Johansson - Stress at Work
Aim: To measure the psychological & physiological stress response in two 'categories' of empolyees.
Methodology: A Quasi experiment using an independant measures design.
Participants: 24 workers at a Swedish Sawmill...
- 14 workers, called finishers, were in the productions line with a set pace etc. (high risk)
- 10 workers who were cleaners/maintanance workers (control group).
- Each pp. had to give urine samples when they arrived at work and at 4 other times during the day.
- They also gave a self report of mood & alertness - plus caffeine and nictotine consumption.
Baseline measurements were taken at the same time on a day when the workers were at home:
- Adrenaline levels measured in urine.
- Body temperature (at the same time as urine collection).
- Self rating scales
Johansson - continued...
The first urine samples of the day showed...
- The finishers group had adrenaline levels twice as high their baselines.
- The finishers adrenaline also continued to increase throughout the day.
- The control group had a peak level of 1.5 times baseline level in the mornings & this declined during the rest of the day.
The self report showed...
- The finishers group felt more rushed & irritated than the control group.
- The finishers also rated their well being as lower than the control group.
The repetative, machine-paced work which was demanding in attention to detail & highly mechanised, contributed to the higher stress levels in the finishers/high risk group.
Johansson - Evaluation
- Standardised procedure - all participants had the same measurments, at the same time etc.
- Used biological methods for measuring stress/adrenaline levels & body temperature - can be easily replicated (can can't be faked!)
- Low poplation validity - (ethnocentric) Small sample, only from Swedish Sawmill so results can't be generalised to other countries/jobs.
- Has concurrent validity - Use of biological methods/tests confirmed the validity of participant's self report answers/results.
- High ecological validity/mundane realism - Quasi experiment which took place in participants 'natural'/normal work setting.
*No control over extaneous variables - Other 'stressors' not taken into account e.g. family problems.
*Participant variables - Different 'types' of personalities cope different with stress i.e. some may naturally get more stressed than others.
Kanner - Daily Hassles
A 'Daily Hassle' is a minor event that arises in the course of a normal day. They're usually short lived but may linger if unresolved. The 'after-effect' of unresolved issues may intensify with subsequent hassles.
Aim: To compare the Hassles & Uplifts Scale and the Berkman Life Events Scale as predictors of psychological symptoms of stress.
Methodology/design: A repeated design as participants completed both the hassles & life events scale.
Psychological symptoms of stress was measured/assessed using the Hopkins Symptom Checklist (HSCL) & the Bradburn Morale Scale.
- 100 people who had previously taken part in a health survey in 1965
- Mostly white, from California, protestant, with adequate or above income & at least 9th grade education.
- Initially 216 contacted, 109 agreed but then 9 dropped out.
Kanner - continued...
Procedure: All tests were sent out 1 month before study began
Participants were asked to complete...
- The Hassles Rating Scale every month for 9 months
- The HSCL & Bradburn Morale Scale every month for 9 months
- The Life Events Scale after 10 months
- Hassles were consistant from month to month.
- Men's life events correlated positively with hassles but negatively with uplifts.
- For women, the more life events they they reported, the more hassles & uplifts they reported.
- Hassle frequency correlated positively with psychological symptoms on HSCL - i.e. the more hassles reported, the more negative the psychological symtoms.
- Hassles also corrleated more positively with psychological symptoms than life events did.
Conclusion: Hassles are a more powerful predicitor of psychological symptoms than life events - Hassles contribute to psychological symptoms, whatever life events have happened.
Kanner - Evaluation
Correlations: Can infer/establish some sort of cause-effect relationship HOWEVER can be misinterpreted - cause & effect not possible!
Rating Scale: People will interpret these differently & some people may lie/show social desireability bias to look more stressed than they actually are for sympathy.
Standardised procedure: All given same tests, at same time etc. so can be easily repeated to look for consitancies in results - reliable.
Concurrent Validity: Used more than one (2) self rating/measure for stress.
Ethnocentric: All participants from California.
Sample: All volunteers - had previously volunteered - only people who are actually bothered by stress (as they may be perceive themselves to be stressed).
Geer & Maisel - Lack of Control
Aim: To see if perceived or actual control can reduce stress reactions to aversive stimuli.
Methodolgy: A Lab. experiment using an independant measures design.
- 60 undergraduates - enrolled on a psychology course
- From New York University
All Pp.s saw the same photographs of dead car crash victims
Participants were randomly assigned to one of the 3 conditions...
- Group 1 - Actual Control. Could press a button to terminate the photographs (max. 35 seconds). A 'tone' told them when the photograph would proceed to the next one.
- Group 2 - No control but had predictability. Saw photograph for exactly same time as group 1 (yoked to group 1). Warned that photographs would be 60 seconds apart & for how long they'd see it. Told a 10 second warning tone would precede each photograph.
- Group 3 - No control or predictability. Told from time to time they would see photographs & hear tones.
Geer & Maisel - continued...
- Each pp. was seated in a sound-shielded room & wired up to a GSR and a heart monitor.
- The machine was calibrated for 5 minutes while the participants relaxed & then a baseline measurement was calculated.
- Instructions were read over the intercom.
- After a minutes rest, the stimuli were presented (see conditions for each group).
- The GSR analysis were taken at the onset (start) of the tone, during the second half of the tone & in response to the photograph.
- The heart rate monitor proved inaccurate & so data from these were discarded.
- Group 2 showed most stress.
- Group 1 showed least stress.
Conclusion: Participants showed less GSR reaction, indicating less stress, when they had control over the length of time they looked at the disturbing photographs - It is likely that being able to terminate aversive stimuli reduces their stressful impact.
Geer & Maisel - Evaluation
Low Ecological Validity - & Mundane Realism, as in real stress situations you wouldn't be wired up to a heart monitor or GSR. Also in real life, seeing a 'real' crash victim would be much more stressful than a picture/have higher fear arousal (includes sounds & smells etc).
- Differences in personality 'types' (whetehr they were type A, B or C) - some pp.s may get naturally more stressed than others.
- Perhaps some pp.s had had bad experiences with car accidents (been in an accident, known friend or family who had been in an accident) so would get more stressed than others despite which group they were in.
Data collection method - used biological methods to measure stress (i.e. heart monitor, GSR) which can be easily repeated and results can't be faked/altered (reliable). However, this method could make pp.s more stressed also.