Strengths of general adaptation syndrome:

• Supporting evidence: Selye’s (1936) experiments on rats demonstrated the same physiological sequence in response to completely different stressors. This supports general adaptation syndrome as a single and complete explanation of stress.

Weaknesses of general adaptation syndrome:

• Conflicting evidence: Mason (1971) conducted experiments similar to Selye’s (1936) but on monkeys instead of rats. In these experiments, rather than exhibiting the same physiological response to the different stressors, the monkeys’ physiological responses differed depending on the stressor. For example, exposure to extreme cold appeared to increase levels of cortisol whereas exposure to extreme heat decreased cortisol levels. This suggests that the explanation of stress as general adaptation is not accurate and is instead specific to different stressors.
• Methodological concerns: As Selye’s (1936) physiological explanation of stress was derived from experiments on rats, general adaptation syndrome may not be valid when explaining the physiological process of stress in human beings.

Strengths:

• Supporting evidence:
  • Immunosuppression: See Kiecolt-Glaser et al (1984) above for a link between stress and immunosuppression.
  • Cardiovascular disorders: As well as Yusuf et al (2004) above, the studies linking the stress-prone type A personality and cardiovascular disorders described below further support a link between stress and cardiovascular disorders.
• Practical applications: Understanding the role of stress in illness has resulted in new and better ways to prevent and treat disease. For example, ABCD.

Weaknesses:

• Conflicting evidence: Some studies suggest that certain types of stress can actually boost immunity rather than causing immunosuppression. For example, Schedlowski (1993) measured natural killer cell activity in 45 first-time parachute jumpers 2 hours before their jump, immediately after their jump, and 1 hour after their jump. The researchers found that natural killer cell activity increased immediately after the parachute jumps and then fell below baseline 1 hour post-jump. This suggests that acute (short-term) stress can initially increase immune response before resulting in immunosuppression.
• Methodological concerns: Much of the research linking stress and illness is correlational. However, as always, correlation does not necessarily mean causation. It could be, for example, that stressed people are more likely to take up smoking and that it’s the cigarettes that cause increased levels of illness rather than stress itself.

See the AO3 evaluation points for the SSRS for more evaluation of the link between life changes and stress.

Strengths:

• Supporting evidence: Rahe et al (1970) got 2664 navy sailors to complete the social readjustment ratings scale prior to a 6-8 month tour of duty during which their medical records were recorded. The researchers observed a positive correlation (+0.118) between life changes (as measured by LCUs on the SRRS) and illness during the tour of duty. This supports a link between life changes, stress, and stress-related illness.
• Practical applications: Understanding the role of life changes in stress could help prevent and reduce stress-related illness. For example, a doctor could get patients to complete a self-report survey of life changes (such as the SSRS) and if the patient scores highly, the doctor could prescribe or advise treatments to manage stress.

Weaknesses:

• Other factors: Kanner et al (1981) compared the accuracy of the SRRS at predicting stress-related illness with their own hassles and uplifts scale, which is a measure of daily hassles. The researchers found the hassles and uplifts scale was a more accurate predictor of stress-related illness than the SRRS, which suggests that daily hassles are a more important cause of stress than life changes (which tend to be rare by comparison).
• Some life changes are relaxing, not stressful: The SSRS assumes all life changes are stressful, when many may actually be the exact opposite. For example, items on the SSRS include ‘vacation’ (13 LCUs) and ‘major holiday’ (12 LCUs), but many people find these life changes relaxing rather than stressful.
• Methodological concerns: A methodological concern with drawing conclusions about stress caused by life changes from the social readjustment ratings scale is that the SRRS and stress is simply correlational. As always, correlation does not necessarily mean causation and so does not prove that life changes cause stress or stress-related illness. For example, illness could be what causes life changes (e.g. time off work from being ill leads to the life change of losing a job), in which case the direction of causation is the other way round.

See the AO3 evaluation points for the hassles and uplifts scale for more evaluation of the link between daily hassles and stress.

Strengths:

• Supporting evidence: Kanner et al (1981) compared the ability of their hassles and uplifts scale to predict stress-related illness against the SRRS, which is a measure of stress caused by life changes. The researchers found a stronger correlation between hassles and stress-related illness than life changes and stress-related illness, which suggests that daily hassles are a more important cause of stress than life changes (which tend to be rare by comparison).
• Practical applications: Understanding the role of daily hassles in stress could help prevent and reduce stress-related illness. For example, a doctor could get patients to complete a self-report survey of daily hassles (such as the hassles and uplifts scale) and if the patient scores highly, the doctor could prescribe or advise treatments to manage stress.

Weaknesses:

• Methodological concerns: A methodological concern with drawing conclusions about stress caused by daily hassles from the hassles and uplifts scale is that even if scores are correlated with stress-related illness, this does not necessarily prove that these hassles cause stress or stress-related illness.

Strengths:

• Supporting evidence: Several studies support Karasek’s (1979) demand-control model. For example:
  • Workload: Kivimäki et al (2006) conducted a meta-analysis of 14 studies that included at a total of 83,104 employees and found that workload was correlated with coronary heart disease. The authors estimate that workplace stress makes employees 50% more likely to suffer from coronary heart disease.
  • Control: The ‘Whitehall Studies’ conducted by Marmot et al (1997) followed 10,308 employees in the UK civil service over a period of 5 years. The researchers found that employees with low levels of control over their work were significantly more likely to suffer from heart attacks compared to employees with high levels of control over their work (even when other factors, such as diet, were controlled for).
• Practical applications: A practical application of findings from the Whitehall Studies above is that giving employees greater control over their work may reduce their risk of suffering from a heart attack.

Weaknesses:

• Conflicting evidence:
  • Workload: Although the Whitehall Studies (Marmot et al (1997)) found a correlation between low levels of job control and heart disease, they found no such correlation between level of workload and heart disease. This contradicts the findings of Kivimäki et al (2006) described above and suggests higher workload does not, in fact, cause increased stress.
  • Control: Meier et al (2008) tested Karasek’s (1979) demand-control model by measuring stress levels and levels of job control in 96 service employees. The researchers found that greater job control did reduce the effects of workplace stress – but only among those with an internal locus of control. For employees with an external locus of control, greater levels of job control was actually correlated with increased levels of stress-related illness.
• Methodological concerns: A methodological concern with the Whitehall Studies is that they did not control for confounding variables. The researchers focused on workload and control, but there are many other factors that may explain higher levels of stress-related illness. For example, jobs with lower levels of control may also be paid less, and these lower levels of pay may also increase the risk of stress-related illness. Such confounding variables may skew the results and decrease the validity of this study’s findings.

Strengths of the social readjustment ratings scale:

• Supporting evidence: Rahe et al (1970) got 2664 navy sailors to complete the SRRS prior to a 6-8 month tour of duty during which their medical records were recorded. The researchers observed a positive correlation (+0.118) between life changes (as measured by LCUs on the SRRS) and illness during the tour of duty. This supports the prediction of the SRRS that LCUs would be correlated with stress-related illness, and also suggests the SRRS is a valid measure of stress.
• Practical applications: The SSRS could potentially be used to prevent stress-related illness. For example, a doctor could get patients to complete the SSRS and if the patient scores more than 300 LCUs, the doctor could prescribe or advise treatments to manage stress.

Weaknesses of the social readjustment ratings scale:

• Conflicting evidence: Kanner et al (1981) compared the ability of their hassles and uplifts scale to predict stress-related illness against the SRRS. The researchers found a stronger correlation between scores on the hassles and uplifts scale compared to the SRRS, which suggests that the hassles and uplifts scale is a more accurate measure of stress than the SRRS.
• Doesn’t distinguish between positive and negative life changes: The SSRS assumes all life changes are stressful, when many may actually be the exact opposite. For example, items on the SSRS include ‘vacation’ (13 LCUs) and ‘major holiday’ (12 LCUs), but many people find these life changes relaxing rather than stressful.
• Methodological concerns: As always, correlation does not necessarily mean causation. Even if scores on the SRRS are correlated with illness, it does not necessarily prove that life changes recorded by the SRRS caused those illnesses. For example, illness could be what causes life changes (e.g. time off work from being ill leads to the life change of losing a job), in which case the direction of causation is the other way round.
• Less scientific: As a self-report method, the SSRS relies on the participant’s subjective interpretation of events (e.g. “trouble with in-laws” is open to interpretation and may differ between people). As such, the SSRS can be said to be less scientific than physiological measures such as skin conductance response, which produce objective and quantifiable data.

Strengths of the hassles and uplifts scale:

• Supporting evidence: Further to the evidence above, Kanner et al (1981) compared the correlation between stress-related illness and scores on their hassles and uplifts scale with the correlation between scores on the SRRS and stress-related illness. They found a stronger correlation with stress-related illness from scores on the hassles and uplifts scale compared to the SRRS, which suggests that the hassles and uplifts scale is a more accurate measure of stress than the SRRS.
• Practical applications: The hassles and uplifts scale could potentially be used to prevent stress-related illness. For example, doctors could get patients to complete the hassles and uplifts scale and if they score highly, the doctor could prescribe or advise treatments to manage and cope with stress.

Weaknesses of the hassles and uplifts scale:

• Conflicting evidence: Although there is fairly good evidence of a link between hassles and stress-related illness, there is little research supporting a link between uplifts and reduced stress. For example, even Kanner et al (1981) above does not support a link between uplifts and reduced stress in men, only in women. Similarly, DeLongis et al (1982) gave 100 adults aged 45-64 the hassles and uplifts scale and found uplifts made little difference to stress-related illness.
• Methodological concerns:
  • Retrospective: The hassles and uplifts scale is retrospective, meaning respondents must remember hassles and uplifts over a period of time (e.g. the last month). However, if a person has a bad memory, their responses may not be valid.
  • Correlation does not necessarily mean causation: Even if scores on the hassles and uplifts scale are correlated with illness, it does not necessarily prove that the hassles recorded by the hassles and uplifts scale caused stress or stress-related illnesses.
  • Ambiguous: Further, many of the items on the hassles and uplifts scale are open to interpretation (e.g. what one person considers “relating well with friends” another person may not) and so the results may not be valid.
• Less scientific: As a self-report method, the hassles and uplifts scale relies on the participant’s subjective interpretation of events (e.g. what counts as “relating well with friends” may differ from person to person). As such, the hassles and uplifts scale can be said to be less scientific than physiological measures of stress such as skin conductance response, which produce objective and quantifiable data.

Strengths of skin conductance response:

• Supporting evidence: Several studies have demonstrated strong correlations between levels of stress and skin conductance. For example, Harrison et al (2006) measured skin conductance in infants when they were undergoing a stressful procedure (blood sampling) and when they were undergoing a non-stressful procedure (feeding or having a nappy changed). Skin conductance increased significantly during the stressful procedure but not during the non-stressful procedure, which supports the validity of skin conductance as a physiological measure of stress. Another example is Kurniawan et al (2013), who found that trained polygraph operators were able to accurately determine (using skin conductance response) whether a person was experiencing high stress or low stress 70% of the time.
• Practical applications: Skin conductance (as measured by polygraph testing) is commonly used as a lie detector test. This can be used by law enforcement, as demonstrated in the Chris Watts police interview above, to help determine a suspect’s guilt.
• More objective and scientific: Skin conductance response uses objective and quantifiable data to measure stress. As such, it can be said to be more scientific than self-report measures of stress, which rely on a person’s subjective interpretation of events.

Weaknesses of skin conductance response:

• Conflicting evidence: Although there is lots of evidence supporting the correlation between skin conductance and stress, it is not a perfect measure. For example, even though Kurniawan et al (2013) found polygraph operators were able to accurately distinguish between high and low stress 70% of the time, this still leaves 30% of the time when they got it wrong. This demonstrates that skin conductance response is not a perfectly accurate measure of stress.
• Misuse: Although lie detection can be a useful application of skin conductance response, polygraph testing is not 100% accurate. Because of this, use of skin conductance response to determine whether someone is lying can lead to negative consequences. A potential example of this is Steve Dymond, who committed suicide after failing a polygraph test on the Jeremy Kyle talk show. Before his death, Dymond insisted he was telling the truth.

A general point with these links between personality types and stress-related illness is that the risks are relative: Type A personalities are more likely to get heart disease than Type B personalities, but this does not mean it is likely (in absolute terms) that type A personalities will get heart disease. (Illustrative example: Someone with a 2 in a million change of getting ill, say, is twice as likely to get ill as someone with a 1 in a million chance of getting ill – but 1 in a million is still a very low chance in absolute terms!)

Strengths of personality types and individual differences in stress:

• Supporting evidence:
  • Types A and B: Rosenman et al’s (1976) Western Collaborative Group Study determined the personality type (A or B) of 3,154 employed men who were not suffering from heart disease. At the follow-up 8.5 years later, 257 had developed heart disease. Type A personalities were approximately twice as likely to be among these 257 men than type B personalities, which supports a link between type A personality and heart disease.
  • Type C: A review by Temoshok (1987) looked at cancer rates and personality traits associated with type C, such as a tendency to repress emotions (particularly anger). She found that these personality traits were indeed correlated with an increased risk of cancer.

Weaknesses of personality types and individual differences in stress:

• Conflicting evidence: Ragland and Brand (1988) followed up with the 257 subjects from Rosenman et al’s (1976) Western Collaborative Group Study who had suffered from heart disease. Of the 231 subjects who initially survived a heart attack, type B personalities were more likely to have died at the follow up than type A personalities. This contradicts the original findings by Roseman et al (1976) and suggests type A personalities may actually have lower risk of stress-related illness and death in some circumstances.
• Classification issues: Several researchers have argued that the general category of type A personality is too broad. For example, Matthews and Haynes (1986) and Dembroski et al (1989) found that hostility alone predicts heart disease just as accurately as type A personality does. This suggests that some components of the type A personality (such as being competitive) are not relevant to differences in stress and stress-related illness.
• Methodological concerns: Correlation vs. causation: All the studies discussed here are correlational – they look for correlations between certain personality traits and certain illnesses. But, as always, correlation does not prove causation.

Strengths of hardiness and individual differences in stress:

• Supporting evidence: Maddi (1987) studied 400 employees at an American phone company during a stressful time when many people were being made redundant. He observed that two-thirds (66%) of employees suffered stress-related illness during this time, but one-third (33%) did not. The one-third who did not suffer from stress-related illness were much more likely to score highly for hardiness (i.e. commitment, challenge, and control) than the two-thirds who did get ill. This supports the idea that hardiness protects against the effects of stress.
• Practical applications: Maddi (1998) hardiness training ABCD.

Weaknesses of hardiness and individual differences in stress:

• Relative importance of different components: Some studies suggest that some of the 3 C’s of hardiness – commitment, challenge, and control – are more important than others.
• Vague: The concept of hardiness is difficult to quantify objectively. Funk (1992) points out that several different scales are used to measure hardiness and that there is little research comparing these different scales. Further, it may be that what some of these scales measure is actually neuroticism – and that it is low levels of neuroticism, not hardiness, that prevents stress-related illness.
• Methodological concerns: The usual correlation vs. causation argument applies here. Even if having a hardy personality is correlated with lower levels of stress-related illness, this does not prove that having a hardy personality causes lower levels of stress-related illness. It may be, for example, that people with hardy personalities are more likely to take part in physical exercise, in which case it is likely to be the physical exercise that causes lower levels of stress-related illness.

Strengths of gender and individual differences in stress:

• Supporting evidence: Taylor et al’s (2000) finding – that the female response to stress is more tend and befriend and that the male response to stress is more fight or flight – has been replicated in other studies. For example, ABCD.
• Practical applications: ABCD

Weaknesses of gender and individual differences in stress:

• Conflicting evidence: ABCD.
• Questions of causal mechanisms: ABCD
• Methodological concerns: ABCD.

Strengths of drug therapy for managing stress:

• Supporting evidence:
  • Benzodiazepines: ABCD.
  • Beta blockers: ABCD.

Weaknesses of drug therapy for managing stress:

• Side effects: Both benzodiazepines and beta blockers produce side effects that must be weighed against the benefits of stress reduction:
  • Benzodiazepines: ABCD.
  • Beta blockers: ABCD.
• Addictive: Benzodiazepines are highly addictive (but beta blockers are not addictive).
• Mechanism: Drug therapy does not tackle the underlying cause of stress, but just masks the physiological symptoms.

Strengths of stress inoculation therapy for managing stress:

• Supporting evidence: A meta-analysis by Saunders et al (1996) concluded that stress inoculation therapy was effective at reducing anxiety and improving performance under stress.
• No side effects: Unlike drug therapy, stress inoculation therapy does not have a risk of side effects.

Weaknesses of stress inoculation therapy for managing stress:

• Individual differences: Stress inoculation therapy is time-consuming and takes a lot of effort on behalf of the patient. As such, many patients may quit without seeing results.
• Cost and difficulty: ABCD

Strengths of biofeedback for managing stress:

• Supporting evidence: A randomised controlled trial by Lemaire et al (2011) found biofeedback resulted in significant and lasting stress reductions (as measured by self-report questionnaire) that were not observed in the control group.
• No side effects: Unlike drug therapy, biofeedback does not have a risk of side effects.

Weaknesses of biofeedback for managing stress:

• Conflicting evidence: Greenhalgh et al (2010) reviewed randomised controlled trials of biofeedback for reducing high blood pressure (an objective measure), but found no difference in results between biofeedback and placebo. This could suggest biofeedback only reduces feelings of stress (e.g. as measured by self-report measures like in Lemaire et al (2011) described above) without actually reducing the risk of stress-related illness.
• Cost and difficulty: Biofeedback can be expensive and may not work for all patients.

Strengths of social support for managing stress:

• Supporting evidence: Orth-Gomér et al (1993) conducted a 6-year longitudinal study of 736 Swedish men aged 50. The researchers found smoking and lack of social support were the 2 factors most strongly correlated with heart disease during the period of the study, which supports the idea that social support is effective at managing stress and reducing stress-related illness.

Weaknesses of social support for managing stress:

• Methodological concerns: As always, correlation (e.g. as demonstrated in Orth-Gomér et al (1993) above) does not prove causation. It’s not clear how social support may reduce stress or stress-related illness.
• Individual differences: Not all social support may be helpful. For example, instrumental support in the form of financial gifts may make a person feel guilty, increasing stress rather than decreasing it. Or, if a person has friends who deal with stress by turning to alcohol or drugs, social support could be counterproductive rather than helpful.