Content uploaded by Wendy Macdonald
Author content
All content in this area was uploaded by Wendy Macdonald on May 29, 2020
Content may be subject to copyright.
Associated with these changes, the expectation that
employees should be able to “do more with less” has be-
come widespread (see Figure 1). Work demands have inten-
sified, and increasing numbers of people report that work
demands are negatively affecting their health (Australian
Centre for Industrial Relations Research and Training
[ACIRRT], 1998; Australian Council of Trade Unions OHS
Unit, 1998; O’Donnell, 1997; Paoli & Merllie, 2001; Wright
& Lund, 1996). Not surpringly, job demands are among the
most frequently cited occupational stressors for full-time
employees, many of whom are experiencing increased
levels of work-related fatigue and stress (Charyszyn
& Tucker, 2001; Cooper et al., 2001; Cox, Griffiths, & Rial-
Gonzalez, 2000; Paoli & Merllie, 2001; Wooden, 2001b).
High stress levels generated by excessive job demands can
result in decreased performance quality, physical “overuse”
injuries and stress-related illness (Cox et al., 2000;
Devereux & Buckle, 2000a, 2000b; Edwards, Caplan,
& Van Harrison, 1998; Ellis, 2001; Houtman & Kompier,
1995; Karasek & Theorell, 1990; Kawakami & Haratani,
1999; Kristensen, 1996; Karasek, Dewe, & O’Driscoll,
1998; Stansfeld, Fuhrer, Shipley, & Marmot, 1999).
In this situation, managers face an increasingly difficult
challenge to maintain their organisation’s economic viability
while also protecting the physical and psychological wellbeing
of their workforce. An optimal balance between these two
objectives is not easy to achieve, particularly in view of the
unbalanced quantity and quality of information about them that
is usually available. Managers continually deal with highly
salient information about production-related issues that are
centrally important to economic viability. In contrast, informa-
tion about occupational health and safety (OHS) issues
is usually intermittent, and often both ambiguous and unreli-
able as a guide to actions (Reason, 1990). Nevertheless, the
need to give appropriate weight to OHS issues cannot be
ignored. In addition to the obvious moral considerations, legis-
lation in many jurisdictions deems employers to be legally
responsible for ensuring safe and healthy “systems of work”,
regardless of particular enterprise or workplace agreements.
Conceptually, it is clear that “system of work” includes
the nature and level of job demands with which employees
are required to cope, and therefore that issues such
as staffing levels, work rates and “workloads” more gener-
ally have potential OHS and related legal implications.
Some recent court decisions are beginning to reflect this
position. For example, in a case before the Australian
Industrial Relations Commission (cited by ACIRRT, 2002)
The Impact of Job Demands and Workload
on Stress and Fatigue
WENDY MacDONALD
La Trobe University, Australia
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
VOLUME 38 NUMBER 2 pp. ??–??
1
Workload” is a hypothetical construct which has been
developed and is widely applied within the domain
of human factors (HF) psychology, and various
workload measurement techniques are typically used
to evaluate equipment or work systems in terms of the
workload experienced by people using them. This workload
construct emerged from extensive, task-specific research on
the capacities and limitations of the human information
processing system; it reflects the perceived margin between
task demands and an individual’s motivated coping capacity.
In the domain of occupational stress, however, workload
is equated with job demand, which is simply one of a hetere-
ogeneous set of “psychosocial hazards” which may
contribute to the development of stress, related illness
or injury. In a recent empirical study, workload in the
HF psychology sense was demonstrated to be a key deter-
minant of stress and fatigue levels among employees
performing repetitive, manufacturing work tasks. It is argued
that application of this conceptual framework to the measure-
ment and management of job demands would serve to delin-
eate more clearly the separate effects of employee
capacity-limited and motivation-limited factors on their work
performance and associated affective states such as stress.
This approach would enhance the ability of managers
to monitor and manage workload levels as part of a proactive
approach to stress management within the broader context
of occupational health and safety.
In recent years, a high rate of change has become the norm
in many workplaces, driven by technological developments
and an increasingly competitive global marketplace.
Changes include the introduction of more automated work
processes and other new technologies, organisational
restructuring and downsizing, multiskilling of employees
and related modifications to many workplace practices.
The percentage of people working a “standard” day has
decreased over the past decade, with more people working
part-time and in some countries, higher proportions of full-
time employees working very long hours (Australian
Bureau of Statistics, 2002; Heiler, 2001; US Department
of Labor, 2000; Wooden, 2001a). Among Australians
working longer than 49 hours per week, the majority are
doing so because of “open ended workload”: that is, more
people now feel that they must work until they have
finished particular tasks, rather than being able to work set
hours (Heiler, 2001; Queensland Government, 2001).
Address for correspondence: ?? Email: w.macdonald@latrobe.edu.au
“
the Commission upheld a claim of unfair dismissal on the
basis that the employee (whose poor performance had led
to his dismissal) had been working “unconscionably long
hours with the approval of his employer”, and that
“a system that allowed him to work these hours almost as
a matter of course is flawed” (M Aggenbach v TXU
Networks Pty Ltd 2001, P 902343). The England and Wales
Court of Appeal (EWCA; 2002) recently issued a set of
guidelines for courts on matters relating to stress-induced
psychiatric illness which specify factors likely to be relevant
in deciding whether the illness was both foreseeable and
attributable to work-related stress. The first listed of these
relevant factors is:
The nature and extent of the work done by the employee.
Is the workload much more than is normal for the particu-
lar job? Is the work particularly intellectually or emotion-
ally demanding for this employee? Are demands being
made of this employee unreasonable when compared with
the demands made of others in the same or comparable
jobs? Or are there signs that others doing this job are
suffering harmful levels of stress? Is there an abnormal
level of sickness or absenteeism in the same job (EWCA
Civ 76, para. 43–5, 2002, Sutherland v Hatton).
However, Australian employer organisations opposing the
formal regulation of “excessive” working hours argued that
“The question would … arise as to how the Commission
or a court of construction would approach an assessment
of … whether or not one or more employee’s ‘workload’
is unreasonable or whether ‘staffing levels’ are unreason-
able” (Australian Industry Group [AIG], 2001, p. 20). It is
certainly true that the vagueness of the notion of “reason-
able workload” presents significant difficulties. Despite
increasing recognition of the role that excessive workload
can play in generating stress and related health problems,
there is currently a lack of both a conceptual framework and
the measurement tools that would enable managers to assess
related workload levels, as part of a proactive approach
to stress management. These deficiencies underlie current
difficulties in addressing questions such as: How will the
introduction of a more automated work process affect
employee workload? How should appropriate staffing levels
be determined when organisations are restructured? What
limits should be placed on total hours worked, per shift
or over longer periods of weeks or months, and how should
such limits take account of the nature and level of work
demands? How should rest break regimes be modified
to take account of varying workload levels?
This paper argues that these questions about workload,
together with some more general issues concerning the
prevention of occupational stress and related health
problems, would be more effectively managed within
a “personenvironment fit” conceptual framework that
focuses specifically on the fit between work demands and
people’s capacities to cope with them. The argument draws
particularly on the concept of “mental workload”.
The Nature of “Workload”
The “mental workload” construct, also referred to simply
as “workload”, has been developed within the domain
of ergonomics or human factors (HF) psychology over the
past few decades (see Gopher & Donchin, 1986; Hancock
& Desmond, 2001; Hancock & Meshkati, 1988; Huey
& Wickens, 1993; Moray, 1979 & 1988; Tsang & Wilson,
1997; Wickens & Hollands, 2000; Xie & Salvendy, 2000).
Central to this concept is the fit — or more specifically,
the size of the gap — between task demands and a person’s
ability, when motivated, to cope with these demands.
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
2
WENDY MacDONALD
FIGURE 1
Bill: a victim of organisational ‘downsizing’ and the intensification of job demands. From The Age newspaper, Melbourne, xxxx
(permission to be obtained)
The concept initially developed in the context of task-
specific research on the capacities and limitations of the
human information processing system (e.g., Broadbent,
1958; Fitts & Posner, 1967; Kahneman, 1973; Mowbray,
1953; Welford, 1967). Gopher and Donchin (1986) defined
mental workload as “… the difference between the capaci-
ties of the information-processing system that are required
for task performance to satisfy expectations and the capacity
available at any given time” (pp. 41–43). Similarly, Jex
(1988, p. 11) defined it as “ … the operator’s evaluation of
the attention load margin (between their motivated capacity
and the current task demands) while achieving adequate
task performance …”
Jex’s “motivated capacity” is similar to Kalsbeek’s
(1968) earlier notion of a “willing to spend” capacity,
acknowledging that an individual’s absolute capacity
in some physical sense is approached only under conditions
of maximum possible motivation. According to this view,
it is the size of the margin or gap between the level of task
demands and the individual’s motivated capacity — influ-
enced by the degree of effort that they are willing to expend
in that situation — which determines their workload level.
Motivation is also closely linked to variations in capacity
level in Kahneman’s influential model (1973; see Figure 2).
Kahneman equated the
use of attentional resources
or “capacity” with effort expenditure, which is centrally
important to the experience and personal costs of workload.
According to his model, capacity level varies with physiolog-
ical arousal (within the lower range of arousal levels), and
both capacity and arousal levels are positively correlated with
changes in the perceived demand levels of current activities.
Motivation influences both the evaluation of task demands
(entailing perceptions of desirable performance objectives,
as well as task difficulty), and the way in which attentional
resources are allocated between different activities.
Kantowitz (1987, p. 97) emphasised the multidimen-
sional nature of workload, defining it as “a subjective experi-
ence caused by … motivation, ability, expectations, training,
timing, stress, fatigue, and circumstances in addition to the
number, type and difficulty of tasks performed, effort
expended, and success in meeting requirements”. This very
broad definition is reflected in the International Standards
Organisation ISO Standards 10075-1 (1991) and 10075-2
(1994) concerning “Ergonomics principles related to mental
workload”, which subdivides “situational influences” into
“task requirements” (sustained attention, information
processing, responsibility, duration and temporal pattern
of action, task content, danger), “environmental conditions”
(lighting, climatic conditions, noise, weather, odours),
“organisational factors” (type of organisation, organisational
climate, group factors, leadership, conflicts, social contacts)
and “factors external to the organisation” (social demands,
cultural standards, economic situation). Although referring
to mental workload, these standards are explicitly intended
to apply to all kinds of work, including work that would be
seen as primarily physical in nature (see Nachreiner, 1995).
A somewhat anomalous characteristic of these broader
conceptualisations of workload, as depicted in Figure 3,
is that they specifically encompass some of the costs
incurred as people attempt to meet task demands. Kantowitz
(1987) included effort expended, fatigue and stress, and
some others focus more directly on the personal costs
of performance as defining characteristics of workload.
Thus, Hart and Staveland (1988) defined it as “a hypotheti-
cal construct that represents the cost incurred by a human
operator to achieve a particular level of performance”
(p. 140), and Matthews, Davies, Westerman, and Stammers
(2000) noted that “workload … refers to people’s experi-
ences of cognitive task performance as effortful and fatigu-
ing” (p. 87). The level of such costs may be moderated by
the perceived adequacy of performance, and by perceived
consequences of possibly failing to cope. Someone who
is expending a high level of effort in an attempt to cope with
demands, who perceives their performance as probably
inadequate and who feel that this matters, is likely to feel
anxious, or frustrated or depressed; that is, they are likely
to experience stress (Burke, 1993; Carayon, 1993; Cox
et al., 2000; Marsella, 1994; Salvendy & Smith, 1981;
Siegrist & Peter, 1994). On this basis, stress and related
negative emotions have been included by some researchers
as a component of workload (Hart & Staveland, 1988;
Kantowitz, 1987; Wierwille Rahimi, & Casali, 1985).
1
Wickens and Hollands (2000) concluded that workload
is inversely related to reserve capacity so long as capacity
is not exceeded by task demands, but when demands exceed
capacity (eliminating all reserve capacity) such that perfor-
mance level is bound to deteriorate, workload is more likely
to vary with performance level. They also noted evidence
that the process of allocating attentional capacity between
several concurrent activities is itself demanding of attention
or effort, in addition to that required for each of the activi-
ties themselves. Some implications of the HF psychology
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
3
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
FIGURE 2
Kahneman’s model depicting of attentional capacity. From
Kahneman (1973). (permission to be obtained)
FIGURE 3
A multidimensional model of workload. Based on the defini-
tion of Kahneman (1987).
concept of workload, as outlined above, are discussed
in later sections of this paper in relation to the assessment
of workload and its role as an occupational stressor.
There is a complex relationship between workload and
stress (e.g., see Gaillard, 2001). Stress is typically associ-
ated with excessively high workload levels — even being
defined as part of the workload experience, as noted above.
On the other hand, people who perceive their work perfor-
mance as successful, despite high levels of perceived
demands and associated high levels of effort expenditure,
are likely to experience considerable satisfaction — perhaps
concurrently with significant levels of stress. We certainly
cannot assume that high workloads are necessarily stressful,
nor that people with high workloads are dissatisfied with
their jobs (MacDonald & Upsdell, 1996; Payne & Morrison,
1999). People in such jobs might be dissatisfied, for various
possible reasons including high levels of stress or excessive
workloads. In addition — or alternatively — they might
experience feelings of considerable satisfaction, engage-
ment, enthusiasm or even exhilaration (Leiter, 1999;
MacDonald & Upsdell, 1996; McIntosh, 1995; Warr, 1990).
Further, negative feelings may be experienced due to very
low workload, most typically when task performance
demands little thought and holds little interest, leading
to boredom. Such circumstances are naturally de-arousing
so there is a tendency to decrease effort expenditure or to
cease performance altogether, as depicted in the Kahneman
model (1973). In these circumstances, a requirement
to maintain good performance — either in terms of a high
quantitative output or a low error rate (such as in a monitor-
ing task) or both, can be experienced as unpleasantly stress-
ful (Cox, 1978; Hancock & Warm, 1989; Huey & Wickens,
1993; Welford, 1976).
In summary, within the domain of HF psychology there
is broad consensus that workload is more than simply the
amount of work that has to be done, although the quantity
of required work is an important factor. Rather, workload
is closely related to the demand experienced and effort
expended during work performance, and to the perceived
adequacy of performance; it is also affected by motiva-
tional factors. Within the domain of HF psychology, some
definitions of workload also incorporate affective dimen-
sions related to the costs of coping and to the perceived
(in)adequacy of performance. However, workload level is
primarily a function of task demands in relation to personal
coping capacity.
Task Demands
Task performance requires some degree of effort expenditure
— usually drawing on both mental and physical resources —
which results in a progressive increase in fatigue and associ-
ated decrease in performance capacity. The greater the level
of task demands, the greater the required effort expenditure
and consequent decrease in capacity (see Gaillard, 2001; Job
& Dalziel, 2001), which will tend to increase workload as
depicted in Figure 3. It is therefore important that managers
are aware of the wide variety of task demands that influence
workload. The expertise of HF psychologists in identifying
and assessing task demands has the potential to make a
significant contribution within the domain of occupational
stress and OHS more generally.
The following list of task demand factors is based on a
review of information about the human information process-
ing system and the demands of both psychological and
physical work (e.g., see Proctor & Van Zandt, 1994;
Wickens et al., 1998; Wilson & Corlett, 1995).
Sensory/perceptual factors. Sensory modalities used;
sensory quality and number of different sources of stimulus
information; stimulus uncertainty (number of possible
stimuli, and their probabilities); stimulus timing (frequency,
temporal uncertainty); form of stimulus information coding;
compatibility between stimulus information coding and the
person’s associated mental model(s)
Central processing factors. Working memory load
(required number of items, retention durations); decision
complexity (number and clarity of relevant factors and their
inter-relationships); required decision rate and overall infor-
mation processing rate; required recall of information from
long-term memory; amount of attention-sharing required;
degree of compatibility between concurrent activities;
requirement for concentrated attention; frequency and
salience of distractions; perceived consequences of perfor-
mance quality, errors; repetitiveness (cycle time, monotony).
Psychomotor factors. Response uncertainty (number
of possible types of responses, and their relative probabili-
ties); response timing requirements; required response
precision (target “tolerance”, distance moved); compatibil-
ity between required response and relevant mental model;
required response rate.
Physical factors. Requirements for: significant force
(e.g., lifting, pushing, pulling); significant local force
(e.g., gripping, squeezing); physical effort causing faster
breathing; awkward or effortful postures (e.g., twisting,
bending); static postures (e.g., standing still; sitting; holding
arms in set positions such as at a keyboard); repetitiveness
(cycle time, number of different action components).
Affective factors. Requirement to: hide own feelings
(e.g., dealing with situations that would normally be very
distressing, or dealing with people who are hurt, upset
or angry); requirement to demonstrate particular feelings
(e.g., happy and friendly when dealing with “the public”).
Affective or emotional demands have been the focus
of some attention in relation to professional “burnout” –
a concept that is closely akin to stress (Schaufeli, Maslach,
& Marek, 1993), but this type of task demand has received
little attention within HF psychology (an exception
is Myrtek et al., 1994). Such demands are obviously impor-
tant in work that entails significant interactions with other
people (e.g., health care, teaching, or dealing with customer
complaints as in many telephone call centres); they are
included here for completeness.
Measuring Workload
A wide range of workload measurement methods have been
developed within HF psychology, including physiological
indices (e.g., pupillary diameter; specific aspects of heart rate
variability; the P300 component of electroencephalogram
output); performance on secondary or loading tasks performed
concurrently with the “primary task” for which workload
is being assessed; primary task performance measures; and
subjective rating methods, of which the most widely used now
is probably the NASA-Task Load Index or “TLX” (Hancock
& Desmond, 2001; Tsang & Wilson, 1997).
The TLX was developed by researchers at the U.S.
National Aeronautical and Space Administration (NASA )
(Hart & Staveland, 1988), and a great deal of research has
established its validity in contexts such as the development
or evaluation of equipment and related systems of work,
usually with the aim of minimising demands on operator
capacity, or more broadly, aiming to maximise performance
quality and system efficiency (Tsang & Wilson, 1997).
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
4
WENDY MacDONALD
People rate the workload associated with their task perfor-
mance using six rating scales, of which three represent
different aspects of task demands (physical, mental, tempo-
ral); one represents the amount of effort expended in coping
with the demands; one represents perceived adequacy
of their performance; and the last represents negative
feelings such as frustration and stress – reflecting the multi-
dimensional nature of workload as discussed above.
MacDonald and co-workers have trialled and developed
some modifications to the TLX workload scales to fit them
better for use in various Australian workplaces. In the first
of several studies, they used the standard scales as part
of a battery of measures to identify factors responsible for
an elevated rate of overuse injuries among assembly line
workers (O’Bryan, MacDonald, & Evans, 1991). TLX
ratings identified significant differences between two of the
tasks in terms of three of its six dimensions: physical
demand, temporal demand and effort; these differences
were validated using a range of objective analysis and
measurement methods including heart rate, and Modapts
which is a widely used and validated Predetermined Motion
Time System (PMTS) for setting or evaluating work rates
(Heyde, 1976; Hoffmann, 1992; Hoffmann, MacDonald,
& Almond, 1993).
MacDonald (2001a) developed a modified set of TLX
scales based on initial interviews with members of the train
controller target population, for use in assessing controller
workload levels for automated versus non-automated
control systems. The standard TLX Performance scale was
omitted since in this safety-critical work it was unlikely that
ratings would vary below a “reasonably good” level, and
previous studies (O’Bryan et al., 1991; So & Mert-Iljin,
1990) had found very little variation on this scale. The
scales used were as follows:
•Perceptual Demands (task information able to be noticed
easily when needed; able to be seen or heard easily)
•Mental Demands (thinking and planning; decision-
making; switching attention between different aspects
of the work while maintaining task priorities; remember-
ing to do things; recalling information when needed)
•Importance of Avoiding Errors (awareness of conse-
quences of errors)
2
•Physical Demands (physical discomfort or tiredness
related to posture; reaching and twisting; effects of seat-
ing and work station layout; adequacy of opportunities to
walk around)
•Time Pressure (work rate demands; time available to deal
with things)
•Effort Required (concentration level; how much attention
needed)
•Frustration Experienced (includes stress, annoyance,
irritation).
Qualitative effects of varying automation levels were identi-
fied in terms of the relative importance of the different
scales. Ratings on these scales also highlighted an enforced
change in speed-accuracy performance strategy among
controllers in the non-automated task condition, where very
high workload conditions were associated with a reduction
in the relative importance of avoiding errors. Relative
differences in overall workload levels for the two systems
were quantified using uni-dimensional magnitude estima-
tion ratings (Alteras-Webb & Dekker, 1994; Gescheider,
1976; MacDonald, 1979), and results were used in assessing
the need for changes in controller staffing levels when the
new system was implemented. Separate task analyses of key
tasks were used to identify task demands and potential
overload situations related to competing needs for atten-
tional resources, based on Wickens’s multiple resource
model (Wickens, Gordon, & Liu, 1998).
So and Mert-Iljin (1990) used semi-structured interviews
and TLX scale ratings to explore workload dimensions
for university technical staff varying in seniority level.
The TLX scales showed both quantitative and qualitative
differences in workload between junior and more senior
staff, and between three different departments. However,
the interview results suggested that these six scales were not
sufficient to cover all dimensions of workload; in particular,
the demands of work-related relationships with other
people, and lack of role clarity, were identified as signifi-
cant omissions. Upsdell (1994) and MacDonald & Upsdell
(1996) reported use of the TLX in conjunction with
measures of occupational stress, burnout, and job satisfac-
tion, to identify predictors of stress-related illness associated
with absence from work, and to explore correlations
between these measures. They found that high ratings on the
TLX Mental Demand scale were associated with lower
levels of stress, higher levels of job satisfaction and self-
reported health. Such relationships are a salutary reminder
that, as noted earlier, demanding jobs need not necessarily
be stressful (also see Hart & Cooper, 2001).
Overall, the above group of studies demonstrated that
standard NASA-TLX scales fail to address some types
of job demand which would be expected, based on the
workload model presented above, to influence people’s
experience of workload related to their overall job perfor-
mance. In the interests of greater conceptual clarity regard-
ing the constructs that are being assessed, it seems
preferable that measures of affective state (e.g., different
types of “stress” and “satisfaction”) should be clearly delin-
eated from measures of workload (despite stress being
included within some definitions of workload). This
is likely to be particularly important in cases where the
primary purpose of workload assessment concerns
employee health and wellbeing, rather than the evaluation
of equipment or work systems.
In their concluding discussion of current stress measure-
ment methodology and related research needs, Cooper et al.
(2001) suggested that there is a need for greater specificity
in assessing “strain” (i.e., stress states), suggesting that the
nature of the experience may differ in accord with different
types of causal factors. MacDonald (2001a) reported
evidence supporting this suggestion. Interviews with train
controllers found that they interpreted “stress” as anxiety
related to performing safety-critical work under conditions
of very high task demands, in contrast to “frustration”
which they related to work that was more demanding
or unpleasant than otherwise due to a problem that could
have been avoided by management actions such as replac-
ing faulty equipment. Such distinctions are not reflected
in the TLX scale labelled “frustration”, which is opera-
tionally defined in terms of the following mixture of both
positive and negative states: insecure, discouraged, irritated,
stressed, annoyed versus secure, gratified, content, relaxed,
complacent. The meanings of these words do not vary along
a single continuum; in fact an earlier version of the TLX
had separate scales for frustration and stress, which in this
later version were combined. For some applications,
however, distinctions between different affective states may
well have diagnostic value, and should be preserved.
Job Demand and other Psychosocial Hazards
Moving from task demands to the broader category of job
demands, we move from the domain of HF psychology and
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
5
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
ergonomics to that of occupational stress. Here, task
demands are subsumed within “job demand”, which in turn
is categorised as one of several types of work-related “stres-
sors” or “psychosocial hazards” (the latter two terms are
used synonymously; see Rick, Briner, Daniels, Perryman,
& Guppy, 2001, for a recent review). Thus, the concept
of “demand”, which is centrally important within the HF
view of workload, is subsumed within a much broader
category. Cox and Griffiths (1996, pp. 129–130) defined
psychosocial hazards as “those aspects of work design, and
the organization and management of work, and their social
and organizational contexts, which have the potential for
causing psychological or physical harm”.
Psychosocial hazards — including job demand — are, by
definition, potential predictors of negative effects on employ-
ees, including stress and health problems, and there is ample
empirical evidence of such relationships (e.g., Cooper et al.,
2001; Cox et al., 2000; Eriksen & Ursin, 1999; Ganster,
Hurrell, & Thomas, 1987; Karasek & Theorell, 1990; Kasl,
1998; Paoli & Merllie, 2001; Theorell, 1998). In the case of
injuries such as musculoskeletal disorders, particularly
“overuse” injuries, most research has until recently focused
on physical risk factors (e.g., forces, postures) but relation-
ships between the commonly recognised psychosocial
hazards and physical overuse injuries have now been estab-
lished (e.g., Andries et al., 1996; Bernard, 1997; ??Cox
& Griffith, 1996; Devereux & Buckle, 2000a, 2000b; Evans
& Patterson, 2000; Karasek et al., 1998; Randall et al., 2002).
The widely accepted argument of Cox and co-workers
(e.g., ??Cox & Griffith, 1996) is that psychosocial hazards
may damage people’s physical and/or psychological health
via “a psychophysiological, stress-mediated mechanism”.
Further, they point out that purely physical hazards — for
example, exposure to organic solvents — can damage health
via two different paths: directly via a physico-chemical
pathway, and indirectly via the same stress-mediated
mechanisms as psychosocial hazards, since exposure to
physical hazards can cause fear of being harmed, which in
turn may produce a stress response with potential damage to
health. However, the causal mechanisms underlying associ-
ations between psychosocial hazards, stress and physical
injury are poorly understood (??Huang et al., 2002). There
is a need for more multidisciplinary research in which valid
and sensitive measures of both physical and psychological
work demands, as well as a comprehensive range of other
potential risk factors, are employed.
Job Demand Factors
“Job demand” is a poorly defined construct, both conceptu-
ally and in operational terms. Cox and co-workers (e.g., Cox
& Griffith, 1995, 1996; Cox & Rial-Gonzalez, 2002)
include various types of job demand within their standard
set of “psychosocial hazards of work”, including work
overload or underload, time pressures, high work pace, short
work cycle times and high uncertainty. Similarly,
commonly used stressor measurement scales include at least
one construct representing job demands, although terminol-
ogy and content vary considerably. Construct labels include
“workload” (referring to the amount and difficulty of work;
Williams & Cooper, 1998); “work pressure” (Carayon
& Zijlstra, 1999); “job pressure” (Vagg & Spielberger,
1998); “cognitive demands”, “quantitative job demands”
(Hurrell & McLaney, 1988); “psychological demands and
mental workload”, “physical demands” (two constructs;
Karasek et al., 1998).
Probably the most comprehensive in its coverage of job
demand is the widely used Job Content Questionnaire (JCQ;
Karasek, 1985), based on the Demand-Control model (see
Karasek & Theorell, 1990). JCQ items within “psychologi-
cal job demands” are summarised as “work hard”,
“no excessive work”, “enough time”, “conflicting
demands”, “intense concentration”, “tasks interrupted”,
“hectic job”, “wait on others”; items within the JCQ
construct “physical job demands” are “much physical
effort”, “lift heavy loads”, “rapid physical activity”,
“awkward body posture” and “awkward arm positions”.
It can be seen that some of these items reflect some aspects
of the task demand factors identified above, while other
items reflect a more holistic experience of work as more
generally difficult, or demanding excessive amounts
of work in relation to time constraints. The JCQ does not
specifically include any affective task demands, although
some elements of such demands are present within both
“supervisor social support” and “co-worker social support”
(e.g., hostile supervisor or coworkers).
A more recently developed scale — the Pressure
Management Indicator (PMI; Williams & Cooper, 1996,
1998), based on the earlier Occupational Stress Indicator
(OSI; ?? Cooper et al., 1988) — includes a “workload”
construct within a broader group of other, significantly
correlated “sources of pressure”. Other sources of pressure
(showing in brackets their correlations with “workload”
as reported by ??Williams & Cooper, 1997) are “home-
work balance” (.59), “managerial role” (.56), “daily hassles”
(.56), “personal responsibility” (.53), “organization climate”
(.48), “relationships” (.44) and “recognition” (.23). The PMI
was developed to overcome problems with the construct
validity of the earlier OSI, particularly its “sources of
pressure” scale which included “workload” or “work
demands”, “managerial responsibility” and “pressures in the
role of employee” (see Rick et al., 2001, pp. 50–55).
Another recent scale — the Job Stress Survey (Spielberger
& Vagg, 1998; Vagg & Spielberger, 1998) was developed
from earlier scales that focused specifically on police stress
and on teacher stress. “Job pressure” is one of its two
constructs (the other being “lack of support”). “Job
pressure” items, in order of their factor weights, are
“meeting deadlines”, “frequent interruptions”, “excessive
paperwork”, “frequent change”, “critical on-the-spot
decisions”, “assigned increased responsibility”, “assignment
of new duties”, “insufficient personal time”, “dealing with
crisis situations”, and “perform tasks not in job description”.
While these various scales have much content in common,
none appears to provide comprehensive coverage of either
“job demand” or “workload” in the HF psychology sense.
Current research (Miller & MacDonald, 2003) addresses
this issue.
It is beyond the scope of this paper to provide a compre-
hensive review of all occupational stressors or psychosocial
hazards. However, other psychosocial hazards can be
expected to influence workload, and some of these are
discussed in the following sections.
Job Control and Participation Level
“Control” has equal status with “demand” within Karasek”s
job content model; level of control or participation is likely
to influence both workload and stress. There is consensus
that stress is more likely when people “… are constrained in
the way they carry out their work and cope with its
demands: they have little control over their work or how
they cope with it” (Cox & Griffiths, 1995, p. 790). That is,
a higher level of control enables people to take actions
to reduce actual or potential stressors (Frese, 1989).
Hockey, Briner, Tattersall, and Wiethoff (1989) presented
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
6
WENDY MacDONALD
a control model of stress regulation which explained
this process in terms of people gaining greater control over
their workload level. Greater control might provide access
to a wider range of alternative performance strategies,
enabling people to decrease demand and thus avoid stress-
fully high workloads, or to increase demand and thus avoid
stressful boredom; or it might provide access to additional
coping resources of various possible kinds, again to avoid
stressfully high workloads.
This view of the relationship between control and
workload is consistent with more recent research (Carayon
& Zijlstra, 1999) showing that when the effects of different
types of control are analysed separately, higher levels
of instrumental control (relating to influence over task
factors such as amount and rate of work, which would
directly influence workload) are related to lower levels
of “work pressure”, whereas higher levels of decision
control (relating to influence over organisational processes,
procedures and policies, which might be associated with
greater responsibilities without any counteracting decrease
in demand) was related to higher “work pressure”. ??Van
Der Doef and Maes (1999) concluded from their major
review of evidence that “Only aspects of job control that
correspond to the specific demands of a given job moderate
the impact of high demands on well-being.” Such findings
emphasise the importance of identifying and assessing
the demands specific to particular tasks and jobs, in order
to manage workload, stress and wellbeing more effectively.
Both the type and extent of control can vary consider-
ably between different types of work, and one group
of researchers (Jackson, Wall, Martin, & Davids, 1993;
Mullarkey, Jackson,Wall, Wilson, & Grey-Taylor, 1997)
have developed two scales that are intended for use
in manufacturing environments. These relate to two differ-
ent aspects of control: method control (how the work
is done) and timing control (work pace, task order).
Machine pacing such as by a moving assembly line imposes
a high level of external control over the pace of work,
minimising opportunities for employees to decide when
to speed up and when to slow down. Performance may also
be paced by the operation of production systems in which
subsequent stages, performed by different people, cannot
commence until work on a previous stage has been
completed, thus imposing a series of performance deadlines;
the product of each phase of the system must be in time
to meet the need of the next phase. A specified daily
production target imposes much less constraint than
a moving assembly line, but it imposes more than a weekly
production target. A high level of external pacing is gener-
ally associated with increased stress (see Salvendy & Smith,
1981). For example, comparison of machine-paced and self-
paced jobs in a sawmill showed that machine pacing
was associated with higher levels of catecholamines —
a physiological indicator of increased stress (Johansson,
Aronsson, & Lindstrom, 1978). All forms of external pacing
impose a specific rate of work, on average, but it is impor-
tant to note that this rate might be high or low. That is,
a high level of external pacing does not necessarily impose
a high work rate: that is, control is not necessarily corre-
lated with demand.
There is evidence that participation in setting the work
rate may to some extent counteract the negative effects
of machine or external pacing (Johansson, 1981; Shikdar
& Das, 1995), and the benefits of employees participating
in the general management of their own work and work-
place have been well documented, particularly in terms
of scores on the “control” dimension of the Karasek model
(e.g., see Karasek et al., 1998); there is also evidence
of a more qualitative nature (e.g., Bertone et al., 1998; Sen,
1987; Wilson, 1995). Apart from the potential for positive
health effects, optimising levels of employee control and
participation can improve job satisfaction and morale, yield-
ing a range of organisational benefits (e.g., Blewett & Shaw,
1995; Marmot, 1994; Parker & Wall, 1998).
Support
The third major category of stressors or psychosocial
hazards are those affecting “support”. The original demand-
control model of Karasek and co-workers was modified
by addition of a third dimension: “social support” (Johnson
& Hall, 1988), based on the hypothesis (confirmed in some
but not all studies) that “jobs which are high in demands,
low in control, and also low in social support at work carry
the highest risk of illness” (??Karasek et al., 1998).
The importance of support mechanisms in relation to stress
is demonstrated by results from a factor analysis of res-
ponses to the Job Stress Survey (Vagg & Spielberger,
1998), from which two factors emerged: “job pressure”
(representing demands) and “organisational support”.
Increased support can increase coping capacity by providing
additional resources, whether by means of instrumental
or affective support (e.g., Greller, Pasons, & Mitchell, 1992;
Karasek et al., 1998; Theorell, 1998). Instrumental support
would tend to reduce workload by, for example, improving
the availability and/or quality of various types of supportive
resources such as equipment, adequate staffing level,
helpful supervisors and management policies that facilitate
work performance.
Further, affective or interpersonal support would tend
to increase morale, which might increase an individual’s
willingness to expend effort, which would increase their
coping capacity and hence decrease their workload level
(as defined earlier). It might also be associated with
workload reduction due to the provision of informal, instru-
mental support, particularly during brief peaks in demand
levels. Provision of performance feedback can function both
as affective or interpersonal support, and as instrumental
support in that it conveys information that can be used
to “fine tune” performance strategies which might result
in more efficient (reduced) effort expenditure and hence
reduced workload, as shown in the performance of a labora-
tory vigilance task for which workload was measured by the
TLX (Becker, Warm, & Dember, 1991).
In terms of the “effort-reward” stress model (Siegrist,
1996), it is also likely that the personal reward value
of some types of support might reduce imbalance in the
effort/reward equation, thus reducing stress levels. This could
result in the adoption of a somewhat different performance
strategy which might either increase or decrease workload
level. More generally, the satisfaction experienced in relation
to performance of jobs with high “motivating potential”
(Hackman & Oldham, 1980) might also have some reward
value in this sense, although such jobs can also be more
demanding, to the extent that their motivating potential
is derived from high levels of task variety and autonomy.
There is a need for better empirical data regarding effects
on perceived workload when similar levels of high effort are
accompanied by positive versus negative affect.
Job Role Clarity, Ambiguity and Conflict
These are potential stressors (e.g., Osipow & Spokane,
1987) which have the potential to affect workload.
Uncertainties concerning performance requirements are
likely to influence performance strategies and related effort
expenditure levels, perhaps to an extent determined
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
7
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
by factors such as job insecurity. For example, someone
who is uncertain about how much work they are expected
to complete, and to what standard, and who is anxious about
the possibility of losing their job, may expend more effort
in producing a larger amount and/or higher quality of work
than otherwise. Also, anxiety associated with increased
uncertainty is itself stressful.
Work Schedule: Working Hours, Shift System,
Rest Breaks
The effects of shift systems and associated working hours
have been the topic of a large amount of research (e.g.,
Charyszyn & Tucker, 2001; Paoli & Merllie, 2001; Rosa,
1995; Wallace & Greenwood, 1995; Wooden, 2001b),
which cannot be reviewed here. Overall level of job demand
is determined by total hours worked — representing
the duration of exposure to task and job demands — as well
as by the level of demands during this time. Increased
working hours are associated with higher fatigue levels,
particularly if opportunities for recovery from accumulated
fatigue are inadequate, and people’s capacity to cope with
work demands decreases as fatigue increases, with conse-
quent increases in workload experienced (since this is a
function of the size of the margin between demands and
capacity), increasing the probability of stress. Due to
chronic lack of sleep, shift workers are likely to be most
affected by fatigue and experience higher stress levels.
Changes in performance strategy: for example, working
more slowly or deliberately reducing performance quality,
may offset such effects, at least partially. Currently, one
of the most important workplace issues related to fatigue
management concerns the interacting effects of working
hours, shift systems and workload levels. To date there has
been very little research directly addressing this issue
(MacDonald & Bendak, 2000). It is argued in the final
section of this paper that this is largely due to a lack of both
the conceptual basis and the measurement tools required
to assess workload.
Workplace Environment and Other Influences
A wide variety of both physical and psychosocial environ-
mental factors can affect workload, directly or indirectly.
For example, hot environments can reduce performance
capacity (optimal temperature level varies with the nature
of required performance) and hence workload, if the same
performance level is maintained. Employee perceptions
of organisational values and expectations regarding accept-
able work practices and the quantity and quality of work
performance, as well as more general social norms concern-
ing “a fair day’s work”, are likely to affect motivation and
performance strategies, with consequent effects on effort
expenditure and workload. Motivational issues are particu-
larly difficult to manage when the work is machine-paced.
Such work is often simple and repetitious with short cycle
times, which many people experience as highly monotonous
and unsatisfying. Employees in manufacturing environ-
ments commonly experience some social isolation due
to high noise levels, which would be expected to reduce
their coping capacity and resistance to stress. Such isolation
means that they are less likely to receive social support from
colleagues, which also effectively reduces their capacity
to cope and has been associated with increased stress
(e.g., Cox & Griffith, 1995; Karasek & Theorell, 1990;
Sauter, Hurrell, Murphy, & Levi, 1997).
Effects of Task Demands
and Workload on Fatigue
and Stress in Manufacturing Industry
Results from a recent project are presented here to illustrate
the significant effects of task demands and workload,
as well as level of control and related motivational factors,
on employee fatigue and stress levels (MacDonald, Evans,
Nolan, & Pratt, 1999; MacDonald, 2001b, 2001c). The pro-
ject was conducted to document and evaluate methods used
to set standard work rates for repetitive tasks, with particu-
lar focus on whether sufficient allowance had been made for
task demands and associated workload. There was some
prior evidence that PMTS calculations take insufficient
account of mental demands and of some aspects of physical
demand (Addison & MacDonald, 1993; Hoffmann, 1992;
Hoffmann et al., 1993; O’Bryan et al., 1991; Sendapperuma,
MacDonald, & Hoffman, 1991). Data were collected during
the late 1990s about 82 work tasks in 22 Melbourne
workplaces. In Stage 1 of the project, information about
tasks was collected by direct observation and by question-
naire-based interviews with production managers, OHS
managers, and 186 workers on the specific tasks under
investigation. Stage 2 focused in more depth on a subset
of these tasks; formal task analyses based on videotapes
of task performance and related physical measurements
(including heart rates) were used to quantify required forces,
postural demands and metabolic demands, and further infor-
mation was obtained from employee interviews.
Work Pacing and Work Rate Determinants.
One or more of the following factors determined or strongly
influenced the pace at which work was performed: machine
operating or production process time (affected 62.3%
of tasks, including 13.5% of all tasks where there was liter-
ally a moving line speed, and 16% classified as “end
of line”); production targets (affected 70% of tasks); pro-
duction orders and related deadlines (affected 46% of tasks).
It was reported that some form of “standard” times, based
either on industry standards or using a PMTS such as
Modapts, had been used in setting the standard work rate for
13.5% of tasks in the whole sample.
3
The researchers used
Modapts to calculate a standard performance time for each
Stage 2 task. Actual performance times clustered close
to Modapts standard times, but 45% of tasks were perform-
ed faster than the standard (including a 15% “fatigue”
allowance, as is usual).
Task Demands and Workload
Objective task demand measures. Based on observations
of task performance, the researchers scored each task (0, 0.5
or 1) on each of four physical demands: heavy loads
(lift/push/pull); local loads (manual pinch, squeeze grips);
static postures; metabolic demands; and four mental
demands: perceptual load; motor control, precision; working
memory load; concentration, mental effort. Contrary
to common assumptions about such work, mental demands
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
8
WENDY MacDONALD
TABLE 1
Workload Dimension Ratings (Modified TLX Scales)
Workload Dimension Mean Standard Deviation
Getting things right 74 24 (new dimension)
Mental Demand 70 23 (similar to standard TLX)
Effort 70 22 (similar to standard TLX)
Physical Demand 65 26 (similar to standard TLX)
Time Pressure 63 25 (similar to standard TLX)
Frustration 42 27 (similar to standard TLX)
tended to be higher than physical demands, as shown in
Figure 4.
Employee ratings of task difficulty. Tasks were rated on
a 10-point scale (1 = extremely low; 10 = extremely high).
Mean rating was 5.4 (SD = 2.4). The correlation between
these ratings and researcher-generated task demand scores
(total of mental and physical demand scores) was 0.41
(p= .002). A correlation of 1 was not expected since these
scores reflected variety in types of demand more than
overall demand level (maximum demand level = 1 for each
of 8 demand types).
Workload dimension ratings (modified TLX scales).
Employees also rated Stage 2 tasks on different workload
dimensions, using 0–100 scales similar to standard NASA-
TLX scales. Mean values (and SDs) were as shown in Table 1.
The non-standard TLX scale — “working carefully
to avoid errors” — had been previously developed for
another project to assess subjects’ concerns about their own
performance in a way that was less personally threatening
than the standard TLX “own performance” scale.
The relatively high rating on this scale reflected employee
perceptions of the need to maintain high quality standards
rather than simply to maintain a high output or work rate;
questionnaire responses indicated that many employees felt
pressure to maximise both speed and output quality, creat-
ing some conflict. Mental Demand was rated higher than
Physical Demand, consistent with researcher demand
scores. The relatively low value for Time Pressure suggest-
ed that, on average, work rates were not excessive for most
tasks (confirmed by Modapts analyses).
Employee Fatigue, Stress and Arousal
Fatigue is seen as an outcome of effort expenditure and was
therefore expected to be correlated with workload; it was
measured on a single scale (0–100), similar to a TLX scale.
Mean fatigue level was 56 (SD = 24). Stress and Arousal
were measured using the Stress Arousal Checklist (SACL;
Gotts & Cox, 1990). These are well established constructs
representing people’s mood at work (e.g., Cox & Griffith,
1995; Gotts & Cox, 1990; Mackay et al., 1978).
Van Katwyk, Fox, Spector, and Kelloway (2000) indepen-
dently identified, and confirmed the validity of these
two dimensions, describing them as pleasure-displeasure,
and degree of arousal. Mean Stress score was 6.5 (SD = 4.0)
and Arousal was 7.4 (SD = 3.1); both values are within one
SD of norms (??Cox & Griffith, 1995).
Results from multiple regression analyses to identify the
main predictors of Fatigue ratings, Stress and Arousal
scores are shown in tables 2 to 4, respectively.
As expected, fatigue ratings increased with Workload
(mean of TLX scales), and also when required work rate
was too fast (which is closely linked, conceptually, to high
workload).
4
There was a negative relationship between
fatigue and job satisfaction (General Satisfaction scale from
the Job Diagnostic Survey, Hackman & Oldham, 1980
5
).
Since working to meet deadlines might be seen as stressful
or fatiguing, it is interesting that fatigue ratings were lower
for tasks whose work rates were influenced by the need
to meet orders or deadlines; this is discussed below
in relation to Arousal level.
It was expected that Stress score would increase with
high workloads (confirmed by its association with mean
TLX scales), due to the small margin between task demands
and coping capacity. Stress was also expected to increase
with highly repetitive work (confirmed by its association
with short cycle times), due to arousal level tending
to decrease and thus make it difficult to sustain performance
at the required level, and due to boredom. Low Motivating
Potential Score (MPS; from the Job Diagnostic Survey,
Hackman & Oldham, 1980) was also associated with high
Stress. The MPS represents a set of job characteristics
which facilitate both coping capacity and job satisfaction,
based on reasonable autonomy and feedback, some variety,
and the opportunity for some sense of “ownership” and
pride in the work performed. The last factor in the regres-
sion model was whether or not work performance was
externally paced; as expected, Stress was higher when
performance was tied to process/machine times or line
speed. This variable had one of the stronger bivariate
associations with Stress; its contribution to the above model
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
9
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
FIGURE 4
Distribution of physical and mental task demand scores for all Stage 1 tasks.
was reduced by its relationship with task cycle time.
However, the hypothesis that Stress would be lower with
higher employee ratings of their amount of participation
or control was not confirmed, probably because it was
generally rated as low, with relatively little variation.
Arousal was expected to increase with overall workload.
However, it was better predicted by some specific task
demands: employee ratings of “effort” and “working care-
fully”, together with two researcher-generated measures:
total task demand score and postural stress score. The latter
two factors made little contribution to the final regression
model but were retained because they had the highest and
second highest (respectively) bivariate correlations with
Arousal. In addition, a high Arousal score was strongly
predicted by the presence of orders or deadlines as a factor
influencing variation in work rates at different times. High
Motivating Potential Score was also a strong predictor of
Arousal. It seems that working to meet deadlines might
often have been experienced as an enjoyable challenge
rather than an unpleasant stressor; this interpretation is
strengthened by the negative relationship between Arousal
and Fatigue, which is suggestive of people being “too busy
to feel tired”, in a positive sense. However, tasks where
performance was influenced by deadlines also tended to
have longer cycle times and higher MPS scores, so interpre-
tation is difficult.
One of the key questions addressed by this research was:
did required work rates make adequate allowance for the
difficulty of the work? The evidence on this question was
clear: employees performing the more demanding tasks
experienced higher levels of stress, and were more likely
to report the required work rate as too fast. It was evident
from other analyses that this was more likely to occur when
tasks were mentally rather than physically demanding; that
is, higher levels of mental task demands appeared to have
been inadequately taken into account by those responsible
for determining standard work rates. Employees’ satisfac-
tion with their overall amount of say in the workplace was
also recorded; this factor represented perceived influence
at the general, organisational level, rather than direct, instru-
mental control (see Carayon & Zijlstra, 1999). It was found
that employees who were dissatisfied with their amount
of “say” were more likely to perceive their required work
rate as too fast.
Overall, these results demonstrate the significant role
of task demands and related workload as determinants
of employee stress and fatigue. However, motivational
factors were also important; these related to work design
(levels of external pacing and repetitiveness), to job design
(Motivating Potential Score, varying work pace to meet
deadlines) and to broader, organisational factors (amount
of general “say” that employees feel they have). In view
of the factors that were not measured in this project
(e.g., personal and non-work factors), the amount of varia-
tion explained in the above regression models is substantial.
Workload Measurement Revisited
It is argued that “workload” is an important determinant
of workplace fatigue and potentially, stress. However,
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
10
WENDY MacDONALD
TABLE 2
Results from Multiple Regression Analysis to Identify the Main Predictors of Fatigue Ratings: F= 19.32, p< .000;
Adjusted R
2
= .63.
Bivariate Correlations
Variables Fatigue WKLD GENSAT DEDLIN
β
p
Workload (mean TLX): WKLD .63 .44 .000
General Satisfaction: GENSAT –.55 –.16 –.39 .000
Orders, deadlines: DEDLIN –.34 –.12 .14 –.22 .033
Work rate too fast: FAST .48 .43 –.30 .11 .19 .089
TABLE 3
Results from Multiple Regression Analysis to Identify the Main Predictors of Stress Scores
(Stress Arousal Checklist; Gotts & Cox, 1990): F= 7.57, p<.000; Adjusted R
2
= .32
Bivariate Correlations
Variables Stress WKLD MPS CYCLE
β
p
Workload (mean TLX): WKLD .39 .42 .000
Motivating Potential Score: MPS –.31 .10 –.28 .019
Task Cycle Time: CYCLE –.30 .04 .10 –.21 .106
Rate set by Process, Line speed: PACED .34 .02 –.30 –.52 .14 .299
TABLE 4
Results from Multiple Regression Analysis to Identify the Main Predictors of Arousal Scores (Stress Arousal Checklist; Gotts &
Cox, 1990): F= 6.402, p<.000; Adjusted R
2
= .36.
Bivariate Correlations
Variables Arousal DEDLN MPS EFFORT POSTR TOTDM
β
p
Orders, deadlines: DEDLN .36 .32 .009
Motivating Potential Score: MPS .30 –.08 .23 .048
TLX EFFORT .46 .08 .16 .24 .101
Posture Score: POSTR .32 .25 .06 .21 .13 .279
Total Demand Score: TOTDM .42 .36 .24 .47 .38 .07 .624
TLX Work Carefully: CARE .35 –.10 .17 .55 .14 .14 .19 .156
it appears that workload is not being well managed in most
workplaces, largely due to a lack of both the conceptual
basis and the measurement tools required to assess it.
Within the domain of HF psychology, the validity
of TLX workload measurement scales is well established
for purposes such as assessing system usability and related
safety or efficiency of operations; modified TLX scales
have also been found useful in a variety of workplace appli-
cations. However, these scales do not include a range
of important factors that are either components or modera-
tors of the overall workload of a job, such as various types
of control and support, lack of role clarity, job insecurity,
and so on. Neither do they adequately assess the affective
states that are associated with (or, according to some defini-
tions, that are a part of) the experience of workload, and that
are centrally important in mediating effects on workplace
health. Nevertheless, the theoretical basis of HF psychology
workload measurement is more coherent than that of the
scales currently available to assess workplace job demands.
This is not surprising, given that the primary focus of work
within the latter domain has been on stress itself, rather than
on the work demands or other factors that may generate it.
Accordingly, scales relating to work demands have been
developed primarily to assess their contribution to stress.
Cox and Rial-Gonzalez (2002, p. 4) asserted that “Most
researchers agree on exactly which factors of the work
environment can cause employees work-related stress.
We can broadly divide these factors into physical hazards
… and psychosocial hazards.” On the other hand, ??Hurrell
et al. (1998) observed in their review of measurement
methods for “job stressors and strains” that there was a need
for “careful examination of the adequacy of existing
constructs and measures for capturing the demands
of contemporary work” (p. 368). Further, Rick et al. (2001)
commented in their comprehensive review of such scales
that “many forms of validity and, in particular, content and
construct validity, are very seriously compromised by the
limited and weak theory which underlies some of the
[psychosocial] hazard measures reviewed” (p. 79).
Viewed from the perspective of HF psychology, the
absence of any conceptual distinction between job demands
and the various other factors which currently are grouped
together as psychosocial hazards or stressors is surprising.
(Also surprising is the omission from many stressor scales
of physical demands, reflecting the focus of much stress
research on “white collar” work.) Conceptually conflating
psychological task demands and other potential stressors
makes it more difficult to identify the most appropriate
types of intervention to improve system performance, since
it is easily assumed that psychosocial factors are, by their
nature, best managed by strategies to influence employees’
motivation or morale. However, if the perceptual and cogni-
tive demands of a set of work tasks require more time than
is made available for their performance, people will experi-
ence excessive workload, for which the most direct solution
is to reduce level of demand. Such problems need to
be distinguished from those where the issue concerns
motivation more than overload, for example due to poor job
design (e.g., lack of variety), or lack of supervisor feedback,
or inadequate opportunities for participation in decision-
making, because the required interventions are very differ-
ent. The absence of a conceptual framework to support such
distinctions helps to sustain the viewpoint whereby exces-
sive physical work demands are recognised as a legitimate
cause for concern and are to some degree controlled
by legislation, while in contrast, other types of task demands
(perceptual, cognitive, psychomotor, affective) are under-
estimated or ignored (Hoffmann et al., 1993; MacDonald
et al., 1999; MacDonald, 2000b, 2000c). If their work is not
physically demanding, employee complaints about exces-
sive workloads are easily interpreted as due to motivational
problems when in fact the work demands might simply
be too great for tasks to be completed in the available time.
It cannot be assumed that psychosocial factors are,
by their nature, best managed by strategies to influence
employees’ motivation or morale, without consideration
of issues relating to people’s performance capacities and
workloads. More highly motivated people are likely
to perform at levels closer to their maximum capacity,
which is desirable — up to a point. The difficulty is in
identifying that point. Humans are highly adaptable when
confronted by impending overload; there is often scope for
them to adopt performance strategies that achieve less than
optimal performance but demand less attention, thus freeing
some attentional resources to enable performance to continue.
For these reasons, human performance typically “degrades
gracefully” under overload conditions, and for many
workplace tasks there is no clear specification of optimal
performance — or even a minimum acceptable performance
level. Therefore, there is often no clearly delineated point at
which a person becomes “overloaded”, such as might
provoke management action to reduce workload. However,
employees themselves are usually aware of their own perfor-
mance deterioration, and many experience it as a particular
source of stress. For example, this was observed by
MacDonald et al. (1999) among manufacturing workers,
many of whom reported conflict between the need to
maintain both high performance quality and a high work rate.
The fact that people continue working despite being
overloaded for much of the time, is not of itself evidence
that the “system of work” is safe. For example, use of piece
work payment or incentive systems that are closely linked
to work output can increase the risk of employee injury
or illness (NIOSH, 1996; Quinlan & Bohle, 1991).
Similarly, it is strongly arguable (and supported by the court
findings and guidelines cited earlier) that employers who
routinely and knowingly operate with sub-standard staffing
levels are exposing their employees to an unacceptable
hazard — that of chronically excessive workload. A parallel
might be drawn with the now unacceptable practice of
employers paying “danger money” to employees who were
knowingly exposed to hazardous situations, rather than
acting to remove or significantly reduce the hazard and
employee exposure to it.
However, the means of solving such problems are not
readily available, since relationships between workload,
performance, satisfaction, stress and health are not well
understood. For example, a recent European Union survey
(Paoli & Merllie, 2001) found that the association between
stress levels and hours worked was not a close one; clearly,
other factors such as job security, level of control and
various other sources of satisfaction are also important
influences. It is a common belief that people who enjoy
their work — for a range of possible reasons — are to some
degree protected from what might otherwise be the ill
effects of excessively high workloads. However, there
is increasing evidence of the negative effects on health
of excessive working hours. For example, recent Japanese
research found that weekly working hours were related
to increased odds ratios of acute myocardial infarction, with
a twofold increased risk for 61 hours or more worked per
week, compared with 40 hours or less (Liu & Tanaka,
2002). In the absence of any evidence that high workloads
can in some circumstances be free of significant risk, the
weight of current evidence suggests that employers should
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
11
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
act to limit workloads, if only to protect themselves from a
possible charge of failing to provide a safe system of work.
To support managers in both safeguarding the health and
wellbeing of their employees and maximising productivity,
there is a need for better methods of identifying the factors
that influence workload, and of monitoring relationships
between workload levels and employees’ wellbeing and
performance. More generally, ??Hurrell et al. (1998)
concluded in similar vein that: “It’s abundantly clear that
there is a need for new and improved measures in the field of
occupational stress”, arguing that the wider use of objective
measures of job-related stressors is necessary “to advance
our knowledge regarding job-stress-related ill health”, and
further, that there is “a need to carefully consider the
adequacy of existing job-stressor constructs and the instru-
ments that are currently used to measure them.”
All too often, it appears that job-stress researchers choose
measures more on the basis of the names of the constructs
of interest rather than on a careful evaluation of the
requirements of the research situation, the characteristics
of the measures, and the availability of alternative
measures (??Hurrell et al., 1998, p. 385).
In particular, they noted that “the measurement of cognitive
aspects of work seems to offer an important challenge
to job-stress researchers”. Also, the wider use of objective
measures of work demands and other stressors, where
feasible, is often recommended to complement the usual
self-report methods (??Hurrell et al., 1998; ??Kasl, 1998;
??Spector et al., 1988). A few such measures have been
developed within the domain of occupational stress
(e.g., ??Elo & Vehvilaienen, 1983; ??Greiner & Leitner,
1989; ??Greiner et al., 1997). However, there is consider-
able scope for more widespread workplace application
of HF psychology methods, particularly in assessing physi-
cal and cognitive work demands (e.g., see Kirwan
& Ainsworth, 1992; Salvendy, 1997; Wilson & Corlett,
1995). In summary, it is argued that application of the HF
psychology workload concept to the measurement of job
stressors would support such developments. It would serve
to delineate more clearly the separate effects of employee
capacity-limited and motivation-limited factors on work
performance and associated affective states such as stress,
thus enabling managers to monitor and manage workload
levels more effectively, as part of a proactive approach to
stress management.
Endnotes
1Conversely, “workload” or job demands are a major determi-
nant of occupational stress, as discussed later.
2This scale was particularly relevant given the nature of the
expected impact of automation; some earlier researchers have
found this to be a significant determinant of workload level
(e.g., Herbert, 1974; Moray & King, 1984)
3This figure over-represents the population average because
companies using such methods were deliberately sought for
inclusion in the study.
4The majority of employees (65%) rated their target or line
speed as “about right”. Of the others, most responded that they
were too high (30%) rather than too low (5%). The strongest
predictors of work rate acceptability included researcher
mental task demand scores and employee ratings of overall
task difficulty.
5Mean score for General Satisfaction was very close to the
normative mean for
“
process workers.
References
Addison, M., & MacDonald, W.A. (1993). Use of subjective and
objective techniques in identifying injury risk in a light repeti-
tive task — A case study. Paper presented to the biennial
conference of the Australia and New Zealand Modapts
Association, Melbourne, Victoria.
Alteras-Webb, S., & Dekker, D.K. (1994). Measuring perceived
task difficulty using magnitude estimation: A demonstration
and replication. Proceedings of the 38th Annual Meeting of the
Human Factors and Ergonomics Society (pp. 335–339). Santa
Monica: California: Human Factors and Ergonomics Society.
Andries, F., Kompier, M.A.J., & Smulders, P.G.W. (1996). Do you
think that your health or safety are at risk because of your
work? A large European study on psychological and physical
work demands. Work and Stress 10, 104–118.
Australian Bureau of Statistics. (2002). Industry overview:
Changes in hours worked by industry, 1999–00 (5204.0).
Canberra: Author.
Australian Centre for Industrial Relations Research and Training.
(1998, November). Work/time/life: Reclaiming the working
time agenda — An issues paper for the Australian Trade Union
Movement. Prepared for participants at the Work/Time/Life
Conference, Melbourne.
Australian Centre for Industrial Relations Research and Training.
(2002). Reasonable Hours of Work Test Case. Retrieved 7th
May 2002 from www.acirrt.com/research/schoolreashours.htm.
Australian Council of Trade Unions. (1988). ACTU Code of
Conduct on Twelve Hour Shift Work. Health and Safety
Bulletin 58. Melbourne: Author.
Australian Council of Trade Unions Occupational Health & Safety
Unit. (1998). A Report on the ACTU 1997 National OHS
Survey on ‘Stress at Work’. Melbourne: Author.
Australian Industry Group. (2001). Reasonable Hours Test Case:
Outline of Contentions. Australian Industry Group,
Engineering Employers’ Association, South Australia: Author.
Becker, A.B., Warm, J.S., & Dember, W.N. (1991). Effects
of feedback on perceived workload in vigilance performance.
In Proceedings of the Human Factors Society 35th Annual
Meeting (pp. 1491–1494). Santa Monica: California: Human
Factors and Ergonomics Society.
Bernard, B.P. (Ed.). (1997). Musculoskeletal Disorders (MSDs)
and Workplace Factors. Cincinnati: National Institute
of Occupational Safety and Health.
Bertone, S., Brown, M., Cressy, P., Keating, C., Morris, A.,
& Worland, D. (1998). Developing Effective Consultative
Practices: Case Studies of Consultation at Work. Melbourne:
South Pacific Publishing.
Blewett, V., & Shaw, A. (1995). OHS best practice column:
integrating OHS through self-managed teams. The Journal
of Occupational Health and Safety — Australia and New
Zealand, 11, 15–19.
Broadbent, D.E. (1958). Perception and communication. London:
Pergamon Press.
Burke, R.J. (1993). Organizational-level interventions to reduce
occupational stressors. Work and Stress, 7, 77–87.
Carayon, P. (1993). A longitudinal test of Karasek’s job strain
model among office workers. Work and Stress 7, 299–314.
Carayon, P., & Zijlstra, F. (1999). Relationship between job
control, work pressure and strain: Studies in the USA and
in The Netherlands. Work and Stress, 13, 32–48.
Charyszyn, S., & Tucker, P. (2001, May). New light on working
hours. The Ergonomist, 7.
Cooper, C.L. (1998). Introduction. In C.L. Cooper (Ed.), Theories
of organizational stress (pp. 1–5). Oxford: Oxford University
Press.
Cooper, C.L., Dewe, P.J., & O’Driscoll, M.P. (2001). Organ-
izational stress: A review and critique of theory, research, and
applications. London: Sage Publications.
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
12
WENDY MacDONALD
Cox, T. (1978). Stress. Basingstoke, Hants: Macmillan.
Cox, T., & Griffiths, A. (1995). The nature and measurement
of work stress: theory and practice.In J.R.Wilson,
& E.N.Corlett (Eds.), Evaluation of human work: A practical
ergonomics methodology (pp. 783–803).Taylor & Francis:
London.
Cox, T., & Griffith, A. (1996). Assessment of psychosocial hazards
at work. In M.J. Schabracq, J.A.M. Winnubst, & C.L. Cooper
(Eds.), Handbook of work and health psychology (pp. 127–143).
Chichester, UK: Wiley.
??Cox, T., Griffiths, A., & Rial-Gonzalez, E. (2000). Research
on work-related stress. European Agency for Safety & Health
at Work. Retrieved from http://agency.osha.eu.int/publications/
reports/stress
Cox,T., & Rial-Gonzalez, E. (2002). Work-related stress:
The European picture. Working on Stress: Magazine of the
European Agency for Safety and Health at Work, 5, 4–6.
Devereux, J.J., & Buckle, P.W. (2000a). Adverse work stress —
A review of the potential influence on work related muscu-
loskeletal disorders. Proceedings of the IEA 2000/HFES 2000
Congress 5 (pp. 457–460). Santa Monica, California: Human
Factors and Ergonomics Society.
??Devereux, J.J., & Buckle, P. (2000b). The risk of neck, shoulder
and upper limb musculoskeletal disorders due to interactions
between physical and psychosocial work risk factors.
Proceedings of the IEA 2000/HFES 2000 Congress 5
(pp. 597??). Santa Monica, California: Human Factors and
Ergonomics Society.
??Edwards, J.R., Caplan, R.D., & Van Harrison, R. (1998). Person-
Environment fit theory. In C. L. Cooper (Ed.), Theories
of organizational stress. Oxford: Oxford University Press.
Ellis, N. (2001). Work and health management in Australia and
New Zealand. Melbourne, Vic: Oxford University Press.
England and Wales Court of Appeal (Civil Division)
Decisions. (2002, February). Sutherland v Hatton [2002]
EWCA Civ 76.
Eriksen, H.R., & Ursin, H. (1999). Subjective health complaints:
Is coping more important than control? Work and Stress, 13,
238–252.
Evans, O., & Patterson, K. (2000). Predictors of neck and shoulder
pain in non-secretarial computer users. International Journal
of Industrial Ergonomics, 26, 357–365.
Fitts, P.M., & Posner, M.I. (1967). Human performance. Belmont,
California: Brooks Cole.
Frese, M. (1989). Theoretical models of control and health. In S.L.
Sauter, C.L. Cooper, & J.J. Hurrell (Eds.), Job control and
worker health (pp. 107–128). Chichester, U.K: Wiley & Sons.
Gaillard, A.W.K. (2001). Stress, workload, and fatigue as three
biobehavioral states: a general overview. In P.A. Hancock
& P.A. Desmond (Eds.), Stress, workload, and fatigue
(pp. 623–639). NJ: Lawrence Erlbaum Associates.
Ganster, D.C., Hurrell, J.J., & Thomas, L.T. (1987). Development
of a scale to assess occupational cognitive demands. In G.
Salvendy, S.L. Sauter & J.J. Hurrell (Eds.), Social, ergonomic
and stress: Aspects of work with computers (pp. 231–238).
Amsterdam: Elsevier.
Gescheider, G.A. (1976). Psychophysics method and theory. NJ:
Lawrence Erlbaum Associates.
Gopher, D., & Donchin, E. (1986). Workload — An examination
of the concept. In K.R. Boff, L. Kaufman, & J.P. Thomas
(Eds.), Handbook of perception and human performance
(pp. 41.1–41.49). NY: Wiley & Sons.
Gotts, G., & Cox, T. (1990). Stress and Arousal Checklist:
A manual for its administration, scoring and interpretation.
Melbourne, Vic: Swinburne Press.
Greller, M.M., Pasons, C.K., & Mitchell, D.R.D. (1992). Additive
effects and beyond: Occupational stressors and social buffers
in a police organisation. In J.C. Quick, L.R. Murphy, & J.J.
Hurrell (Eds.), Stress and well-being at work: Assessments and
interventions for occupational mental health (pp. 33–47).
Washington DC: American Psychological Association.
Hackman, J.R., & Oldham, G.R. (1980). Work redesign. Reading,
MA: Addison-Wesley.
Hancock, P.A., & Desmond, P.A. (2001). Stress, workload and
fatigue. London: Lawrence Erlbaum Associates.
Hancock, P.A., & Meshkati, N. (Eds.). (1988). Human mental
workload. Amsterdam: North Holland.
Hancock, P.A., & Warm, J.S. (1989). A dynamic model of stress
and sustained attention. Human Factors, 31, 519–537.
??Hart, P.M., & Cooper , C.L. (2001). Occupational stress: toward
a more integrated framework. In N. Anderson, D.S. Ones, H.K.
Sinangil, & C.Viswesvaran (Eds.), Handbook of industrial,
work and organizational psychology, volume 2. London: Sage.
Hart, S.G., & Staveland, L.E. (1988). Development of NASA-TLX
(Task Load Index): Results of empirical and theoretical
research. In P.A. Hancock & N. Meshkati (Eds.), Human
mental workload (pp. 139–183). Amsterdam: North-Holland.
Heiler, K. (2001, August). How effectively do we regulate exces-
sive hours of work in Australia? Draft discussion paper
prepared for the Victorian Government. University of Sydney,
Australian Centre for Industrial Relations Research and
Training.
Herbert, A. (1974). Factors in the perception of work difficulty.
Report No. 53. Institute of Applied Psychology, University
of Stockholm.
Heyde, C. (1976). The sensible taskmaster. Sydney: Heyde
Dynamics.
Hockey, G.R., Briner, R.B., Tattersall, A.J., & Wiethoff, M.
(1989). Assessing the impact of computer workload on opera-
tor stress: the role of system controllability. Ergonomics,
32, 1401–1418.
??Hoffmann, E.R. (1992). Predetermined Motion Time Systems
for estimation of task times: capabilities, limitations and possi-
ble alternatives. In M.A Regan (Ed.), Proceedings of
a Seminar Conducted by the Victorian Branch of the
Ergonomics Society of Australia, Canberra, ACT: Ergonomics
Society of Australia.
Hoffmann, E.R., MacDonald, W.A., & Almond, G.C. (1993).
Quantification of the cognitive difficulty of mail sorting.
International Journal of Industrial Ergonomics, 11, 83–98.
Houtman, I.L.D., & Kompier, M.A.J. (1995). Risk factors and
occupational risk groups for work stress in the Netherlands.
In S.L. Sauter & L.R. Murphy (Eds.), Organizational risk
factors for job stress (pp. 209–225). Washington, DC:
American Psychological Association.
Huang, G.D., Feuerstein, M., & Sauter, S.L. (2002). Occupational
stress and work-related upper extremity disorders: Concepts
and models. American Journal of Industrial Medicine, 41,
298–314.
Huey, B.M., & Wickens, C.D. (Eds.). (1993). Workload transition:
Implications for individual and team performance.
Washington, DC: National Academy Press.
Hurrell, J.J., & McLaney, M.A. (1988). Control, job demands, and
job satisfaction. In S.L. Sauter, J.J. Hurrell, & C.L. Cooper
(Eds.), Job Control and Worker Health (pp. 97–103).
Chichester, UK: Wiley & Sons.
International Standards Organisation (1991). ISO 10 075
Ergonomic principles related to mental workload — General
terms and definitions. Geneva: Author.
International Standards Organisation (1994). ISO 10 075–2
Ergonomic principles related to mental workload — Part 2:
Design principles. Geneva: Author.
Jackson, P.R., Wall, T.D., Martin, R., & Davids, K. (1993). New
measures of job control, cognitive demand and production
responsibility. Journal of Applied Psychology ,78, 753–762.
Jex, H.R. (1988). Measuring mental workload: problems, progress
and promises. In P.A. Hancock & N. Meshkati (Eds.), Human
mental workload (pp. 5–39). Amsterdam, North-Holland.
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
13
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
??Jex, S.M. (1998). Stress and job performance: Theory, research
and implications for managerial practice. California: Sage
Publications.
Job, R.F.S., & Dalziel, J. (2001). Defining fatigue as a condition
of the organism and distinguishing it from habituation, adapta-
tion, and boredom. In P.A. Hancock & P.A. Desmond (Eds.),
Stress, workload, and fatigue (pp. 466–475). NJ: Lawrence
Erlbaum Associates.
Johansson, G. (1981). Psychoneuroendocrine correlates of unpaced
and paced performance. In G. Salvendy & M.J. Smith (Eds.),
Machine pacing and occupational stress (pp. 277–286).
London: Taylor & Francis.
Johansson, G., Aronsson, G., & Lindstrom, B.O. (1978). Social
psychological and neuroendocrine stress reactions in highly
mechanised work. Ergonomics, 21, 583–599.
Johnson, J.V., & Hall, E.M. (1988). Job strain, workplace social
support, and cardiovascular disease: A cross-sectional study
of a random sample of the Swedish working population.
American Journal of Public Health, 78, 1336–1342.
??Jung, H.S., & Jung, H-S. (2001). Establishment of overall
workload assessment technique for various tasks and
workplaces. International Journal of Industrial Ergonomics,
28, 341–353.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ:
Prentice-Hall.
Kalsbeek, J.W.H. (1968). Measurement of mental work load and
of acceptable load. International Journal of Production
Research, 7, 1, 33–45.
Kantowitz, B.H. (1987). Mental workload. In P.A. Hancock (Ed.),
Human factors psychology (pp. 81–121). Amsterdam: North-
Holland.
Karasek, R. (1985). Job Content Questionnaire and users’ guide.
Los Angeles: University of Southern California, Department
of Industrial and Systems Engineering.
Karasek, R., Brisson, C., Kawakami, N., Houtman, I., Bongers, P.,
& Amick, B. (1998). The Job Content Questionnaire (JCQ):
An instrument for internationally comparative assessments
of psychosocial job characteristics. Journal of Occupational
Health Psychology, 3, 322–356.
Karasek, R., & Theorell, T. (1990). Healthy work: Stress, produc-
tivity and the reconstruction of working life. USA: Basic
Books.
Kasl, S.V. (1998). Measuring job stressors and strains: Where
we have been, where we are, and where we need to go. Journal
of Occupational Health Psychology, 3, 390–409.
Kawakami, N., & Haratani, T. (1999). Epidemiology of job stress
and health in Japan: Review of current evidence and future
direction. Industrial Health, 37(2), 174–186.
Kirwan, B., & Ainsworth, L.K. (Eds.). (1992). A guide to task
analysis. London: Taylor & Francis.
Kristensen, T.S. (1996). Job stress and cardiovascular disease: A
theoretic critical review. Journal of Occupational Health
Psychology, 1(3) 246–260.
Leiter, M. (1999, March). The impact on staff members of a hospi-
tal merger in Dublin. Paper presented to APA-NIOSH
Conference on Work, Stress, and Health ’99: Organization
of Work in a Global Economy, Baltimore. Washington DC:
American Psychological Association.
Liu, Y., & Tanaka, H. (2002). Overtime work, insufficient sleep,
and risk of non-fatal acute myocardial infarction in Japanese
men. Occupational and Environmental Medicine, 59, 447–451.
MacDonald, W.A. (1979). The measurement of driving task
demand. Unpublished doctoral dissertation, University
of Melbourne, Victoria.
MacDonald, W.A. (2001a). Train controllers, interface design and
mental workload. In J.M. Noyes & M.L. Bransby (Eds.),
People in control (pp. 239–258). Stevenage: IEE Press.
??MacDonald, W.A. (2001b). Workload, fatigue and job satisfac-
tion in repetitive blue-collar work. Keynote address to the
Annual Meeting of the Europe Chapter of the Human Factors
and Ergonomics Society. In Dick de Waard, C. Weikert, J.
Hoonhout, & J. Ramaekers (Eds.), Human system interaction:
Education, research and application in the 21st Century,
Maastricht, NL: Shaker Publishing BV.
MacDonald, W.A. (2001c, December). Stress, arousal and fatigue
in repetitive assembly line work: A case study. Proceedings
of the NOHSC Symposium on the OH&S Implications of Stress
(pp. 98–123). Canberra, ACT: National Occupational Health
and Safety Commission.
MacDonald, W., & Bendak, S. (2000). Effects of workload and
8- versus 12-h workday duration on test battery performance.
International Journal of Industrial Ergonomics, 26, 399–416.
MacDonald, W., Evans, O., Nolan, G., & Pratt, L. (1999). An
evaluation of current practice in setting work rates. Report to
the National Occupational Health and Safety Commission,
Canberra, ACT.
MacDonald, W., & Upsdell, T. (1996). Measuring job-related
mental workload — Or should that be stress? Proceedings
of the 1996 National Occupational Stress Conference: Health
and Well-Being in a Changing Work Environment. Brisbane:
Australian Academic Press.
Mackay, C.J., Cox, T., Burrows, G.C., & Lazzerini, A.J. (1978).
An inventory for the measurement of self-reported stress
and arousal. British Journal of Social and Clinical Psychology,
17, 283–284.
Marmot, M. (1994). Work and other factors influencing coronary
health and sickness absence, Work and Stress, 8, 191–201.
Marsella, A.J. (1994). The measurement of emotional reactions
to work: Conceptual, methodological and research issues.
Work and Stress, 8, 153–176.
Matthews, G., Davies, D.R., Westerman, S.J., & Stammers, R.B.
(2000). Human performance: Cognition, stress and individual
differences. Hove, East Sussex: Psychology Press.
McIntosh, N.J. (1995). Exhilarating work: An antidote for danger-
ous work? In S.L. Sayter & L.R. Murphy (Eds.),
Organizational risk factors for job stress (pp. 303–316).
Washington, DC: American Psychological Association.
Miller, P., & MacDonald, W. (2003). Relationships between job
demands, workload and employee wellbeing. Proceedings of
Work, Stress, and Health: New challenges in a Changing
Workplace: The Fifth Interdisciplinary Conference on
Occupational Stress & Health. Convened by the American
Psychological Association and the National Institute for
Occupational Safety and Health, Toronto, March.
Moray, N. (Ed.). (1979). Mental workload: Its theory and measure-
ment. NY: Plenum.
Moray, N. (1988). Mental workload since 1979. International
Reviews of Ergonomics, 2, 123–150.
Moray, N., & King, B. (1984). Error as a cause and effect
of workload: Mental workload as a closed loop system.
Department of Industrial Engineering, Working Paper 84–11,
Toronto, Ontario: University of Toronto.
Mowbray, G.H. (1953). Simultaneous vision and audition: the
comprehension of prose passages with varying levels of diffi-
culty. Journal of Experimental Psychology, 46, 365–372.
Mullarkey, S., Jackson, P.R., Wall, T.D., Wilson, J.R., & Grey-
Taylor, S.M. (1997). The impact of technology characteristics
and job control on worker mental health, Journal
of Organizational Behaviour, 18, 471–489.
Myrtek, M., Deutschmann-Janicke, E., Strohmaier, H.,
Zimmermann, W., Lawerenz, S., Brugner, G., & Muller, W.
(1994). Physical, mental, emotional and subjective workload
components in train drivers. Ergonomics, 37, 1195–1203.
Nachreiner, F. (1995). Standards for ergonomics principles relating
to the design of work systems and to mental workload. Applied
Ergonomics, 26, 259–263.
??National Institute of Occupational Safety and Health. (1996,
May). Ergonomics Health Hazard Evaluation. Report
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
14
WENDY MacDONALD
to Hanover Shoe Company, West Virginia: Author. Retrieved
?? from http://www.cdc.gov/niosh/02452577.html
O’Bryan, S.J., MacDonald, W.A. & Evans, O.M. (1991).
A comparison of some workload analysis techniques.
Proceedings of the 27th Annual Conference of the Ergonomics
Society of Australia (pp. 139–147). Canberra: Ergonomics
Society of Australia.
O’Donnell, M. (1997, May). Work intensification, female NESB
workers, public hospital and contracting out. Paper presented
to seminar Work Intensification: The Darker Side
of Workplace Reform, Melbourne, Victoria: Workplace
Studies Centre, Victoria University.
Osipow, S.H., & Spokane, A.R. (1987). Occupational Stress
Inventory (OSI). Odessa, Florida: Psychological Assessment
Resources.
??Paoli, P., & Merllie, D. (2001). Third European Survey
on Working Conditions 2000. European Foundation for the
Improvement of Living and Working Conditions Dublin, 18,
Ireland:??
Parker, S., & Wall, T. (1998). Job and work design: Organizing
work to promote well-being and effectiveness. London: Sage
Publications.
Payne, R.L, & Morrison, D. (1999). The importance of knowing
the affective meaning of job demands revisited. Work and
Stress, 13, 280–288.
Proctor, R.W., & Van Zandt, T. (1994). Human factors in simple
and complex systems. Boston, MA: Allyn and Bacon.
??Queensland Government. (2001). Reasonable Hours Test Case.
Supplementary Submission to Australian Industrial Relations
Commission. Retrieved ?? from http://www.ir.qld.gov.au/
reports&submissions/2001sub-reasonablehourscase.pdf
Quinlan, M., & Bohle, P. (1991). Managing occupational health
and safety in Australia: A multidisciplinary approach.
Melbourne: Macmillan.
Reason, J. (1990). Human error. Cambridge: Cambridge
University Press.
Rick, J., Briner, R., Daniels, K., Perryman, S., & Guppy, A. (2001).
A critical review of psychosocial hazard measures. Prepared
by the Institute for Employment Studies, University of Sussex,
UK, for the Health and Safety Executive (HSE), Contract
Research Report 356/2001, Norwich, UK: HMSO.
Rosa, R.R. (1995). Extended workshifts and excessive fatigue.
Journal of Sleep Research 4, 51–56.
Salvendy, G. (Ed.). (1997). Handbook of human factors and
ergonomics. NY: John Wiley.
Salvendy, G., & Smith, M.J. (Eds.). (1981). Machine pacing and
occupational stress. London: Taylor & Francis.
Sauter, S., Hurrell, J., Murphy, L., & Levi, L. (1997). Psychosocial
and organizational factors. In J.M. Stellman (Ed.),
Encyclopaedia of occupational health and safety 1 (pp.
34.1–34.77). Geneva: International Labour Office.
Schaufeli, W.B., Maslach, C., & Marek, T. (Eds.). (1993).
Professional burnout: Recent developments in theory and
research. Philadelphia, PA: Taylor & Francis.
Sen, T.K. (1987). Participative group techniques. In G. Salvendy
(Ed.), Handbook of human factors (pp. 453–469). NY: Wiley
& Sons.
Sendapperuma, L., MacDonald, W.A., & Hoffmann, E.R. (1991).
PMTS: Are they applicable to visual inspection tasks?
In Proceedings of the 27th Annual Conference of the
Ergonomics Society of Australia, Coolum (pp. 269–275).
Canberra, ACT: Ergonomics Society of Australia.
Shikdar, A.A., & Das, B. (1995). A field study of worker produc-
tivity improvements. Applied Ergonomics, 26, 21–27.
Siegrist, J.A. (, 1996). Adverse health effects of high-effort/low-
reward conditions. Journal of Occupational Health
Psychology, 1, 27–41.
Siegrist, J., & Peter, R. (1994). Job stressors and coping character-
istics in work-related disease: Issues of validity. Work and
Stress, 8, 2, 130–140.
So, C., & Mert Iljin, V. (1990). Measuring mental workload in an
occupational setting. Unpublished research project conducted
in partial fulfilment of the requirements of the Graduate
Diploma in Ergonomics, School of Human Biosciences,
La Trobe University, Melbourne, Victoria.
Spielberger, C.D., & Vagg, P.R. (1998). Manual for the Job Stress
Survey (JSS): Research Edition. Odessa, FL: Psychological
Assessment Resources.
Stansfeld, S.A., Fuhrer, R., Shipley, M.J., & Marmot, M.G. (1999)
Work characteristics predict psychiatric disorder: prospective
results from the Whitehall II study. Occupational and
Environmental Medicine, 56, 302–307.
Theorell, T. (1998). Job characteristics in a theoretical and practical
health context. In C.L. Cooper (Ed.), Theories of organiza-
tional stress (pp. 205–219). Oxford: Oxford University Press.
Tsang, P., & Wilson, G.F. (1997). Mental workload. In G.
Salvendy (Ed.), Handbook of human factors and ergonomics
(pp. 417–449). NY: Wiley & Sons.
Upsdell, T. (1994). Occupational burnout, stress, mental workload
and job satisfaction among rehabilitation counsellors.
Unpublished research project conducted in partial fulfilment
of the requirements of the Graduate Diploma in Ergonomics,
La Trobe University, Melbourne, Victoria.
US Department of Labor. (2000). Are managers and professionals
really working more? Issues in Labor Statistics. USA: Bureau
of Labor Statistics, US Department of Labor, Summary 00–12.
Vagg, P.R., & Spielberger, C.D. (1998). Occupational Stress:
Measuring job pressure and organizational support in the
workplace. Journal of Occupational Health Psychology,
3, 294–305.
Van Der Doef, M., & Maes, S. (1999). The Job Demand-Control
(-Support) Model and psychological well-being: A review
of 20 years of empirical research. Work and Stress, 13, 87–114.
Van Katwyk, P.T., Fox, S., Spector, P.E., & Kelloway, E.K.
(2000). Using the Job-related Affective Well-being Scale
(JAWS) to investigate affective responses to work stressors.
Journal of Occupational Health Psychology, 5, 219–230.
Wallace, M., & Greenwood, K. (1995). Editorial: Twelve-hour
shifts. Work and Stress, 9, 105–108.
Warr, P.B. (1990). Decision latitude, job demands, and employee
well-being. Work and Stress, 4, 285–294.
Welford, A.T. (1967). Single-channel operation in the brain.
Acta Psychologica, 27, 5–22.
Welford, A.T. (1976). Skilled performance: Perceptual and motor
skills. Glenview, Ill: Scott, Foresman & Co.
Wickens, C.D., Gordon, S.E., & Liu, Y. (1998). An introduction
to human factors engineering. NY: Longman.
Wickens, C.D., & Hollands, J.G. (2000). Engineering psychology
and human performance. NJ: Prentice Hall.
Wierwille, W.W., Rahimi, M., & Casali, J.G. (1985). Evaluation
of 16 measures of mental workload using a simulated flight
task emphasizing mediational activity. Human Factors,
27, 489–502.
Williams, W., & Cooper, C.L. (1996). Pressure Management
Indicator. Harrogate, England: RAD.
Williams, S., & Cooper, C.L. (1998). Measuring occupational
stress: Development of the Pressure Management Indicator.
Journal of Occupational Health Psychology, 3, 306–321.
Wilson, J.R., & Corlett, E.N. (Eds.). (1995). Evaluation of human
work. London: Taylor & Francis.
Wilson, J.R. (1995). Ergonomics and participation. In J.R.Wilson
& E.N.Corlett (Eds.), Evaluation of human work: A practical
ergonomics methodology (pp. 783–803). London: Taylor
& Francis.
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
15
THE IMPACT OF JOB DEMANDS AND WORKLOAD ON STRESS AND FATIGUE
Wooden, M. (2001a). Are non-standard jobs sub-standard jobs?
Australian Social Monitor, 3, 65–70.
Wooden, M. (2001b). The growth in ‘unpaid’ working time.
Economic Papers, 20, 29–43.
Wright, C., & Lund, J. (1996). Best practice Taylorism: ‘Yankee
speed up’ in Australian grocery distribution. Journal
of Industrial Relations, 38, 196–212.
Xie, B., & Salvendy, G. (2000). Review and reappraisal of model-
ling and predicting mental workload in single- and multi-task
environments. Work and Stress, 14, 74–99.
JULY 2003 ψAUSTRALIAN PSYCHOLOGIST
16
+WENDY MacDONALD
