them to the scores of Greek civil servants before the onset of the economic crisis.

In spite of the above limitations, the current findings, when used with caution, can have useful implications for future research and be used as a baseline for future comparisons. A retest in later time may reveal the course of the Greek civil servants’ mental health and its relationship to the financial situation. In addition, it would be interesting to study how management styles (authoritarian vs democratic) and personality traits, could moderate this relationship. Moreover, civil servants’ mental health is a novel area of research, therefore qualitative investigation could provide depth and suggest topics not yet examined.

This paper offers a deeper understanding of the psychological determinants affecting public sector‘s services, a fact that is necessary in order to maximize the potential of the public sector, and to guarantee the employees’ well-being. We believe that it is essential to fully investigate how psychological factors interact together with this economical crisis on a person’s working behaviour, and to highlight the importance of counseling as a means of change.

The results of the present study may also have important implications for the development of intervention strategies. Screening for the factors influencing the civil servants’ mental health, could help civil sector policy makers discover not only the circumstances that are liable to cause depression, job insecurity or burnout, but also the employees that are liable to experience the above. Therefore, it can allow the development of interventions on organizational level, as well as counseling interventions that could empower employees to find meaning in their work and express greater empathy and respect to colleagues and clients. The National Centre for Public Administration and Local Government (EKDDA) that has been founded to upgrade the development of Human Resources in the Public Administration and Local Government offers standalone packages that include counseling support provisions to groups of civil servants participating in its programs. We propose creating programs that optimize research findings and apply not to every public servant, but are targeted to each organization separately, in order to improve effectiveness and allow for evaluation of these programs. Moreover, counselors could work closely with HR managers, trade unions and government representatives, in order to suggest initiatives and organizational changes. Even though it is likely that in times of economical crisis such programs could be at risk, it is undeniable that counseling services are important and financially sensible.

Funding [TOP]

The authors have no funding to report.

Competing Interests [TOP]

The authors have declared that no competing interests exist.

Acknowledgments [TOP]

The authors have no support to report.

References [TOP]

  • Abdul Kadir, N. B., & Bifulco, A. (2013). Insecure attachment style as a vulnerability factor for depression: Recent findings in a community-based study of Malay single and married mothers. Psychiatry Research, 210(3), 919-924. doi:10.1016/j.psychres.2013.08.034

  • Adkins, C. L., Werbel, J. D., & Farh, J. L. (2001). A field study of job insecurity during a financial crisis. Group & Organization Management, 26(4), 463-483. doi:10.1177/1059601101264004

  • Ahola, K., Honkonen, T., Isometsä, E., Kalimo, R., Nykyri, E., Aromaa, A., & Lönnqvist, J. (2005). The relationship between job-related burnout and depressive disorders—Results from the Finnish Health 2000 Study. Journal of Affective Disorders, 88(1), 55-62. doi:10.1016/j.jad.2005.06.004

  • Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. (1971). Individual differences in strange situation behavior of one-year-olds. In H. R. Schaffer (Ed.), The origins of human social relations (pp. 17-57). New York, NY, USA: Academic Press.

  • Akbag, M., & Imamoglu, S. E. (2010). The prediction of gender and attachment styles on shame, guilt, and loneliness. Educational Sciences: Theory and Practice, 10(2), 669-682.

  • Amarantidou, S., Mantis, K., & Koustelios, A. (2009). Relation of job security and job satisfaction among physical education teachers in Greece. International Journal of Physical Education, 2009(1), 20-23.

  • Bakker, A. B., Schaufeli, W. B., Demerouti, E., Janssen, P. P. M., Van der Hulst, R., & Brouwer, J. (2000). Using equity theory to examine the difference between burnout and depression. Anxiety, Stress, and Coping, 13, 247-268. doi:10.1080/10615800008549265

  • Bank of Greece. (2011). Annual Reports. Athens, Greece: B.G. Printing Office.

  • Bank of Greece. (2013). Annual Reports. Athens, Greece: B.G. Printing Office.

  • Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social and Personal Relationships, 7, 147-178. doi:10.1177/0265407590072001

  • Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61(2), 226-244. doi:10.1037/0022-3514.61.2.226

  • Bellani, M. L., Furlani, F., Gnecchi, M., Pezzotta, P., Trotti, E. M., & Bellotti, G. G. (1996). Burnout and related factors among HIV/AIDS health care workers. AIDS Care, 8(2), 207-221. doi:10.1080/09540129650125885

  • Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation. New York, NY, USA: Basic Books.

  • Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss. New York, NY, USA: Basic Books.

  • Bowlby, J. (1982). Attachment and loss: Vol. 1. Attachment (2nd ed.). New York, NY, USA: Basic Books.

  • Bridges, S., & Disney, R. (2010). Debt and depression. Journal of Health Economics, 29(3), 388-403. doi:10.1016/j.jhealeco.2010.02.003

  • Burgard, S. A., Brand, J. E., & House, J. S. (2009). Perceived job insecurity and worker health in the United States. Social Science & Medicine, 69(5), 777-785. doi:10.1016/j.socscimed.2009.06.029

  • Butterworth, P., Rodgers, B., & Windsor, T. D. (2009). Financial hardship, socio-economic position and depression: Results from the PATH Through Life Survey. Social Science & Medicine, 69(2), 229-237. doi:10.1016/j.socscimed.2009.05.008

  • Cherniss, C. (1980). Professional burnout in human service organizations. New York, NY, USA: Praeger.

  • Collins, J., Gibson, A., Parkin, S., Parkinson, R., Shave, D., & Dyer, C. (2012). Counselling in the workplace: How time-limited counselling can effect change in well-being. Counselling and Psychotherapy Research Journal, 12(2), 84-92. doi:10.1080/14733145.2011.638080

  • Conradi, H. J., & de Jonge, P. (2009). Recurrent depression and the role of adult attachment: A prospective and a retrospective study. Journal of Affective Disorders, 116(1-2), 93-99. doi:10.1016/j.jad.2008.10.027

  • Cooley, E. L., Van Buren, A., & Cole, S. P. (2010). Attachment styles, social skills, and depression in college women. Journal of College Counseling, 13(1), 50-62. doi:10.1002/j.2161-1882.2010.tb00047.x

  • Economou, M., Madianos, M., Peppou, L. E., Patelakis, A., & Stefanis, C. N. (2013). Major depression in the era of economic crisis: A replication of a cross-sectional study across Greece. Journal of Affective Disorders, 145(3), 308-314. doi:10.1016/j.jad.2012.08.008

  • Ferrie, J. E., Shipley, M. J., Stansfeld, S. A., & Marmot, M. G. (2002). Effects of chronic job insecurity and change in job security on self reported health, minor psychiatric morbidity, physiological measures, and health related behaviours in British civil servants: The Whitehall II study. Journal of Epidemiology and Community Health, 56(6), 450-454. doi:10.1136/jech.56.6.450

  • Galanakis, M., Moraitou, M., Garivaldis, F. J., & Stalikas, A. (2009). Factorial structure and psychometric properties of the Maslach Burnout Inventory (MBI) in Greek midwives. Europe’s Journal of Psychology, 4, 52-70.

  • Gerlsma, C., & Luteijn, F. (2000). Attachment style in the context of clinical and health psychology: A proposal for the assessment of valence, incongruence, and accessibility of attachment representations in various working models. The British Journal of Medical Psychology, 73(1), 15-34. doi:10.1348/000711200160273

  • Green, F. (2011). Unpacking the misery multiplier: How employability modifies the impacts of unemployment and job insecurity on life satisfaction and mental health. Journal of Health Economics, 30(2), 265-276. doi:10.1016/j.jhealeco.2010.12.005

  • Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511-524. doi:10.1037/0022-3514.52.3.511

  • Hellgren, J., & Sverke, M. (2003). Does job insecurity lead to impaired well-being or vice versa? Estimation of cross-lagged effects using latent variable modelling. Journal of Organizational Behavior, 24(2), 215-236. doi:10.1002/job.184

  • Holmes, J. (2009). John Bowlby and the theory of attachment (G. Athanasiou & Th. Athanasiou, Trans.). Athens, Greece: Ellinika Grammata. [in Greek]

  • Iacovides, A., Fountoulakis, K. N., Kaprinis, S., & Kaprinis, G. (2003). The relationship between job stress, burnout and clinical depression. Journal of Affective Disorders, 75(3), 209-221. doi:10.1016/S0165-0327(02)00101-5

  • Kafetsios, K. (2005). Attachment, emotion and close relationships. Athens, Greece: Typothito-Giorgos Dardanos. [in Greek]

  • Kafetsios, K. (2012). Relationship Questionnaire. In A. Stalikas, S. Triliva, & P. Roussi (Eds.), The psychometric tools in Greece (p. 711). Athens, Greece: Pedio. [in Greek]

  • Kleftaras, G. (1991). Les événements et les activités agréables et désagréables chez les personnes âgées en fonction du niveau des sentiments dépressifs. Revue Européenne de Psychologie Appliquée, 41(2), 107-112.

  • Kleftaras, G. (1997). Control-related beliefs and depressive symptomatology in a semi-institutionalized population of French elderly persons. European Review of Applied Psychology, 47(3), 225-229.

  • Kleftaras, G. (2000). Interpersonal problem-solving of means-ends thinking, frequency and strength of pleasant and unpleasant activities and symptoms of depression in French older adults. Physical & Occupational Therapy in Geriatrics, 17(4), 43-63. doi:10.1080/J148v17n04_04

  • Kleftaras, G. (2004). La dépression: Approche cognitive et comportementale. Paris, France: L’Harmattan.

  • Kleftaras, G. (2006). Psychosocial factors and later life depressive symptoms in a Greek community sample. Physical & Occupational Therapy in Geriatrics, 25(2), 13-32. doi:10.1080/J148v25n02_02

  • Kleftaras, G., & Katsogianni, I. (2012). Spirituality, meaning in life, and depressive symptomatology in individuals with alcohol dependence. Journal of Spirituality in Mental Health, 14, 268-288. doi:10.1080/19349637.2012.730469

  • Kleftaras, G., & Psarra, E. (2012). Meaning in life, psychological well-being and depressive symptomatology: A comparative study. Psychology, 3(4), 337-345. doi:10.4236/psych.2012.34048

  • Kleftaras, G., & Tzonichaki, I. (2012). Questionnaire d’Auto-Evaluation de la Symptomatologie Dépressive. In A. Stalikas, S. Triliva, & P. Roussi (Eds.), The psychometric tools in Greece (pp. 140-141). Athens, Greece: Pedio. [in Greek]

  • Koustelios, A., Kouli, O., & Theodorakis, N. (2003). Job insecurity and job satisfaction among Greek fitness instructors. Perceptual and Motor Skills, 97, 192-194. doi:10.2466/pms.2003.97.1.192

  • Lambie, G. W. (2006). Burnout prevention: A humanistic perspective and structured group supervision activity. The Journal of Humanistic Counseling, Education and Development, 45(1), 32-44. doi:10.1002/j.2161-1939.2006.tb00003.x

  • Linz, S. J., & Semykina, A. (2008). How do workers fare during transition? Perceptions of job insecurity among Russian workers, 1995–2004. Labour Economics, 15(3), 442-458. doi:10.1016/j.labeco.2007.05.003

  • Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Organizational Behavior, 2(2), 99-113. doi:10.1002/job.4030020205

  • Maslach, C., & Jackson, S. E. (1986). Maslach Burnout Inventory (manual) (2nd ed.). Palo Alto, CA, USA: Consulting Psychologists Press.

  • Maslach, C., & Leiter, M. P. (2008). Early predictors of job burnout and engagement. The Journal of Applied Psychology, 93(3), 498-512. doi:10.1037/0021-9010.93.3.498

  • McKnight, J. D., & Glass, D. C. (1995). Perceptions of control, burnout, and depressive symptomatology: A replication and extension. Journal of Consulting and Clinical Psychology, 63(3), 490-494. doi:10.1037/0022-006X.63.3.490

  • Mcleod, J. (2008). Counseling in the workplace: A comprehensive review of the research evidence (2nd ed.). Lutterworth, United Kingdom: British Association for Counseling and psychotherapy.

  • Miana, B. S., González-Morales, M. G., Caballer, A., & Peiró, J. M. (2011). Consequences of job insecurity and the moderator role of occupational group. The Spanish Journal of Psychology, 14(2), 820-831. doi:10.5209/rev_SJOP.2011.v14.n2.29

  • Näswall, K., Sverke, M., & Hellgren, J. (2005). The moderating role of personality characteristics on the relationship between job insecurity and strain. Work and Stress, 19(1), 37-49. doi:10.1080/02678370500057850

  • Papadatou, D., Anagnostopoulos, F., & Monos, D. (2012). Maslach Burnout Inventory. In A. Stalikas, S. Triliva, & P. Roussi (Eds.), The psychometric tools in Greece (p. 485). Athens, Greece: Pedio. [in Greek]

  • 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:10.1016/j.jvb.2011.09.007

  • Pichot, P., Boyer, P., Pull, C. B., Rein, W., Simon, M., & Thibault, A. (1984). Un questionnaire d’auto-évaluation de la symptomatologie dépressive, le questionnaire QD2, I. Construction, structure factorielle et propriétés métrologiques. Revue de Psychologie Appliquée, 34, 229-250.

  • Pickett, K. E., & Wilkinson, R. G. (2010). Inequality: An underacknowledged source of mental illness and distress. The British Journal of Psychiatry, 197(6), 426-428. doi:10.1192/bjp.bp.109.072066

  • Pines, A. M. (1986). Who is to blame for helper’s burnout? Environmental impact. In C. D. Scott & J. Hawk (Eds.), Heal thyself: The health of health care professionals. New York, NY, USA: Bruner/Mazel Publications.

  • Schutte, N., Toppinen, S., Kalimo, R., & Schaufeli, W. (2000). The factorial validity of the Maslach Burnout Inventory – General Survey (MBI-GS) across occupational groups and nations. Journal of Occupational and Organizational Psychology, 73(1), 53-66. doi:10.1348/096317900166877

  • Sharma, E. (2012). Employee counselling clinics. International Journal of Human Development and Management Sciences, 1(1), 23-33.

  • Sverke, M., & Hellgren, J. (2002). The nature of job insecurity: Understanding employment uncertainty on the brink of a new millennium. Applied Psychology, 51(1), 23-42. doi:10.1111/1464-0597.0077z

  • Theofilou, P. (2009). Burn-out syndrome in the field of health. E-Journal of Science & Technology (e-JST), 4(4), ). Retrieved from [in Greek].

  • Weiss, D., Dawis, R., England, G., & Lofquist, L. (1967). Mannual fοr the Minnesota Satisfaction Questionnaire. Minnesota, MN, USA: Work Adjustment Project Industrial Relations Center, University of Minnesota.

  • Wentland, D. M. (2009). Organizational performance in a nutshell. Charlotte, NC, USA: Information Age Pub.

  • Whiffen, V. E., Kallos-Lilly, A. V., & MacDonald, B. J. (2001). Depression and attachment in couples. Cognitive Therapy and Research, 25(5), 577-590. doi:10.1023/A:1005557515597

  • Witte, H. D. (1999). Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology, 8(2), 155-177. doi:10.1080/135943299398302