Validation of the Greek Version of the “ Job Stress Measure ”

The aim of the present study is the validation of the “Job Stress Measure” in the Greek population. “Job Stress Measure” along with ASSET scale was finally distributed to 238 individuals, working in public and private organizations. The Cronbach’s alpha coefficient of the “Job Stress Measure scale” was found to be 0.868. Three main circumstantial factors explain the 55% of the variance of the phenomenon: a) characteristics of work, b) clarity of objectives and c) without specific concept. The study showed positive correlation between stress and workload, job insecurity and difficulties in working relationships, physical and mental condition, communication and beliefs about work. Conclusively the “Job Stress Measure” can be used as a reliable validated tool for measuring job stress in the Greek population.


Introduction
The most accepted and popular definition of stress is determineds as: "a state in which homeostasis is actually threatened or perceived to be so; homeostasis is re-established by a complex repertoire of behavioral and physio-logical adaptive responses of the organism."(Chrousos, 2009).Working environment is considered as a major source of stress.Specifically, work-related stress is experienced mainly when the demands of the work environment exceed the workers' ability to cope with (European Agency for Safety and Health at Work, 2002).It is a condition that is frequently experienced by the individual employee as a result of increased working demands.This condition can cause many health effects on the workers (Doherty & Tyson, 1998), or even may cause disturbances on working behaviour and in general lifestyle (Wheatley, 2000).
A large number of individual self-report scales (Table 1) are available, in order to assess both work-related stress and various job stressors.Such psychometric tools when used appropriately are useful, practical and irreplaceable.The most known tools are the following (Table 1).
The aim of this study is, "Job Stress Measure" (Judge et al., 1994) validation, a measuring tool of work-related stress and evaluation of its psychometric properties, on a population based survey (Greece).
Greek studies attempting to address either construction or validation of job stress measures are limited.Therefore, it is important to have available, reliable and valid assessment instruments."Job Stress Measure" is our choice as an abridged, short and effective tool of measuring job stress, as it is more practical to use and time saving, particularly in busy working environments.

Subjects
The sample consisted of 238 participants (Table 2).

Measures
The "Job Stress Measure" consists of sixteen self-report items rated on a Likert scale from 1 -5, depending on the extent of the work related stress produced by each of the items.On this scale, "1" indicates that the item produces no stress, "2" produces little stress, "3" produces some stress, "4" produces quite a bit of stress and "5" produces a great deal of stress.
A double translation of the English questionnaire to Greek was made by 2 bilingual psychologists, followed by translation to English.
The construct validity of the Job Stress Measure was evaluated by examining its divergent and convergent validity with the ASSET scale.The ASSET scale is used to measure job stress and is divided into four subscales.The first subscale refers to the causes of job stress "Perceptions of your work", the other two refer to the effects of stress ("Attitudes towards your organization" and "Your Health") and the last one includes additional information (social-demographic characteristics).We used two subscales of the ASSET scale ("Perceptions of your work" and "Your Health").Based on the bibliography, this measurement tool has reliability a = 0.74 to a = 0.91 (Cartwright & Cooper, 2002;Nikolaou & Tsaousis, 2002).

Data Collection
The study took place in Athens from September 2013 to April 2014.Participants were approached by the authors in their workplace.After informed consent has been obtained, questionnaires were distributed to the participants.
There were no ambiguities and questions about the items.The average time of filling out the questionnaires was 25 minutes.250 questionnaires were distributed, and eventually 238 questionnaires were returned.Our final sample consisted of 230 individuals, since we excluded those individuals that did not answer all of the 16 items on the scale, in order to have more reliable results.
Data collected were introduced and processed by statistical software SPSS (Statistical Package for the Social Sciences), Version 21 (SPSS, Inc., Chicago, IL, USA).

Scores
The mean and standard deviation of each of the items in the "Job Stress Measure" are shown in Table 3.
According to  question.This can be explained by changes in the quality of jobs (civil servants, official rank, few meetings, and small requirement for a business trip).

Reliability
The internal consistency of the "Job Stress Measure" was examined by Cronbach's coefficient alpha (Cronbach, 1951), and the item-total correlations (Table 4).The Cronbach's alpha of the scale was found to be 0.868.Moreover we performed item analysis to ascertain the possibility of further enhancing Cronbach's coefficient alpha by deleting unnecessary questions (Table 5) with negative results, keeping intact the initial structure of the questionnaire.According to Table 4, we found statistically significant, positive correlation between correlations of the items to the total score of the questionnaire (r = 0.296 ** -0.841 ** ).This indicates that questions refer to a central meaning (stress).

Construct Validity
The construct validity of the "Job Stress Measure" was evaluated by examining its divergent and convergent validity with the ASSET scale (Table 6 and Table 7).We found statistically significant, positive correlation with the following subscales: Workrel

Factor Analysis
Finally we performed factor analysis of the "Job Stress Measure" (Table 8).The results showed 3 major components with eigenvalues over 1.Further factor loadings of the "Job Stress Measure" items (Table 9) revealed that questions belonging to the first component (indicative of characteristics of work) are the following:  The number of projects and/or assignments I have 0.813 The amount of time I spend at work 0.508 The amount of time I spend in meetings 0.847 The number of phone calls and office visits I have during the day 0.412 The degree to which politics rather than performance affects organizational 0.791 The inability to clearly understand what is expected of me on the job 0.658 The volume of work that must be accomplished in the allotted time 0.756 The extent to which my position presents me with conflicting demands 0.514 The amount of red tape I need to go through to get my job done 0.577 The time pressures I experience 0.790 The lack of job security I have 0.551 The amount of responsibility I have 0.579 The scope of responsibilities my position entails 0.682 The degree to which my career seems "stalled" 0.736 The opportunities for career development I have had 0.736 The amount of traveling I must do 0.823 Note: Extraction method: Principal component analysis.Rotation method: Varimax with Kaiser normalization.
The number of projects and/or assignments I have; The amount of time I spend at work; The number of phone calls and office visits I have during the day; The volume of work that must be accomplished in the allotted time; The time pressures I experience; The lack of job security I have; The amount of responsibility I have; The scope of responsibilities my position entails; The degree to which my career seems "stalled"; The opportunities for career development I have had; Questions belonging to the second component (nonspecific) are the following: The amount of time I spend in meetings; The amount of traveling I must do; Finally, questions belonging to the third component (indicative of the clarity of objectives) are the following: The degree to which politics rather than performance affects organizational decisions; The inability to clearly understand what is expected of me on the job; The extent to which my position presents me with conflicting demands; The amount of red tape I need to go through to get my job done.

Discussion
Factor analysis revealed that the first circumstantial factor refers to the characteristics of work, and the third circumstantial factor refers to the clarity of objectives.As far as the second circumstantial factor is concerned, the configuration may be more a co product of different subgroups of the main sample (public/private) where those 2 items (travel, meetings) are occurring at specific levels and job roles.However, we decided to maintain this second factor in case of workers who either make business trips or participate in long meetings.Those 2 professional items seem to exacerbate stress caused by job characteristics and unclear objectives.It has been shown that work-related stress has severe emotional and physical health effects (Ferracci et al., 2003;Smith et al., 2005;Robin & Leslie, 2006;Yu et al., 2006;Brent, 2008;Schneider & Irastorza, 2010).The positive correlation between stress and 1) job insecurity 2) beliefs about work, has been confirmed byseveral studies, that agree with the fact that threats of career development, including threats against dismissal and promotion prospects with fuzzy criteria, produce stress to workers (Nelson & Burke, 2000;Burchell, 2002;Sverke et al., 2002;Hirsch & De Soucey, 2006).The economic pressure leads companies and organizations to undergo restructures, mergers and acquisitions (Hirsch & De Soucey, 2006).The result of this policy is the reduction of the workforce and a more flexible labor management (e.g.part-time job).An important factor that contributes to the poor physical health is the job insecurity (Burchell, 2002).The positive correlation between stress and 1) workload 2) employees' physical and mental health status, has been confirmed by several studies (Nishitani & Sakakibara, 2010;Burchell, 2002).A recent study suggests that a heavy workload increases stress levels (Nishitani & Sakakibara, 2010).
Finally, regarding health issues, the majority of the participants did not have a serious health problem in the last 6 months.Personality characteristics of the individuals, may explain this phenomenon, as the majority of the participants were more or less optimistic.There have been only a small number of studies investigating the influence of optimism on work-related stress.Researches showed that optimistic individuals used more problemfocused coping strategies outperforming pessimistic individuals in the work environment (Strutton & Lumpkin, 1992).Optimism could be a powerful ally and a very positive source of force in the workplace.
As we previously mentioned, work-related stress affects not only the health of employees, but also the productivity of organisations (Leka & Griffiths, 2003).
Limitations: 1) One limitation is that the survey was conducted exclusively in urban population; 2) Another limitation refers to the age group.Specifically, the majority of the participants (87.4%) were 21 to 55 years old; 3) The last limitation refers to the type of job.The majority of the participants (71.4%) said that they were working as administrators/clerks.

Conclusion
Job stress arises not only from situations people encounter on a daily basis, but also from people's own perceptions regarding their perceived ability to deal with such situations.The "Job Stress Measure" can be used as a tool of measuring job stress in the Greek population.The occupational stress has many implications on the individual's health and proves to be detrimental to the economy of the country.Therefore, future researches in Greece should focus on locating and assessing risk stressors in the workplace, through work-related stress psychometric tools and interviews.

Table 3 ,
all items have sufficient dispersion.Exception to this, is the third, fourth and sixteenth

Table 2 .
Demographic information of the sample.

Table 4 .
Correlations of the items to the total score of the questionnaire.

Table 5 .
Item analysis of the "Job Stress Measure".

of Squared Loadings Rotation Sums of Squared Loadings
Factor analysis based on orthogonalrotation.