Vous êtes sur la page 1sur 12

Conservation of Resources

A New Attempt at Conceptualizing Stress

Stevan E. Hobfoll

ABSTRACT: Major perspectives concerning stress are presented with the goal of clarifying the nature of what has proved to be a heuristic but vague construct. Current conceptualizations of stress are challenged as being too phenomenological and ambiguous, and consequently, not given to direct empirical testing. Indeed, it is argued that researchers have tended to avoid the problem of defining stress, choosing to study stress without reference to a clear framework. A new stress model called the model of con- servation of resources is presented as an alternative. This resource-oriented model is based on the supposition that people strive to retain, protect, and build resources and that what is threatening to them is the potential or actual loss of these valued resources. Implications of the model of conservation of resources for new research directions are discussed.

There are few areas of contemporary psychology that re- ceive more attention than stress (Hobfoll, 1986, 1988; Kaplan, 1983; Lazarus & Folkman, 1984; Milgram, 1986). This literature reflects researchers' beliefthat stress

is a major factor affecting people's lives, is intimately tied

with mental health, and is very possibly linked with many problems of physical health. The interest in stress has also caught the attention of the popular press, illustrating that stress is of concern to the lay public as well as the academic community ("Stress," 1983).

With all this great breadth of interest in stress, there has been a surprising paucity of work on related theory (Kaplan, 1983; Lazarus & Folkman, 1984). Although initial stages of any research direction may be strictly observational, this phase should give way to a more the- ory- or model-based stage that provides a web of insights and directions to guide research (Popper, 1959). Without

a clear theoretical backdrop, it is difficult to create a true body of knowledge because there are no defined borders of theory to be challenged (Cook & Campbell, 1979). Newton's theory was generally accepted for some 200 years because of such a lack of rival theories. As stated by Kurt Lewin, "There is nothing as practical as a good theory." In this article, I will outline some of the stress models that have guided research and thinking on stress histor- ically. This will lead to a discussion of current models that are relied on in framing the topic of stress for re- search. However, it will be argued that only a weak link exists between current stress models and the actual re-

March 1989 • American Psychologist

Copyright 1989by the American PsychologicalAssociation, Inc. 0003-066X/89/$00.75 VOl.44, No. 3. 513-524

Kent State University

search that is conducted. In addition, it will be argued that current models are tautological and so can do little to move stress investigators toward new horizons of re- search. Yet, because they are tautological, they can never be rejected either. Finally, a new stress model will be pre- sented called the model of conservation of resources. It is proposed that this model is more directly testable, comprehensive, and parsimonious than previous ap- proaches and that it provides a dearer direction for future research on stress and stress resistance.

The Cannon-Selye Tradition

The term stress is loosely borrowed from the field of physics. Humans, it is thought, are in some ways analo- gous to physical objects such as metals that resist mod- erate outside forces but that lose their resiliency at some point of greater pressure. The analogy to humans is ob- vious, albeit inexact. Walter Cannon (1932) was probably the first modern researcher to apply the concept of stress to humans in these kinds of terms. Cannon was principally concerned with the effects of cold, lack of oxygen, and other envi- ronmental stressors on organisms. He concluded that al- though initial or low level stressors could be withstood, prolonged or severe stressors lead to a breakdown of bio- logical systems. Cannon's emphasis on stress as response was carried on by Hans Selye (1950, 1951-1956). Selye depicted stress as an orchestrated defense operated by physiological sys- tems designed to protect the body from environmental challenge to bodily processes. He called this the General Adaptation Syndrome. Specifically, he felt that there was a common reaction to outside stressors following the se- quence of alerting response, resistance response, and ex- haustion. Selye has been criticized on two levels. First, the idea that the reactions of humans to stress is so uniform can be challenged by a wealth of data (Appley & Trum- bull, 1986; Lazarus & Folkman, 1984). How people re- spond to challenges from their environment can be seen as a function of their personality, constitution, percep- tions, and the context in which the stressor occurs (Mei- chenbaum, 1977; Moos, 1984; Sarason, 1975; Spielberger, 1972; Zuckerman, 1976). Second, Selye has been criti- cized for employing somewhat illogical deductive reason- ing. He depicted stress in terms of outcome, such that an organism could be seen as under stress only when a phase of the general adaptation sequence was occurring. This

513

viewpoint precludes the possibilityof prospectively iden-

tifying the cause of stress, because we are forced to wait until the outcome to know when stress will occur.

Stimulus Definitions of Stress

A less well-articulated view of stress may be identified in

terms of depicting stress from the nature of the stimulus, as opposed to the response. Elliot and Eisdorfer (1982), for instance, have focused on stressors, or that which is likely to cause stress, as the object of interest. They out-

fined four kinds of stressors: (a) acute, time-limited stressor& such as a visit to the dentist, a wasp entering the car while one is driving, or a woman's awaiting breast biopsy; (b) stressor sequences, such as divorce, bereave-

ment, or job loss; (c) chronic, intermittent stressors, such

as examinations for students, meetings with business as-

sociates one dislikes, or a regimen of visits to a physician for painful treatments; and (d) chronic stressors, such as debilitating illness, prolonged marital discord, or exposure

to occupation-related dangers.

Events in this case are considered stressful on the basis of whether they normally lead to stress reactions. That is, if the stimulus usually leads to emotional upset, psychological distress, or physical impairment or dete- rioration, then the stimulus is said to be a stressor. This thinking loosely follows from the important work of Ger- aid Caplan (1964) and Eric Lindemann (1944), who were

among the first to introduce a psychological view of stress,

as opposed to the physiological view advanced by Selye.

They also saw it as important to emphasize that psycho- logical distress was not necessarily the product of deep- seated personological disturbance, as psychodynamic theorists would have it, but rather that it could occur as the product of confrontation with especially stressful events. In particular, their work focused on normative life crises and extreme challenges to people's self or social world. The study of the process of bereavement is one commonly cited offshoot of their approach (Parkes, 1970, 1972), and the early work on stress as change that requires adaptation may also be seen as a product of this approach (Holmes & Rahe, 1967). A normative view of stressors as suggested by Elliott and Eisdorfer is a good starting point because it outlines those events that are likely to lead to stress responses. This limits the world of events that one would otherwise have to observe in each and every case in order to pro- spectively study the stress process. Perceptions may be important determinants of what is stressful, but these perceptions are far from idiographic, that is, there is broad agreement as to what is stressful (Dohrenwend, Krasnoff,

I wishto thank LeonardJason, BenjaminNewberry,AngelaBridges, and the anonymousreviewersfor their insightfulcommentson earlier

dralLsofthisarticle.Thanksare alsoextendedto KennethHeller,Rudolf Moos, and Irwin Sarasonfor their commentson the conceptualization

of someof the central ideasregardingthe stressmodelpresented.I, of

course,assumeall responsibilityforthe finalproduct. Correspondenceconcerningthisarticleshouldbe addressedto Ste- van E. Hobfoll,AppliedPsychologyCenter,KentStateUniversity,Kent, OH 44242.

514

Askenasy, & Dohrenwend, 1978; Holmes & Rabe, 1967). Indeed, if this were not the case, how would it be possible to construct finite fists of minor irritations and hassles (Kanner, Coyne, Schaefer, & Lazarus, 1981), the most perception-bound kinds of stressors? Without a broad sense of agreement about what is stressful in an objective sense there would be little agreement between people on which environmental occurrences were found to be stressful. Perhaps more important, if we do not begin with an environmental view of stressors, then the division be- tween stress responding and neurotic symptoms is lost (Dohrenwend, Dohrenwend, Dodson, & Shrout, 1984). In other words, if stress is only that which is in the eye of the beholder, then we return to a psychology of internal processes. If, as Dohrenwend et al. (1984) pointed out, stressors are indistinguishable from symptoms, then the environmental component is removed from stress re- search. It is not clear that this would push the study of stress beyond where the study of psychopathology has already taken us. By creating a taxonomy of stressful events we set an anchor point by which the differences in how people react can be compared and through which the nature of the events themselves can be further categorized (see Baum, Singer, & Baum, 1981). Kessler (1983) suggested that this may best be accomplished through longitudinal study of such events. Unfortunately, this lead has seldom been followed by stress researchers, as few studies have com- pared reactions to different categories of events. Even if this approach was adopted, however, it is clear that the stimulus is only one facet of the stress phenomena.

Event-Perception Viewpoints

Another perspective that was influential to stress re- searchers is a viewpoint that focuses both on the category of stressor events and the individual differences in the appraisal of those events. This approach should not be confused with "stimulus only" perspectives or appraisal perspectives because it emphasizes both the event and the individual's reaction to the event. The research of Spielberger (I 966, 1972) is illustra- tive of this viewpoint. Spielberger suggested that certain events are stressful if they are thought to be threats to the physical self or the phenomenological self. He called these physical threats and ego-threats, respectively. Although individuals with different personalities responded some- what uniformly to physical threats, people's responses to ego-threats were related to personality traits. In particular, Spielberger noted that people with high trait anxiety tended to react with state elevations in anxiety to ego- threat, whereas those who were low in trait anxiety tended to be comparatively impervious to ego-threats. In this way, it is neither the stimulus nor the appraisal that is important, but rather their particular interaction. Another line of research is also instructive in this regard. Specifically, research on test anxiety has led to one of the most complete understandings of responding to one kind of stressful stimulus. The work of I. Sarason

March 1989 American Psychologist

typifies this direction of investigation. Sarason (1972, 1975) has illustrated that examinations constitute a class of environmental event that are very commonly found to be stressful. However, he and other test anxiety re- searchers also have suggested that relative sensitivity to stress is a product of personality. This perspective is im- portant because it illustrates both that certain events are commonly viewed as stressful and that individuals differ in their degree of reactivityto normatively stressful events. It further indicates that such sensitivity is a fairly stable personality trait and that although related to sensitivity to other stressors, it may also exist independently of other sensitivities. So, for example, a test-anxious person might be resilient in the face of threatening interpersonal inter- actions. Although both Sarason and Spielberger highlighted the role of perception in their theorizing and research, their approach is more complex than this. Its complexity derives from a three-part emphasis on appraisal, known environmental threats, and personality traits. Broadly speaking, it integrates appraisal- and stimulus-based models, adding the vital third element of individual char- acteristics. This three-part approach represented a con- ceptual leap for stress researchers that has not often been followed. Instead, many investigators have returned to models that emphasize appraisal to such an extent as to all but ignore actual environmental occurrences or per- sonality traits, others have focused on personality to the exclusion of more idiographic perceptions, and still others have remained tied to stimulus-only models.

Homeostatic and Transactional Models of Stress

Probably the most commonly adopted model of stress employed by stress investigators today is a homeostatic model of stress presented in detail by McGrath (1970), but based in large part on the work of Lazarus (1966). McGrath defined stress as a "substantial imbalance be- tween environmental demand and the response capability of the focal organism" (p. 17). More recently, Lazarus and Folkman (1984) have defined stress as "a particular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being" (p. 19). Implied in these definitions is that stress is not the product of imbalance between objective demands and response capacity, but of the perception of these factors. Second, the consequences of failure to cope must be per- ceived as important to the individual. Finally, McGrath suggested that imbalance may take the course of under- load rather than overload, that is, too little demand vis- fi-vis coping capacity. Few researchers, however, have adopted or followed up on the underload notion, and it

has garnered littlesupport. Despite the widespread adoption of the balance model by stressresearchers(Cohen, Kamarck, & Mer- melstein, 1983;Gentry & Kobasa, 1984;Kendall, 1983; Wilcox & Vernberg, 1985), I would argue that it is tau-

March 1989 • American Psychologist

tological, overly complex, and not given to rejection. To be scientific, the case for rejecting or accepting a model must be clear. These points will now be reviewed in detail.

First, the balance model is tautological because it does not separately define demand or coping capacity, the two sides of the model. Demand is that which is offset by coping capacity. Yet, coping capacity is that which offsetsthreat or demand. Clearly, this reasoning is circular and evolves from the sole emphasis on perceptions. The balance perspective seems to lose in this way the basic measuring stone that stimulus-based models could pro- vide. So, for example, an understanding of how people who possess varying coping resources respond to a known threat, including the degree to which they find the stim- ulus threatening, could help anchor the perceptive com- ponent because the stimulus is a known entity. However, ifboth the demand and the resources mobilized to combat that demand exist only in the perceptual world, no such anchor point exists. An additional problem with balance perspectives is that demand and coping capacity are conceptualized post hoe. We only know that a resource aids coping capacity after it is observed to counteract some demand. We do not know if it will continue to be a resource in the future because we must always wait to see if individuals perceive

it to be such. So, if optimism aids adjustment, it is ret-

rospectively said to be a resource, and if it impairs ad- justment, it is retrospectively labeled a negative coping style. Certainly, there is room for process in developing

a model of stress. However, if everything is process, there is an absence of marker flags, standards of comparison, and other points of reference that can aid in organizing

a taxonomy and a prediction of behavior that develops from interactions of different factors. By deemphasizing the objective environment, some

demands may even go unnoticed because individuals are succeeding in coping with them in the natural course of events. So, for example, persons with strong personality hardiness (Kobasa, 1979) may fail to even notice demands being placed on them because they see such demands as exciting challenges. In fact, they may be so rich in coping resources as to be unaware of these environmental oc- currences. Other individuals may feel overwhelmed by these same events. Has no demand been placed on the hardy individuals because they failed to pay it heed, or

is it a demand that their strong resources quickly over-

came?

Even if coping resources and demands can be iden- tified, the researcher is confronted with a methodological problem of measuring them in meaningful units that may

be balanced. In other words, to test the model, the units of coping resources must be compared to the units of demands for balance or imbalance to be judged. No at- tempt has been made to develop such a system of equiv- alent units, no doubt because it would be an extremely difficult task. Without such units, however, the model remains a general conceptual framework, but one that may never be directly tested. Trumbull and Appley (1986) provided a variation

515

on the balance theme that is more compatible with the

empirical study of stress. First, they suggested that im- balance can occur when either the real or perceived de-

mands outstrip real or perceived coping capacity. This seems to allow more equal emphasis for environmental

and personal factors than do the perception-focused stress models. Second, Trumbull and Appley (1986) made an effort to integrate the overlap of findings on physiological, psychological, and social stress and suggested ways in which biological, psychological, and environmental pro- cesses could be studied side by side, rather than blurred together. The problem of circularity in defining demand and coping capacity, however, remains. It could be argued that Lazarus and Folkman (1984) have emphasized the interaction between the environment and the individual in their transactional model of coping (Folkman & Lazarus, 1985; Folkman, Lazarus, Dunkel- Schetter, DeLongis, & Gruen, 1985). However, even what they termed the environment is really the individual's appraisal thereof. This has led to criticisms of the cir- cularity of their approach (Dohrenwend et al., 1984; Kasl, 1978). This circularity, it has been argued, follows from their overemphasis on perception and their lack of em- phasis on environmental contingencies. This may be seen

in their definition of stress, in which there is no stress

without perception, and may be contrasted to Trumbull

and Appley's (1986) definition, in which both objective and subjective stress is acknowledged. Kasl (1978) argued that when perceptions are used

to establish independent and dependent variables, as in

the transactional model, the two variables "are sometimes

so close operationally that they appear to be simply two

similar measures of a single concept" (p. 13). Lazarus

and Folkman (1984), in opposition to such criticisms, went so far as to argue that the study of major stressors that are not confounded with psychological outcome (e.g., loss of loved ones, severe illness, or combat) "must not

be allowed to seduce us into settling for a simplistic con-

cept of stress as environmentallyproduced" (p. 19). These are exactly the kinds of events that Dohrenwend et al. (1984) found to be the most likely to represent environ- mental occurrences. Lazarus and Folkman (1984) instead chose to study those "garden variety" stressors for which

one can "no longer pretend that there is an objective way

to define stress at the level of environmental conditions

without reference to the character of the person" (p. 19).

The jury is not yet in on the contrasting viewpoint promoted by the more environmentally or the more cog- nitively oriented camps. On one hand, there is actually great overlap between the two positions. On the other hand, the difference between the two positions was con- sidered to be great enough for Lazarus and Folkman

(1984) to have stated that "at stake is the viability of all

cognitive,

The Conservation of Resources:

A New Stress Model

I would like to present a new stress model that I believe more closely reflects current understanding of the ubiq-

relational conceptualizations of stress" (p. 64).

516

uitous stress phenomena and perhaps bridges the gap be- tween environmental and cognitive viewpoints. I will argue that the proposed model is clearly testable, com- prehensively explains behavior during stressful cir- cumstances, and is more parsimonious than the balance or transactional model, while still encompassing the rel- ative importance and complexity of cognitions. The model's basic tenet is that people strive to retain, protect, and build resources and that what is threatening to them is the potential or actual loss of these valued resources. I

have termed it the model of conservation of resources

(Hobfoll, 1988). The view that individuals actively seek to create a world that will provide them pleasure and success is a

long-standing one in psychology, but one that has been relatively ignored in stress theory. Freud ( 1900/1913) in- troduced the pleasure principle, the notion that humans instinctually seek that which is pleasurable. Maslow (1968) also proposed that people seek physical resources, then social resources, then psychological resources, in a hierarchical manner. Pearlin, Lieberman, Menaghan, and MuUen ( 1981) have similarly proposed "that the protec-

. are fundamental goals

tion and enhancement of self

after which people strive" (p. 340, emphasis added). Most germane to this article, social learning theory proposes that people actively engage their environment in order to increase the chances of obtaining positive reinforcement (Bandura, 1977). There are two basic ways to achieve these goals. First, people act to enhance the likelihood of situational (i.e., here and now) reinforcement (Swarm & Read, 1981). Success is more likely, however,if individuals seek to create and maintain personal characteristics (e.g., mastery or self-esteem) and social circumstances (e.g., tenure or intimacy) that will increase the likelihood of receipt of reinforcement and to avoid the loss of such characteristics and circumstances (Wicklund & Goll- witzer, 1982). The model of conservation of resources rests on this second strategy. The definition of stress is derived directly from the model and the above mentioned basic tenet: Psychological stress is defined as a reaction to the environment in which there is (a) the threat of a net loss of resources, (b) the net loss of resources, or (c) a lack of resource gain fol- lowing the investment of resources. Both perceived and actual loss or lack of gain are envisaged as sufficient for producing stress. Resources, then, are the single unit necessary for understanding stress. Resources are defined as those ob- jects, personal characteristics, conditions, or energies that are valued by the individual or that serve as a means for attainment of these objects, personal characteristics, con- ditions, or energies. Examples of resources include mas- tery (Pearlin & Schooler, 1978), self-esteem (Rosenberg, 1965), learned resourcefulness (Rosenbaum & Smira, 1986), socioeconomic status (Worden & Sobel, 1978), and employment (Parry, 1986).

or cause a depletion of people's resources. They may threaten people's status, position, economic stability, loved ones,

Environmental circumstances

often

threaten

March 1989 • American Psychologist

basic beliefs, or self-esteem. These losses are important on two levels. First, resources have instrumental value to people, and second, they have symbolic value in that they help to define for people who they are (Brown & Andrews, 1986; Cooley, 1902; Erikson, 1968; James, 1890).

Behavior During Stressful and Everyday Circumstances

The model of conservation of resources goes beyond pre- vious models in that it inherently states what individuals do when confronted with stress and when not confronted with stress. Specifically, when confronted with stress, in- dividuals are predicted by the model to strive to minimize net loss of resources. This prediction is not inconsistent with Lazarus and Folkman's (1984) coping model, but they did not specify the goal of coping other than as an attempt to limit stress. Other stress theories do not predict psychological or behavioral action when people are not confronted with stressors. According to the conservation of resource model, when not currently confronted with stressors, people strive to develop resource surpluses in order to offset the possibility of future loss. This phenomenon re- ceived support when studied under the rubric of self-ac- quisitive or self-assertive styles (Schlenker, 1987; Thibaut & Kelley, 1959). When people develop resource surpluses, they are likely to experience positive well-being (eustress), as reviews of studies of personal and social resources attest (Cohen & Edwards, in press; Cohen & Wills, 1985). Where individuals are ill equipped to gain resources, in contrast, they are likely to be particularly vulnerable (Rappaport, 1981). Such individuals lean toward preven- tion of resource loss, or what some have termed self-pro- tective styles (Arkin, 198 l; Cheek & Buss, 198 l; Thibaut & Kelley, 1959). People may also enrich resources by investing other resources, such as when people give aid to kith or kin. Such resource investment is not necessarily tit for tat (Clark, 1983), but rather, the model suggests that people take a long-term outlook toward the conservation of re- sources (see also Dodge & Martin, 1970). This discussion begs the question of what people employ to offset resource loss or to gain resources. The answer to this question is that they employ resources that they possess or they call on resources available to them from their environment. Individuals, for instance, invest their love and affection to receive a return of the same. Often individuals invest their time and energy, two im- portant resources, in attempts to translate them to other more highly prized resources, for example, power and money. Moulton (1980) argued that women are vulner- able to stress because they are often challenged with new expectations prior to the acquisition of relevant resources.

Kinds ofResources

As mentioned earlier, the model identifies four kinds of resources whose loss and gain result in stress or eustress (i.e., well-being), respectively. Object resources are valued

because of some aspect oftheirphysicalnature or because

March 1989 • American Psychologist

of their acquiring secondary status value based on their rarity and expense. A home has value because it provides shelter, whereas a mansion has increased value because it also indicates status. Objects have seldom been consid- ered in stress research, but are linked to socioeconomic status, which has been shown to be an important factor in stress resistance (Dohrenwend, 1978). Conditions are resources to the extent that they are valued and sought after. Marriage, tenure, and seniority are examples of these. Conditions have been studied in- frequently vis-a-vis their stress mediating effect, but Pear- lin (1983) has suggested that roles inherent in being sub- ject to certain conditions (e.g., wife, employee, or partner) are critical to an understanding of people's stress resis- tance capacity. Vachon (1986) provided evidence that just living with someone (a condition) resulted in decreased mortality rates for women with cancer, and Henderson, Byrne, and Duncan-Jones (1981) found that being mar- ried is a resistance resource. Others, however, argue that conditions may need to be qualified; a bad marriage or poor social relationship is unlikely to have salutary effects (Rook, 1984; Thoits, 1987). Clearly, this area of resources holds much promise for future work. The conservation of resource model suggests that measuring the extent to which conditions are valued by individuals or groups may provide insight into their stress-resistance potential. Personal characteristics are resources to the extent that they generally aid stress resistance. Antonovsky

(1979) coined the term general resistance resources and

suggested that one's personal orientation toward the world is the key; specifically, this means seeing events as pre- dictable and generally occurring in one's best interest. Investigations of various personal resources suggest that many personal traits and skills aid stress resistance (Cohen & Edwards, in press; HobfoU, 1985b). In addition, social support's effect seems to hinge on its value in promoting or supporting a positive sense of self and a view that one can master or at least see through stressful circumstances (Cohen & Wills, 1985; Pearlin et al., 1981). Energies are the last resource category and include such resources as time, money, and knowledge. These resources are typifiednot by their intrinsic value so much as their value in aiding the acquisition of other kinds of resources. Energy resources have not been studied in North America to my knowledge, but have received some attention within the German action orientation model of stress research (Schtnpflug, 1985). Indirect evidence for the importance of energy resources may be found in net- work research by Wellman (1981), who illustrated that a large social network becomes valuable when information (an energy resource) that requires numerous sources (e.g., contacts for employment) is required. Finally, the reader might note that social support does not fit in any one category above. Rather, social relations are seen as a resource to the extent that they provide or facilitate the preservation of valued resources, but they also can detract from individuals' resources. This notion is consistent with research that finds social support beneficial when it provides for situational needs (Cohen

517

& Wills, 1985; Hobfoll, 1985b) and harmful or benign when it does not (Hobfoll & London, 1986; Riley & Eck- enrode, 1986; Rook, 1984).

The Concepts ofLoss

To assert that stress only concerns a loss or the potential loss of resources may seem on first blush to be an over- statement, but this is what the data on stress seem to support. Early discussion of the idea that loss is central to threat may be attributed to such authors as Lindemann (1944) and Parkes (1970). In studying bereavement, they found that loss of a loved one constituted social loss, po- tential loss of status and economic stability, loss of a way of life, as well as loss of the loved individual. For loss to be central to a comprehensive theory of stress, however, loss would have to be central to all psy- chological stressors, not just bereavement. In reviewing any of a number of stressful event surveys (Dohrenwend et al., 1978; Holmes & Rahe, 1967; Sarason, Johnson, & Siegel, 1978), it becomes clear that most items are ob- viously loss events. These include death of a spouse, di- vorce, marital separation, being fired at work, retirement, foreclosure of a mortgage or loan, and so on. These events are given the strongest severity weighting when weights are assigned to items (Holmes & Rahe, 1967). This sup- ports the notion that items that clearly reflect loss are the most psychologically threatening. Furthermore, if the loss does not directly affect the individual, it is less salient (Hobfoll, London, & Orr, 1988; Swindle, Heller, & Lakey, 1988). Brown and Andrews (1986) reported that other than in cases of depression where the disorder is likely to be dispositional, loss events are responsible for approxi- mately 90% of the cases of depression they studied. Other events are more ambiguous in terms of why they might cause stress. These include such events as business readjustment, change in health of a family member, receiving a large mortgage, or graduation from college. Researchers have consistently found that if the dimension of"undesirability" is partialed out of the cor- relation between the endorsement of the event and out- come, the relationship between the experience of these events and strain (i.e., the outcome of stress) approaches zero (Mueller, Edwards, & Yarvis, 1977; Thoits, 1983; Vinokur & Selzer, 1975). The question is what is meant by the vague concept "undesirable." In reviewing event items, it is readily noted that the positive interpretation of an item, such as business readjustment, indicates some increase in acquisition of something the individual values. Obversely, the negative possibility would suggest a loss of something the individual values (e.g., money, status, or flexibility). Consequently, when individuals are endorsing the item, "business readjustment," they are either indi- cating adjustment that denotes a gain in power, finances, or positive challenge, or they are indicating adjustment that denotes their loss. Only the loss alternative seems to be stressful and then only when it threatens cherished role involvements (Swindle et al., 1988). Others have suggested that positive events also are stressful because change itself is stressful. Again, there is

518

little evidence supporting this supposition and a good deal of evidence challenging it (Thoits, 1983). According to the model of conservation of resources, many changes signify a gain of resources, and these should aid stress resistance, not hinder it. Indeed, as already mentioned, studies that have separated positive and negative events have found that positive events have a stress-limiting effect (see Thoits, 1983, for a review). Transitions have also been seen as potentially stress- ful. Felner, Farber, and Primavera (1983) made a strong case, however, for seeing transitions as periods requiring adaptation--a process that is challenging but not nec- essarily stressful (Kobasa, Maddi, & Courington, 1981). Wilcox (1986) also argued that transitions are series of linked events. When these chains of events entail multiple loss events (e.g., divorce leading to income loss, break- up of other relationships, and childcare difficulties), they are likely to prove stressful. However, when they are made up of positive events or challenges that are successfully met, they are more likely to produce stress inoculations (Meichenbaum & Jaremko, 1983). Again, the evidence suggests that stress is likely to ensue only when loss is evidenced: Change, transitions, and challenge are not of themselves stressful.

Resource Replacement

The model of conservation of resources also suggests that

although loss of resources is stressful, individuals may em- ploy other resources to offset net loss (cf. Pearlin et al., 1981). Replacement is the most direct way this is accom-

plished. Following divorce,

course is remarriage (Burgess, 1981). Following miscar- riage, women are told by supporters to quickly get pregnant again (Hobfoll & Leiberman, 1987). Kessler, Turner, and House (1987) also found that reemployment following prolonged unemployment largely counteracts loss-related depression, anxiety, and somatization. When direct re- placement is not possible, symbolic replacement or re- placement through indirect means may be possible. Nu- merous studies illustrate, for instance, that following loss of self-esteem people attempt to directly alter potentially harmful conditions in order to facilitate positive feedback (relativelydirect), compensate by regaining esteem in other related areas (relatively indirect), or artificially acquire support for their desired identity through superficial in- terpersonal manipulations (more indirect; see Schlenker, 1987, for detailed discussion of these points). Employing resources for coping is also stressful in itself. In some elegant experiments, Schtnpflug (1985) illustrated that individuals employ resources in the coping process and that such employment often depletes these resources. Energy is expended, favors are used up, and self-esteem is risked, all in the service of offsetting loss of other potential loss. If the resources expended in coping outstrip the resultant benefits, the outcome of coping is likely to be negative. Illustrating this point, studies have found that people who were placed in a position in which they were required to give support, at a time when they themselves needed support, experienced increased psy-

for instance, the most common

March 1989 • American Psychologist

chological distress (Hobfoll & London, 1986; Riley & Eckenrode, 1986). I think it would yield important insights about stress if the cognitive and behavioral strategies people employed in coping were tested following this process of replace- ment, substitution, or investment. Work on energization following failure (Ford & Brehm, 1987) suggests that people roughly judge their potential losses, determine what they stand to lose by expending other resources, and analyze the likelihood of succeeding or offsetting losses if they choose to employ a given coping strategy. The extent to which they make rational, efficient, and suc- cessful decisions about how to cope would be of great interest. No doubt such strategies are clouded by emotions and the inherent complexity of real human problems, but research should nonetheless be capable ofuncovering important insights. There is a hint here of blaming the victims, that is, "If they were only to employ their resources well, they could overcome stress." However, resources are not dis- tributed equally, and those people who lack resources are most vulnerable to additional losses (Dohrenwend, 1978). Loss spirals develop because they lack the resources to

offset loss. If resources are used to prevent loss of other resources, such loss would be predicted to lead to further decreases in the likelihood of possessing necessary re- source reserves (see Billings & Moos, 1981; Eron & Pe-

terson,

their spouse's death, many women are too economically disadvantaged to upgrade their education, and so they are also likely to be subjected to economic stressors. The model also predicts that those lacking the op- tions made possible by possessing many resources will attempt loss-control strategies that have a high cost and poor chance of success. When the stakes are high, indi- viduals have little choice but to attempt gain strategies that are likely to fail, but have some short-term payoff (Ford & Brehm, 1987). If they do not, they risk lapsing into a sense of helplessness and despair (Alloy, Peterson, Abramson, & Seligman, 1984) and further exposing themselves to loss. Mitchell and Hodson (1986), for ex- ample, found that battered women who lack personal re- sources such as education, self-earned income, and oc- cupational status use passive coping. Such inappropriate forms of coping can best be explained by the fact that passive coping strategies are the most potent ones in the women's repertoire. In the past, instead, it has often been interpreted as support for these women's masochistic tendencies. Another study noted that Black women in a traditional Black community will seek counterproductive types of support from elders because more appropriate support is unavailable (Dressier, 1985). These studies support the hypothesis that those who lack resources at- tempt to employ what resources they have, often pro- ducing self-defeating consequences.

Appraisal of Resources

Until this point, emphasis has been placed on objective loss and shared social standards of what constitutes loss.

1982; Menaghan, 1983). So, for example, following

March 1989 • American Psychologist

The model also proposes an important role, however, for appraisal (Lazarus & Folkman, 1984). It should be un- derscored, however, that even when perception is impor- tant, normative tendencies regarding how resources are evaluated and what constitutes loss guide individuals' as- sessments of their environments and their selves. In this regard, Rokeach (1973) provided ample evidence for common (i.e., normative) values within like cultures or groups, and scaling of life events provides evidence that an agreed-on set of weights can be applied to events rang- ing from the most benign to the most extreme (Holmes & Rahe, 1967).

One way individuals

Shifting thefocus of attention.

may conserve resources is by reinterpreting threat as challenge (Kobasa, 1979; Kobasa et al., 198 I). Thus,

people may focus on what they might gain, instead of what they might lose, in light of a particular situation. I have argued previously, however, that this type of trans- formation should not be romanticized as occurring for all stressors (Hobfoll, 1985a). Many tragic stressors have little redeeming value (Lehman, Wortman, & Williams, 1987). Still, many everyday stressors are neither clearly positive nor negative and so are most likely to be open to personal appraisal. The takeover of one's employer by a larger corporation may, for instance, be appraised as an opportunity for a quick rise in the executive ranks, or it

may be perceived

death of a spouse or child, in contrast, has much more straightforward meaning in terms of its consequences.

In addition to refocusing

attention as to whether and which resources are likely to be lost or gained, people may combat their sense of loss by reevaluatingthe value of resources that are threatened or that have been lost. So, for example, the stress of school failure can be mitigated by lowering the value placed on education. Similarly, a social rejection can be lightened by lowering the value placed on the lost relationship.

This appears to be the most simple course for people, because rather than combating the stressor or enduring the stress, people could merely alter their interpretation of events and their consequences. Indeed, many stress theories have suggested that appraisal is the key to stress resistance (Bulman & Wortman, 1977; Goodhart, 1985; Johnson & Sarason, 1978; Kobasa, 1979; Lazarus & Folkman, 1984; Spielberger, 1972). However, this suggests that appraisal of the values placed on internal and external resources are more pliable than they may, in fact, be. Devaluation of resources in order to counteract the impact of their loss challenges basic notions concerning the self and the world. Following examination failure, students may devalue the worth of education. However, if their self-esteem was based, in part, on their past aca- demic excellence and if they were raised in a society that highly esteems education, there will be impediments to making such reassessments. Because those things that people value and the ways in which they perceive the world are basic to their sense of self (Erikson, 1963; May, 1950; Rokeach, 1973), perceptions may not be so pliable. This may be one reason why so few interventions aimed

as spelling a rash of job lay-offs. The

Reevaluating resources.

519

at changing how people make transformations of this kind have even been attempted. Phrasing this as an empirical question, I would sug- gest that people's resource assessments are derived, in part, from their basic values (Rokeach, 1973) and devel- opmental history, that is, what they have learned through experience to be valuable to them. Hence, it would be hypothesized that although minor reappraisals may allow individuals to buffer the brunt of stressors, reappraisal of more basic aspects of the self and the environment are more likely to backfire against the individual--resulting in a sense of insecurity and despair--than they are to have stress-moderating effects. Kaplan (1983) made a similar argument, as follows:

To the extent that individuals are born with or acquire attributes •

and are circumstantially required to perform behaviors

that threaten their basic values or reflect the inabilityto achieve these values, individuals will experience a degree of subjective distress. Insofar as the basic value in question is positive self- evaluation, a history ofcircumstances in which individuals come

to experience themselves as possessing highly disvalued attri- butes and performing disvalued behaviors will evoke highly dis- tressful self-rejecting feelings. (p. 211)

So, what Kaplan was suggesting is that when people must behave or experience themselves in a way dissonant with their basic view of the self or the world, they are likely to experience psychological distress. Rollo May (1980) also suggested that such threats to underlying or basic values are counterproductive.

The distinctive qualit~ of human anxiety arises from the fact that man is a valuing animal, who interprets life and world in terms of symbols and meanings. It is the threat to these values-- specifically,to some value that the individual holds essential to his existence as a self--that causes anxiety. (p. 241)

Additional research is required concerning norma- tive evaluation of resources and individuals' more stable resource appraisals. Certainly, situational changes in per- ceptions are important, but they seem to have been stud- ied to the exclusion of more basic, stable self and world views. More attention should be drawn to how stress is influenced by socialization (Kaplan, 1983), roles (Merton, 1957; Pearlin, 1983), and social development (Bandura, 1977) than models that overwhelmingly emphasize ap- praisal have done.

The Expectation of Net Gain of Resources

The model of conservation of resources also suggestswhat, in part, motivates people's behavior when they are not currently experiencing taxing stressors. Specifically, in- dividuals are motivated to gain resources. This motivation drives people to invest resources in order to enrich their resource pool. This serves both to shelter them from fu- ture losses and contributes to enhanced status, love, pos- sessions, or self-esteem, depending on individuals' goals and the direction of their investment. Investment of resources may be observed in good marriages. In such marriages, both partners are constantly contributing from what they have to each other and to

520

the relationship. There is a long-term expectation, how- ever, that their investment will produce a payoffin terms of returned love, esteem, affection, and security (Clark, 1983). Clearly, people's financial investments also follow this principle. People are willing, in this regard, to invest greater resources for greater payoffor for increased odds of payoff(i.e., low risk). A small gain is poor compensation for a large risk. The model of conservation of resources predicts that when such investment does not provide a good return, people will experience this as loss. The loss is the loss of the expected or envisioned gain. Schfnpflng (1985) also suggested that the cost of expended internal and external resources must be added to negative outcome or sub- tracted from positive outcome, in order to predict people's stress reactions, and he provided some laboratory evi- dence for this phenomenon. Other research has also in- dicated that depression is most likely to follow when in- vestment of resources fails to resolve a conflict (Brown

& Andrews, 1986). Obviously, more research is required

to explore this proposed aspect of the stress phenomenon.

Future Research on Stress

The model of conservation of resources helps structure the sphere in which research on stress may advance. First, the model emphasizes that resources have both objective and subjective components. Current reliance on self-re- port increases the potential for confounding with self- reported measures of outcome. Future research should attempt to study objectively observable resources as well. This can be done by measuring behavioral (e.g., sup- portive behavior) or structural (e.g., belonging to a co- hesive group) operationalizations of resources (Hobfoll & Lerman, 1988; Solomon, Mikulincer, & Hobfoll, 1987) or by seeking assessments by clinicians (e.g., evaluation of self-esteem; Brown & Harris, 1978). This will not be possible for all resources, but where it is possible it would help dispel arguments that resources are themselves symptoms (Depue & Monroe, 1986; Dohrenwend et al.,

1984).

It also follows that the field of stress would benefit from investigations of resources that people find helpful

in light of differing kinds of losses. Individual traits, such as hardiness (Kobasa, 1979), locus of control (Lefcourt, Martin, & Selah, 1984), personal self-consciousness (Suls

& Fletcher, 1985), optimism (Carver & Scheirer, 1983),

absence of chronic psychopathological disorder (Depue

& Monroe, 1986), and low negative affectivity (Watson

& Tellegen, 1985) have been studied, but only as general

stress mediators. Similarly, social support has been studied regarding its effect on stress in general, not some specific kind or category of stressors (Cohen & Wills, 1985; Sar- ason & Sarason, 1985). By studying the specificity of re- sources for counteracting the effects of different kinds of losses, the potency (i.e., effect size) and robustness (i.e., degree of potency for varied kinds of losses) may be iden- tified. In our own research, for example, we have found self-esteem to be a generally robust resource (Hobfoll & Leiberman, 1987; Hobfoll & London, 1986), whereas the

March 1989 • American Psychologist

benefit of social support was found to be limited to sit- uations in which social interaction is possible and in which social interaction does not add to already experi- enced stress (e.g,, stress contagion; Riley & Eckenrode, 1986). Furthermore, social support was found to have a time-limited effect, whereas the influence of self-esteem was not time limited. If resources are only beneficial when they help counteract loss or aid net increase of resources, it also follows that a resource may, indeed, be detrimental in certain instances (Hobfoll, 1985b). We have adopted the term "resources" in our own work, despite this problem, because so many other authors have used this term. How- ever, there is a scientific concern when research concen- trates on the case in which investigators search for in- stances that prove their point or support the robustness of the resource that most interests them. We are all guilty of this tendency, and it is important that we test the notion of fit of resources with situational demands by illustrating both when a given resource aids adjustment and when it interferes with adjustment or is neutral. Only by this multimethod, multitrait approach (i.e., indicating when different resources do and do not contribute to stress re- sistance; Campbell & Fiske, 1959) may we differentiate between the actual active ingredients of coping and such problems as shared method variance and self-fulfilling prophesy. The model of conservation of resources also de- mands that the popular concept of "fit" (Caplan, 1983; French, Rodgers, & Cobb, 1974) be spelled out such that we may learn what aspects of the resource "fit" the de- mands resulting from the loss. Fit has typically been mea- sured in terms of self-report as to whether individuals globally judge their resources as fitting demands (i.e., "how good is the fit of your resources with job demands?). More recent applications of the concept have attempted, in contrast, to predict a priori what properties of certain resources offset the effect of specific demands (Hobfoll & Leiberman, 1987; Hobfoll & Lerman, 1988; Hobfoll & London, 1986; Keinan & Hobfoll, in press; Mitchell & Hodson, 1986; Parry, 1986). Moos (1984) has been in- strumental in this regard in urging researchers to consider the ecology of stressful events, although he has not pro- posed a specific theory about the properties of events that are stressful. Future research will need to more carefully delineate how certain demands are met by specific re- sources, and the concepts of substitution, replacement, and investment of resources may provide one avenue for such refinement. Because loss spirals and investment of resources are predicted to occur and because events themselves are likely to evolve over time, it is also important to focus on how the interplay between resources and situational needs changes over time as stressor sequences unfold. In addi- tion, because both of these processes result in potential resource depletion, it will be important to illustrate not only the effect of resources on outcome, but also of out- come on resources (Hobfoll & Lerman, 1988; Pearlin et al., 1981; SchSnpflug, 1985). The transactional model of

March 1989 • American Psychologist

stress also makes this point regarding the recursive nature of stress (Lazarus & Folkman, 1984), but does not ex- plicitly state that coping itself potentially depletes re-

sources.

In outlining the process by which resources operate, the model of conservation of resources suggests a specific set of behaviors and cognitions that may be observed in order to support, clarify, or disconfirm the model. So, for example, rather than merely showing that social support or a particular style of coping results in a stress-buffering effect, to confirm the present model it would have to be shown that cognitions or events caused real or perceived losses and that social support or coping style limited these losses or enabled gain of other resources. Prior research has focused on outcome (e.g., direct versus stress-buffering effects) and not process, but outcomes alone can neither disqualify nor buttress any particular stress model. Overall, research on stress must move from its pres- ent acceptance of studying stress as a vague, general con- cept and move toward direct testing of stress models. Lacking precise measurement instruments and physical laws that allow absolute standards, the social sciences must proceed through the cumbersome process of hy- pothesis testing of comparative models. However, the stress literature may be criticized as having many hy- potheses but lacking overarching models that create competing viewpoints. The conservation of resources model provides a per- spective that may better reflect the current state of knowl- edge concerning stress. In addition, it leads to many test- able directions that lend the model both to potential con- firmation or rejection. Comparison to alternative models will be useful and may encourage researchers to adopt a more theory-based approach to their investigations. Only in this way can we possibly avoid a science that merely reflects how we wish to see things. Nowhere is this point more clear than in the study of stress.

REFERENCES

Alloy, L. B., Peterson, C., Abramson, L. Y., & Seligman, M. E. P. (1984).

Attributional style and the generality

of learned helplessness. Journal

of Personality and Social Psychology, 46, 681-687.

Antonovsky, A. (1979). Health, stress, and coping. San Francisco: Josscy- Bass. Appley, M. H., & Trumbull, R. (Eds.). (1986). Dynamics ofstress:Phys-

iological psychological, and social perspectives. New York: Plenum Press.

Arkin, R. M. (1981). Self-presentational

styles. In J. T. Tedeschi (Ed.),

Impression management theory and socialpsychologicalresearch(pp.

311-333). New York: Academic Press. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Pren- tice-Hall. Baum, A., Singer,J. E., & Baum, C. S. ( 1981). Stress and the environment.

Journal of Social Issues, 37, 4--35.

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.

Brown, G. W., & Andrews, B. (1986). Social support and depression. In M. H. Appley & R. Trumbull (Eds.), Dynamics of stress: Physi-

New

ological, psychological and social perspectives (pp. 257-282).

York: Plenum Press.

Brown, G. W., & Harris, T. (1978). The social origins of depression:

The study ofpsychiatric disorder in women. New York: Free Press.

521

Bulman, R. J., & Wortman, C. B. (1977). Attributions of blame and coping in the "real world": Severe accident victims react to their lot.

Journal of Personalityand Social Psychology,35, 351-363.

Burgess, R. L. (1981). Relationships in marriage and the family. In S.

Duck & R. Gilmour (Eds.), Personalrelationships 1:Studyingpersonal

relationships (pp. 179-196). New York: Academic Press. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. PsychologicalBul-

letin, 56, 81-105.

Cannon, W. B. (1932). The wisdom of the body (2nd ed.). New York:

Norton.

Caplan, G. (1964). Principlesofpreventivepsychiatry. New York: Basic

Books.

Caplan, R. D. (1983). Person-environment fit: Past, present, and future. In C. L. Cooper (Ed.), Stress research(pp. 35-77). New York: Wiley. Carver, C. S., & Scheirer, M. E (1983). A control theory model of normal behavior and implications for problems in self-management. In P. C.

Kendall (Ed.), Advances in cognitivebehavioralresearchand therapy

(Vol. 2, pp. 127-194). New York: Academic Press. Cheek, J. M., & Buss, A. H. (1981). Shyness and sociability. Journal of

Personalityand Social Psychology, 41, 330-339.

Clark, M. S. (1983). Reactions to aid in communal and exchange rela- tionships. In J. D. Fisher, A. Nadler, & B. M. DePaulo (Eds.), New directions in helping(Vol. 1, pp. 281-304). New York: Academic Press.

Cohen, S., & Edwards, J. R. (in press). Personality characteristics as moderators of the relationship between stress and disorder. In

R. W. J. Neufeld (Ed.), Advances in the investigation ofpsychological

stress. New York: Wiley. Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure

of perceived stress. Journal of Health and Social Behavior, 24, 385-

396.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering

hypothesis. PsychologicalBulletin, 98, 310-357.

Cook, T. D., & Campbell, D. 1".(1979). Quasi-experimentation:Design

and analysis issuesforfield settings. Boston: Houghton Mifflin. Cooley, C. H. (1902). Human nature and the social order. New York:

Scribner.

Depue, R. A., & Monroe, S. M. (1986). Conceptualization and mea- surement of human disorders in life stress research: The problem of

chronic disturbance. PsychologicalBulletin, I, 36-51.

Dodge, D. L., & Martin, W. T. (1970). Social stress and chronic illness.

University of Notre Dame, IN: University of Notre Dame Press.

Dohrenwend, B. S. (1978). Social status and responsibility for stressful life events. In C. D. Spielberger & L G. Sarason (Eds.), Stress and anxiety (Vol. 5, pp. 25-42). New York: Wiley. Dohrenwend, B. S., Dohrenwend, B. P., Dodson, M., & Shrout, P. E. (1984). Symptoms, hassles, social support, and life events: Problem

R. J. (1985). Dynamics of a stressful encounter. Cngnitive appraisal,

coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992-1003.

Ford, C. E., & Brehm, J. W. (1987). Effort expenditure following failure.

In C. R. Snyder & C. E. Ford (Eds.), Coping with negativelife events

(pp. 81-104). New York: Plenum Press. French, J. R. P., Rodgers, W. L., & Cobb, S. (1974). Adjustment as person-environment fit. In G. V. Co¢lho, D. A. Hamburg, & J. E. Adams (Eds.), Copingand adaptation (pp. 316-333). New York: Basic

Books. Freud, S. (1900). The interpretation of dreams. New York: Macmillan. (Anthological translation by A. A. Brill, 1913) Gentry, W. D., & Kobasa, S. C. (1984). Social and psychological resources mediating stress-illness relationships in humans. In W. D. Gentry

New York:

Guilford Press. Goodhart, D. E. (1985). Some psychological effects associated with pos- itive and negative thinking about stressful events: Was Pollyanna fight?

(Ed.), Handbook of behavioral medicine (pp. 87-113).

Journal of Personalityand Social Psychology, 48, 216-232.

Henderson, S., Byrne, G. O., & Duncan-Jones, P. (1981). Neurosis and

social environment. Sidney, Australia:

Academic Press.

Hobfoll, S. E. (1985a). The limitations of social support in the stress process. In I. G. Sarason & B. R. Sarason (Eds.), Social support:

Theory, research, and application (pp. 391--414). The Hague, The

Netherlands: Nijhoff. Hobfoll S. E. (1985b). Personal and social resources and the ecology of stress resistance. In P. Shaver (Ed.), Review of personality and social psychology (Vol. 6, pp. 265-290). Beverly Hills, CA: Sage.

Hobfoll S. E. (Ed.). (1986). Stress, social support, and women. Wash-

ington, DC: Hemisphere. Hobfoll, S. E. (1988). The ecologyofstress. Washington, DC: Hemisphere. Hobfoll S. E., & Leiberman, Y. (1987). Personality and social resources in immediate and continued stress resistance among women. Journal

of Personalityand Social Psychology,52, 18-26.

Hobfoll

attitudes, and stress resistance: Mothers' reactions to their child's ill-

ness. American Journal of Community Psychology,, 16, 565-589.

Hobfoll S. E., & London, P. (1986). The relationship of self concept and social support to emotional distress among women during war.

S. E., & Lerman, M. (1988). Personal relationships, personal

Journal of Social and ClinicalPsychology, 12, 87-100.

Hobfoll S. E., London, P., & err, E. (1988). Mastery, intimacy, and

stress-resistance during war. Journal of Community Psychology, 16,

317-331.

Holmes T. H., & Rahe, R. H. (1967). The social readjustment rating

scale. Journal of PsychosomaticResearch, 11, 213-218.

James, W. (1890). Principles ofpsychology (Vols.

Johnson

1-2). New York: Holt.

J. H., & Sarason, I. G. (1978). Life stress, depression, and

of confounded

measures. Journal ofAbnormal Psychology,, 93, 222-

anxiety: Internal-external control as a moderator variable. Journalof

230.

PsychosomaticResearch, 22, 205-208.

Dohrenwend, B. S., Krasnoff, L., Askenasy, A. R., & Dohrenwond,

B. P. (1978). Exemplification of a method for scaling life events: The

PERI life events scale. Journal of Health and Social Behavior, 19,

205-229.

social support

Dressier, W. W. (1985). Extended family relationships,

and mental health in a Southern Black community. Journal ofHealth

and Social Behavior, 26, 39-48.

Elliot, G. R., & Eisdorfer, C. (1982). Stress and human health. New York: Springer.

Erikson, E. H. (1963). Childhoodand society(2nd ed.). New York: Nor-

ton.

Erikson, E. H. (1968).

Eron, L., & Psterson, R. A. (1982). Abnormal behavior: Social ap-

Identity: Youthand crisis. New York: Norton.

proaches. Annual Review of Psychology,,33, 231-264.

Felner, R. D., Farber, S. S., & Primavera, J. (1983). Transitions and

stressful life events: A model for primary prevention. In R. D. Felner,

L. A. Jason, J. N. Moritsugu, & S. S. Farber (Eds.), Preventivepsy-

New York:

Plenum.

chology: Theory, research and practice (pp.

199-215).

Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process:

Study of emotion and coping during three stages of a college exami-

nation. Journal of Personalityand Social Psychology, 48, 150-170.

Folkman, S., Lazarus, R. S., Dunkel-Schctter, C., DeLongis, A., & Gruen,

522

Kanner, A. D., Coyne, J. C., Schaefer, C., & Lazarus,

R. S. (1981).

Comparisons of two modes of stress measurement: Daily hassles and

uplifts versus major life events. Journal of BehavioralMedicine, 4, 1-

39.

Kaplan, H. B. (1983). Psychological distress in sociological context: To- ward a general theory of psychosocial stress. In H. B. Kaplan (Ed.),

Psychosocialstress: Trendsin theoryand research(pp. 195-264). New

York: Academic Press. Kasl, S. V. (1978). Epidemiological contributions to the study of work stress. In C. L. Cooper & R. Payne (Eds.), Stress at work(pp. 3-48). New York: Wiley. Keinan, G., & Hobfoll, S. E. (in press). Stress, dependency, and social support: Who benefits from husband's presence in delivery? Journal

of Social and ClinicalPsychology.

Kendall, P. C. (1983). Stressful medical procedures: Cognitive-behavioral strategies for stress management and prevention. In D. Meichenbaum

& M. E. Jaremko

(Eds.), Stress reduction and prevention (pp. 159-

190). New York: Plenum Press. Kessler, R. C. (1983). Methodological issues in the study of psychological

stress. In H. B. Kaplan (Ed.), Psychosocialstress: 7~ends in theory

and research (pp. 267-341). New York: Academic Press. Kessler, R. C., Turner, J. B., & House, J. S. (1987). Unemploymentand

ill health in southeastern Michigan: An overview of resultsfrom a

March

1989

American

Psychologist

community survey. Paperpresented at the National Institute of Mental

Health meeting for prevention intervention research centers, Wash- ington, DC. Kobasa, S. C. (1979). Stressful life events, personality, and health: An

inquiry into hardiness. Journal of Personality and Social Psychology,

37, 1-11. Kobasa, S. C., Maddi, S. R., & Courington, S. (1981). Personality and constitution as mediators in the stress-illness relationship. Journal of

Health and Social Behavior, 22, 368-378. Lazarus, R. S. (1966). Psychological stress and the coping process. New

York: McGraw-Hill.

Lazarus,

New York: Springer. Lefcourt, H. M., Martin, R. A., & Selah, W. E. (1984). Locus of control and social support: Interactive moderators of stress. Journal of Per-

R. S., & Foikman, S. (1984). Stress, appraisal

and coping.

sonality and Social Psychology, 47, 378-389.

Lehman, D. R., Wortman, C. B., & Williams, A. E (1987). Long-term effects of losing a spouse or child in a motor vehicle crash. Journal

of Personality and Social Psycholog£, 52, 218-231.

Lindemann, E. (1944). The symptomatology and management of acute

grief. American Journal of Psychiatry, 101, 141-148.

Maslow, A. H. (1968). Toward a psychology of being. New York: Van Nostrand Reinhold. "May, R. (1950). The meaning of anxiety. New York: Ronald Press. May, R. (1980). Value conflicts and anxiety. In I. L. Kutash & L. B. Sehlesinger (Eds.), Handbook on stress and anxiety (pp. 241-248). San Francisco: Jossey-Bass. McGrath, J. E. (1970). A conceptual formulation for research on stress.

In J. E. McGrath (Ed.), Social and psychologicalfactors in stress (pp.

10-21). New York: Holt, Rinehart & Winston.

Meichenbaum, D. (1977). Cognitivebehaviormodification:An integrative

approach. New York: Plenum Press. Meichenbaum, D., & Jaremko, M. E. (Eds.). (1983). Stress reduction and prevention. New York: Plenum Press. Menaghan, E. G, (1983). Individual coping efforts: Moderators of the relationship between life stress and mental health outcomes. In H. B.

Kaplan (Ed.), Psychosocial stress: Trends in theory and research (pp.

157-191). New York: Academic Press.

Merton, R. K. (1957). Social theory and social structure: Toward the codification of theory and research. New York: Free Press.

Milgram, N. A. (1986). An attributional analysis of war-related stress:

Modes of coping and helping. In N. A. Milgram (Ed.), Stress and

coping in time of war (pp.

9-25). New York: Brunner/Mazel.

Mitchell, R. E., & Hodson, C. A. (1986). Coping and social support among battered women: An ecological perspective. In S. E. Hobfoll (Ed.), Stress, social support, and women (pp. 153-168). Washington, DC: Hemisphere. Moos, R. H. (1984). Context and coping: Toward a unifying conceptual

framework. American Journal of Community Psychology, 12, 5-25.

Moulton, R. (1980). Anxiety and the new feminism. In I. L. Kutash & L. B. Schlesinger (Eds.), Handbook on stress and anxiety (pp. 267- 284). San Francisco: Jossey-Bass. Mueller, D. P., Edwards, D. W., & Yarvis, R. M. (1977). Stressful life events and psychiatric symptomatology: Change or undesirability.

Journal of Health and Social Behavior, 18, 307-316.

Parkes, C. M. (1970). "Seeking" and "finding" a lost object: Evidence from recent studies of the reaction to bereavement. Social Science

and Medicine, 4, 187-201.

Parkes, C. M. (1972). Bereavement. New York: International Universities Press. Parry, G. (1986). Paid employment, life events, social support, and mental

health in working class mothers. Journal ofHealth and Social Behavior,

27, 193-208. Peadin, L. I. (1983). Role strains and personal stress. In H. B. Kaplan

(Ed.), Psychosocial stress: Trends in theory and research (pp. 3-32).

New York: Academic Press. Peadin, L. I., Lieberman, M. A., Menaghan,

E. G., & Mullah,

J. T.

( 1981). The stress process. Journal of Health and Social Behavior, 22,

337-356.

Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal

of Health and Social Behavior, 19, 2-21.

March

1989

American

Psychologist

Popper, K. R. (1959).

Books.

The logic of scientific discovery. New York: Basic

Rappaport, J. (1981). In praise of paradox: A social policy of empow-

erment over prevention. American Journal of Community Psychology,

51, 770-778. Riley, D., & Eckenrode, J. (1986). Social ties: Subgroup differences in

costs and benefits. Journal of Personality and Social Psychology, 51,

770-778.

Rokeach, M. (1973). The nature ofhuman values. New York: Free Press. Rook, K. S, (1984). The negative side of social interaction: Impact on

psychological well-being. Journal ofPersonality and Social Psycholog£,

46, 1097-1108. Rosenbanm, M., & Smira, K. B. (1986). Cognitive and personality factors in the delay of gratification of hemodiaiysis patients. Journal of Per-

sonality and Social Psychology, 51, 357-364. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton,

NJ: Princeton University Press. Sarason, I. G. (1972). Experimental approaches to test anxiety: Attention and the uses of information. In C. D. Spielberger (Ed.), Anxiety: Cur- rent trends in theory and research (Vol. 2). New York: Academic Press. Sarason, I. G. (1975). Test anxiety, attention, and the general problem of anxiety. In C. D. Spielberger & I. G. Sarason (Eds.), Stress and anxiety (Vol. 1, pp. 165-187). Washington DC: Hemisphere. Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978). Assessing the impact of life changes: Development of the Life Experiences Survey.

Journal of Consulting and Clinical Psychology, 46, 932-946.

Sarason, I. G., & Sarason, B. R. (1985). Social support: Insights from assessment and experimentation. In I. G. Sarason & B. R. Sarason

(Eds.), Social support: Theory, research and applications (pp. 39-51).

The Hague, The Netherlands: Nijhoff. Schlenker, B. R. (1987). Threats to identity: Self-identification and social stress. In C. R. Snyder & C. E. Ford (Eds.), Coping with negative life events (pp. 273-322). New York: Plenum Press. SchiSnpflug, W. (1985). Goal-directed behavior as a source of stress:

Psychological origins and consequences of inefficiency. In M. Frese

& J. Sabini (Eds.), The concept of action in psychology (pp. 172-188).

Hillsdale, NJ: Erlbaum.

Selye, H. (1950). The physiology and pathology of exposure to stress.

Montreal: Acta. Selye, H. (1951-1956). Annual report of stress. New York: McGraw- Hill. Solomon, Z., Mikulincer, M., & Hobfoll, S. E. (1987). Objective versus subjective measurement of stress and social support: Combat related

reactions. Journal of Consulting and Clinical Psychology, 55, 577-

583.

Spieiberger, C. D. (1966). Anxiety and behavior. New York: Academic Press.

Spielberger, C. D. (Ed.). (1972). Anxiety: Current trends in theory and

research (Vols. 1-2). New York: Academic Press. Stress: Can we cope? (1983, June 6). Time, pp. 44-52. Suls, J., & Fletcher, B. (1985). Self-attention, life stress and illness: A prospective study. Psychosomatic Medicine, 47, 469--481. Swann, W. B., Jr., & Read, S. J. (1981). Self-verification processes: How

we sustain our self-conceptions. Journal of Experimental Social Psy- chology, 17, 351-372.

Swindle, R. W., Jr., Heller, K., & Lakey, B. (1988). A conceptual reori- entation to the study of personality and stressful life events. In L. H.

Cohen (Ed.), Research on stressful life events: Theoretical and meth-

odological issues (pp. 237-268). Beverly Hills, CA: Sage.

Thibaut, J. W., & Kelley, H. H. (1959). The socialpsychology ofgroups. New York: Wiley.

Thoits, P. A. (1983). Dimensions of life events that influence

distress: An evaluation and synthesis of the literature. In H. B. Kaplan

(Ed.), Psychosocial stress: Trends in theory and research(pp. 33-103 ).

New York: Academic Press. Thoits, P. A. (1987). Gender and marital status differences in control and distress: Common stress versus unique stress explanations. Journal

psychological

of Health and Social Behavior, 28, 7-22.

Trumbull, R., & Appley, M. H. (1986). A conceptual model for the examination of stress dynamics. In M. H. Appley & R. Trumbull

(Eds.), Dynamics of stress: Physiological psychological and social

perspectives. New York: Plenum Press.

523

Vaehon, M. L. S. (1986). A comparison of the impact of breast cancer and bereavement: Personality, social support and adaptation. In S. E. Hobfoll (Ed.), Stress, socialsupport and women (pp. 187-202). Wash- ington, DC: Hemisphere. Vinokur, A., & Seizer, M. L. (1975). Desirable versus undesirable events:

Their relationship to Stl~SSand mental distress. Journal ofPersonality

and Social Psychology, 32, 329-337.

Watson, D., & Tellegen, A. (1985). Toward the structure of affect. Psy-

chological Bulletin, 98, 219-235.

Wellman, B. ( 1981). Applying network analysis

to the study of support.

In B. H. Gottlieb (Ed.), Social networks and social support (pp. l 71-

200). Beverly Hills, CA: Sage. Wieklund, R. A., & Gollwitzer, P. M. (1982). Symbolic self-completion. Hillsdale, NJ: Erlbaum.

524

Wilcox, B. L. (1986). Stress, coping, and the social milieu of divorced

women. In S. E. Hobfoll (Ed.), Stress, socialsupport, and women (pp.

115-133). Washington, DC: Hemisphere. Wilcox, B. L., & Vernberg, E. M. (1985). Conceptual and theoretical dilemmas facing social support. In I. G. Sarason & B. R. Sarason

(Eds.), Social support: Theory, research and applications (pp. 3-20).

The Hague, The Netherlands: Nijhoff. Worden, J. W., & Sobel, H. J. (1978). Ego strength and psychosocial adaptation to cancer. Psychosomatic Medicine, 40, 585-592. Zuckerman, M. (1976). Sensation seeking and anxiety, traits and states, as determinants of behavior in novel situations. In I. G. Sarason & C. D. Spielberger (Eds.), Stress and anxiety (Vol. 3, pp. 141-170). New York: Wiley.

March

1989

American

Psychologist