Chapter prepared for:
Keywords: Coping instruments; Coping measurement; Coping assessment;Coping strategy; Psychometric scales
As we go through life, we can resist temptation, exercise to deal with depression, reinterpret loss, talk ourselves through challenges, avoid confronting an opponent, seek help, etc. Coping with an adversity includes innumerous ways of dealing with diverse person-environment transactions. Thus, coping does not represent a homogeneous concept. Instead, it is a diffuse umbrella term. Coping can be described in terms of strategies, tactics, responses, cognitions, or behavior. Actual coping is a phenomenon that can be noticed either by introspection or by observation, and it includes internal events as well as overt actions. It has been broadly defined as "...cognitive and behavioral efforts to manage specific external or internal demands (and conflicts between them) that are appraised as taxing or exceeding the resources of a person (Lazarus, 1991, p. 112).
There are at least three important ingredients that should be considered in the conceptualization of coping, namely (a) coping need not be a completed "successful" act, but an effort has to be made; (b) this effort need not be expressed in actual behavior, but can be directed to cognitions as well; and (c) a cognitive appraisal of the taxing situation is a prerequisite of initiating coping attempts.
A serious problem, however, is that cognitive coping and cognitive appraisal can be confounded. Appraising a situation, for instance as a threat, may trigger coping, i.e., further thoughts or defenses that imply a reappraisal of the same situation as being more or less threatening. In such cases, a distinction between appraisal and coping cannot be practically made, but it remains at least of heuristic value. Lazarus (1991) tries to disentangle this overlap somewhat, stateing: "... coping refers to what a person thinks or does to try to manage an emotional encounter; and appraisal is an evaluation of what might be thought or done in that encounter" (p. 113).
Other conceptual problems arise when coping is to be separated from coping resources (e.g., hardiness, dispositional optimism, self-efficacy, sense of coherence, social support, etc.) Resources can be personal, social or other antecedents of appraisals and coping. An optimistic attitude towards life may result in a more favorable appraisal of a taxing situation, in adopting an efficient problem-solving strategy, or in creating optimistic coping self-talk. The existence of a social network may result in successful support-seeking behaviors when need arises. Although in reality coping resources and actual coping may be difficult to disentangle, it is important to make this distinction in theory and research. Resources are relatively static antecedents, whereas coping is a process that depends on these resources. If, for example, an optimistic statement is made by a coper, it may mainly reflect a personality trait, or it may have just been generated as a product of effortful stress management.
The measurement of coping is complicated by the above-mentioned conceptual issues. Furthermore, there are problems that will be discussed under the headings of stability, generality, and dimensionality of coping.
The assessment of coping can be a description of the cognitions and behaviors of a person dealing with a stressful encounter. This approach does justice to the fact that coping is a process, and it allows the identification of contingencies between changing situations and changing actions-be it by time sampling or event sampling. For example, one can assess whether a person always applies and reapplies the same set of strategies, or whether she applies a broad range of tactics that are well-adapted to changing encounters. This idiographic approach to stability and change can be adopted for single cases in clinical settings, but it is not common in research. Rather, the focus in empirical studies is on individual differences. Stability then refers to the pattern similarity of interindividual differences at multiple points in time. If, for example, some persons cope in a vigilant manner while others do so in a more avoiding manner, and if this reoccurs at later observations, one is inclined to attribute stable coping preferences to these individuals. When we measure coping with standardized instruments, we therefore imply that people can be characterized by some preferred ways of coping with adversity, and that they continue to apply the same kind of strategies over time. This dispositional implication helps to reduce the complexity of coping assessment, but it does so at a high price: It assumes namely that uniqueness of situation-specific coping responses only represents a negligible aspect.
A further difficulty that makes stability a crucial issue is the fact that people usually go through stages when managing a taxing demand. For example, someone confronted with surgery has to proceed from the preparation stage to the confrontation stage, and then to the recovery stage. A coping strategy that was adaptive in the first stage may not be so in the second, and a completely different approach might be practical in the third. Therefore, if stages can be identified, it is important to assess coping within each of them and to group individuals in terms of them. This remains an issue for further research.
Closely related to stability is another problem, namely the consistency of coping responses across different situations. Do people apply the same strategy when they face an exam, the bereavement of a loved one, or an argument with their spouse? They may not show exactly the same responses, but maybe they can be characterized by a general tendency to select appropriate behaviors either from the class of avoidance or from the class of confrontation strategies. If all responses could be explained by the challenging events, this would reflect a pure situation determinism. In contrast, the common person-situation interaction perspective would consider joint influences from both sources. A moderate amount of generality implies that people construct a series of person-dependent strategies for a number of situations. The measurement of coping can only be fruitful under the assumption that individuals generalize across situations to a certain degree and come up with a limited set of strategies that they reapply at different occasions.
These sets of strategies can be grouped, for example according to their purpose, meaning, or functional value. Since the number of specific responses is endless, it appears to be useful to classify them in one way or the other. Empirically, this can be accomplished by factor analysis, but factor solutions typically differ from sample to sample and from stressor to stressor. There are many attempts to reduce the total of possible responses to a more parsimonious set of dimensions. Some researchers have come up with two basic dimensions-such as instrumental, attentive, vigilant, or confrontative coping on the one hand, in contrast to avoidant, palliative, and emotional coping on the other (for an overview see Krohne, 1993; Laux & Weber, 1993; Suls & Fletcher, 1985).
The number of dimensions that have been established theoretically or found empirically also depends on the level within a hierarchy of coping concepts. Krohne (1993) distinguished a behavioral from a conceptual level, each consisting of two subclasses. Reactions and acts constitute the behavioral level at the bottom of this hierarchy. A coping reaction is considered as a single behavioral element, for instance tuning to a music channel instead of an information channel in a laboratory stress experiment. Several similar reactions can be grouped into an act, such as executing a specific problem-solving behavior. At the conceptual level, researchers can identify a set of acts reflecting a particular strategy, for example making use of social resources or turning to religion. Strategies, in turn, can be grouped into superstrategies, two of which are vigilance and cognitive avoidance. According to Krohne's coping theory, these two superstrategies constitute orthogonal dimensions of attention orientation and can be understood as personality dimensions. "Vigilance refers to those strategies which are characterized by intensified intake and processing of threatening information. Cognitive avoidance is marked by a turning away from the threat-related cues" (Krohne, 1993, p. 21).
Another conceptual distinction has been suggested between assimilative and accomodative coping, the former aiming at an alteration of the environment to oneself, and the latter aiming at an alteration of oneself to the environment (Brandtstädter, 1992). This pair has also been coined "mastery versus meaning" (Taylor, 1983, 1989) or "primary control versus secondary control" (Rothbaum, Weisz, & Snyder, 1982). These coping preferences may occur in a certain time order when, for example, individuals first try to alter the demands that are at stake, and, after failing, turn inward to reinterpret their plight and find subjective meaning in it.
A well-known approach has been put forward by Lazarus and Folkman (1984), who discriminate between problem-focused and emotion-focused coping. The first is seen as being action-centered in the sense that the troubled person-environment relationship is changed by instrumental actions. These actions need not necessarily be successful and may have even detrimental side effects; however, it is the attempt that counts. In contrast, the second kind of coping includes mainly cognitive coping strategies that do not directly change the actual situation, but rather help to assign a new meaning to it. They are not passive, but may require an internal restructuring and may cost considerable effort.
A different view has been suggested by Klauer, Filipp, and Ferring (1989) in the context of coping with chronic disease. The authors start out with a multidimensional structure to describe actual coping. Three dimensions have been established and empirically confirmed: (a) focus of attention that differentiates whether attention is directed at the disease or elsewhere, (b) sociability that describes whether the patient turns toward others or withdraws, and (c) response level that describes whether coping is reflected by overt behaviors or by cognitive (intrapsychic) responses. This results in eight possible coping strategies.
There are many other attempts to conceptualize coping dimensions, and those mentioned above may serve as examples. They are a prerequisite of coping measurement because a pure inductive collection of many single responses that have been factor-analyzed would result in an unstable solution and could hardly be replicated in further studies. The high degree of situational or intraindividual variability in coping obviously implies a multilevel conceptualization with a few stable dimensions at higher levels that are theoretically linked to a variety of specific strategies and acts at lower levels. The following sections survey the coping inventories used frequently and scrutinize their contributions to the stability, generality, and dimensionality issues.
One basic dimension of coping is attentional style. The question is whether individuals under stress tend to tune into the threat, search for information, and visualize scenarios to control the situation, or whether they tune out from the threat, distract themselves, avoid further information, and postpone instrumental action. Adopting a vigilant attentional style was coined "monitoring", and adopting an avoiding attentional style was called "blunting" by Miller (1987), who conceived these as typical individual differences in a stress situation. To assess these styles, she developed the Miller Behavioral Style Scale (MBSS), an instrument consisting of four hypothetical situations, two of them referring to physical threat and two to ego-threat. One of the physical threat scenarios is phrased "Imagine that you are afraid of flying and have to go somewhere by plane." Following these situations are eight coping options, half of them reflecting a monitoring and half a blunting attentional style. Monitoring is expressed in items such as "I would carefully read the information provided about safety features in the plane." Blunting is signified in items such as "I would watch the in-flight film even if I had seen it before." Since each of the four stress situations is followed by eight coping responses, the inventory consists of 32 items. Scores are supplied by adding up the corresponding items yielding a monitoring and a blunting sum score.
The MBSS was originally validated by two experiments (Miller, 1987). In the first, a physical threat situation was created by announcing an electrical shock. Individuals characterized by high monitoring and low blunting were searching for information about the imminent shock (that was never given). On the other hand, high blunters who were also low monitors tended to distract themselves. In the second experiment, an ego-threat situation was provided where students underwent an academic achievement test. They could obtain on-line information about their current achievement status. The monitors made frequent use of this opportunity, whereas the blunters did so less often. Additional studies have confirmed the diagnostic value of the MBSS (see Miller, Combs, & Kruus, 1993).
Attentional style has proven a useful dimension of individual coping differences, and its measurement with the MBSS has stimulated research on this topic. The main limitation of this inventory lies in its range of content because it restricts itself to only those stress situations that elicit anxiety. Other stress situations that are appraised as challenge, harm, or loss, for example, are not suitable for it.
According to the two-dimensional coping theory of Krohne (1993) mentioned briefly in the introduction, an instrument has recently been developed to assess vigilance and cognitive avoidance in threat situations (Krohne, Schuhmacher, & Egloff, 1992). Eight hypothetical situations are selected that supposedly elicit anxiety, half of them emulating physically threatening situations and the other half ego-threatening situations. One of the four physical threat situations is phrased "Imagine that you are riding in a car as a front-seat passenger next to an obviously inexperienced driver. Road conditions are poor due to snow and ice." One of the four ego-threat situations is phrased "Imagine that you have to take an important exam three weeks from now." Researchers who consider adopting this instrument may select only one of its two parts depending on whether their study is restricted to physical threat or to ego-threat.
Each situation is followed by 18 coping acts that are subdivided into those reflecting vigilance and those reflecting cognitive avoidance. For the physical threat situation listed above (car scenario), the nine vigilant coping acts include items such as "I watch the driver carefully and try to tell in advance when he is going to make a mistake." The nine avoidant coping acts include items such as "I tell myself 'Thank goodness, he is not driving that fast.'" A true-false response format is provided for each coping act. The nine vigilant coping acts refer to: (1) recall of previous negative events, (2) self-pity, (3) information search, (4) social comparison, (5) planning, (6) flight tendencies, (7) informational control, (8) anticipation of negative events, and (9) situation control. The nine avoidant coping acts refer to: (1) general bagatellisation, (2) self-enhancement, (3) bagatellisation by reinterpretation, (4) distraction, (5) playing down by incompatible reactions, (6) denial, (7) focus on own strength, (8) focus on positive aspects, (9) confidence. Whether these are coping acts or coping strategies instead seems to be a matter of definition, and the author uses both expressions (Krohne, 1993). Most of these expressions can be found in other coping inventories, some linked to different dimensions. For example, situation control can also be labelled mastery, and reinterpretation can also be labelled meaning.
Sum scores can be created separately for the type of situation and type of coping, e.g., there are subscales for (a) vigilance in ego-threat situations, (b) cognitive avoidance in ego-threat situations, (c) vigilance in physical threat situations, and (d) cognitive avoidance in physical threat situations. The focus of the assessment lies on these four sum scores, which represent the two dispositional superstrategies vigilance and cognitive avoidance for two distinct classes of situations. The 18 coping acts do not represent dimensions on their own, but merely serve as test items to generate scores for the superstrategies.
Krohne reports favorable psychometric properties of the instrument. In particular, the subdivision into the two hypothesized superstrategies vigilance and cognitive avoidance was confirmed by factor analysis in a sample of women responding to ego-threat as well as to physical threat. For men, however, the structure could be confirmed only in the physical threat situation. Since this is a very recent instrument, further validation studies need to be conducted.
The limitation of this approach lies in its mere focus on threat, in the same manner as for the monitoring and blunting concept. Krohne's coping mode theory stems from his research on anxiety and personality and, thus, represents a dispositional approach to threat management. Another limitation of the instrument lies in its validation within German samples only. The research has been published in English, but the scale itself awaits cross-cultural validation.
In the context of research on chronic stress and coping in families, Billings and Moos (1981) have designed a measure to assess coping with life stress. After indicating a recent life crisis, 194 families (men and women) responded to 19 coping statements with "yes" or "no". Inspired by the stress and coping theory of Lazarus (1966), the items were intuitively grouped according to the method of coping and the function of coping. The methods were (a) active cognitive, (b) active behavioral, and (c) avoidance. The functions were (a) problem-focused and (b) emotion-focused coping. (The authors use the term "focus" instead of "function", the latter being the term preferred by Lazarus). Each of the 19 items fitted into one of the methods and one of the functions. The internal consistencies are partly satisfactory. This instrument is based on a rational construction and is in line with other theories, but empirically its structure has not been confirmed.
The authors have followed up their work in subsequent reseach where they applied an extended version of their instrument in a study with 424 depressives (Billings & Moos, 1984), in which they used 32 items and preferred a four-point Likert scale response format. After an item analysis, 28 items remained and formed five scales that were grouped into three major forms of coping. Appraisal-focused coping consisted of a four-item scale that referred to a logical analysis of the stress at hand, with items such as "considered several alternatives for handling the problem". Problem-focused coping included two subscales, (a) information seeking with seven items, such as "tried to find out more about the situation", and (b) problem solving with five items, such as "made a plan and followed it." Finally, emotion-focused coping also consisted of two subscales, (a) affective regulation with six items, such as "tried to see the positive side of the situation", and (b) emotional discharge with six items, such as "let my feelings out somehow."
In terms of theory, this instrument conforms well with the majority of current measures. Empirically, the scales show unsatisfactory internal consistencies. In addition, there is not sufficient evidence about time stability, and, in particular, the intuitive dimensional structure requires a rigid test in various samples before it can be generalized. The value of this inventory lies in its theoretical perspective and in the stimulation it has provided at a time when almost no satisfactory coping scales were available.
Already in the 1970s, the stress and coping research group of Lazarus developed the Ways of Coping Checklist (WCC) in line with the transactional phenomenological stress theory that suggested two main functions of coping: problem solving and emotion regulation (cf. Lazarus, 1991). From a pool of 68 items with a yes-no response format, 40 items formed the problem-solving subscale, and 24 the emotion-focused subscale (Folkman & Lazarus, 1980). Since this classification did not reflect the complexity and richness of coping processes, a series of factor analyses with different data sets were carried out, generating over time the current version of the instrument now called the Ways of Coping Questionnaire (WCQ). The WCQ consists of 50 items (plus 16 fill items) within eight empirically derived scales (Folkman & Lazarus, 1988). A difficulty with the instrument has always been that the number of extracted factors changed from sample to sample or from stressor to stressor (Parker & Endler, 1992). But this seems to be a general problem with most coping measures, reflecting the unresolved disposition versus situation issue.
Following are the eight scales with sample items. The response format has been changed to the four-point Likert rating scale. Responses are made after a real-life stress situation that had been experienced, for example during the last week, was written down.
1. Confrontive coping (6 items)
"Stood my ground and fought for what I wanted." 2. Distancing (6 items)
"Went on as if nothing had happened." 3. Self-controlling (7 items)
"I tried to keep my feelings to myself." 4. Seeking social support (6 items)
"Talked to someone to find out more about the situation." 5. Accepting responsibility (4 items)
"Criticized or lectured myself." 6. Escape-avoidance (8 items)
"Hoped a miracle would happen." 7. Planful problem solving (6 items)
"I made a plan of action and followed it." 8. Positive reappraisal (7 items)
"Changed or grew as a person in a good way."
The intercorrelations among these scales are rather low, confirming their desired distinctiveness. Internal consistencies are not always satisfactory, and test-retest reliabilities are not reported. According to the theory, a high stability is not desired because individuals are expected to adjust their actual coping responses to the requirements of each specific situation. A problem with this measure, as well as with all measures that are based on many factors, is that theoretical cross-linked relationships between scales are not considered. For example, mobilizing social support is seen here as a distinct strategy, but actually it can serve a number of purposes, such as solving a problem, obtaining information, calming down, or distracting oneself. Carver, Scheier, and Weintraub (1989) have accounted for this problem by establishing separate social support factors for problem-focus and emotion-focus. Parker and Endler (1992) argue that social support should not be conceived of as a distinct coping dimension but as social resources that may be available for a number of different coping strategies. A different view is to divide all coping strategies into social and nonsocial acts, by this doubling the amount of coping options (Klauer, Filipp, & Ferring, 1989).
Another difficulty with a high number of extracted factors is that they do not appear to be all of the same weight or of the same theoretical level. Some may be closer to a higher-order factor or to a general factor, accounting for a larger amount of variance, whereas others may be rather peripheral. It remains undetermined in what way the eight factors are embedded into the initial dimension of problem-focused and emotion-focused functions. There seems to be no empirical evidence for testing such a hierarchy with confirmatory factor analysis.
The authors of the questionnaire encourage researchers to adjust the WCQ to the specific study context in order to achieve a close match between the stress experience and the coping statements. This has been done, for example, by Dunkel-Schetter, Feinstein, Taylor, and Falke (1992), who have developed a version for cancer patients. It consists of 49 WCQ items, some of them slightly reworded, that were given to 603 cancer patients. A factor analysis yielded five factors: seek and use social support (11 items), cognitive escape-avoidance (9 items), distancing (12 items), focus on the positive (8 items), and behavioral escape-avoidance (9 items). This version has also been used by Stanton and Snider (1993).
In contrast to the majority of coping inventories that aim at rather dispositional coping styles, McCrae (1984) has favored a situation-specific approach. He argued that the nature of the stressor would fundamentally determine how people cope with it. He asked 255 persons to name a recent life event, and this was later classified as being either a challenge (75 events), a threat (114 events), or a loss (66 events), according to the stress appraisals suggested by Lazarus (1966). Afterwards, participants responded to a broad range of 118 coping items, 68 taken from the WCQ and 50 added by the author. By factor analysis, scales were constructed in a "rational" way, i.e., the 34-factor solution was used only to guide the assignment of items to 28 scales. Of these scales, many consisted of only one or two items, and part of the internal consistencies were unsatisfactory. In a second study, 151 persons were asked to recall either a challenge, threat, or loss that was experienced within the last six months, that is, they had to elicit the cognitive appraisal themselves. The author concludes that there was no difference between assessing the nature of the stressor and the explicit cognitive appraisal. In both cases, subsets of the instrument received situation-specific responses.
This approach adds an important aspect to the measurement of coping. The nature of the stressful encounter can be considered a key element in actual coping responses. According to Lazarus' (1991) stress theory, its cognitive appraisal should be the critical antecedent of coping but, unfortunately, this relationship is not well researched. McCrae (1984) is one of the few who followed up this line of thinking and deserves much credit for it. The resulting instrument, however, hardly seems to be acceptable due to its methodological limitations (see also McCrae & Costa, 1986).
Commonly, scale development starts out in a rational way and continues in an empirical way. An example of this approach is the Coping Strategy Indicator (CSI) presented by Amirkhan (1990). The author collected 161 coping behaviors from existing scales and from previous own research, and then reduced this accumulation step by step in a series of factor analyses with independent large samples. The participants were solicited at public sites, such as supermarkets and health clinics. The first sample of 357 respondents were asked to recall a critical event that was experienced within the last six months ("a problem you think was important and caused you to worry"). Then they responded to the 161 coping items on a three-point scale. By factor analysis, 3 out of 17 dimensions were associated with eigenvalues above unity, and manifested the basis for all further inquiry. The three subscales of the CSI were implanted at this early stage, although they only accounted for 21% of the variance: problem solving, seeking support, and avoidance. At the next stage, the 63 best factor indicators were given to 520 respondents. Another factor analysis yielded the same factors as before, backed up by confirmatory factor analysis. This procedure was repeated at Stage 3 with 954 respondents, who received the 36 best factor indicators. The final version consisted of 33 items, 11 for each subscale. Problem solving is denoted by items such as "tried to solve the problem"; seeking support is marked by items such as "confided your fears and worries to a friend or relative"; and avoidance is indicated by items such as "daydreamed about better times". In another sample of 92 persons, good psychometric properties of the instrument were determined.
It is desirable that different diagnostic instruments arrive at similar major dimensions of coping behavior. In the present case, this may be partly due to the fact that preexisting measures were taken to provide the raw material for the new one. It is also desirable that large independent samples are employed to identify those dimensions, and there is no better statistical procedure than confirmatory factor analysis. In spite of this, the present results are not convincing because all factor solutions were characterized by poor goodness-of-fit indices, and, in addition, only 21%, 33%, or 37% of variance were accounted for at three stages of the instrument development. It would have been necessary to conduct more fine-grained analyses with hierarchical structures that would have resulted in lower-level subscales. In the validation section, it was surprising to find the "seeking social support" scale of Folkman and Lazarus (1988) to be more closely associated with Amirkhan's problem solving scale (r = .55) than with his seeking support scale (r = .46). The enormous effort to construct this instrument is admirable, but the product is hardly convincing, neither in theoretical nor in empirical terms.
Children may experience stress in a different manner than adults. Research with age-specific stress and coping inventories, therefore, is appropriate. Dise-Lewis (1988) has presented an instrument that has been generated and validated within samples of junior high school students in the age range from 11 to 14 years. Initially, 104 students were interviewed with open-ended questions to explore their experience of stress ("events that produce stress") and their ways to handle it (what they "can do to cope with or reduce stress"). The item pool was refined and enriched with the help of 90 students, resulting in a checklist of 125 life events and 42 (later 49) coping strategies. Normative data were then gathered in a sample of 502 students who checked those events they had experienced within the recent year, and who rated their stressfulness and the appropriate coping strategies. Retest reliabilities were determined for a subsample of 85 students, and concurrent validity data for a subsample of 198 students.
A factor analysis with oblique rotation of the 49 coping strategies resulted in a five-factor solution that accounted for half of the variance. The first factor was labelled "aggression" and included seven items, such as "hit someone or hurt someone physically." The second factor "stress recognition" with 13 items contained behaviors such as "to cry." The third one was coined "distraction" with 11 items such as "do a hobby or something I enjoy." Fourth, eight items, for example cigarette smoking, was called "self-destruction". Finally, nine items ("just hold in") referred to what was called "endurance". Distraction and endurance were the most frequently selected coping strategies by children.
Because of the stepwise instrument development with large and independent samples, this inventory includes an abundant and well-established source of coping strategies for a specific age group. However, the construction was mainly empirical, and it is not well balanced with rational procedures. The final five factors do not conform well with theory and research by others. The factor analysis is not cross-validated and will probably fail to show up again in different samples, which is a common experience in this field of study. The coping strategies are applied to many different stressful encounters. Also, we do not know whether there are stressor-specific coping preferences in children.
Coping with life stress may be different at different stages of life. In adolescence, girls and boys face challenging developmental tasks, such as fitting into a peer group, differentiating from the family, and advancing identity formation. In order to identify the major coping strategies and behaviors in dealing with developmental stress in adolescence, Patterson and McCubbin (1987) have designed an inventory that covers a variety of such behaviors. The instrument was generated in an inductive way. First, a sample of 30 high school students in grades 10 to 12 were interviewed. They described what they did to manage the hardships and to relieve the discomforts of (a) personally experienced stress, (b) stress experienced by other family members, and (c) critical life changes in general. The responses resulted in a 95-item inventory that was administered to a different sample of 467 students who responded to these coping behaviors on a five-point frequency scale. Those 27 items that yielded almost no variance were excluded. The remaining 68 items were factor analyzed repeatedly, resulting in a final set of 54 coping behavior items that loaded on 12 factors (labelled "coping patterns" by the authors). Most of the reliabilities (internal consistencies) of these 12 subscales are in the 70s. The factors are:
1. Ventilating feelings:
Six coping behaviors that focus on the expression of tensions and frustrations (swearing, letting off steam, yelling at people, etc.) 2. Seeking diversions:
Eight coping behaviors that reflect how to keep oneself busy or escape from sources of tension (sleeping, watching a movie, playing video games or using drugs). 3. Developing self-reliance and optimism:
Six coping behaviors that include efforts to be more organized and in charge of the situation (trying to make one's own decision, figuring out how to deal with the problem, trying to think of the good things in life, or working harder at a job). 4. Developing social support:
Six coping behaviors focused on staying connected with other people through expression of affect or mutual problem solving (apologizing or talking to a friend about one's feelings). 5. Solving family problems:
Six coping behaviors directed at working out difficult issues with family members) talking to one's father, trying to reason with parents, going along with parents' requests). 6. Avoiding problems:
Five coping behaviors that reflect how adolescents use substances as a way to escape or to avoid persons difficult to deal with. 7. Seeking spiritual support:
Three coping behaviors, namely going to church, praying, and talking to a minister. 8. Investing in close friends:
Only two coping behaviors: being close with someone you care about, and being with a boyfriend or girlfriend. 9. Seeking professional support:
Only two coping behaviors: getting professional counseling and talking to a teacher. 10. Engaging in demanding activity:
Four coping behaviors that involve challenges to excel at something or achieve a goal
(working hard on school projects, gettaing body in shape, etc.) 11. Being humorous:
Only two coping behaviors: keeping a sense of humour and trying to be funny. 12. Relaxing:
Four coping behaviors directed at ways to alleviate tensions (eating, daydreaming, riding around in a car, and so on).
In a further sample of 709 adolescents, this inventory was validated within a longitudinal study on health-risk behaviors. The frequency rank order of these coping patterns showed that relaxing, developing social support, investing in close friends, and developing self-reliance were used most often, whereas avoiding problems and seeking professional support were reported least often. Girls in particular preferred to turn to their social bonds. Substance use was more pronounced in those who were inclined to ventilate their feelings and who invested in close friends, whereas less substance use was reported by those who tried to solve family problems.
This inventory is a good example for the measurement of coping with general life stress at a specific developmental stage. It seems to be appropriate for research on adolescent stress and health risk behaviors. Its main weakness lies in its inductive development. It has not been influenced by major theories of stress and coping, but relies basically on the responses of the 30 adolescents who assembled the first item pool. It is likely that other researchers using the initial set of 95 coping behaviors with other samples would come up with different factors, in particular when using theory-guided confirmatory factor analysis. Therefore, this inventory is not well compatible with the leading coping instruments. Instead, it represents a unique and stimulating approach but may be restricted to the original study context.
Based on principal coping inventories, Feifel and Strack (1989) developed their Life Situations Inventory (LSI) to assess three forms of coping: problem-solving, avoidance, and resignation. Similar factors, that had been identified by various studies, conform with leading theories. The approach aimed at coping with real-life circumstances in middle-aged and elderly men.
The authors began with a collection of more than 70 coping behaviors found in other instruments. By a rational selection procedure they eliminated those items that appeared to be redundant or did not fit the three theoretical coping forms well enough. They ended up with 28 items which they administered to 182 men. The stress situations were not presented in the questionnaire, but were individually generated by the participants. However, five conflict areas were given as a guideline for the recall of personal problems: (a) decision making, (b) defeat in competition, (c) a frustration-producing situation, (d) difficulty with an authority figure, and (e) general disagreement with a peer. Subjects were asked to write down a significant conflict they had experienced recently within each area. Then they responded to how they had coped with it by answering the 28 LSI items. Thus, these 28 items were repeated five times, once for each stress situation. The instruction for the decision-making problem, for example, was: "Take a few minutes and think about a situation in which you had trouble making a decision. That is, you had to make a choice, and you were not sure what to do. It might have been a decision involving just yourself, or involving your job, family, or friends. Be sure that the decision ... is one that was important to you" (Feifel & Strack, 1989, p. 28). Participants came up with decision problems, such as finding a new place to live or investing in the stock market. Each of the 28 LSI questions was presented with a four-point response format indicating the degree of endorsement. Examples are "How much did you try to settle things as soon as possible?" (problem solving), "How much did you try not to think about the situation?" (avoidance), and "How much did you feel that the final decision was beyond your control?" (resignation).
The 5-situations-by-28-items data set was examined by using item analysis, resulting in three scales after eliminating seven items. The problem-solving scale consisted of eight items, the avoidance scale also of eight items, and the resignation scale of five items. The pooled within-situations correlations between the scales was r = .01 for problem-solving with avoidance, r = .22 for problem-solving with resignation, and r = .51 for avoidance with resignation.
The development of this inventory represents a well-balanced approach where rational and empirical steps were taken and where scales emerged that are in line with current thinking. On the other hand, these scales are not very innovative. A unique feature lies in the self-generation of the five stress situations. This distinguishes the instrument from others which offer hypothetical scenarios that may be alien to the respondents. The disadvantage of real-life situations is, however, that comparisons between persons are not accurate since individual differences of stress experiences in the first place may result in subsequent coping differences. Not personal coping style is responsible then for choosing one strategy or the other, but the match between the situation and the behavior. Another problem is that within each conflict area different cognitive appraisals can be made. Loss appraisals across different areas might correspond with each other more closely than, for example, threat and loss appraisals within one area. Instead of content, the cognitive appraisal might be a better categorization of stress situations. Further, the specific stage in which a conflict is examined appears to be important: Whether one recalls the anticipation stage, the confrontation stage, or the recovery stage does make a difference for appraisal and coping.
The weakness of most of the measures lies in their unsatisfactory psychometric properties, unstable factor structures, and lack of cross-validation. These shortcomings have been overcome with the Coping Inventory for Stressful Situations (CISS), that was developed in an accurate and rigorous way (Endler & Parker, 1990a, 1990b), and in a balanced approach, including rational and empirical steps. Coping behaviors were compiled that fitted the two generally accepted coping functions problem solving and emotion regulation. A 70-item pool based on existing instruments and on additional collections was presented to 559 undergraduates who responded on five-point frequency scales. A general instruction was used, aiming at a "typical" coping response ("...how much you engage in these types of activities when you encounter a difficult, stressful, or upsetting situation"). Factor analyses, performed for men and women separately, yielded three factors: task-oriented, emotion-oriented, and avoidance-oriented coping. A revised 66-item version was given to 394 college students and to 284 adults. For each subsample, divided by sex, factor analyses were conducted that replicated the three factors, producing the final 48-item inventory with 16 items per scale. The avoidance dimension could be further subdivided into a Distraction scale and a Social Diversion scale. Very good psychometric properties were identified in several validation samples. Construct validity was documented by appropriate correlations with the Ways of Coping Questionnaire (Folkman & Lazarus, 1988) and various personality traits.
Recent studies have used the CISS to predict state anxiety before an academic exam (Endler, Kantor, & Parker, in press; Zeidner, in press). As a trait measure, its predictive value for a situation-specific emotion obviously has to be very limited. Rather, coping dispositions play a role as distal antecedents in more complex path models that also include situation-specific coping responses.
Technically, the CISS is a state-of-the-art inventory based on stable factors that were replicated across various samples and that met the congruency test criteria. It represents a fine model of classical test construction. However, it is not better than others for resolving today's critical issues in coping assessment. It is limited to three factors, although other researchers have shown that this small range does not reflect the complexity and heterogeneity of actual coping. Further elaboration is necessary to identify subgroups of coping strategies that remain stable across individuals and situations. The instrument is disposition-oriented and, therefore, covers only one facet of coping. A more situation-oriented CISS version is required; this is what is currently being developed (Endler et al., in press).
The distinction between problem-focused coping and emotion-focused coping is widely acknowledged, but many authors, including Carver et al. (1989), felt that this was too simple. They believe that both coping functions have to be subdivided because there are a variety of distinct ways to solve problems or to regulate emotions. The COPE inventory has gone through several generations in its development, and a number of theoretically important factors were identified. By using rational construction procedures, diverse scales were composed, which were then tested in a sample of 978 undergraduates. The students were asked what they usually do when they are under considerable stress. A factor analysis produced almost the same structure as was hypothesized. (It is not reported how many items were initially used, and where the 13 scales came from after only 11 meaningful factors had been extracted). The final version of the COPE inventory (Carver et al., 1989) contains 13 scales with four items each (the authors now distribute a version with two additional experimental scales).
Following is a list of the scale labels together with one example each. The first five
categories were established as subdimensions of problem-focused coping, and the next five
as subdimensions of emotion-focused coping. 1. Active coping
"I do what has to be done, one step at a time." 2. Planning
"I make a plan of action". 3. Suppression of competing activities
"I put aside other activities in order to concentrate on this." 4. Restraint coping
"I force myself to wait for the right time to do something." 5. Seeking social support for instrumental reasons
"I talk to someone to find out more about the situation." 6. Seeking social support for emotional reasons
"I talk to someone about how I feel." 7. Positive reinterpretation and growth
"I learn something from the experience." 8. Acceptance
"I learn to live with it." 9. Turning to religion
"I put my trust in God." 10. Focus on and venting of emotions
"I let my feelings out." 11. Denial
"I refuse to believe that it has happened." 12. Behavioral disengagement
"I just give up trying to reach my goal." 13. Mental disengagement
"I daydream about things other than this."
In addition to these 52 items, one coping behavior was kept as a single item that pertained to alcohol or drug use. After the analysis, this was extended to another four-item scale, along with one four-item scale concerning joking about the stressor (Carver et al., 1989, p. 280). Most of the above scales were found to have satisfactory psychometric properties, and evidence for validity is provided (cf. Carver & Scheier, 1993; Carver, Scheier, & Pozo, 1992).
The authors argue that this is a theoretically based approach, and certainly, it is more rational than many others. But the use of exploratory factor analysis discounts this strategy because it is hardly suitable to test a theory, in particular when used in a default manner with the eigenvalue criterion. Instead, a hierarchical confirmatory factor analysis would have been appropriate. Theoretically, five of the factors were established as subdimensions of problem solving, and five more as subdimensions of emotional coping. This makes good sense, but requires a test of the two levels. A second-order factor analysis did not replicate this hypothesized structure (Carver et al., 1989, p. 274; see also Zeidner & Hammer, 1992).
The COPE scale is conceived of as a more fine-grained dispositional measure of individual differences in coping than previous instruments, and it reflects a balanced view about the disposition versus situation issue. In a separate study, a situational version of the instrument was administered to 128 students together with the dispositional version (Carver et al., 1989). The situational version was rephrased to denote a recent real-life experience instead of a generalized response to all kinds of stress. Subjects were asked to think about their most stressful event of the past two months. Low to moderate correlations between the two versions were found, which was expected and which underscores the role of situation variability. However, in order to decide whether a "trait version" or a "state version" of the inventory is most suitable it is necessary to examine the predictive power of both versions in a series of studies with different stressful encounters. The same remains to be answered for the possible influence of personality dispositions on adaptation. For example, monitoring and blunting were almost unrelated to dispositional coping, but optimism was closely linked to it. It may be that the direct or indirect influence of such traits accounts for more outcome variance than situation-specific actual coping. The COPE instrument is a good tool to continue research on these issues.
Inspired by the transactional stress theory (Lazarus, 1966, 1991), Stone and Neale (1984) attempted to develop an instrument to assess daily coping for use in longitudinal studies. In a pilot study, subjects were asked how to handle a recent problem and were asked to respond to 87 coping items. This was reduced to a 55-item checklist in the subsequent study. Eight categories were established that were labelled (a) distraction, (b) situation redefinition, (c) direct action, (d) catharsis, (e) acceptance, (f) social support, (g) relaxation, and (h) religion. Since the psychometric properties repeatedly turned out to be unsatisfactory, the authors gave up their intention to construct psychometric rating scales with multiple items, and decided to apply the eight categories directly with an open-ended response format. It was filled out by 120 married individuals for 21 consecutive days. Participants checked the appropriate categories and wrote down descriptions of their coping behaviors where applicable. The authors claim content validity for this measure and argue that this approach has advantages over traditional ones. In particular, they question the usefulness of internal consistency in coping measurement, of retrospective assessment, and of representing coping processes that change over time by one static value for a specific coping strategy.
The fact that Stone and Neale (1984) failed to develop reliable and valid psychometric coping scales and had to resort to a written structured interview can be considered as a blessing today because their article has become influential and has sparked an ongoing debate about the merits of situation-oriented coping assessment (Folkman, 1992; Stone & Kennedy-Moore, 1992). It is argued, for example, that inter-item covariation does not make sense when one strategy has been used at the expense of others within the same subcategory, e.g., after "did what I had to do" has been selected successfully, it is no longer necessary to "think about different solutions to the problem." Within a traditional psychometric scale designed to tap problem solving it would be expected that these items are highly correlated. In an actual coping situation, however, they may be unrelated because the first one is sufficient and the second may then be neglected. Another issue pertains to the duration of the stressful encounter and the opportunity for coping acts. If subjects recall an event experienced within the last year, those who report an ongoing long term event could come up with more coping strategies than those who report a very brief event or one that had just started. In addition, recall and labelling of coping efforts could be influenced by its outcome. If "thinking about the problem" is followed by failure it may be labelled "rumination", if followed by success it could be interpreted as "planning" or "instrumental problem solving" (Stone & Kennedy-Moore, 1992). It has also been found that the inapplicability of part of the items within a specific stress context biases the scores (Ben-Porath, Waller, & Butcher, 1991; Stone, Greenberg, Kennedy-Moore, & Newman, 1991). For example, the item "learned to know myself better" may be applicable when harm or loss is experienced or when patients suffer from a fatal disease, as opposed to the item "made a person change her mind" that appears to be more applicable in demanding interpersonal relationships. When the entire item pool of a coping scale is presented, part of the items turn out to be inapplicable to the situation, and are not only useless, but may even bias the results. These are merely a few critical points put forward by those favoring a situation-oriented approach. Due to space limitations, this highly important issue cannot be discussed here in more detail.
Inspired by the transactional stress theory of Lazarus (1991), Swiss researchers have designed a unique rationale for the assessment of stress and coping (Reicherts & Perrez, 1991; for an English version see Perrez & Reicherts, 1992). Its aim is to measure coping within the time course of various hypothetical stressful encounters as it relates to cognitive appraisals, emotions, and attributions. The instrument consists of 18 such episodes, tapping two classes of stress: (a) aversive or ambiguous situations, and (b) loss or failure situations. Each episode is subdivided into three stages as the encounter unfolds. An example for failure is the following episode:
At Stage 1, the instruction is phrased "Try to clearly imagine the following situation: Previous relations with your boss have been quite complicated. Now your boss gives you a task which you are supposed to work on for the next two days. This job is very inconvenient for you because you have a lot of routine work to do at the moment." Next comes a two-page questionnaire that requires answers about emotions, judgments, intentions, and actions.
At Stage 2, the instruction is phrased: "Imagine the situation carries on in the following way: Your boss tells you that your routine work also has to be done. As you begin to work on the new task, it becomes evident how difficult and time consuming it really is. It seems that you will only finish it if you ignore your other work, and even then you may have to do overtime on it." Again, a two-page questionnaire has to be completed.
At Stage 3, the instruction says: "Imagine the situation finally turns out as follows: You didn't carry out the job in the assigned time. Also, a lot of routine work remained unfinished." Now, the third two-page questionnaire asks for feelings, attributions, and intentions.
An important aspect of this instrument lies in its distinction between coping intentions and coping strategies. It is assumed that individuals act according to their intentions or goals. This is more specific than merely stating that there are different coping functions, as Lazarus and Folkman (1984) do. This point is also made by Leventhal, Suls, and Leventhal (1993) and by Laux and Weber (1993). Four intentions are provided, each of them with four-point rating scales, (a) ... to actively confront the other and to clear up what is at stake, (b) ... to maintain a friendly atmosphere and to prevent an argument with the other, (c) ... to remain calm and composed, and (d) to keep my self-esteem. In other words, the authors believe that there are four basic intentions associated with coping. One focuses on the person-environment adaptation, subdivided into mastery-orientation (assimilative) and meaning-orientation (accomodative). The third is to keep one's emotional balance, and the fourth is to maintain self-esteem.
In terms of coping behaviors, a distinction is made between self-directed and environment-directed strategies, the latter again subdivided with respect to either aversive/ambiguous or to loss/failure situations. Self-directed strategies are search for information, suppression of information, reevaluation, palliation, self-blame, and other-blame. Environment-directed behaviors are instrumental actions, avoidance, hesitation, prevention, reorientation, etc.
Some validation studies with small samples have been conducted yielding satisfactory psychometric properties, in particular high retest reliabilities. The newness of the inventory and its presentation in the German language may have impeded a large-scale evaluation up to now. In addition, this inventory is designed in a very complicated manner requiring not only an effort for the interviewee, but also for the researcher in analyzing and interpretating the data. No other measure is as closely composed in line with transactional stress theories, at the expense that it is not as parsimonious as other measures. A great advantage is the consideration of coping intentions and of different stages. However, the latter is of limited value since this remains a scenario approach with hypothetical situations that change in a hypothetical way. If the aim is to tap transactional processes it might be better to conduct research in real-life situations. Nevertheless, this is a unique diagnostic approach that challenges other assessment tools.
The present survey of instruments is not exhaustive. It is arbitrarily limited to some examples that were published within the last 10 years and that exposed certain features of interest. Our portrayal has to be brief and cannot pay attention to all aspects. As a result of this overview, we will delineate a number of critical issues that have not been resolved sufficiently by the available operationalization.
The procedure to generate a coping inventory can be more deductive (i.e., based on theoretical assumptions), or more inductive (i.e., starting with observations). Both directions are needed but, unfortunately, a firm balance between them is rarely found (Burisch, 1984; Gough & Bradly, 1992). In addition, the conceptual and technical quality of these steps is not always satisfactory, a point well made by Parker and Endler (1992). Some authors collect items from preexisting questionnaires fitting to some theoretical distinctions, add a few items of their own, compute a factor analysis, and construct scales that match their judgment rather than the empirical factor solution. Others do not collect their items in light of theory, but they compile unsystematically a data base that may have a range too narrow or too broad or that is otherwise biased; they then run sophisticated statistical analyses and value the resulting coefficients more than may be justified. In particular, some authors tend to pay lip service to a stress and coping theory that focuses on context specifity, reciprocal processes and dynamics of the unfolding encounter, but what they actually come up with is a kind of coping style inventory that assesses a stable personality trait instead. A much clearer distinction between personal coping resources, coping behaviors, and coping outcomes is often needed.
Assessing either dispositional or episodic coping has indeed been the main issue in recent years (for example, see Cohen, 1987; Folkman, 1992; Stone & Kennedy-Moore, 1992). Do individuals make rigid or flexible use of habitual coping strategies across situations and across time points? Or is each coping episode highly idiosyncratic and chiefly determined by the actual situation? The question can be reframed by asking whether we prefer to study interindividual or intraindividual differences. Another perspective would be to focus either on trait coping or on state coping. These alternatives have implications for the development of measures and for the psychometric properties that we can ask for. Dispositional measures should be constructed according to high psychometric standards, but this is a trade-off because such inventories cannot account for the array of coping responses on a microanalytic level when dealing with the complexity of a stressful encounter (Folkman, 1992). "Psychometric rigor may prove to be a strait-jacket which can confine research and ignore the study of process" (Leventhal et al., 1993, p. 95). High internal consistency, for example, can be exptected when all coping responses within the same category (such as avoidance) are likely to be used, not when only one of them is preferred which may render the others redundant (Stone & Kennedy-Moore, 1992). High test-retest reliability, for example, is a contradiction to the variability inherent in continuous coping within a stress episode. However, it would be somewhat more applicable when comparing individuals who cope with the same kind of stress at two distinct points in time. Thus, coefficients should not be evaluated regardless of the measurement context. If we could trade in reliability for validity, we would gain something valuable, but this can only be accepted if there is empirical proof of it.
However, it could be that we are not interested in an accurate description of the coping process. Science has to aim more towards prediction and explanation than towards mere description. If adaptation outcomes are better predicted by dispositional coping strategies than by situation-oriented measures, then the first ones deserve more scientific attention.
The nature of a stressful encounter and its cognitive appraisal influence the way people cope with it. Imagining a possible future situation may provoke similar reactions to recalling a past situation experienced as being stressful. Asking all respondents to imagine the same fictitious stress situation has the obvious research advantage of improved internal validity. In contrast, people who recall a prior real life situation experience something that has a unique personal significance for them but maybe not for others, which complicates comparability between situations. This can be somewhat alleviated by classifying the recalled events into distinct cognitive appraisals (such as challenge, threat, or loss) or to elicit certain types of events in the first place (McCrae, 1984). But this has been rarely done. Real-life events are often taken as "generic" stress, and then it becomes impossible to relate coping to its context. If the research goal is to identify dispositional coping strategies, one may resort to hypothetical scenarios as prototypical stress situations, but if the aim is to describe actual coping, the situation parameters have to be carefully studied. Sometimes life stress is defined by the selection of the sample, for example when cancer patients, refugees, or rape victims are studied. The issue here is whether population-specific/domain-specific inventories should be developed or whether general coping inventories should be preferred in order to allow for group comparisons. Again, the need of a yardstick is a matter of the research question.
When context-specific research is being conducted, further distinctions have to be made in terms of episode, domain, and stage. An episode is a stressful encounter that has a beginning and an ending (such as an academic exam or a dentist appointment). A domain refers to the area of human functioning where the problem arises (career, health, interpersonal relationships, etc.). Stages subdivide an episode and are in particular important because individuals do not cope in the same manner all the time. It may be that a patient undergoing surgery is vigilant during the preparation stage, avoiding during the confrontation stage, and seeking support during the recovery stage. If the researcher aims at an accurate description of the actual coping process, it is indispensable to identify the domain, the episode, and the stage, and to assess coping at multiple points in time. Retrospective reports of coping may pertain to any stage, but most likely they will be biased towards the most recent one.
Some reports come up with two or three, others with 8, 13, or even 28 coping scales, no matter whether they were developed in a more rational or in a more empirical way. Dispositional instruments tend to include less, situation-oriented measures tend to have more dimensions. There is agreement about some major factors, such as either problem focus and emotion focus, or vigilance and avoidance, or a combination of both sets. Obviously, these are conceptually at a high level of abstraction, whereas others are more proximal to the coping responses. Several authors, among them Krohne (1993) and Leventhal et al. (1993), have suggested to establish hierarchies to account for this evidence. With this purpose in mind, some researchers have conducted second-order factor analyses, but the solutions have not been found stable across samples. Carver et al. (1989), for example, extracted for their COPE scales four second-order factors; Zeidner and Hammer (1992) found only two such factors based on a modified version of the COPE inventory. Another example is the Ways of Coping Questionnaire, which consists of eight strategies originally expected to be related to problem-focused and emotion-focused coping. However, this has not been demonstrated empirically. One likely reason for the common failure to link two levels of coping strategies to each other lies in the technique of exploratory factor analysis that is unsuitable for such a design. Instead, item sets should be constrained to fit into a multidimension-multistrategy model to be tested by confirmatory factor analysis.
Most inventories include a scale to measure the seeking or mobilization of social support, but it has been found that this is not a stable, replicable factor. This has to do with the phenomenon that social support is a multidimensional construct in itself (Schwarzer & Leppin, 1991). Instrumental support assists in problem solving, emotional support may comfort the person, and so on. It is therefore not surprising that Carver et al. (1989) have found two support factors. The more general question is whether this construct has been conceptualized at the appropriate level within a system of coping concepts. Parker and Endler (1992) argue that social resources can be used as part of many strategies, and they suggest to drop it from the strategy list and add it to the resource list (see also Hobfoll, 1988; Stephens, Crowther, Hobfoll, & Tennenbaum, 1990). Klauer et al. (1989) have divided all coping attempts into social and nonsocial ones (see also Filipp, Klauer, & Ferring, 1993). Directing one's coping efforts towards a social network represents one medium of action that is distinct from other media of action.
A related issue lies in the meaning of coping dimensions. Problem solving and emotion regulation, for example, are understood as coping functions (Folkman, 1992; Lazarus, 1991), vigilance and cognitive avoidance are interpreted as attentional styles (Krohne, 1993, and this volume; Miller et al., 1993). However, it has hardly been recognized that one specific coping act can serve different functions. By seeking information, for example, one can calm down and reduce threat, but also prepare subsequent action. Furthermore, individuals pursue a variety of goals that may differ from time to time and from situation to situation. Identifying personal coping intentions, therefore, is a prerequisite for a judgment about certain coping efforts (Laux & Weber, 1993; Perrez & Reicherts, 1992). In establishing a hierarchy of coping acts, one should top it with "generalized coping intentions" (Leventhal et al., 1993), followed by strategies, acts, and even more specific behaviors. The assessment of coping should then take place as a multilevel process while the stressful encounter unfolds. Closest to this suggestion come Perrez and Reicherts (1992), who have developed a computer-assisted data collection procedure (called COMES) that allows to input self-reported coping information during stressful interactions. The trade-off is that such data bases are not only complicated to obtain, but also difficult to score, analyze, and interpret. In light of this disadvantage, traditional psychometric scales will continue to serve a purpose and will remain attractive in future research.
If one acknowledges the need to combine multidimensionality with hierarchy, one remains open to consider all kinds of coping measures within one single conceptual framework. The guideline for future research would be to establish a multilevel assessment rationale according to the researcher's specific coping theory. Rather than lumping together heterogeneous measures in an eclectic way, it is superior to relate, for example, dispositional coping to situational coping assessment, to describe actual coping with one part of the inventory, and to identify stable coping preferences with the other. Hypothetical scenarios may be used to identify one's personal inclination for coping, but nothing can replace real-life events for ecological validity of the assessment procedure. We need more attention to the passage of time and to the accumulation of coping experience as prerequisites for actual coping responses. A hierarchy of coping instruments, therefore, must include some parts that are administered at multiple points in time, which in turn poses demands on the parsimony of the tools. It is of less importance to what degree rational or empirical construction techniques are being used, compared to the necessity that both perspectives have to operate reciprocally in a spiral-shaped process. Coping research has to escape from narrowminded empiricism. It can only make a significant leap forward when multilevel instruments that match the complexity of coping are embedded in a multiwave design, analyzed by causal modelling approaches.
Amirkhan, J. H. (1990). A factor analytically derived measure of coping: The Coping Strategy Indicator. Journal of Personality and Social Psychology, 59, 1066-1074.
Ben-Porath, Y. S., Waller, N. G., & Butcher, J. N. (1991). Assessment of coping: An empirical illustration of the problem of inapplicable items. Journal of Personality Assessment, 57, 162-176.
Billings, A. G., & Moos, R. H. . (1981). The role of coping responses and social resources in attenuating the stress of life events. Journal of Behavioral Medicine, 4, 139-157.
Billings, A. G., & Moos, R. H. (1984). Coping, stress, and resources among adults with unipolar depression. Journal of Personality and Social Psychology, 46, 877-891.
Brandtstädter, J. (1992). Personal control over development: Implications of self-efficacy. In R. Schwarzer (Ed.), Self-efficacy: Thought control of action (pp. 127-145). Washington, DC: Hemisphere.
Burisch, M. (1984). Approaches to personality inventory construction. A comparison of merits. American Psychologist, 39(4), 214-227.
Carver, C. S., & Scheier, M. F. (1993). Vigilant and avoidant coping in two patient samples. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 295-320). Seattle: Hogrefe & Huber.
Carver, C. S., Scheier, M. F., & Pozo, C. (1992). Conceptualizing the process of coping with health problems. In H. S. Friedman (Ed.), Hostility, coping, and health (pp. 167-187). Washington, DC: APA.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56 (2), 267-283.
Cohen, F. (1987). Measurement of coping. In S. V. Kasl & C. L. Cooper (Eds.), Stress and health: Issues in research methodology (pp. 283-305). New York: Wiley.
Dise-Lewis, J. E. (1988). The Life Events and Coping Inventory: An assessment of stress in children. Psychosomatic Medicine, 50, 484-499.
Endler, N. S., Kantor, L., & Parker, J. D. A. (in press). State-trait coping, state-trait anxiety and academic performance. Personality and Individual Differences.
Endler, N. S., & Parker, J. D. A. (1990a). Coping Inventory for Stressful Situations (CISS): Manual. Toronto: Multi Health Systems.
Endler, N. S., & Parker, J. D. A. (1990b). Multidimensional assessment of coping: A critical evaluation. Journal of Personality & Social Psychology, 58, 844-854.
Feifel, H., & Strack, S. . (1989). Coping with conflict situations: Middle-aged and elderly men. Psychology and Aging, 4(1), 26-33.
Filipp, S.H., Klauer, T., & Ferring, D. (1993). Self-focused attention in the face of adversity and threat. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 267-295). Seattle, WA: Hogrefe & Huber.
Folkman, S. . (1992). Improving coping assessment: Reply to Stone and Kennedy-Moore. In H. S. Friedman (Ed.), Hostility, coping, and health (pp. 215-223). Washington, DC: APA.
Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-234.
Folkman, S., & Lazarus, R. S. (1988). Manual for the Ways of Coping Questionnaire. Palo Alto, CA: Consulting Psychologists Press.
Gough, H. G., & Bradly, P. (1992). Comparing two strategies for developing personality scales. In M. Zeidner & B. Most (Eds.), Psychological testing: An insider view (pp. 215-246). Palo Alto, CA: Consulting Psychologists Press.
Hobfoll, S. E. (1988). The ecology of stress. Washington, DC: Hemisphere.
Klauer, T., Filipp, S.H., & Ferring, D. (1989). Der "Fragebogen zur Erfassung von Formen der Krankheitsbewältigung" (FEKB): Skalenkonstruktion und erste Befunde zu Reliabilität, Validität und Stabilität [The "Inventory to assess forms of coping": Scale development and first results pertaining to reliability, validity, and stability]. Diagnostica, 35, 316-335.
Krohne, H. W. (1993). Vigilance and cognitive avoidance as concepts in coping research. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 19-50). Seattle: Hogrefe & Huber.
Krohne, H. W., Schumacher, A., & Egloff, B. (1992). Das Angstbewältigungs-Inventar (ABI) (The Mainz Coping Inventory (MCI)] (Mainzer Berichte zur Persönlichkeitsforschung). Mainz: Johannes-Gutenberg-Universität, Psychologisches Institut.
Laux, L., & Weber, H. (1993). Emotionsbewältigung und Selbstdarstellung [Coping with emotions, and self-presentation]. Stuttgart: Kohlhammer.
Lazarus, R. S. (1966). Psychological stress and the coping process. New York: McGraw-Hill.
Lazarus, R. S. (1991). Emotion and adaptation. London: Oxford University Press.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Leventhal, E. A., Suls, J., & Leventhal, H. (1993). Hierarchical analysis of coping: Evidence from life-span studies. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 71-100). Seattle: Hogrefe & Huber.
McCrae, R. R. (1984). Situational determinants of coping responses: Loss, threat, and challenge. Journal of Personality and Social Psychology, 46, 919-928.
McCrae, R. R., & Costa, R. T. (1986). Personality, coping, and coping effectiveness in an adult sample. Journal of Personality, 54, 385-405.
Miller, S. M. (1987). Monitoring and blunting: Validation of a questionnaire to assess styles of information seeking under threat. Journal of Personality and Social Psychology, 52 (2), 345-353.
Miller, S. M., Combs, C., & Kruus, L. (1993). Tuning in and tuning out: Confronting the effects of confrontation. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 51-70). Seattle: Hogrefe & Huber.
Parker, J. D. A., & Endler, N. S. (1992). Coping with coping assessment: A critical review. European Journal of Personality, 6 (5), 321-344.
Patterson, J. M., & McCubbin, H. I. (1987). Adolescent coping style and behaviors: conceptualization and measurement. Journal of Adolescence, 10, 163-186.
Perrez, M., & Reicherts, M. (1992). Stress, coping, and health. Seattle/Toronto: Hogrefe & Huber Publishers.
Reicherts, M., & Perrez, M. (1991). Fragebogen zum Umgang mit Belastungen im Verlauf (UBV) [Coping process questionnaire]. Bern: Huber.
Rothbaum, F., Weisz, J. R. & Snyder, S. (1982). Changing the world and changing the self: A two-process model of perceived control. Journal of Personality and Social Psychology, 42, 5-37.
Schwarzer, R., & Leppin, A. (1991). Social support and health: A theoretical and empirical overview. Journal of Social and Personal Relationships, 8, 99-127.
Stephens, M. A. P., Crowther, J. H., Hobfoll, S. E., & Tennenbaum, D. L. (1990). Stress and coping in later-life families. New York: Hemisphere.
Stone, A. A., Greenberg, M. A., Kennedy-Moore, E., & Newman, M. G. (1991). Self-report, situation-specific coping questionnaires: What are they measuring? Journal of Personality & Social Psychology, 61, 648-658.
Stone, A. A., & Kennedy-Moore, E. (1992). Assessing situational coping: Conceptual and methodological considerations. In H. S. Friedman (Ed.), Hostility, coping, and health (pp. 203-214). Washington, DC: APA.
Stone, A. A., & Neale, J. M. (1984). New measure of daily coping: Development and preliminary results. Journal of Personality and Social Psychology, 46, 892-906.
Suls, J., & Fletcher, B. (1985). The relative efficacy of avoidant and nonavoidant coping strategies: A meta-analysis. Health Psychology, 4, 249-288.
Taylor, S.E. (1983). Adjustment to threatening events. A theory of cognitive adaptation. American Psychologist, 38, 1163-1171.
Taylor, S. E. (1989). Positive illusions: Creative self-deception and the healthy mind. New York: Basic Books.
Zeidner, M. (in press). Personal and contextual determinants of coping and anxiety in an evaluative situation: A prospective study. Personality and Individual Differences.
Zeidner, M., & Hammer, A. L. (1992). Coping with missile attack: Resources, strategies, and outcomes. Journal of Personality, 60, 709-746.