Norbert K. Semmer and Laurenz L. Meier
University of Bern, Switzerland
Despite many problems of individual studies, links between work stress and strain (i.e. physical and psychological health) are now quite well established (Sonnentag & Frese, 2003). With about 10% variance explained, the relationships found are typically not very strong, which is unsurprising, given the complex etiology of stress symptoms (Semmer, Zapf & Greif, 1996).
There can be no doubt that associations between stressors and strain do not hold for everyone in the same way (Spector, 2003). However, in current stress research there is a tendency to emphasize individual differences to the point where stress is being reduced to nothing but idiosyncratic appraisals and coping styles, rendering such concepts as ‘environmentally induced stress’ useless, as Lazarus and Folkman (1986, p. 75) asserted. This view tends to equate ‘interpretation’ with ‘confined to the individual’, and ‘environment’ with ‘physical environment’, and to neglect that the social environment is a powerful reality, where people in the same culture share ‘rules of appraisal’ (Averill, 1986) and ways of dealing with the world (Semmer, McGrath & Beehr, 2005; cf. Kahn & Byosiere, 1992). Thus, it should be kept in mind that not all individual differences found are only differences between individuals, but often differences between the (sub-) cultures they belong to.
This chapter will first discuss mechanisms that may be responsible for individual differences in stress reactions. It will then deal with the question of what makes stressors ‘stressful’, concentrating on goals people have. Individual differences in resources (‘resourceful belief systems’) will be covered next. Finally, we will deal with individual differences in coping. Rather than representing an exhaustive review of the literature, the chapter tries to tie together various ‘literatures’ under common themes. This will inevitably lead to oversimplifications. Yet, we are convinced that all too often we tend to dwell on differences and difficulties, and sometimes it seems worthwhile to see if there is some forest emerging behind the many trees.
Stress will be used as a rather general term in this chapter, denoting a state of tension that is experienced as aversive. Stress, therefore, involves negative emotional states such as anxiety, frustration, and the like (Lazarus, 1999). Stressors are characteristics of the environment that tend to elicit such emotional states in a given population (cf. Semmer et al., 2005). Note that this does not require that each and every individual will experience stress vis à vis a stressor. All it requires is that in a given population certain characteristics tend to induce stress, on the basis of shared meaning in a given culture (Kahn & Byosiere, 1992; Semmer et al., 2005).
If personality plays a role in the stress experience, it somehow must ‘translate’ into stressful experiences. We will discuss four mechanisms, which refer to (i) encountering stressors; (ii) appraising stressors; (iii) reacting to stressors; and (iv) coping tendencies. These mechanisms are not mutually exclusive, and may combine into a ‘cascade’ (Suls & Martin, 2005). Furthermore, we will briefly comment on personality profiles.
The most important factor determining the probability to encounter stressors is probably socioeconomic status (Adler et al., 1999). However, personality also plays a role, which is reflected in the generally higher risk of encountering stress for people high in neuroticism (Suls & Martin, 2005), or the higher involvement in accidents of people low in agreeableness (Clarke, 2006). Furthermore, people may actively select or avoid situations associated with certain stressors, such as the risk of failure in the case of promotion vs. prevention orientation (Brockner & Higgins, 2001). Finally, they may involuntarily ‘produce’ stressors, as when hostile people induce conflict (Smith et al., 2004). People’s well-being may influence future working conditions (or the perception of these conditions), although these effects tend to be weaker than the effect of stressors on well-being (Zapf, Dormann & Frese, 1996). For instance, people high in negative affectivity are more likely to be in high stress jobs (Spector, Jex & Chen, 1995) and people high in self-esteem as well as those with an internal locus of control are more likely to end up in jobs they like (Brockner, 1988; Furnham, 1992).
Many personality variables are linked to the way people appraise situations. Examples include the tendency of threatening appraisals when anxiety is high (Suls & Martin, 2005), of hostile attributions when hostility is high (Berkowitz, 1998), of interpreting failure as ‘self-diagnostic’ when self-esteem is low (Brockner, 1988).
Reacting more strongly to events that are appraised in a similar way constitutes another mechanism for the translation of personality variables into stress experiences. Even with appraisal held constant, people may display greater reactivity (see Suls & Martin, 2005, for anxiety; Grebner et al., 2004, for neuroticism). In fact, an important component of the very general traits of negative and positive affectivity (neuroticism, extraversion) can be seen in physiologically routed differential responsivity to negative and positive stimuli, respectively (Gray, 1987; Larsen & Ketelaar, 1991).
Finally, there is some stability in the way people cope with stressful situations. This aspect will be discussed in more detail below (Section 6.5).
The general personality characteristic that seems most important for all mechanisms is neuroticism (Semmer, 2006). However, the few studies investigating profiles of personality traits show that the combination of high neuroticism and low conscientiousness is especially problematic (and the combination of low N and high C especially resilient) in terms of stress and coping (Vollrath & Torgersen, 2000).
Basically, stress has to do with appraisals of threat and/or loss (Lazarus, 1999). Threat or loss are appraised with regard to goals (cf. the dimension of ‘goal conduciveness’ in appraisal theories; Ellsworth & Scherer, 2003). An event or condition therefore should be worse to the extent that the goal it threatens is important for the person. Such goals may refer to different levels of generality, such as general motives (McClelland, 1987), (professional) identities (Semmer, Jacobshagen, Meier & Elfering, 2009), personal projects (Emmons, 1996) or current task goals (cf. the concept of performance constraints – see Leitner & Resch, 2005; Spector & Jex, 1998; Semmer et al., 1996). Since people differ with regard to the goals they pursue (Cropanzano, James & Citera, 1993), goals should be an important source of individual differences with regard to (occupational) stress (cf. Semmer, 2006).
The first aspect that seems important is simply having goals and being committed to them. Pursuing goals in general is associated with higher well-being (Emmons, 1996), and commitment to personal work goals is associated with higher job satisfaction (Roberson, 1990). However, not all goals are created equal. Thus, seeking positive goals is associated with better well-being than avoiding negative goals (Emmons, 1996), and the proportion of positive/negative work-related goals is associated with job satisfaction (Roberson, 1990). Furthermore, ‘extrinsic’ goals (e. g, money, good looks) are associated with well-being only to the extent that they are instrumental for achieving intrinsic goals (Diener & Fujita, 1995).
At the same time, however, being committed to a goal may increase vulnerability (Lavallee & Campbell, 1995). The two processes together may imply that people are better off on average when they are committed to a goal, yet suffer more when that goal is threatened. In line with this reasoning, Reilly (1994) reports that nurses who were more committed to their profession showed lower mean levels of emotional exhaustion (the core component of burnout). At the same time the relationship between the frequency of experienced stressors and emotional exhaustion was stronger for the more committed. Similarly, research by Brockner, Tyler and Cooper-Schneider (1992) shows that people who were highly committed to their organization reacted in an especially negative way to perceived unfair treatment. On the positive side, the importance placed on a given facet of one’s work (e.g., opportunity for promotion, amount of decision making) moderated the relationship between the extent to which this facet is perceived to be present and overall job satisfaction in a study by Rice, Gentile and McFarlin (1991).
Motives represent rather general, and quite stable, goal strivings. There is some research on the implications of motives for stress, well-being and health. Thus, power strivings are associated with lower well-being (Emmons, 1991), and with stronger reactions to stressful events (Jemmott, 1987). Affiliation/intimacy strivings tend to be associated with positive indicators of well-being but with a stronger reaction to interpersonal stressors (Emmons, 1991, 1996). In line with such findings, women, who tend to be more committed to interpersonal goals than men, have been found to be more vulnerable to stress experienced by significant others (Kessler & McLeod, 1984). The main effect of need for achievement (nAch) is controversial (Veroff, 1982 vs. Emmons, 1991, 1996), which may be due to nAch having a ‘toxic’ and a ‘non-toxic’ component (Birks & Roger, 2000). Again, however, people high in nAch react more strongly to achievement-related events (Emmons, 1991). A high need for control (which is characteristic of people with the type A behaviour pattern) may induce people to set too high goals (Ward & Eisler, 1987) and to react strongly to control being threatened (Edwards, 1991).
A special case refers to multiple goals, which implies the possibility that these goals come into conflict with one another (Kahn & Byosiere, 1992). Multiple goals have often been studied in relation to the combination of work and family roles, in many cases referring to women participating in the labour force, in addition to their role as spouse and parent. As with goal pursuit per se, the main effect of multiple roles on well-being typically is positive rather than negative, provided that the attitude towards the multiplicity of roles is positive (Amstad & Semmer, 2009). At the same time, work–family conflict is a well-established predictor of strain (Amstad et al., 2009).
The conclusion from these considerations is that differences in vulnerability to stressful experiences often are to be found in people’s goals. It might be hypothesized that one way to reduce vulnerability is to reduce one’s goals. And, indeed, being able to reduce or give up unrealistically high goals is important for a person’s well-being (Wrosch et al., 2003). At the same time, however, reducing one’s aspirations is a double-edged sword (Hobfoll, 2001). Recall that the main effect of pursuing goals typically is positive. Giving up goals may therefore imply forgoing possible challenges that are experienced in a positive way and may entail positive consequences. Furthermore, giving up may have detrimental consequences. Especially if the goal in question is highly valued not only individually but also socially (e.g., achievement goals), giving up may have far-ranging negative social consequences. It is probably due to such negative consequences that people sometimes reduce goals in a resigned and resentful way, acquiescing themselves to the inevitable more than really giving up the goal. Wrosch et al. (2003) characterize this mechanism as ‘giving up effort, but remaining committed to the goal’ (cf. the concept of ‘resigned job satisfaction’; Büssing, 1992).
In line with these considerations, person-environment (P-E) fit research shows that ‘fit’ at low levels (e.g., having, and ‘wanting’ low complexity) is associated with more distress than fit at high levels (Edwards & Van Harrison, 1993). Frese (1992) reports similar results for people who reject, rather than aspire, control at work. The role of ‘fit’, therefore, cannot be assessed without considering absolute levels.
Thus, it seems that aspirations and expectations cannot be reduced ad lib. Rather, they may in many cases be indicators of a ‘failed person–work interaction’ (Büssing, 1992, p. 254). Note that this implies that employing ‘stress-resistant’ people may be counterproductive if their resilience is based not on high resources but on low commitment to certain goals (especially in the interpersonal domain; cf. Cobb, 1973).
People differ in their beliefs about the world and their relationship with it, especially possibilities to deal with it. Candidates for this kind of variables are hardiness (e.g. Kobasa, 1988) or sense of coherence (Antonovsky, 1991), locus of control (Rotter, 1966), self-efficacy (Bandura, 1989), optimism (Scheier & Carver, 1992), self-esteem (Brockner, 1988), or hostility (cf. Siegman, 1994a).
Hardiness is conceived of as being composed of three components: commitment, challenge and control (see Maddi, 2002). ‘Commitment is the ability to believe in the truth, importance, and interest value of who one is and what one is doing; and thereby, the tendency to involve oneself fully in the many situations of life... Control refers to the tendency to believe and act as if one can influence the course of events... Challenge is based on the belief that change, rather than stability, is the normative mode of life’ (Kobasa, 1988, p. 101).
From this concept it follows that people high in hardiness should better be able to deal with stressful aspects of life. Research often shows the main effects of hardiness on physical and psychological health (Beehr & Bowling, 2005). Both stress appraisal and coping seem to be mediators of this relationship (Florian, Mikulincer & Taubman, 1995), as implied by the concept. However, evidence on moderator effects is mixed (Beehr & Bowling, 2005). Furthermore, recent longitudinal research also found a reverse relationship, showing that stress can have a negative effect on hardiness (Vogt et al., 2008).
A basic problem with the hardiness construct is the confound with other constructs, such as neuroticism (or negative affectivity). Relations with such constructs are substantial, and associations with third variables usually drop considerably when controlling for these. In a number of studies, however, associations do remain even with these controls (Semmer, 2006).
Sense of coherence (SOC) also is quite a broad construct and overlaps with hardiness. Its three main features are that the environment is perceived as structured, predictable and explicable, and thus as comprehensible, that one perceives to have the resources necessary to deal with one’s environment, thus perceiving manageability, and that the demands posed by one’s environment are interpreted as challenges which are worthy to be taken up, leading to the perception of meaningfulness (Antonovsky, 1991).
Research on SOC shows relationships with a number of indicators of well-being and health (e.g. Antonovsky, 1993; Söderfeldt et al., 2000). Main effects are predominant, but interactions with working conditions also are sometimes found (e.g. Feldt, 1997; Johansson Hanse & Engström, 1999; Söderfeldt et al., 2000). Effects of SOC have also been demonstrated longitudinally for subjective health ratings (Suominen et al., 2001), diabetes (Kouvonen et al., 2008), and reduced mortality (Surtees et al., 2006). Similar to hardiness, rather strong relationships with anxiety (Antonovsky, 1993), depression (Geyer, 1997), and other indicators of well-being (Eriksson & Lindström, 2005) have raised doubts about its distinctiveness from neuroticism, or negative affectivity (see Geyer, 1997).
Locus of control is one of the variables that has very often been shown to be related to well-being (Spector et al., 2002). Locus of control may also be a moderator of the interaction proposed by Karasek (Karasek & Theorell, 1990). Thus, Meier et al. (2008) found that job control had a buffering effect only among individuals with an internal locus of control. Like locus of control, self-efficacy has consistently been shown to be related to well-being (cf. Bandura, 1992) and both are part of a chain of convictions regarding a person’s possibility to cope. The two convictions are not the same, but they do overlap (cf. Ng, Sorensen & Eby, 2006). A number of studies have found self-efficacy to buffer the effects of stressors (e.g., Jex et al., 2001; Jimmieson, Terry & Callan, 2004) or of resources like control (Jimmieson, 2000). Some recent findings suggest that the interaction between demands and control as specified in the Karasek model (Karasek & Theorell, 1990) might be valid only for people high in self-efficacy (e.g., Jimmieson, 2000; Meier et al., 2008; Schaubroeck & Merritt, 1997), resembling the pattern found for internal locus of control (see above). Self-efficacy not only has similarity with locus of control, but also with self-esteem. In its generalized form (general self-efficacy; Jerusalem & Schwarzer, 1992) it seems quite indistinguishable from self-esteem, at least from those parts of self-esteem that are related to a person’s perceived competences (cf. Judge & Bono, 2001).
Both self-efficacy and self-esteem seem especially important for dealing with negative feedback and failure in terms of distress as well as persistence (Bandura, 1989; Brockner, 1988). However, Cohen and Edwards (1989) are very sceptical about the moderating effect of self-esteem, although some more recent studies do show such interactions (Ganster & Schaubroeck, 1991; Jex & Elaqua, 1999; Mäkikangas & Kinnunen, 2003). A number of studies indicate that it is not simply the level of self-esteem that is important but also its stability (Kernis, 2005). Unstable high self-esteem reflects a fragile feeling of self-worth which is associated with more hostility and anger (Kernis, Grannemann & Barclay, 1989) and a heightened responsiveness to self-threatening events, such as negative feedback, or unfair treatment (Greenier et al., 1999). Maintaining self-esteem is an important goal (Lazarus & Folkman, 1984; Semmer et al., 2007) and threats to self-esteem are perceived as stressful. In line with this, two studies of our group showed that individuals with a fragile self-esteem are especially reactive to egothreatening job conditions like unfair treatment (Meier, Semmer & Hupfeld, 2009) and effort–reward imbalance (Meier & Semmer, 2008).
Optimism is distinct from control-related concepts because it does not require that the course of events is influenced by one’s own actions. Rather, it includes the belief that things are likely to turn out reasonably well anyway (thus being related to a belief in a basically benign world). It has been shown to influence stress appraisals, well-being and coping strategies (Carver & Scheier, 1999). Optimists tend to employ more problem-solving strategies under controllable conditions, and more reinterpretation and acceptance under less controllable conditions (Nes & Segersstrom, 2006). Of special importance is the finding that optimists tend to accept failures better, which relates to the ‘circumscribed’ frustration as described by Hallsten (1993) and is indicative of the capability of putting things into perspective.
Hostility is often regarded as the major ‘toxic’ component of the type A behaviour pattern (Adler & Matthews, 1994), but some authors treat hostility and type A as related but independent constructs (Myrtek, 2007). Conceptually, we can distinguish between: (i) a cognitive component, involving hostile beliefs and attitudes about others (cynicism, mistrust, hostile attributions) – this aspect is dominant in the most frequently used Cook–Medley–Ho Scale (cf. Myrtek, 2007); (ii) an emotional component, involving anger; and (iii) a behavioural component, involving physical or verbal assault (Buss & Perry, 1992). The accumulated evidence suggests that hostility is predictive of ill health coronary heart disease (CHD), and all-cause mortality (Miller et al., 1996), although a definitive conclusion cannot be drawn yet (Myrtek, 2007). Hostility is associated with vascular resistance during interpersonal stress (Davis, Matthews & McGrath, 2000) and stronger neuroendocrine, cardiovascular and emotional responses to interpersonal harassment (Suarez et al., 1998). Recent diary studies give interesting insight in the functioning of hostile people. Judge, Scott and Illies (2006) showed that among individuals high in trait hostility, perceived unfair behaviour by their supervisor was more strongly related to state hostility. State hostility, in turn, was negatively related to job satisfaction, and positively related to workplace deviance. However, hostile persons are not only particularly reactive to negative interpersonal but also to positive, supportive interactions: Vella, Kamarck and Shiffman (2008) showed that for people high in hostility (and only for them) instrumental social support increased diastolic blood pressure.
There has been quite some debate on the role of expressing anger as predictor of CHD, with some authors (e.g., Steptoe, 1996) regarding anger-in, and others (e.g., Siegman, 1994b) anger-out as the important component. Evidence seems to be more supportive for anger-out as predictor of CHD (Miller et al., 1996). Note, however, that the implication is not that components of hostility other than anger-out are irrelevant. They are weaker predictors only with regard to CHD, but they are good predictors of mortality from all causes (Miller et al., 1996). Anger-in may be especially important for the development of cancer (Siegman, 1994b) and being low in anger expression may be involved in the development of high blood pressure (Steptoe, 2001).
Expressing or not expressing one’s anger may, however, not be the most important aspect. Rather, it may be what expressing, or not expressing, one’s anger does to the person in terms of ending vs. prolonging the anger. Expressing anger can be constructive (e.g. explaining one’s feelings to a partner) or antagonistic (offending, blaming the partner). Likewise, not expressing the anger may be antagonistic if associated with ruminating, self-pity, dreaming about revenge etc., but it may be non-antagonistic by putting things into perspective, trying to see them from a humorous side, trying to understand the other’s perspective, etc. It may well be that it is the antagonistic vs. constructive way of dealing with anger that is most important, not the question of whether it is expressed or not (Davidson et al., 1999; cf. Semmer, 2006).
Judging from one perspective, the different concepts and the findings related to them are rather confusing. Although there is some overlap between different concepts, it is unclear how many different constructs are involved and how they are hierarchically ordered. Certainly, more studies are needed to investigate the communalities and differences involved.
From another perspective, however, the picture is not so gloomy. There do seem to be some common elements that appear in different studies, and if we look at the ‘great lines’, we might come to a conclusion like the following:
People who are resilient:
All this implies also that people who are resilient do show emotional stability and do not have a tendency to experience negative emotions (neuroticism, negative affectivity).
Theoretically, the concepts mentioned should influence coping strategies which would imply that they should act as moderators in the relationship between stress and outcome variables. Such findings are obtained quite often (see above) but not nearly as consistently as would be expected theoretically.
One reason for this is certainly to be found in methodological difficulties, because moderated regression procedures tend to yield very conservative estimates of interaction effects (cf. Cohen & Edwards, 1989). A further reason for this might be that the resourceful belief system pictured here has an influence at a much earlier point, that is, changes the stress appraisal in the first place, as discussed in Section 6.2.2.
That resources in the sense of the belief systems discussed here are so often found to be directly related to symptoms points to another, very basic question. It is possible that all these measures are really indicators of NA (Watson, Pennebaker & Folger, 1987), or neuroticism (Dembroski & Costa, 1987). In fact, the measures discussed here are often found to correlate with one another, some have been combined to the larger meta-construct of ‘core self-evaluations’ (Judge & Bono, 2001), and controlling for NA often reduces associations between belief systems and symptoms (e.g., Schaubroeck & Ganster, 1991). Indeed, it would be quite strange if there were not a substantial relationship between the broad construct of NA and belief systems that have to do with an environment that is meaningful, basically benign and able to be influenced, and with a self-concept that involves the capability to actually influence this environment in accordance with one’s goals. Also, the etiology for the aforementioned belief system involves experiences of mastery, as well as experiences of failure that can be dealt with; these are conditions that we would also assume to influence NA. And, indeed, chronic stress conditions are found to influence changes in NA over time (cf. Spector et al., 2000).
Therefore, attributing an important role to NA in concepts of a disease (or health) prone personality does not imply that associations between stressors and health can be reduced to reflecting NA. Rather, NA would be seen as a factor that may influence the experience and perception of, as well as the reactions to, stress factors, but at the same time may be influenced by these factors (Spector et al., 2000).
A related issue refers to common method variance. Since most studies on stress at work use self-report, common method variance may account for the associations found (and the common factor behind that may, again, be negative affectivity; cf. Spector, 2006). This has, however, not gone unrefuted (e.g., Chen & Spector, 1991; Spector, 2006; Spector et al., 2000). Analyses with different indicators of job stressors (e.g., self-report and ratings by trained observers) in our research have consistently shown that correlations between stressors and symptoms are, indeed, inflated by common method variance but that substantive associations remain when this is controlled (e.g., Semmer, Zapf & Greif, 1996).
Coping is one of the most important concepts in research on stress. It refers to all attempts to manage a stressful transaction, to make it less stressful (cf. Lazarus & Folkman, 1984). These attempts are based on an appraisal of the situation (primary appraisal) and a person’s possibilities to deal with it (secondary appraisal) and are, therefore, specific to the characteristics of the situation. Nevertheless, it has become clear that people also have person-specific tendencies to use certain coping strategies (cf. Carver, Scheier & Weintraub, 1989). Stable individual differences in coping, however, do not require that certain people will always employ strategy A and others strategy B. There may also be differences on a ‘meta-level’ in that some people are habitually more flexible in their strategies. A recent meta-analysis shows that optimists are more likely to accept uncontrollable situations and more likely to use active coping strategies in controllable situations than are pessimists (Nes & Segersstrom, 2006). In other words: optimists show a tendency towards coping strategies that are adequate for the situation (Carver & Scheier, 1999).
There are many classifications of coping, the most basic one being the dichotomy between problem-focused vs. emotion-focused coping (see Lazarus & Folkman, 1984). A somewhat different approach concentrates on the tendency to seek or avoid information concerning the stressful aspects of the situation. This is most clearly expressed in the coping styles called ‘monitoring’ and ‘blunting’ by Miller (e.g. 1990).
There are many expansions and blends of these approaches and there is by no means consensus over number and kind of the dimensions to be employed. This problem is further aggravated by the fact that the same labels do not necessarily imply the same concept. For example, items referring to alcohol consumption, eating or smoking are sometimes part of an avoidance or denial factor (e.g., Koeske, Kirk & Koeske, 1993; Endler & Parker, 1990; Carver, Scheier & Weintraub, 1989), but sometimes they belong to an emotion-focused factor (e.g., Billings & Moos, 1984).
In light of this confusion it is surprising that nevertheless there are some tendencies where research is converging. Thus, in general (and with many exceptions), the tendency to employ problem-focused coping is associated with better mental (and sometimes, physical) health while emotion-focused coping tends to show the opposite relationship (e.g., Carver et al., 1989). Problem-focused coping has also been found to moderate the relationship between control and demands according to the Karasek model (Karasek & Theorell, 1990), in that people who show ‘active coping’ profit most from control under conditions of high stress (e.g., Ippolito et al., 2005). Avoidance-oriented coping is often found to be beneficial in the short run but detrimental in the long run (Suls & Fletcher, 1985). Also, not surprisingly, avoidance is more beneficial if the problem is uncontrollable whereas approach is more instrumental when something can be done about the situation (Miller, 1990).
One aspect of this research that is somewhat difficult to interpret is the often-reported finding that ‘emotional coping’ tends to be associated with poorer mental health and poorer outcomes (Edwards, 1998). The reason why this is puzzling is that many authors postulate that emotional coping should not be detrimental per se. Rather, highly stressful experiences may require some management of intense emotions before a person is able to deal with the problem in a more active and direct way (e.g., Lazarus & Folkman, 1984; Reicherts & Perrez, 1992). Emotional coping in terms of trying to calm down has been found to predict calming down in a diary study by Elfering et al. (2005).
An especially interesting approach to this problem is presented by Reicherts and Perrez (1992). They formulate a ‘behaviour rules approach’ which specifies which coping strategies should work best under what conditions. In line with many others, they postulate that under conditions of greater controllability there should be more active and less avoidance coping. At the same time, however, they ‘prescribe’ more palliative coping under high stressfulness and more re-evaluation of standards when the probability of re-occurrence of the situation is judged to be high. In taking into account the ‘fit’ between characteristics of the situation and the coping strategies used, and by allowing analyses of how appraisals of and coping with a situation may change over time, this seems an especially interesting approach which can be expected to greatly enhance our understanding of coping effectiveness and of the role of individual differences in coping processes.
There is an additional problem, however, with the conceptualization and, especially, the operationalization of ‘emotion-focused’ coping. As an example, Carver et al. (1989) report a scale they call ‘focus on and venting of emotions’, with items such as ‘I get upset and let my emotions out’; Endler and Parker (1990) report an emotion-focused scale that taps ‘self-blame’. These scales correlate with several scales indicative of NA, such as anxiety, depression or neuroticism. If we define coping as ‘efforts to manage specific external and/or internal demands’ (Lazarus & Folkman, 1984, p. 141), then it is doubtful whether what is being measured here can really be called coping. Rather, items like these seem to measure how strongly a person feels distressed and the inability to concentrate on anything than the distressing thoughts. Based on these considerations, Kälin and Semmer (2008) developed an instrument that aims at assessing ‘true’ emotional, or palliative, coping. Unlike traditional measures of emotional coping, such as the respective subscale of the CISS (Endler & Parker, 1990), it correlates negatively with measures of strain (cf. Elfering et al., 2005).
In general, people who have a tendency to cope by dealing actively with problems tend to be better off. However, where the situation is taken into account, is becomes clear that palliative modes of coping may be beneficial in highly stressful situations where palliation may be instrumental in building up the resources needed for other forms of coping, and in situations that cannot be controlled. The latter also call for a re-evaluation of one’s goals. Such coping strategies do seem to have a habitual component, as does the tendency to adjust one’s strategy to the characteristics of the situation and its changes over time. These strategies, in turn, depend, at least in part, on personality characteristics such as hostility, neuroticism and on resourceful belief systems.
It becomes increasingly clear that resilient people have a certain way of dealing with reality. Coping actively under all circumstances, nourishing illusions over one’s capabilities that are far from reality, or having a naive optimism, are not characteristics of this effectiveness. While individual differences with regard to coping with and suffering from stressors are pervasive, they should not seduce us to reduce everything to idiosyncratic, exclusively subjective, phenomena (see Hobfoll, 2001). In many cases, it is not the objective situation per se but the way people deal with it that decides about outcomes. It should be kept in mind, however, that resiliency itself is, albeit only partly, a product of such circumstances. If one examines the effects of stress on health on the one hand and on the development of resiliency vs. vulnerability on the other, the parallels are striking. Apart from traumatic single experiences, it is chronically stressful conditions that overtax people’s resources, impairing not only their health but also their coping resources. This often leads to the perception of the person being ‘the cause’ of the problems, because he or she seems unable to deal with problems that other people do deal with effectively. This supports an attribution error – for lay people and scientists alike – that induces people to overemphasize individual differences and to underemphasize reality and not to see the vicious circle in which one is reinforcing the other. The picture worsens when some of these cumulative effects refer to characteristics of the person that tend to irritate others (e.g. excessive complaints – Silver, Wortman & Crofton, 1990) or even antagonize them (e.g., uncivil behaviour – Blau & Anderson, 2005). In such a case, the person does, indeed, create new stressors for him- or herself as well as for others, and it becomes very difficult for others to see how much this ‘actor’ is also a ‘victim’ of stressful life circumstances during his or her life.