Johansson et al (1978)

Aims

To investigate the relationship between levels of occupational stress and productivity on workers in a Swedish factory.

The company was mostly interested in whether stress levels could be reduced and productivity increased.

The researchers wanted to compare the psychological and physiological stress responses in 2 different categories of workers.

Procedures

Natural experiment – no manipulation as workers already were in categories.

Independent groups design:

1^st group – skilled workers, responsible for transforming timber into finished product

• The rate they worked determined the wage for everyone on working line

2^nd group – maintenance/cleaners = more flexible conditions

Daily urine sample – 4 times a day, baseline was at home

Body temperature measured, caffeine and nicotine consumption.

Self-rating scales on words such as ‘sleepiness’, ‘well-being’.

Researchers examined records of illness and absence.

Findings

High-risk group:

• First urine sample of the day = x2 adrenaline than baselines – increased during the day.
• Higher levels on work days than rest days
• Felt more rushed and irritated, rated well-being lower
• Higher rates of absenteeism, more psychosomatic illness

Low-risk group:

• urine sample = x1.5 adrenaline than baselines – declined throughout the day.

Conclusions

The finishers experiences high levels of occupational stress due to 4 factors:

• Responsibility for setting the wage rates.
• Skilled but monotonous and repetitive work
• Machine paced work = absence of control
• Working in isolation

Strengths:

• Useful – researchers recommended factories of what to do to make the employee less stressed. Occupational stress was reduced.
• Valid – what was meant to be measured was studied and the results had positive consequences.
• Mundane realism and ecological validity – real workers in work environment.

Weaknesses:

• Correlation – no cause and effect could be determined as many variable were studied. This makes the study a mere correlation, meaning that it will be hard to help other factories to reduce stress levels.
• Self-report = social desirability could have affected the results as the mill workers may have wanted to appear a certain a way to their employers to perhaps improve their working conditions or salaries. 
• Cultural/Gender bias – the study was carried out with Swedish men, meaning that results could have been affected by their culture and gender. Therefore, the study is not highly generalisable or applicable.