Stress in Aviation Ground Staff
Stress in Aviation Ground Staff
Aviation ground crews play a very important role in air transportation. Not only does their work
pressure affect individual health and organizational efficiency, but aviation safety. The aim of this study
is to explore the relationship between various sources of occupational stress and work performance. In
this study, the connection among a variety of occupational stress, coping strategies and work
performance was also explored, using the Structural Equation Modeling (SEM). The result of the
analysis exhibited that occupational stress had a negative impact on work performance, and the coping
strategies were the mediator survivals between occupational stress and work performance. The
findings of the study argued that we can have a better understanding of the characteristic of aviation
ground crews and the causal relationship between occupational stress and work performance, and the
proposed model can be beneficial to improve the practices of human resources management and the
policies of relevant aviation industries.
Key words: Aviation ground crews, occupational stress, coping strategies, work performance.
INTRODUCTION
organizational efficiency, but also it affects aviation safety. perceived threats of job obsolescence. “Organization
There are considerable evidences that professionals are structure and climate” measures stress originating from
increasingly exposed to a working environment that the bureaucratic nature of the organization, communica-
places their health at risks through workplace stress and tion problems and morale in the organizations, and lastly,
difficulties in maintaining an effective work-life balance “home-work interface” measures stress originating from
(Davidson and Cooper, 1983; Greenhaus and Beutell, difficulties in coordinating family responsibilities with
1985; Burke, 1994; Dwyer, 1999). career demands.
Due to the work of aviation ground crews, the activity Clarke and Cooper (2000) focus on the stress of retail
duration of flight is urgent and irregular in shifts. workers. Their study indicated that occupational stress
Therefore, their work pressure is much heavier than other must be negatively perceived and subjected to inade-
industry not only affecting individual health and quate coping to result in negative stress outcome, while
performance, but organizational efficiency and aviation the experience of stress is also moderated by a number
safety. The purpose of this study is to discuss the causal of individual difference variables.
relationship between occupational stress and work
performance and to explore the mediator between them,
with the expectation that the results can be beneficial to Coping strategies
improve the practices of human resources management
and the policies of relevant aviation industries. Coping is the process by which people try to manage the
perceived discrepancy between the demands and resour-
ces they appraise in a stressful situation (Sarafino, 2005),
LITERATURE REVIEW whereas coping research is defined by a plethora of
diverse theoretical models and measurement instruments
Stress and effect (Parker and Endler, 1992; Schwarzer and Schwarzer,
1996; Skinner et al., 2003). In theory, there are three
In the discussion of a previous study that was conducted, main categories of coping strategies, including problem,
it was seen that stress is not the demand or the source of emotion and avoidance specific behaviours. The coping
pressure itself; but it is the perception of that pressure process is not a single event, and as such, it involves the
(Cox, 1978; Lazarus, 1966; McGrath, 1970). Lazarus and ongoing transactions with the environment (Sarafino,
Folkman (1984) indicated that work stress can be defined 2005).
as a relationship between the person and the The coping inventory for stressful situation (CISS)
environments. Gellis et al. (2004) indicated the service measures three dimensions (task, emotion and
coordination nature of the work and dealing with crisis avoidance oriented coping) that are common across most
situations was a major source of stress frequency and conceptions of coping strategies. It was developed as a
intensity. consequence of extending the interaction model of
According to Williams and Cooper (1998), the different anxiety (Endler, 1982) to include coping variables like the
perception of individuals (such as coping and supporting) interactional model of stress, anxiety and coping (Endler,
and the perception of stress outcomes (such as well- 1997). There are personal variables (for example, trait
being and job satisfaction) should also be measured. Hart anxiety, vulnerability, cognitive style, heredity and emo-
and Wearing (1995) made the point that stress cannot be tionality) that may interact with not only one another, but
expressed as a single variable and that elements such as also with the situation variables (for example, life events,
personality characteristics, coping processes, and hassles, crises, pain and trauma), which in themselves
positive and negative work experiences must also be also interact with one another. That is, there is an
considered. interaction between individual and situation variables. In
Cooper and Marshall (1976) have categorized the effect, there is a feedback loop and it is a continuous
sources of occupational stress as: intrinsic to the job, role process. One might assess various aspects of the model
in the organization, relationships at work, career develop- or assess the model as a whole (Endler and Parker,
ment, organizational structure and climate, and home- 1988).
work interface. In the models of occupational stress, Most of the coping instruments are related to the
“intrinsic to the job” included physical aspects of the limitation of unsatisfactory psychometric properties, such
working environment and personal psychosocial aspects. as unstable factor structures and low reliability (Endler et
“Role in the organization” measures how individuals al., 2003). The CISS seems much less plagued by psy-
perceive the expectations that others have of them. chometric limitations than are most of its predecessors
“Relationships at work” measures the stress originating (Schwarzer and Schwarzer, 1996). Rafnsson et al. (2006)
from personal contacts at work such as lack of social investigated 1251 adolescents in Iceland using the CISS
support from superior and office politics. “Career finding that the Cronbach's ranged from 0.82 to 0.92,
development” is concerned with respondents’ perceptions indicating that there are high levels of internal
of their career development, promotion prospects and consistency for CISS.
Sun and Chiou 2867
Figure 1. The model for occupational stress, coping strategies and work performance.
which questionnaire in which 16 items assessed task- Borman and Motowidlo (1993). It was a self-administered correlations analysis was applied to assess the relation of
oriented coping, 16 items assessed emotion-oriented questionnaire, in which 8 items assessed task performance all factors in this study. Finally, we analyzed the connection
coping, and the rest 16 items assessed avoidance-oriented and 15 items assessed contextual performance. Items among occupational stress, coping strategies and work
coping. Each item on a 6-point frequency scale ranged were scored from 1 (never) to 6 (all the time). performance by using the structural equation modeling
from 1 (never) to 6 (all the time). (SEM) (Jöreskog and Sörbom, 1988) and the AMOS 7.0
software package. This method allowed us to identify latent
Analysis procedure variables and structural equation coefficients simultaneous-
Work performance survey ly using covariance structure modeling. We developed and
Initially, the demographic data were displayed and the test a structural equations model to identify the interrela-
The work performance survey, assessed with 23 items reliability analysis with Cranach's coefficient was tionships among the variables, with a set of simultaneous
Adopted from Yu (1996), follows the consequence of performed to identify the reliability of this survey. Next, equations. However, we used this model to formally test
Sun and Chiou 2869
the three hypotheses. structural equation modeling, whereas the basic causal
model, incorporating the hypothesized relationships, is
presented in Figure 1. Before testing the structural model
RESULTS proposed in Figure 1, we assessed the validity of the
multiple-item scales by analyzing the measurement
Data collection model using confirmatory factor analysis (CFA) with
maximum likelihood estimation (Arbuckle, 1999). The
The target population of this study was aviation ground study used AMOS 7.0 to evaluate measurement model
crews working in Taoyuan International Airport, Taiwan. and structural equation modeling; then the overall fit of
The questionnaire was sent to 604 aviation ground the hypothesized model shown in Figure 1 was tested.
crews. Firstly, descriptive statistics was performed. Of all The results tested by the structural equation analysis are
2
these respondents, 342 (male =198 and female = 144) presented in Figure 2. The x divided by the degree of
had completely filled out the self-administered question- freedom is 4.67 (Chin and Todd, 1995; Gefen et al.,
naires, accounting for an effective response rate of 2000). The goodness of fit index (GFI) is 0.91 (greater
56.62%, and their ages ranged from 21 to 64 years with than 0.9) and the adjusted goodness of fit index (AGFI) is
the mean age being 35.97 years (SD = 8.73 years). 0.85, which indicates a good fit between the data and the
proposed model (Hayduk, 1987). The root mean square
error of approximation (RMSEA) is 0.10 (Browne and
Reliability analysis Cudeck, 1993; Hu and Bentler, 1999). Nonetheless, the
standardized path coefficients of the proposed model of
The value of the diagonal in Table 1 was reported on the relationships are shown in Figure 2.
Cronbach's coefficient of each construct and was The completely standardized solution in Figure 2
validated on the internal-consistency reliability. Conse- indicated the effect of the model, and it was seen that all
quently, all values of Cronbach's were greater than 0.7 the values in the model were significant (p<0.01). The
(DeVellis, 1991), indicating a high reliability. direct effect between occupational stress and coping
strategies was 0.23, while the direct effect between
occupational stress and work performance was -0.21 and
Correlations among occupational stress, coping the direct effect between coping strategies and work
strategies and work performance performance was 0.75.
However, the indirect effect between occupational
Correlations among occupational stress, coping stress and work performance by way of coping strategies
strategies and work performance were calculated based was -0.16 (-0.21 × 0.75). The total effect between occu-
on the total valid sample, and the results shown in Table pational stress and work performance was -0.37 [-0.21 +
1 are the coefficients of correlation and their statistical (-0.16)]. Totally, the latent variable “coping strategies”
significance. The factors among occupational stress was a mediator between occupational stress and work
severity scores are correlated with each other signi- performance.
ficantly (p < 0.01) and positively, whilst the coefficient of Table 2 showed the composite reliability (CR) and the
correlation is from 0.73 to 0.87 for occupational stress. average variance extracted (AVE) of the full model. All
Also, the factors among coping strategies severity scores values of CR were greater than 0.5, indicating a high
are correlated with each other significantly (p < 0.01) and reliability, while the values of AVE were greater than 0.5
positively, and it is shown that the coefficient of (except the construct “coping strategies”), indicating that
correlation is from 0.20 to 0.38 for coping strategies. the observed variables can explain the latent variables.
Furthermore, the factors in work performance severity
scores are also correlated with each other significantly (p
< 0.01) and positively, and it is shown that the coefficient DISCUSSION AND CONCLUSION
of correlation between task performance and contextual
performance is 0.70. According to the analysis, all factors Occupational stresses are phenomena that are more
in the survey are correlated significantly (p < 0.01) and common in modern times than it is generally believed. It
positively. After closely examining the relationship has a large impact on both individual and society. An
between occupational stress and work performance, we understanding of the nature and effects of occupational
found that it was positive, but not significant. Therefore, stress suggest that it poses a significant hazard to
we used the structural equation modeling (SEM) to test employees’ health and it also acts as a source of
the relationship between them. business cost (Shanton et al., 2001). The purpose of this
study was to explore the determinants of work perfor-
mance. The paper proposes and empirically evaluates a
Structural equation modeling model which generates a series of hypotheses regarding
stress by using the decisions and variables that are likely
We tested the overall fit of the hypothesized model using to influence work performance.
2870 Afr. J. Bus. Manage.
Table 1. The correlation of occupational stress, coping strategies and work performance.
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
(1) Intrinsic to the job 0.8200
(2) Role in the organization 0.8706** 0.8700
(3) Relationships at work 0.8184** 0.8681** 0.8600
(4) Career development 0.7762** 0.7823** 0.7577** 0.8100
(5) Organizational structure and climate 0.7943** 0.8079** 0.8142** 0.8249** 0.8700
(6) Home-work interface 0.7865** 0.8238** 0.8005** 0.7260** 0.7322** 0.8800
(7) Task-oriented coping 0.1691** 0.1909** 0.1366** 0.1831** 0.1430** 0.0897 0.9100
(8) Emotion-oriented coping 0.3150** 0.2639** 0.2654** 0.2991** 0.2335** 0.2665* 0.2039** 0.8800
(9) Avoidance-oriented coping 0.1123* 0.0884 0.0792 0.2082** 0.1162* 0.0903 0.3519** 0.3809** 0.8500
(10) Task performance -0.0602 -0.0336 -0.0860 0.0719 -0.0245 -0.0941 0.5276** -0.0221 0.1931** 0.9100
(11) Contextual performance -0.0247 -0.0078 -0.0488 0.0283 -0.0328 -0.0927 0.5228** 0.0348 0.2219** 0.7011** 0.9200
Mean 32.2700 38.6700 32.5000 33.5700 40.0200 33.9100 67.2100 56.9800 60.6600 38.8800 69.0100
SD 7.1700 9.4200 8.5300 7.2900 8.9400 9.5400 11.0400 11.3500 11.400 5.3100 9.9200
** p < 0.01; *p < 0.05. The value of the diagonal is Cronbach's .
The results of this study yielded some important occupational stresses and work performance. the model of this study, each of the variables
findings and generally supported previous resear- The present findings are partially consistent with showed a direct effect on aviation ground crews.
ches. The results indicated that occupational previous studies (Clarke and Cooper, 2000) in Therefore, the proposed model can be beneficial
stresses made significant contributions to coping identifying a causal relationship between occupa- to improve the practices of human resources
strategies, and that both occupational stress and tional stress and work performance. In essence, management and policies of relevant aviation
coping strategies made significant contributions to the negative effect of stress can be mediated by industries.
work performance. As such, the fit of the study’s coping. In short, employees who perceive low The findings have several implications for the
hypothesized partial mediation model was tested occupational stress and high coping tend to have managers and researchers of aviation ground
by the SEM approach. Conclusively, the analysis higher work performance. Managers traditionally crews. They suggest that personnel managers in
results showed that occupational stress has a have administered human resources by applying particular need to be aware of important psycholo-
negative impact on work performance (Pincherle, the same policies and methods to all employees, gical characteristics between occupational stress
1972; Jex, 1998), and our findings supported the but the analytical results presented here suggest and work performance, and specifically, that
expected partial mediation effects of coping that employees differ, and that the differences are coping strategies determine how employees are
strategies on the occupational stresses (that is, manifested in perceptions of occupational stress treated in an organization and influence how they
work performance relation). Consistent with the and coping strategies. The mediator roles of co- respond. The long-term consequences of high
transactional (Lazarus, 1984; Fried et al., 2008) ping show the effectiveness of coping strategies in levels of job stress and low coping crews need to
model, stressed employees’ responses to their the workplace (Kitaoka-Higashiguchi et al., 2003). be taken into consideration when calculating the
stressful experience, probably reflect further The study argued that we can have a better short- versus long-term benefits of the psycholo-
damage to their psychological state, which in turn understanding of the characteristic of aviation gical support system building of an organization.
weakens their work performance. These findings ground crews and the causal relationship between The results can be used to help researchers and
are quite consistent with previous studies of occupational stress and work performance. Using managers predict employees’ work performance
Sun and Chiou 2871
2
Variable Observed variable Standard parameter Error R CR AVE
Intrinsic to the job 0.91 8.69 0.83
Role in the organization 0.94 9.91 0.89
Relationships at work 0.91 11.96 0.84
Occupational stress 0.96 0.80
Career development 0.85 14.68 0.72
Organizational structure and climate 0.88 18.18 0.77
Home-work interface 0.86 23.02 0.75
and help companies take employee’s personality reports. Finally, since all variables were measured in the
characteristics into account. Perhaps, the research has same questionnaire, the findings were susceptible to
some limitations. First, we have proposed longitudinal problems associated with the common method of
stability for the causal relationship between occupational variance, that is, they may be distorted due to correlation
stress and work performance in the absence of a longi- inflation. However, the appearance of several low
tudinal design. The relation between cause and effect of variable correlations, adequate sample size and the relia-
the model would therefore benefit from the longitudinal bility levels of the measures suggest that the common
research to show that it is settled. This research also variance methodology used here did not create credibility
benefit from replication using measures other than self problems (Lindell and Whitney, 2001).
2872 Afr. J. Bus. Manage.