Table of Contents

Impact of physical work conditions on productivity .................................................................. 1

Impact of the design over the work environment: Open plan offices ......................................... 4
4.2 Temporal conditions of work environments ................................................................ 8
Rotational shift work ............................................................................................................... 8
Effect of shiftwork on health and accidents ............................................................................ 11
4.3 Health and safety ......................................................................................................14
Accidents at work .................................................................................................................. 14
Reducing accidents at work: Token economy .......................................................................... 16
Monitoring accidents at work ................................................................................................. 17

Organisational work conditions

4.1 Physical work conditions

Impact of physical work conditions on productivity
Workers who are satisfied with their job are often more productive. Bockerman and
Limakunnas (2012) found that increasing workers’ job satisfaction had a positive
relationship to their manufacturing output at work. Workers spend a great deal of time in
the buildings they work in and therefore the environment is very important. Some of the
factors that may affect performance in work are:
1. Temperature: Several studies have shown that temperature may impact physical
health and performance, which may then affect productivity. If you are too hot or
too cold in your workplace then you will become distracted and focus less on your
work. Lorsch and Abdou (1994) found that when air conditioning was introduced
into a too hot workplace, workers felt more comfortable and productivity went up
between five and fifteen per cent. Other studies have suggested that workers
exposed to cold temperatures at work may make up to 44 per cent more mistakes
and be less productive.
2. Noise: Noise issues in the office are something that cannot really be avoided. When
noise is very low productivity increases and errors reduce. Noise can be an ambient
stressor and this is not only due to speech but other sounds such as photocopiers,
 keyboards and chairs scraping on the floor. Smith (1989), however, found that the
effects of intensity of noise are still unclear and that type of noise may be more
relevant.
3. Light: A number of research studies have shown that there is an association between
lighting and work performance, with 80 per cent of workers saying having good
lighting at work is important to them. There is evidence that lighting can have as
much of an effect as any other workplace factor. Poor lighting can give a worker
headaches due to the strain it puts on a worker’s eyes, although too intense artificial
light can have the same effect.
Other factors may include humidity and levels of carbon dioxide.

The Hawthorne effect

The Hawthorne effect is a term which suggests that levels of performance increase when
people know they are part of a research study, such as when they are being observed or
part of an experiment. This could mean that the results obtained in a study may be due to
the presence of the researcher rather than the manipulation of variables.

The original Hawthorne studies (Mayo et al.)

The original studies were conducted in the 1920s, on employees at the Hawthorne plant of
the Western Electrical Company in Chicago, USA. The original aim of the studies was to look
at the effect of levels of lighting on the productivity of the workers.
The ‘illumination experiment’ looked to see whether there was a relationship between the
work environment and productivity, for example, the level of lighting in a factory. The
researchers hypothesised that higher illumination would lead to better productivity. There
were two groups in the study:
1. In the experimental group, the levels of illumination were systematically decreased.
2. In the control group, the levels of illumination were constant.
Both groups showed increased levels of performance even when the lighting got lower, and
it was not until lighting was at moonlight level that participants started to complain they
could hardly see what they were doing and productivity consequently did not increase. The
experiment showed that lighting did not affect productivity and instead another factor was
more important. It was suggested that it was due to employees knowing that they were in
the study. Being observed and paid attention to made them more productive.
The relay test room experiment – leading on from the illumination study, Mayo et al.
conducted a series of experiments where they changed one part of the work environment
such as rest periods, working day or other physical conditions, for a group of five women.
The results of the relay test studies provided support for the illumination experiment and
the Hawthorne effect as whatever was changed (whether fewer or more rest periods, for
example) the women worked harder and more efficiently.

Later studies such as Wikstrom and Bendix (2000) acknowledged that while the original
studies seemed to show the Hawthorne effect, there may be other factors which had
impacted the results such as relief from a strict supervisor, increased attention, having a
different workday and believing they may have an influence in work practices.

Example study Kompier (2006)

HAWTHORNE EFFECT IS A MYTH
To examine the changes in employees’ reactions to work after moving from a conventional
office to an open plan office design. Kompier [2006] in his work, The Hawthorne effect is a
myth, agreed that there are other ways to look at the Hawthorne studies, suggesting five
reasons why the Hawthorne effect itself is a myth, 'that has become a legend’.
1. The myth of scientific worth: The methodological quality of the original case studies
is low. There is a lack of scientific rigour, many uncontrolled variables, and it is
virtually impossible to identify any causal relationships. The number of participants
was also small.
2. The myth of continuous improvement: In many of the original studies it was
concluded that performance improved. For many, however, there were periods of
time where productivity went down and this was ignored or downplayed by the
researchers.
3. The myth of social factors being more important than physical factors and pay: The
researchers suggested that the reason behind improvement was due to change in
supervision, such as the extra attention workers received, and not economic
interest. Some of the studies did not support this: when incentivised pay was
introduced in one study productivity increased and when it was taken away it
promptly dropped, suggesting pay is a key factor.
4. The myth of wholehearted cooperation: There was a suggestion that in the first test
study everyone cooperated with the experiment and with management. Any form of
conflict was denied. However, worker resistance and annoyance towards
management was documented, with some even facing disciplinary action. One study
even finished early due to worker conflict.
5. The myth of the neurotic worker: Whenever conflict within the workforce was even
mentioned it was blamed upon the mental health of the worker or situations outside
of the factory’s control, such as family issues. This then put the blame on the
individual rather than the situation they found themselves in at work. Negative
attitudes regarding the work environment were then dismissed as ‘obsessive
thinking’.
But why does the ‘myth’ of the Hawthorne effect still continue? For Kompier [2006] it may
be that psychologists think the story is just too good to be untrue and it has become
something of a legend. Perhaps there are certain elements that are true? For example, it
may well be that social factors are important in work productivity, but the point is that they
are not the only motivators as suggested by the original researchers. Perhaps social
scientists are just too far removed from the workplace, so they feel like the idea that the
workers have feelings and attitudes is a scientific breakthrough! Perhaps it just fits in nicely
with already established theories such as social influence, so seems credible. Whatever the
reasons are, the ‘myth’ of the Hawthorne effect is still alive and being taught in a school
near you today.

Evaluating the Hawthorne theories/studies

One strength of the Hawthorne studies and consequent idea of the Hawthorne effect is that
it provides a baseline for all future research due to its use of a number of different
techniques and methods. The studies used a mix of quantitative and qualitative methods,
interviews and observations, blood pressure and heart rate monitoring. They also built up a
strong collaboration between a factory and a university, something that has been used in
many different studies since then. This means that the studies were highly useful and this
methodology continues to be used effectively in organisational psychology.

The support for the Hawthorne effect came from one set of field experiments which was
conducted at an electrical plant in Chicago, USA. Although the amount of data collected is a
strength of the study, it was only obtained from one area of the USA. This may mean that
the results obtained were only applicable to Western individualist cultures such as the USA,
or perhaps even that particular type of workplace within the USA. There may be specific
characteristics of that type of work or organisation that meant that the presence of an
outside researcher was going to increase productivity levels. Therefore, there may be a
problem with generalisability in terms of the results.

ISSUES AND DEBATES

Application to everyday life
Another strength is that the studies used to support the Hawthorne effect are applicable to
real life situations and have high ecological validity. All of the original studies took place
within a ‘real life’ electrical plant in Chicago, using workers who were going about their
normal shifts. The only variables that were manipulated were things like lighting and length
of breaks which are conditions that are often changed in a real work environment. The use
of the workers within a real work environment means that the results obtained can be
applied to real life situations, as the study was high in ecological validity. In addition, the
fact that the study was not about researcher presence at all, but lighting in the workplace
means that the Hawthorne effect was an unintended consequence of the study, suggesting
it is high in validity and applicable to real life.

Determinism versus free will

One weakness of the Hawthorne effect is that it is deterministic. The Hawthorne effect is
the idea that our behaviour is determined by social situations and social interactions. In this
study it is suggested that the attention and the real or imagined presence of others,
especially if that social presence is someone seen to be higher in terms of authority, has
caused a behaviour change. However, this is a weakness as it ignores the role of free will in
behaviour, suggesting that the individuals had no choice but to behave in this way. In fact,
Kompier [2006] showed that not all participants followed this pattern of behaviour, some
dissented and complained about the new practices and for some areas productivity did fall.
The idea that not everyone responded to the researchers’ attention in the same way shows
that free will is present in behaviour.

Impact of the design over the work environment: Open plan offices
An open plan office is an office where all employees work on the same floor within the same
open space. The open plan office was designed to get rid of individual rooms or areas, in
favour of a space which was intended to increase collaboration, creativity and productivity
in a world of work which is increasingly complex, and where skill and idea sharing is
important. For some, the ability to network with colleagues will be an advantage, although
for some the ability to work in a private and quiet space of their own would be preferable.

In recent years the open plan office has become more popular. James et al. (2021) identified
three main factors that have influenced the move toward this style of office:
 • The type of work being done: Work has become more knowledge based and
complex.
• Changes in technology: Advancements in portable computers (laptops) and freely
available internet means work can be done anywhere and more people are working
from home, leaving desks unoccupied.
• Cost: Open plan offices are more space efficient by fitting more people in a space,
they are cheaper to build as fewer walls and doors are needed and they have lower
running costs (James et al., 2021).

Research findings about the impact of open plan offices on health, and social relationships
have been contradictory. Some studies such as Oommen et al. (2008) looked at the positive
and negatives of open plan offices and have shown that open plan offices have a number of
positives such as higher levels of communication, high levels of collaboration and increased
flexibility to work in different areas. However, they were also found to have negatives such
as high levels of noise, loss of concentration, issues with privacy and various health issues
such as increased stress, people were more likely to get flu, experience fatigue and have
increased blood pressure. James et al. (2021) conducted a review of thirty-one papers which
looked at the effect on the individual working in an open plan office and found that working
in an open plan office was associated with more negative outcomes on measures such as
health, satisfaction, social relationships and overall stress. Environmental concerns included
a lack of privacy, poor lighting, excessive noise and poor temperature control.

Example study Oldham and Brass (1979)

AIM
To examine the changes in employees’ reactions to work after moving from a conventional
office to an open plan office design.
METHODOLOGY
76 employees at an American newspaper moved from a conventional office to a new open
plan office, with no interior walls, no filing cabinets more than three feet high and no
private offices. These workers completed a questionnaire eight weeks before the move and
twice after the move at nine and eighteen weeks. Their results were compared with those
of two control groups (see Table 9.5).

Table 9.5 The three groups compared by Oldham and Brass (1979)
The questionnaire used seven-point rating scales to measure characteristics of the work
environment including work satisfaction - the degree to which employees were satisfied and
happy with their jobs, interpersonal satisfaction - the degree to which employees were
satisfied with co-workers and supervisors and internal work motivation - the degree to
which an individual experiences positive internal feelings when performing effectively on
the job.
Management and employees were questioned informally to gather extra qualitative data on
their reactions to the open plan office and interviews were used to see whether other
changes in the working environment may have occurred simultaneously. The researchers
also asked about concentration levels.

RESULTS
Quantitative findings Employees’ internal motivation and satisfaction with work and
colleagues declined sharply after moving from the traditional office to the open plan office.
There was largely no difference between the scores of the experimental group tested three
times and the quasi-control group, showing that the decreased motivation and satisfaction
was not a result of order effects resulting from completing the questionnaire before and
then after the move. The non-equivalent control group showed no difference in their scores
before and after the move suggesting that changes in the experimental groups scores were
not related to the move itself but to the difference in office structure.

Qualitative findings Many interviewees described the open plan office as a ‘fishbowl’, ‘cage’
or ‘warehouse’. They discussed their feelings about a lack of privacy and inability to
concentrate due to noise with one referring to it as ‘Grand Central Station’.

CONCLUSION
• Moving from conventional offices to open plan office styles can have a negative
effect on workers’ levels of motivation and satisfaction.
 • Environmental factors such as noise levels and lack of privacy can have negative
effects on an employees’ concentration levels.

Evaluating Oldham and Brass (1979)

One strength of the research by Oldham and Brass (1979) is that it used a standardised
procedure throughout the weeks studied. For example, all the employees were given the
same structured questionnaire whenever it was filled in and the data collection points were
at the same time for every participant dependent on their experimental group. The use of a
standardised procedure is a strength as it allows researchers to repeat the study which will
ensure the study’s results are reliable.

In addition, Oldham and Brass (1979) was a longitudinal study, which involves monitoring a
population over a sustained period of time; in this case three different data points, i.e. eight
weeks before the move and then nine and 18 weeks afterwards. This allowed researchers to
obtain a large amount of data and see changes over time. For example, Oldham and Brass
findings showed that satisfaction with work and colleagues decreased sharply after the
move/over time. This type of result could not be obtained through the use of a snapshot
study, which could only have discovered how the workers felt in any one moment of time.

Another strength is the use of both qualitative and quantitative data to support their
results. For example, questionnaires were used across all three data collection points which
used seven-point rating scales for the outcomes measured. Also, qualitative data was
collected through the use of informal interviews with the employees where they expressed
their feelings about the open plan office. The ability to use both types of data increases the
validity of the results as, for example, the initial results showing that workers had negative
feelings about open plan offices was supported by their comments in the interviews.

One weakness of Oldham and Brass’s study is that it only used data from one company in
the Midwest of the USA. It may well be that the company had particular characteristics that
meant that open plan offices did not suit the workers, or it may be that there were other
issues within that caused the results collected. In addition, it may be that cultural factors
may have contributed to the results and perhaps these results would not be replicated in
other countries. Therefore, the fact it was only done on one company limits the
generalisability of the results of the study.

In addition, the study was done in 1979 when open plan offices were not as popular as they
are today and the type of work within those offices was different; this may lower the
temporal validity of the study. For example, it was in the 1990s when open plan offices
became increasingly popular due to the technological advances of that period and the rise
of portable computers. It may be that these technological advancements meant that open
plan offices would be more favourably looked on by employees as they could see the
flexibility it gives them. However, James et al. (2021) found a similar negative reaction to
open plan offices in his more recent study suggesting perhaps the results would be the same
now as they were in 1979.
4.2 Temporal conditions of work environments

Rotational shift work

Rotational shift work is often used in manufacturing sectors when production needs to be
continuous. Employees work a shift for a set period of time, for example a week, and then
move on to another shift. Rotational shifts allow organisations to run two or three shifts a
day, seven days a week. Workers will take turns working eight to twelve hours a day on the
different shifts. For example, a worker may work an early shift from 6 a.m. till 2 p.m., a late
shift from 2 p.m. till 10 p.m. and then perhaps an overnight shift from 10 p.m. till 6 a.m.
Some workplaces work a two-shift pattern as dedicated staff work permanent night shifts.

• Forward rotation shifts: This is a strategy used to decrease the disruption of

circadian rhythms which govern the sleep-wake cycle. Known as a phase delay, this
type of shift system moves from a morning shift, to evening shift then possibly to a
night shift before the cycle begins again.
• Backward rotation shifts: This is a counterclockwise or phase advance shift system
where a worker will begin with the night shift, then move to the evening shift and
then to the morning shift. This type of shift is known for poorer health outcomes.

There are different types of shift dependent on how fast workers move through the shift
pattern. This can range from a few days to a number of weeks.

Rapid rotation shifts

These involve shifts that change frequently. It may be that a worker will work an early shift
for one week and then the next week change to late shifts the following week and night
shifts the week after this, so on. However, there are a number of rapid rotation shift types
which are quicker than this, where shifts are rotated every two to three days over a seven-
or eight-day period.

Examples
Metropolitan shift pattern This is a fast, forward rotation shift pattern which involves four
working teams and three eight hour shifts to enable an organisation to have 24/7 working.
Each team does two early shifts (6 a.m.–2 p.m.), two late shifts (2 p.m.–10 p.m. shifts) and
two night shifts (10 p.m.–6 a.m.) and then has two days off; free weekends will come once
every eight weeks. One issue with this quick rotation is if people are off sick it is difficult to
cover a shift due to working time arrangements (such as having to have a certain period of
time off before starting another shift).

Continental shift pattern Like metropolitan shift patterns, continental shifts are fast
forward rotation shifts using four teams and allowing 24/7 coverage in the workplace. They
consist of:
• Week 1: Two early shifts, three late shifts, and two nights shifts (full seven days
working).
• Week 2: Two days off, three early shifts, two late shifts.
• Week 3: Two late shifts, three days off, two early shifts.
• Week 4: Two night shifts, three night shifts, two days off.

Continental shifts are still popular and allow for more frequent weekends off (one in four)
than metropolitan shifts, however its main disadvantage is the long seven consecutive
workdays, as seen in Week 1.

Slow rotational shifts This is a shift pattern where there are longer intervals between the
changes of shift. It could be that shift patterns only change once every two weeks, but it
could be even longer than that. For example, workers would work day shifts for six weeks
then change to night shift for six weeks.
Slow rotations provide more consistency and regular patterns for workers, allowing them to
plan time outside of work. In addition, the body will adjust easier to regular schedules.

On-call work patterns

Many professions such as aviation, medicine and engineering use on-call work patterns. This
is where workers have to provide 24-hour coverage, seven days a week and so on 'standby'
to respond to a time critical situation. This is most common in places such as hospitals
where an emergency means that a person must deal with a problem immediately.

There are generally two types of on-call working:

1. You leave your workplace, for example at evenings and weekends, but can be called
back to work during these periods. This means the person is free to do what they
want to in the period at home and only time spent in their place of work is counted
as working hours. However, there are some restrictions that they may need to follow
such as:
a. They need to be able to get to the workplace within a certain time period, such as
fifteen minutes from the time they are notified.
b. They must live in a location within a set distance, for example no more than five
miles away.
c. Alcohol consumption is not allowed when on-call at home.
d. They must be awake at set times, such as between 12 a.m. and 2 a.m. e. They
must always be contactable by phone.
2. You are at your workplace when on call. For some professionals such as junior
doctors a period of time on call means that they will remain at their place of work
but will be given somewhere to sleep. If you remain on site, then the whole of the
period on call will be classed as working time.
Nicol and Botterill (2004) found that on-call work can play a role in increasing stress and
decreasing mental health. It can also decrease the quality and quantity of sleep and leave
people more tired after their on-call periods. The disruption caused by being on call can also
affect social and family life.

Flexi-time
Flexi-time is a flexible way of work where a worker completes a set number of hours per day
but their starting times, lunch times and finishing times are chosen within the limits set by
the employer. This allows workers to fit their work hours around their personal
circumstances and commitments outside of work, such as childcare or transport to school.
Flexi-time therefore does not change the total hours worked or the responsibilities the
workers have.

Some organisations allow workers to start and finish whenever they like but many
employers have core times within the day (usually their peak time for being busy) when
employees have to be at work. For example:
• Arrival at work: Can be between 7 a.m. and 10 a.m.
• Lunch hours: Can be between 12 p.m. and 2 p.m.
• Flexible leaving time: Between 4 p.m. and 6 p.m.
• Core working hours (at desk): 10 a.m.–12 p.m. and 2 p.m.–4 p.m.

Flexi-time’s freedom increases satisfaction and morale amongst the workforce, with
reduced stress and fatigue. Traffic rush-hours can be avoided and people are in the office
for a longer number of hours per day. It also allows people to work whilst having family
commitments. However, flexi-time requires more discipline in workers and also an ability to
track when everyone is working and where!

Evaluating design of work

One issue when looking at different types of working patterns and how they affect
individuals is the idea of individual differences. This is whether our own differences in
physiology and personality may affect the type of work patterns which may suit us best. It is
easy to suggest that certain types of shift patterns benefit the individual more and that
other types of shift patterns damage health, but this is too simplistic and suggests that we
all react the same. This is not the case, and some people may be more suited to working
nights or early mornings so we cannot look for generalisations and need to look more at the
individual.

When researching shift work and the different patterns, researchers tend to use
questionnaires asking questions about health, accidents, near misses and so on.
Questionnaires allow for a larger sample as they are simple to distribute and allow for
quantitative data to be collected and statistically analysed. This will give researchers an idea
about the effects of shift work patterns on the individual and be useful to organisations.
However, questionnaires can be subject to a poor response rate and social desirability bias
which can lower the validity of the research.
Effect of shiftwork on health and accidents

Effects of shiftwork on health

Ansiau et al. (2017) were commissioned by the Institute of Safety and Health (IOSH) to look
at the long-term effects of shift work on the following:
1. Sleep quality: There is a general agreement that shift work affects the quality and
duration of sleep and this is widespread despite individual differences. The
researchers found that the time shift workers spent asleep was reduced by around
two hours a day, but their quality of sleep also declined due to the reduction of
Rapid Eye Movement and stage 2 sleep. They found that those who gave up shift
work at a younger age continued to report poor quality sleep showing longer term
effects.
2. Quality of life: Researchers found that shift workers reported more chronic fatigue
than non-shift workers. There were no real differences in other indicators such as
social isolation and stress.
3. Physical health: The researchers found that shift workers and former shift workers
are more likely to suffer from obesity, peptic ulcers, gastro-intestinal problems,
failure to control blood sugar levels and cardiovascular disease. This was especially
true for those who had worked shifts for over ten years, even controlling for gender
and age.
4. Cognition: There was a significant association between shift work and poorer
cognitive performance including poorer attention and episodic memory. Those who
had quit shift work more than five years previously had no differences in cognitive
performance, suggesting that it was a temporary issue whilst shift work was ongoing.

Cardiovascular disease (CVD) Torquati et al. (2018) reviewed 21 studies which looked at the
link between CVD and shift work and found that the risk of CVD was 17 per cent higher
amongst shift workers than day workers whilst the risk of Coronary Heart Disease (CHD)
morbidity was 26 per cent higher. Abu Farha and Alefishat (2018) looked specifically at night
shift workers in Jordan and found that there was a positive association between night shift
work and CVD and clogged arteries, with more frequent and longer night shifts increasing
into the risk.

Reproductive effects Harrington (2001) found that shift work, and especially night work, is
of higher risk to women of childbearing age. This may be due to disruptions to the
menstrual cycle and increased stress because of the conflicts between night shifts and
family life. Increased risk of abortion, low birth weight and premature birth have all been
seen as outcomes of shift work.

Shift work and accidents

The cause of accidents is complex and it cannot be suggested that they are due to shift work
alone. However, there is a link between shift work and increased fatigue leading to lower
performance levels and higher rates of accidents. Ryu et al. (2017) studied shift workers in
factories in South Korea, finding that current shift workers were 2.7 times more likely to
have a work-related injury, whilst past shift workers were 1.7 times more likely, compared
to workers who have never worked shifts. Shift work duration only affected this level of risk
in women and not men.
The 1986 Challenger disaster happened at night in circumstances where managers had
fallen behind in the planning of the shuttle launch, leading to little more than two hours of
sleep on the night before the launch. Poor judgement and human error were suggested as
part of the cause for this disaster, with actions partially brought about through sleep
deprivation.

Evaluating the link between shift work and poor health/accidents

The link between shift work and health/accidents is purely correlational. There may be
other factors that have caused physical health problems and accidents rather than just shift
work. Individuals may have a poor diet, external problems that have caused high stress or
they may be inexperienced at their job. All of these may cause health problems such as CVD
or errors at work, suggesting that the link is too simplistic and there are many factors that
need to be taken into account.

ISSUES AND DEBATES

Application to everyday life One strength of researching the effects of shift work is that it
has application to real life. If research shows that shift work has negative consequences,
then organisations can adapt working patterns to reduce errors and improve health. For
example, instead of rotational shifts they could do phase delay shifts. However, this is all
dependent on whether the workforce wants and would adapt to those changes, or whether
the type of work can be completed on a different shift pattern.

Determinism versus free will One weakness of suggesting that shift work affects health and
performance negatively is that it is deterministic, suggesting that health/accidents are
caused by external forces over which we have little control. This is too simplistic as the
individual has control over other aspects of their life which may help lessen the impact of
shift work such as good nutrition, ensuring the environment at home is good enough for
sleep and reducing external stress, for example.

Example study Gold et al. (1992)

AIM To examine the impact of work schedule on the sleep schedule, sleepiness and
accident rates of female nurses in a Massachusetts hospital.
METHODOLOGY Questionnaires were distributed to 878 nurses [mean age 33.9] and other
auxiliary hospital staff. The nurses recorded for the current week, the following week and
the previous two weeks their work shift category at the hospital, and any other job. Types of
shift included:
1. Rotator [four days/evening, then four days of nights].
2. Day/evening shifts but no nights.
3. Nights – eight shifts in a month, no days or evenings.
4. Day/evening with occasional nights.
5. Nights with occasional days/evenings.

The nurses were also asked whether their job involved a variable work shift such as a day
shift sometimes and a night shift at other times. Sleep-wake times were recorded
throughout. Information was also collected about sleep quality, alcohol consumption,
medication taken, falling asleep at work or when driving to and from work, accidents, errors
and 'near-miss' accidents in the past year. Sleep and wake times were used to see whether a
nurse obtained 'anchor sleep' which was at least four hours of sleep during work days and
days off.

RESULTS
• Rotators reported fewer hours of sleep, more accidents and more near misses than
those on regular day or evening shifts, as well as night shift workers.
• Falling asleep at work occurred at least once a week in 35 per cent of rotators, 32.4
per cent of night nurses and 20.7 per cent of day/evening nurses who worked
occasional nights.
• Rotators had 3.9 times the odds of falling asleep whilst driving and night nurses had
3.6 times the odds compared to day/evening nurses.
• Adjusting for the effects of alcohol and medication, the odds of reporting any
accident or error were twice as high for rotators as for day/evening nurses and there
were 2.5 times the odds of near-miss accidents.

CONCLUSION Sleep deprivation and disruption of the circadian rhythms experienced in a

rotating shift pattern are associated with lapses of attention and increased error rates on
performance tasks.

Evaluating Gold et al. (1992)

The strength of Gold et al.’s (1992) study was that quantitative data was collected on the
amount of sleep and the numbers of near misses or errors. Quantitative data is objective,
and therefore reliable, and statistical comparisons and analysis can be performed. However,
the use of questionnaires can also cause problems due to social desirability bias. Sleeping
whilst driving or working could cause a negative reaction at work and therefore participants
may not be honest in their answers. This may mean that the answers collected are not
measuring true levels of errors, lowering validity.

However, Gold et al.’s (1992) research is that it only used nurses and other staff from one
hospital in Massachusetts, USA. This leads to doubts about generalisability to other
countries, cultures or even other states in America. However, over 800 nurses were
recruited which is a significant number which may increase generalisability. In addition,
these are physiological reactions to shift work which may be the same wherever the worker
came from, suggesting the use of only one hospital may not be so much of a problem as first
thought.

ISSUES AND DEBATES

Application to everyday life Another strength is that it has high application to real life and
therefore can be useful to organisations. If it is found that a particular shift pattern is
causing illness or potential danger to the worker, or to the clients they are working with,
then the organisation can adjust the shift pattern or put in place support to ensure that this
does not occur.
Idiographic versus nomothetic The study itself takes a more nomothetic approach rather
than idiographic. A nomothetic approach analyses trends to provide general rules of
behaviour and usually uses quantitative methods. In this case a wide variety of nurses was
asked to fill in questionnaires about shift work, meaning that we can suggest that shift work
in general causes health issues. However, it may be worth taking a more idiographic
approach and using qualitative methods such as case studies, as we would be able to gain
more insight into the reasons why shift work causes the individual such problems and
whether there are specific external factors, such as family issues, that are making the
problem worse.

4.3 Health and safety

Accidents at work

Human errors in operator machine systems

Whenever an operator processes information from the environment, makes a decision and
takes action, there is always the possibility that an error can occur. An error can occur
consciously such as misjudging the height of a vehicle at a low bridge, or unconsciously such
as accidentally tripping a switch which turns something off. Errors can be classified in a
number of ways:
• An error of commission: This is when an operator performs an action which is
incorrect/inappropriate or an additional action which should not have been done in
the first place. This could have been pressing a button twice or opening a valve on a
machine that should be closed.
• An error of omission: This is where an operator fails to perform an action which they
were meant to do such as failing to put a lock on to stop a machine being operated
accidentally or failing to press a stop button on a machine. In many cases these
errors occur because someone has forgotten to do something
• An error of sequencing: This is where an operator does a series of actions out of the
order in which they were supposed to be done.
• An error of timing: This is where an operator does an action at the wrong time – too
early or too late.

System errors in operator machine systems

The other type of error that can have negative consequences involves a problem with the
system itself. This could be an error in the controls of the machine or an error with the
display itself.
A display is any device which provides information to the machine operator.
• Visual displays: These include lights, dials and digital readouts. They can show
measures such as temperature, speed or time. Visual displays are better for complex
messaging.
• Audio displays: These can include pre-recorded messages and tones. These are
important if working in a dark room; if the message is time dependent and if the
worker is not static.
For example, on an aircraft a pilot will have a visual display which tells them their altitude
and how fast they are descending or ascending. However, an audio display such as the TCAS
(traffic collision avoidance system) will inform the pilot of a problem, warning them to pull
up if they are too near another aircraft.

In order to reduce system errors, companies should:

• Make sure that the right kind of display is used for the information needed to be
input. For example, a dial is not the best choice if you need an accurate reading – a
digital display is preferable.
• Group displays logically – if more than one display is for a single system then group
them together.
• Put a relevant display next to a relevant control. If a display shows you the speed of
a machine, then put the control which adjusts speed near to it.
• Use colour to enhance the display and warn of danger but do not make it too bright
or vivid, which may cause eye strain.

Controls are anything that a worker uses to operate a system such as a lever, switch, handle,
keyboard or joystick. Companies should reduce likelihood of errors by:
• Grouping the controls logically and arranging them in the order in which they are to
be used and by their frequency of use.
• Making sure controls can be easily reached and operated. Make sure you have a
system to protect any controls where it is critical that they should not be accidentally
operated. In this case, something like turning a key should be used to operate.
• Label each control so its function is clear, such as ‘Emergency shutdown, push down
to operate’. Use colour coding.

Despite all precautions, things do go wrong with automated systems, so it is vital that
backup systems are in place and operators are trained in the correct procedure to follow if
something goes wrong. For all electronic automated systems wherever possible there
should be a manual backup.

Evaluating human and system errors

ISSUES AND DEBATES

Application to everyday life Researching the reasons behind accidents at work has obvious
application to real life. To be able to reduce accidents you need to understand why they
occur in the first place. Once that understanding is there, then you can develop systems to
make their occurrence less frequent, as discussed with visual and audio displays. This
includes having back-up systems for systems failures and regular appraisals and
observations to ensure fewer human errors occur.
Individual and situational explanations When looking at human errors you can discuss the
individual versus situational debate. In this case many of the errors could be due to
situational factors such as poor layout of displays and controls, poor training or distractions.
However, there could be some individual factors occurring, such as lack of self-esteem or a
lack of confidence in their ability to perform the job or a lack of coordination which makes
them trip or fall.

Reducing accidents at work: Token economy

A token economy is a system based upon principles of operant conditioning where specific
behaviours are rewarded with tokens (secondary reinforcers) which are then exchanged at
a later date for a desired reward (primary reinforcer). Tokens can be in the form of trading
stamps and the rewards can be anything from activities to privileges at work.

Example study Fox et al. (1987)

In 1985, there were 500 accidents, and 40,000 work related injuries in mines that meant
workers had to have at least one day off work. Most previous interventions to reduce
accidents had been relatively short term, meaning that the long-term effectiveness was
unknown.

AIM To investigate the use of a token economy in which miners were given tokens for not
having accidents or injuries for specified periods of time.

METHODOLOGY The study was carried out at two open-pit mines, one in Wyoming and one
in Arizona, USA. The two settings used similar mining procedures and accidents had
occurred at both mines. The participants were the employees at the two mines. Both mines
had similar health and safety policies, such as on the job hazard training and yearly
refresher training. Each week the health and safety staff inspected all areas of the mine for
any correctable hazards and violations of health and safety policy.

Workers in each mine were divided into hazard groups based on lost time injuries during the
baseline period. Group 1 jobs were the least hazardous, where workers spent 75 per cent of
their time in the office; Group 4 jobs were the most hazardous, such as electricians,
scrapers, and operators.

Individual employees were given trading stamps at the end of the month with their wages if
they had suffered no lost time injury or compensation injury that required a doctor’s care.
The number of stamps awarded went up depending on the level of hazard of the job. Safety
groups also awarded extra stamps for those employees who came up with safety
suggestions that prevented serious injury or death or prevented accidents or damage to
property. Employees who had had an accident received no stamps for that month if they
had been off for 1–2 days, or for two months if they had been off for 3–4 days, and so on.
Failure to report an accident also resulted in a loss of tokens. The token economies went on
for over a decade at both mines. The trading stamps could be exchanged at neighbouring
stores carrying hundreds of items.
RESULTS At both mines there were substantial decreases in the number of days lost during
the first years of the token economies. In terms of lost time injuries, the trend went down
significantly during the second year of the token economy, but not the first. Costs of injuries
and accidents fell significantly, and the cost-benefit ratio showed that the money saved
from the decline in accidents more than compensated for the cost of the trading stamps.

CONCLUSIONS Behavioural programmes can be effectively administered and can be

maintained for long periods of time.

Evaluating Fox et al. (1987)

One strength of the Fox et al. study is that it uses a longitudinal research methodology over
a number of years. For example, at one mine the token economy programme ran for twelve
years. This is important as it allows researchers to track changes over time, to gather a large
amount of data and also to see whether the interventions implemented are effective in the
long term.

In addition, the study has important applications to real life. Time off work for accidents is a
real issue for organisations in terms of employee morale, loss of money and work time.
Therefore, if interventions such as token economy can be used effectively within a
workplace to decrease accidents there is less chance of negative consequences for the
employee and the cost of the programme will be less than the benefits, making the study
useful.

One weakness of the study is that the sample may not be representative of the wider
population, limiting the study’s generalisability. For example, the two companies were in
the USA and were both mining companies. The effectiveness of reward systems may only be
apparent in certain countries and there may be something about the mining community
which may make them more receptive to this type of behavioural intervention, for example
the element of risk and danger. Therefore, the results of the study may not generalise to
other countries and organisations.

Finally, there may be an issue with the validity of the study. For example, the incentive of a
reward may have stopped employees reporting incidents and therefore the number of
accidents may not have decreased, only the reporting of them. However, a serious accident
is not easily hidden so this may not be a real issue. However, a potentially bigger issue was
the fact that accidents were decreasing at one mine even prior to the start of the study so
there may have been sustained improvements even without the intervention, decreasing
the validity of the study’s findings.

Monitoring accidents at work

Key study: Monitoring of accidents and risk events - Swat (1997)

CONTEXT There is no universal way of monitoring accidents and every country, and even
every organisation, has a different way of monitoring them. Two types of monitoring have
been discussed in terms of accident prevention.
 1. Active monitoring: These are all the formal and informal checking activities that are
carried out by line managers to ensure accidents do not occur.
2. Reactive monitoring: This looks at safety through looking at accidents and incidents
that have occurred. It looks at the incident types and frequency of accidents,
sickness rates, near misses and property damage.

Swat (1997) found that Poland’s system of documenting accidents was ineffective. No detail
on losses resulting from accidents and incidents was available. The medical costs associated
with accidents were just included in the overall health care system costs, and therefore they
did not affect the insurance plans of the affected organisations. Also unplanned events, such
as machine issues, which may interrupt production were not monitored.

AIM To develop an organisationally useful method of recording risk events in organisations

which could help find causes of accidents and consequently prevent them.

METHODOLOGY This was a case study with a longitudinal design which took place between
1994 and 1997 and used already available secondary data on accidents and minor injuries in
four different industrial plants in Poland. Interviews with line managers, safety supervisors
and employees were also conducted.

Sample Four industrial plants from different industrial branches [foundry, machinery, meat
processing and furniture] in Lodr, Poland were chosen. All of the plants chosen were rather
old and had equipment which was a number of years old. A total of 2964 workers were
employed in the plants in 1993.

PROCEDURE The number of accidents recorded in 1993, within each of the four plants, was
the first and most common source of information about risk at work. The 83 accidents
which occurred in 1993 were analysed for frequency, accident severity and their direct and
indirect causes. In the study:

• An accident was defined as a 'case which had been formally reported as an accident
by the... safety supervisor’. In other words, a sudden undesired event which was
connected with the workplace resulting in health impairment in the worker. In
Poland, this would only mean an accident that resulted in sick leave or death.
• Accident frequency was defined as 'the number of injury accidents resulting in sick
leave per 100 employees within a year'.
• Accident severity rate was the number of sick days per accident.
Three types of information were used about accidents: collective accident reports from
safety supervisors, the researchers’ own investigations of accident protocols and interviews
with the safety supervisors and line managers.

The second source of information on risk at work used in the study came from minor
incidents from 1994 which did not result in death, serious injury or serious damage. Only
the meat processing plant was analysed, with an incident being defined as 'any sudden
event which resulted in personal injury'. Data on first aid cases reported to the meat plants
was analysed as well as interviews with 96 employees, who were told to report even small
injuries.
RESULTS
1 Accidents
Table 9.6 Accident frequency rate (per 100 employees)

Table 9.7 Accident severity rate (number of sick days per accident)

Five different accident types were proposed from these results which could be used for
monitoring:
1. Falls and slips
2. Accidents connected with manual work
3. Accidents connected with working parts of machinery
4. Accidents connected with sources of energy
5. Others

Table 9.8 Frequency rates for types of accidents (per 100 employees) on average in all
plants

The study found four essential causes of accidents:

1. Insufficient supervision
2. Poor workplace organisation (such as inadequate personal protection)
3. Technical factors
4. Individual error of the worker

Insufficient supervision was seen in the highest number of accidents (89 per cent), with
technical problems the least (11 per cent). However, there were overlaps between these
causes, with a number of incidents having more than one cause.

Poor housekeeping [maintenance] was also seen frequently as a cause of accidents and
Swat (1997) suggested that this classification should be used when monitoring accidents.
Frequency and type of accident was very much related to the levels of housekeeping within
the plant. Examples of types of housekeeping accidents are:
1. Falls on slippery floors
2. Faulty staircases leading to slips and falls
3. Use of improper tools
4. Unsuitable clothes
Housekeeping contributed, in total, to 65 per cent of the accidents in the foundry, 47.8 per
cent in the meat processing area, 37.5 per cent in the furniture area and 33.3 per cent in the
machinery section, showing that housekeeping needs to be considered as a risk factor in
workplaces.

2 Incidents
Note: The frequency rate of incidents in the meat factory increased from 2.8 to 3.5 between
1993 and 1994 with manual accidents being the main reason.

In terms of incidents in the meat factory in 1994, there were 254 injuries requiring medical
treatment (218 of these were manual injuries and most likely to lead to injuries of the left
hand), 23 cases with sick leave days (remember sick leave days are the only ones usually
reported in Poland so these are the only figures they would normally have) of which five
were serious leading to over 30 sick days each. Interviews with employees suggested the
real total of incidents could be as high as 520, meaning that 95 per cent of incidents are
never reported.

CONCLUSION
• Accidents should be recorded according to the circumstances in which they occurred
and their type, for example where they were related to manual work or to energy
sources. The comparison of frequency indicators for accident types both nationally,
and within the different sections of the plants, may supply information on the extent
of safety management faults.
• Lesser incidents should be noted, especially those needing first aid. That would
supply more data for these types of incidents which would help safety management
more than just knowing about the few serious incidents.
• Poor housekeeping should be noted as a cause. Poor operation is a frequent source
of incidents and this therefore is a central risk factor and should be incorporated into
safety monitoring systems.

Methodological strengths and weaknesses

One strength is that a variety of different data collection methods were used. For example,
the data from the established accident monitoring procedures was used alongside
information from interviews with managers and employees. This is a strength as it means
that organisational data can be compared with the results of the interviews to identify
differences. This case is it showed that the company data was not comprehensive enough
and many incidents were missing, allowing changes to monitoring procedures to be
suggested.

In addition, a longitudinal design was used which allowed the researchers to look at
organisations over a period of three years. Longitudinal studies are often used to look at
trends over time and allow a greater amount of detail to be obtained than shorter studies.
For example, in this case accident rates over a period of a year for three of the plants, and
over two years for the meat processing plant, were compiled and analysed alongside the
interviews. This is a strength, as it leads to a more comprehensive understanding of the
subject, in this case accident monitoring at work.
One weakness is that the research methods used have issues of subjectivity. Although the
information officially collected by the organisations should be said to consist of quantitative
and objective data, it is subject to omissions, where employees may not record or even
report a particular accident or the line manager may not want a number of incidents
reported because it would cause them issues with their managers. In addition, the
researcher used their own analysis to look at the different plants, which again may lead to
researcher bias and subjective interpretation to support their aims.

There are also issues concerning the generalisability of case studies. Although four different
trades were studied, these are a very small proportion of the types of organisations in
countries such as Poland; it may be that there is something specific about those
organisations that caused the accidents recorded. In addition, the fact that only
organisations in Poland were studied would also affect generalisability. However, it is worth
noting the case study very often has no intention of generalising outside of one particular
country and this may be the case with Swat (1997) who was focusing on the particular
monitoring problems within Poland.

Ethics There were no specific ethical issues mentioned within the study, although no
particular businesses were identified and therefore confidentiality was upheld. It may be,
however, that there was only one meat factory within that area in Poland, so they may be
more easily identified. There was no harm from the study and results could have a benefit
to the workers due to increased monitoring of accidents leading to greater accident
prevention, meaning that the results are beneficial and therefore ethical.

Application to everyday life This study has obvious application to real life as it will allow
organisations within Poland and in other countries to formulate a more robust accident
monitoring system. This will allow them to identify how many accidents occur and also to
look at the causes and be able to work on prevention of accidents rather than reaction to
accidents.

ISSUES AND DEBATES

Idiographic versus nomothetic Although this study takes a nomothetic approach by looking
at general trends in accidents which can be applied across a number of different trades and
countries, it may be worth taking a more idiographic approach and looking at the individual,
with more qualitative forms of data collected. This would allow researchers to investigate
individual reasons behind accidents which could then be analysed using thematic analysis
and used for accident prevention.