TERRY HARTIG

Journal of Social Issues, Vol. 59, No. 3, 2003, pp. 611--636

Residence in the Social Ecologyof Stress and Restoration

Terry Hartig∗Uppsala University

Gunn Johansson and Camilla KylinStockholm University

We relate residence to health within a social ecological model of stress and restora-tion. As an isolated setting and in relation to other everyday settings, we discussthe residence in terms of demands, coping resources and responses, and oppor-tunities for restoration. Our model indicates how processes operating above thehousehold level can affect health by modifying the quantity, quality, and distribu-tion of demands, resources, and restoration opportunities within and across thesettings of everyday life, including the residence. We illustrate some of these socialecological dynamics with the case of home-based telework. Concluding, we dis-cuss the utility of the model for environmental interventions intended to alleviatehealth-threatening chronic stress.

How does residence relate to health? Answers to this question reflect on theunderlying conceptions. A conception of health that emphasizes physical, mental,and social well-being rather than the absence of disease and infirmity (World HealthOrganization, 1946) allows for a broader range of outcomes that we might linkwith residence. The same holds when we conceive of residence as more than thelocational, physical, or economic aspects of a person’s housing or neighborhood

∗Correspondence concerning this article should be addressed to Terry Hartig, Institute forHousing and Urban Research, Uppsala University, Box 785, S-801 29 Gavle, Sweden [e-mail:[email protected]].

For their constructive comments on an initial draft of this article, we thank Sherry Ahrentzen, GaryEvans, Roderick Lawrence, Susan Saegert, Susan Smith, and other participants in the workshop on “Theresidential context of health,” held during the European Network for Housing Research conference inGavle, Sweden, June 26–30, 2000. We also thank Irene Frieze and the anonymous reviewers for theirhelpful comments.

611

C© 2003 The Society for the Psychological Study of Social Issues

612 Hartig, Johansson, and Kylin

(the facts of the residence). When we also take into account the activities whichthat person centers in his or her housing and neighborhood (the acts of residence;Hartig & Lawrence, this issue; cf. Hiscock, Macintyre, Kearns, & Ellaway, thisissue; Kemeny, 1992), then we of course have more ways in which to associateresidence with health.

Valid associations between residential and health variables imply processesthat lead from residence to health and/or from health to residence. In this article weconsider how residence relates to health by focusing on such processes, rather thanon specific associations that they may engender. Given the number and variety ofpotential associations allowed by broad conceptions of residence and health, wewould seem to have a bewildering tangle of processes to consider. Yet a modelof how residence relates to health can build on an understanding of how a fewgeneral processes—stress, coping, and restoration—together subsume a varietyof behavioral, psychological, and physiological processes that figure in diversehealth outcomes, for ill or good.

We adopt a social ecological perspective in relating residence to health viastress, coping, and restoration. Researchers working from a social ecological per-spective chart the exchange between people and environment across levels ofanalysis and over time (e.g., Catalano, 1979; Stokols, 1992). In applying this per-spective here, we consider the facts and acts of residence as integral to the spatial,temporal, physical, and social arrangements that people make to manage changein themselves and their environments. People make or come into such arrange-ments not only as individuals, but also together with others, as in households,communities, and societies.

In the first of the following sections we set out some fundamentals of stress,coping, restoration, and their relation to health. Building on these, in the secondsection we sketch a social ecological model of stress and restoration, elaboratingon basic assumptions concerning the spatial and temporal distribution of stress-evoking demands, coping resources and responses, and opportunities for restora-tion. In the third section, we discuss residence as a crucial setting in the ecologyof stress and restoration for individuals and households, acknowledging in this theimportance of social roles. In the fourth section, we use the case of telework toillustrate how processes operating above the level of individuals and householdscan modify demands, coping resources and responses, and restoration opportuni-ties within and across the settings of everyday life, including the residence. In thefinal section, we discuss the model’s practical utility.

Stress, Coping, Restoration, and Health

Most readers will recognize the following fundamentals of stress, coping,and their relation to health. They will find the discussion of restoration a less-familiar though—we trust—helpful complement. In covering these conceptual

Stress and Restoration 613

basics here, we make no pretense of exhaustiveness; we intend only to providesufficient underpinnings for the assumptions of our ecological model and theensuing discussion of stress, coping, and restoration in relation to residence.

Stress

We view stress as a process of responding to an imbalance between de-mands and resources available for meeting those demands (Appley & Trumbull,1986). Whether an imbalance exists depends on appraisal of those demands andresources (Lazarus & Folkman, 1984). Although we usually observe stress re-actions in individuals, and we may observe stressors in the sociophysical en-vironment, the stress process lies in the transaction between person andenvironment.

Individuals face environmental demands of several broad types (Evans &Cohen, 1987). Some, such as daily hassles (Kanner, Coyne, Schaefer, & Lazarus,1981), ambient stressors (Campbell, 1983), and role stressors (Eckenrode & Gore,1990), may be encountered frequently, often at particular times and/or places,and for widely varying durations. Others, such as major life events, may occurrarely, but can retain potency across places and over extended durations. Whetherparticular demands arise rarely or all too commonly, in connection with a partic-ular spatiotemporal location or not, people do not necessarily appraise and copewith them one by one. Rather, people frequently deal with multiple demands,which may have other than additive effects in eliciting stress (Lepore & Evans,1996).

Coping Resources and Responses

To meet these potentially many and varied demands, individuals can nor-mally draw on a variety of resources. These include good health and ample en-ergy; positive beliefs and confidence that things will work out as well as possible(Antonovsky, 1979); competencies in gathering information, solving problems,and getting along with others (Lazarus & Folkman, 1984); emotional, informa-tional, and tangible support from family, friends, and others (Cohen & Wills, 1985);and money and the goods and services it can buy.

The given context or situation influences choices among coping responses(Aldwin, 1994; Holahan, Moos, & Schaefer, 1996). One crucial determinant ofthe choice is the possibility for acting on the conditions eliciting stress. If the personhas the possibility to take action to reduce the stressor exposure, then that personwill more likely do so (instrumental, approach, or problem-focused coping). Ifaction on a stressor is not possible, then the person may try to change the way heor she thinks or feels about the situation (emotional, avoidant, or emotion-focusedcoping). Yet control is not just an objective matter of whether any one person


can do anything to change an exposure. It is also a matter of whether a personbelieves that he or she has the competence needed to change the exposure or wantsto exercise control. Thus, control has to do with both the person and the stressor(e.g., Shapiro, Schwartz, & Astin, 1996).

Restoration

Restoration involves the recovery of functional resources or capabilities di-minished in efforts to meet demands. We may observe restorative processes inindividuals, dyads, or larger groups, with a view to personal as well as social re-sources and functional capabilities. Although we may emphasize restoration hereas the negation of stress, restoration is not limited to recovery from efforts to meetdemands perceived as excessive. A potential for restoration is created wheneverpeople mobilize adaptive resources that can be depleted, whether those resourcesare physical, psychological, or social. Different forms of restoration may proceedin tandem, some requiring more time, some less (e.g., Haynes, Gannon, Orimoto,O’Brien, & Brandt, 1991).

What does restoration require? Take sleep as a basic form of restorative ac-tivity, one through which people renew physical energy and a capacity to remainattentive, among other things. Although people may exercise some control overthe onset, duration, and periodicity of their sleep, during each 24-hour light–darkcycle an evolved circadian timing system will initiate physiological changes thatprepare them for a hard-to-resist period of quiescence (Lavie, 1996). Beyond thoseenvironmental conditions considered appropriate in terms of human evolutionaryadaptedness, as darkness is for sleep, we can think of how well the immedi-ate environment supports the activity, as with having a comfortable place to liedown, and whether sufficient time is available, as for passage through the sev-eral phases of sleep. As stress impedes restoration, stressors such as communitynoise need mitigation lest they disturb sleep (e.g., Carter, 1996). Insomnia fre-quently involves emotionally charged intrusive thoughts, worries, and an inabilityto turn off one’s mind (Espie, 2002); freedom from such internal conditions shouldalso promote restorative sleep. Finally, for someone who sleeps in proximity toanother person, the quality of sleep may depend on that other’s sleep behaviors(e.g., snoring).

As for restoration during waking hours, different accounts converge on two re-quirements, one permitting, the other promoting. First, the person is out of harm’sway and distanced from demands. This permits restoration. Second, the personbecomes engaged by pleasing aspects of the environment or other positive distrac-tions. This promotes restoration. For example, attention restoration theory (Kaplan,1995) holds that we can replenish a capacity for directing attention in situationsthat involve psychological distance from aspects of one’s routines (being away)and effortless attention drawn by objects in the environment or engaged in the


process of making sense of the environment (fascination; cf. Driver & Knopf,1976; Repetti, 1992; Ulrich, 1983).

Stress, Coping, and Restoration Figure in Diverse Health Outcomes

We indicated at the outset that stress, coping, and restoration can figure in asso-ciations between diverse residential and health variables. Chronic stress in particu-lar will have negative impacts on health, more so than short-lived stress reactions totransient demands (e.g., Lepore, 1997). Research has described neuroendocrine,immune, cardiovascular, and musculoskeletal pathways through which chronicstress generates a multiplicity of physical outcomes (Melin & Lundberg, 1997;Steptoe, 1997). For example, studies have linked chronic stress with such dis-parate outcomes as upper respiratory infections (Cohen, Doyle, & Skoner, 1999),very low birth weight (Catalano, Hansen, & Hartig, 1999), and upper back pain(Lundberg et al., 1999).

Research has also described pathways that involve behaviors, such as smoking(e.g., Parrott, 1999) and alcohol consumption (e.g., Sayette, 1999), that appear tohave some emotionally palliative effect in the face of demands, but which can put aperson at greater risk for negative health consequences, not to mention exacerbatestress. In serving emotional coping, such behaviors contrast with instrumental cop-ing behaviors that resolve demands generating stress. Instrumental coping involvesthe exercise of control, which research has linked with stress-related health out-comes in a variety of circumstances (e.g., Karasek & Theorell, 1990; Syme, 1990).Still, emotional coping behaviors like social withdrawal may enable restoration, sothat a person can renew the resources needed to cope instrumentally (cf. Repetti,1992).

Ultimately, stress has negative health effects when coping does not enablesufficient restoration (cf. McEwen, 1998). In coping, then, people must ensureopportunities for restoration. Yet many people choose to forego restoration in theirefforts to cope with everyday demands, particularly related to work. For example,Heslop and colleagues (2002) found self-reported stress inversely associated withsleep duration. They found, also, that men and women who appeared to habituallysleep less than 7 hours each night had a greater risk of death from all causes duringa 25-year follow-up than those who had reported sleeping between 7 and 8 hours.

Conceivably, many people who prioritize work demands over sleep also missopportunities for restoration during waking hours. Research on the health conse-quences of foregoing waking restoration per se must face the challenge of disentan-gling beneficial effects of passive forms of leisure from those of physical activityand exercise (e.g., Lee, Paffenberger, & Hennekens, 1997; Scully, Kremer, Meade,Graham, & Dudgeon, 1998), which may include increased resistance to stress (e.g.,Scully et al., 1998; Steptoe, Kimbell, & Basford, 1998).


The Social Ecology of Stress and Restoration

Stress-eliciting demands, coping resources and response options, restorationopportunities—these do not occur randomly, but to a substantial degree withina spatiotemporal pattern. In articulating a social ecological model of stress andrestoration, we attend to what that pattern involves and how it can change. Attentionto this pattern can aid evaluations of environmental interventions aimed at reducingchronic stress. It might also help in identifying measures that supplement thosedirected at one or another of the locations within it (e.g., the workplace).

We constitute our model here in terms of three assumptions. These situatestress arousal, coping responses, and restoration within the day-to-day flow ofpeople’s activities and in turn within processes operating above the level of indi-viduals and households.

Stress–Restoration Cycles

First, we assume that people cycle through stress arousal and restoration pro-cesses. Stress–restoration cycles involve departure from and return to some con-dition of adaptation. Terms such as “allostasis” (McEwen, 1998; Sterling & Eyer,1988) and “homeostasis” (Cannon, 1932) have been used to describe this condi-tion of adaptation with respect to physiology. The departure that occurs with themobilization of physiological and other adaptive resources potentiates restoration.The magnitude and duration of such departures depend on the characteristics of thedemand or demands (e.g., controllability; Johansson et al., 1978) and the person(e.g., repressive coping style; Weinberger, 1990). The speed and completeness ofrestoration depend on characteristics of the person, the demand (e.g., degree towhich it arouses anger; see Linden, Earle, Gerin, & Christenfeld, 1997; control-lability; see Frankenhaeuser & Johansson, 1986), and strategies for restoration,including entry into an environment with positive restorative value (e.g., presenceof positive distractions). Given incomplete restoration and/or frequent stress reac-tions, the point of departure can shift toward a condition of relatively unsustainableor problematic adaptation (McEwen, 1998; Sterling & Eyer, 1988).

Activity Cycles

Second, we assume that stress–restoration cycles are regulated by activity cy-cles, or regular patterns of activities performed within allocated periods of time.We emphasize daily and weekly activity cycles here. They are both routinizedand planful; people proceed with some understanding about how time will be di-vided among different activities (e.g., Axhausen & Garling, 1992; Garling, Kalen,Romanus, Selart, & Vilhelmson, 1998). Factors such as gender and age shape the


activity cycles that individuals follow (e.g., Chapin, 1974), in part through theirinfluence on the set of social roles that an individual can assume (Pearlin, 1989).

Activity cycles incorporate multiple settings, or places “characterized by re-curring patterns of behavior and by widely recognized place meanings (e.g., func-tional orientation)” (Stokols & Shumaker, 1981, pp. 483–484; see also Schoggen,1989). Settings are arranged and furnished in particular ways to support particu-lar activities by particular people who interact in fulfilling particular roles. Theyalso have temporal features; people occupy them at particular times for particulardurations (cf. Hagerstrand, 1970).

Among meanings or values that people attach to a setting, some may reflecton its stressfulness or capacity for supporting restoration. Thus, settings may cometo have stress or restoration valences (cf. Lewin, 1951). Such valences presumablyhave some grounding in a setting’s characteristic array of social and physical envi-ronmental demands and opportunities for restoration: deadlines at the workplace,and coffee breaks with colleagues; a leaky dishwasher at home, and a chat witha friendly neighbor. At some times and for some people, demands may predomi-nate in a given setting, while at other times and/or for other people, one or morerestoration possibilities may become open.

To the extent they incorporate multiple settings, activity cycles involve move-ments between settings. Movements are relevant in at least four ways. First, somemovements, notably the commute, constitute a type of setting, with attendantstress or restoration valences that vary with such factors as distance, hassle, andpredictability of the commute conditions (e.g., Kluger, 1998; Stokols & Novaco,1981). Second, movements can take on an affective character that reflects therelative affective values assigned to the point of departure and the destination.Anticipating an angry encounter at the destination can make the trip less enjoy-able than anticipating recreation in a pleasant place. Leaving a favorite recreationsetting while thinking of the next day’s tension-filled meetings might also drawone’s feelings down (cf. Hammitt, 1980). Third, individuals’ activity cycles con-verge and diverge with movement. When people converge—at homes, workplaces,pubs, playing fields, and so forth—their stress–restoration cycles also converge,and in ways that color their interpersonal exchanges (e.g., Repetti, 1994). Thoseexchanges may be influenced by the conditions each has left behind as well asthe circumstances that each anticipates entering as their respective activity cyclesproceed. Fourth, some settings become linked through an opposition of stress andrestoration valences that is reinforced with alternations between them (e.g., cityvs. country; Hartig, 1993).

Higher-Level Processes

Third, we assume that social, economic, technological, and other higher-levelprocesses influence individuals’ activity cycles. They affect, among other things,


the times of the day, days of the week, and weeks of the year allocated for work andrest; the constellations of settings that individuals move among and the distancesbetween them; the stressfulness and restorativeness of those settings; the frequencyand character of the individuals’ movements among those settings; choices amongtransportation modes for moving among settings; the average level of demandsfaced across settings; and the stress and restoration values assigned to particu-lar settings. This influence may be mediated by social roles, urban planning andenvironmental protection measures, political measures that affect the distributionof wealth, technological innovations, and so forth. It follows that, by influenc-ing activity cycles, processes above the individual and household level influencestress–restoration cycles.

Residence in the Social Ecology of Stress and Restoration:Individual and Household Aspects

Having overviewed some fundamental aspects of stress, coping, restoration,and their temporal and spatial distribution, we turn now to situate residence withinthe social ecology of stress and restoration. In this, and in line with our use of theterm “setting,” we see the residence constituted not only of housing, location, andneighborhood attributes (the facts of the residence), but also of the activities thatpeople perform in and around their housing—daily routines, social interactionswith neighbors, expressions of attachment, and so forth (the acts of residence).

In the following, we indicate some of the reasons why the residence canassume a crucial status within a person’s ecology of stress and restoration. Weelaborate on four aspects of the residence in this regard. The first three involve,respectively, adaptational demands, coping resources and responses, and restora-tion possibilities. The fourth has to do with the location of the residence relativeto other settings in the activity cycles of household members.

We mentioned earlier that factors such as gender and age shape an individual’sactivity cycles through their influence on the set of roles that person can assume.Because the set of roles influences the activities carried out in and around theresidence (and other settings within activity cycles), factors such as gender andage have a bearing on stress, coping, and restoration in relation to residence. Toacknowledge the influence of such factors and the importance of social roles, weend this section by commenting on working mothers in dual-income families.

Adaptational Demands in Residence

Demands faced in the residence differ from those faced in other settings inquantitative and qualitative respects. Because demands can disallow or underminerestoration, in the following we implicitly give an account of possible constraintson restoration.


People ordinarily and routinely perform basic, obligatory personal and house-hold activities in and around the residence, including cooking, eating, cleaning, lau-ndry, care for others, repairs, grooming, and waste disposal (Robinson & Godbey,1997). They may enjoy some of these activities, or they may only endure thembecause they simply must be done (see, e.g., Michelson & Tepperman, this issue).Whether an activity offers enjoyment or evokes stress, planning and executing itrequires coordination with other activities, which itself constitutes a demand.

In conjunction with a person’s activities, the locational, physical, and socialcharacteristics of the housing and/or its surroundings will influence the likelihoodof frequent and severe daily hassles (e.g., troublesome neighbors, concerns aboutaccidents, fears of confrontation); the intensity and duration of exposure to ambientstressors such as air pollution and noise (e.g., Evans, Jacobs, Dooley, & Catalano,1987); and the likelihood of experiencing catastrophes such as floods, earthquakes,and toxic accidents (e.g., Margai, 2001).

Some stressful events acquire or gain potency because of a link to the residence.Some catastrophic events damage physical, psychological, social, and economicresource values that people attach to their homes. Thus, adaptational demandsincrease suddenly and dramatically just as important coping resources are under-mined (cf. Brown & Perkins, 1992). Similarly, some events—a child leaving home,the death of one’s spouse—involve household members that one may depend on forsupport. Life events like relocation, separation, and divorce may involve changesboth in the location of the residence and in the people associated with it.

Stokols’ (1976) distinction between primary and secondary environments alsospeaks to the added potency that stressors may acquire in the residential context. Aperson “spends much of his time, relates to others on a personal basis, and engagesin a wide range of personally important activities” in primary environments, suchas residential settings (p. 73). Secondary environments “are those in which one’sencounters with others are relatively transitory, anonymous, and inconsequential”(p. 73). Stokols proposed that people satisfy more of their basic needs (e.g., phys-ical comfort, protection-dependency, love and affection, dominance) in primaryenvironments. His analysis, which focuses on crowding, also hypothesizes strongerstress responses in primary environments. The loss of control that he views as aprecipitant of crowding stress should typically not occur in a primary environmentand can negatively affect the fulfillment of a greater variety of needs.

Coping Resources and Responses in Residence

We can also distinguish the residence from other everyday settings in terms ofcoping resources and responses. As just indicated, many people will hold relativelyhigh control expectations with regard to residence. Possibilities for exercisingcontrol can enable instrumental coping measures directed at the residence itself.These can include, for example, modifications to reduce stressor exposures and/or


improve restoration possibilities (e.g., adding insulation to reduce sound intru-sions) and community activism to prevent or reduce stressor exposures and/or pre-serve or improve restoration possibilities (e.g., Brown & Masterson-Allen, 1994).

If measures directed at the present residence do not suffice, then a house-hold may have the option to relocate. Certainly, involuntary relocation can evokedeep and lasting strain (e.g., Brown & Perkins, 1992). Yet an inability to relo-cate can also impair well-being, as when residents cannot escape stressors, suchas crowding, possibly brought on by changes in the household (e.g., Stokols& Shumaker, 1982; Wolpert, 1966). When searching for a new residence, peo-ple commonly base their evaluation of alternatives in part on attributes of thehousing and neighborhood in and of themselves, aside from matters of rela-tive location (cf. Lindberg, Hartig, Garvill, & Garling, 1992; Garling & Friman,2002). Some of those attributes refer to ambient stressors and daily hassles. Arethere enough rooms? Enough places for parking? What are the noise sources?In weighing such attributes when seeking a better alternative than the presentone, people reveal not only an attempt to cope with present demands, but alsoan attempt to prevent future stressor exposures. Similarly, in evaluating alterna-tives, people will commonly consider some attributes that refer to future restora-tion possibilities. Do the windows afford pleasing views? Is it possible to havea garden?

The set of alternative residential locations depends on the material and personalresources available to the household. For some people, notably those who live inpoverty, all available alternatives may involve significant exposures to stressorsor hazards (e.g., Evans & Kantrowitz, 2002), and they may lack high-qualitypossibilities for restoration. Still, many households spend large proportions of theirincome on even poor-quality housing (Saegert & Evans, this issue). Clearly, wecan view simply maintaining access to housing as a fundamental form of coping.Relocation into homelessness translates into increased exposure to environmentalconditions that may directly and indirectly affect health in a variety of ways (e.g.,Matte & Jacobs, 2000).

In a basic sense, “homelessness” means not having housing. Yet a commondistinction drawn between “housing” as a physical construction and “home” asa social and psychological construct (Tognoli, 1987) encourages a deeper lookwhen considering what the home means in coping beyond the provision of shelter.Meanings commonly attached to the home include security, control, permanenceand continuity, relatedness, and refuge from the outside world (e.g., Despres, 1991;Somerville, 1997), all of which have a plausible connection to coping resourcesor responses. For example, the “relatedness” meaning of the home seems to re-flect the fact that many people center a substantial proportion of their significantrelationships (and so their social resources) in the residence. To take another ex-ample, consider Saegert’s (1985) observation that the notion of home as refuge orhaven implies a desire to escape from the outside world. In this, the home serves


emotional coping; it exists as a place to retreat to from things that one can’t domuch about for the time being. Thus, various acts of residence confer upon housingcoping values that become captured in the notion of “home.”

Restoration Opportunities in Residence

Of all settings organized in regular activity cycles, the residence normally pro-vides the most significant opportunities for regular restoration. Sleep is an essentialrestorative activity, one sanctioned in the residence and for which the residenceis usually best furnished relative to other settings. For U.S. census purposes, thedefinition of residence has since 1790 rested in part on the notion that a personsleeps there most of the time (U.S. Census Bureau, 2002).

People also typically spend the largest part of their leisure time in their res-idence (e.g., Glyptis & Chambers, 1982), and they frequently equip it to supportleisure activities. However, while leisure activities in and around the residencemay aid restoration (cf. Kelly & Kelly, 1994), those activities appear to varyconsiderably in restorativeness. It appears that the greatest part of leisure timespent at home is spent watching television, an activity that may yield relativelylittle enjoyment in comparison to other passive leisure activities (e.g., reading,conversing) or to more active leisure (Robinson & Godbey, 1997). However,the choice of an activity like watching TV may reflect on constraints imposedon other activities at a given time. For example, for people who return homefrom work in the evening, a lack of light (and with it increased concern forpersonal safety) may hinder outdoor activities in the neighborhood (e.g., Keane,1998).

For one person, participation in and the restorative quality of a leisure activitymay hinge on the activities of others in the residence. Repetti’s (1989) study ofend-of-workday interactions between male air traffic controllers (ATCs) and theirwives illustrates this point. Repetti found that support from the wives moderatedthe relationship between the husbands’ workload during the day and their socialbehavior at home; when support was high, the ATCs were more likely to withdrawafter a heavy workday (see also Bolger, DeLongis, Kessler, & Wethington, 1989).Repetti viewed the ATCs’ withdrawal as an effective short-term strategy for re-ducing arousal; the ATCs could more freely engage in activities that drew theirattention from the day’s work (see also Repetti, 1992; cf. Kaplan, 1995; Driver &Knopf, 1976; Saegert, 1985).

Some residential activities may bear on both future and present restoration.Household members and neighbors can improve their own and others’ futurerestoration possibilities by modifying their homes and common spaces in theirneighborhood (cf. Cooper Marcus & Sarkissian, 1986). Whether alone or together,participation in such activities (e.g., interior decorating, gardening) may itselfyield restorative benefits. However, we should not ignore the risk that an activity


may, over time, shift from pleasant diversion into obligatory undertaking (see, forexample, the discussion of lawn care in the USA by Jenkins, 1994).

Residential Location in Relation to Other Settings

The residence can also assume a crucial status in a person’s ecology of stressand restoration because of its spatial and temporal location relative to other settingsorganized in the given set of activity cycles (for that person alone or for the multi-person household to which that person belongs). In choosing among residentialalternatives, people may have latitude to consider how they are located relativeto other settings encompassed by the given set of activity cycles—workplaces,schools, shopping, and so on. Their choice will have implications for the fre-quency, duration, length, and character (e.g., stressfulness) of movements amongsettings and the manner in which the activity cycles of household members (ifmore than one) converge and diverge. As with the weighting of attributes of theresidential alternatives in and of themselves (e.g., of the extent to which theyinvolve stressor exposures), the weighting of relative location attributes can in fa-vorable circumstances represent instrumental coping that is both reactive (i.e., thehousehold moves away from some undesirable constellation of locations) and pre-ventive (i.e., the household moves to a constellation of locations that imposes fewerdemands and/or offers better restoration opportunities (cf. Golledge & Stimson,1997; Wolpert, 1966).

Another aspect of the location decision reflects, also, on the crucial status ofthe residence. Households can ordinarily exercise greater control in locating theresidence relative to other settings in activity cycles than they can in locating thosesettings at convenient distances from the residence. Many households will find iteasier, for example, to move the residence closer to the workplace or schools thanto move those settings closer to the residence.

Similarly, members of households may have more control at the residencerelative to other settings with respect to modifying furnishings, the structure, andperhaps some features of the immediate surroundings to reduce stressor exposuresand enhance restoration possibilities (here, though, we must acknowledge the dis-tinction between homeowners and renters, and its relation to socioeconomic status;see Hiscock et al., this issue). Further, members of households may have (and, atleast with respect to adults, are expected to have) better possibilities for arrang-ing schedules so they conform with the scheduling dictates of other settings, thusmanaging various capability, coupling, and authority constraints (cf. Hagerstrand,1970). In short, as a setting, the residence has physical, social, spatial, and tem-poral features relatively amenable to manipulation by household members. Assuch, it serves as a setting on which household members can concentrate effortsto balance or coordinate demands, resources, and restoration possibilities acrosssettings within one or more activity cycles.


The Influence of Social Roles: The Case of Working Mothersin Dual-Income Families

Just how crucial residence becomes in a person’s ecology of stress and restora-tion depends to a significant degree on characteristics of that person. Gender, age,and other characteristics are determinants of the social roles that a person assumes.As those roles are manifest in activity cycles, they are key determinants of stress,coping, and restoration in relation to residence.

Social roles shape activity cycles in numerous ways (cf. Pearlin, 1989). Theyspecify activities and settings. They specify with whom certain activities willbe carried out, when, how frequently, and for how long. They shape a person’sexpectations regarding, for example, control and responsibility regarding givenactivities and settings. They shape the constellation of settings encompassed in anactivity cycle, as well as characteristics of the convergences with other people inthose settings. And so on.

We can use the case of working mothers in dual-income families to illustratesome of the ways in which social roles can, through their influence on activitycycles, affect stress, coping, and restoration in relation to residence. Over the pastseveral decades, in Sweden, the United States, and elsewhere, the proportion ofmarried or otherwise partnered women who work outside the home while theirchildren are still young has increased tremendously (e.g., Barnett & Hyde, 2001;Matthews & Rodin, 1989). Yet neither the mother’s role as primary caregiver tochildren nor the division of domestic labor in two-parent households has changedas quickly. Women still assume more responsibility for domestic work, includingchild care (e.g., Lundberg, 1996). This cultural lag (Michelson, 1988) has had avariety of consequences.

One notable consequence is a gender differential in total workload. As thenumber of children in a household increases, the combined burden of paid andunpaid work also increases, and more so for mothers than for their partners (e.g.,Lundberg, Mardberg, & Frankenhaeuser, 1994). Consider what this means in termsof activities within the residence. Time-use studies provide details of how thespatial and temporal distribution of domestic demands and restoration possibilitiescan differ between working mothers and fathers. Ahrentzen, Levine, and Michelson(1989) examined the time that adult members of Canadian households spent in thekitchen, bathroom(s), bedroom(s), and living room, alone or with other familymembers, over a 24-hour period during the workweek. Patterns of activity andco-occupancy within rooms of the house varied with gender in those couples withboth members employed full-time. During their time at home, full-time employed(FTE) married women in the sample spent twice as much time alone in the kitchenand more time with children in bedrooms and bathrooms than did the FTE marriedmen. Looking to the combination of location and activity, the men spent more timein passive leisure in the living room than did the women. For their part, the women


spent more time working in the kitchen and more time caring for children in thebedrooms and bathrooms.

In line with this result, Rydenstam’s (1992) analyses of Swedish time-budgetdata revealed that women’s leisure time was divided into shorter periods thanmen’s, and their leisure activity episodes were more frequently broken off to dowork in the home. As with Ahrentzen et al.’s (1989) results, these findings suggestthat women in dual-income families with children at home have their attempts atrestoration in the home frustrated more frequently than do their partners, just asthey take on a greater burden of domestic work.

A relatively large burden of domestic work also affects the way in whichsome working mothers approach the convergence in the home after the workday.While men may prepare to unwind when heading home after work, some womengear up for the next “shift” (Hochschild, 1989). For example, Frankenhaeuser andcolleagues (1989) reported that female managers in their Swedish sample did notunwind like the male managers or clerical workers of both genders; their systolicblood pressure after work was like that measured during work, and they secretedmore norepinephrine, signaling preparation to deal with further demands (seealso Lundberg & Frankenhaeuser, 1999). They also experienced greater conflictbetween demands from their paid jobs and demands from other duties.

Such conflicts exemplify the reach that role stressors can have across differentsettings. They often stem from constraints that role-related activities in one settingimpose on activities related to some other role in some other setting (see, e.g.,Eckenrode & Gore, 1990). Moreover, in that roles imply interactions with particularpeople in particular settings at particular times, the constraints that cause conflictsfor one person can originate in an activity cycle of someone else. A workingmother, for example, may have to leave work to pick up a child at day care byclosing time, even though an important meeting at her workplace has not yet ended.In general, we would find it difficult to separate the daily activity cycle of a workingmother in a dual-income family from the activity cycles of her young children andpartner. Depending on its location relative to other major settings in the set of theiractivity cycles (her workplace, his workplace, their day care or schools, etc.), theresidence can assume decisive importance for a working mother’s experience ofrole conflicts, and for her ecology of stress and restoration in general.

Residence in the Social Ecology of Stress and Restoration:The Interplay of Household and Higher-Order Processes

To this point we have focused on individual and household aspects of resi-dence, indicating a variety of reasons why that setting can attain crucial importancein a person’s ecology of stress and restoration. Those reasons involve distinctionsbetween the residence and other settings in terms of, for example, demands from


obligatory activities; control expectations; access to social support; suitability forsocial withdrawal; opportunities for sleep and leisure; and possibilities for improv-ing relative locations within a constellation of settings. We have also discussed howthe social roles that different members of a household assume will help determinejust how crucial the residence becomes as a setting in their respective ecologiesof stress and restoration. Social roles weight the residence in importance by, forexample, specifying activities performed, time spent, and social contacts madethere, as well as relations between it and other settings.

What we have not done to this point is directly examine how social, economic,cultural, and other higher-order processes can affect residence through their influ-ence on activity cycles. Mediated, for example, by social roles or technologicalinnovations, such processes can modify the constellation of settings encompassedby the activity cycles of the one or more individuals comprising a household, aswell as the quantity, quality, and distribution of demands, coping resources andresponses, and restoration opportunities within and across those settings.

Although we have not yet directly examined the influence of processes operat-ing above the household level, we have touched upon that influence. In discussingworking mothers in dual-income families, we noted that a cultural lag betweenchange in women’s work roles and change in their domestic roles is manifest ina gender difference in total workload in some households. However, rather thanattending to how higher-order processes contributed to that cultural lag in the firstplace by opening up work roles for women, we simply took it as a given, andfocused instead on what the greater total workload of working mothers means interms of their stress, coping, and restoration in relation to residence.

In the following, we shift our perspective to how the operation of higher-orderprocesses can translate into significant changes in activity cycles. Rather thanfocusing on how a given set of circumstances is expressed in activity cycles, weconsider what happens when new means to manage those circumstances emergefrom higher-order processes. In this, we do not give a detailed account of theprocesses generating those means. Instead, we look at how the adoption or rejectionof those means reflects concerns about stress, coping, and restoration in existingversus potential activity cycles. We also consider how adoption of those meansaffects stress, coping, and restoration in the residence, and in turn feeds back intohigher-order processes.

We use the case of telework as a window into these phenomena. In somecountries, the shift from industrial production, increasing suburbanization, andthe tightening congestion of urban transportation systems have over the past fewdecades generated both new allowances and new incentives for moving workfrom dedicated workplaces into residences. Over the same period, advances ininformation and communication technologies (ICT) have provided some workerswith the means to perform a significant fraction of their paid work in locationsoutside the employer’s main facilities. These technological innovations thus give


some individuals and households the means to redistribute work and nonworkactivities across times and settings within one or more activity cycles.

We define “telework” as ICT-supported paid work in the residence, a satelliteoffice, or a neighborhood center. Some researchers equate telework with telecom-muting (for a review, see Ellison, 1999). Others assign a narrower meaning totelecommuting, specifying the substitution of ICT-supported remote work for com-muting (see Shin, Sheng, & Higa, 2000). Still others further specify that the workbe performed during normal office hours (e.g., Belanger, 1999). In referring totelework rather than telecommuting, we explicitly acknowledge motivations asidefrom reduced commuting, as well as paid ICT-supported remote work that occursboth within and outside of normal office hours. We do, however, draw on studiesof telecommuting narrowly defined, including some by transportation researchersand geographers whose interest in activity spaces matches our interest in activitycycles and the settings they encompass.

In line with transportation research on telecommuting, we expect that aworker’s decision to telework rests in part on possibilities for reducing demands,maintaining or improving access to coping resources, and/or maintaining or im-proving opportunities for restoration. In deciding, the worker considers the relativelocations of the residence, other available workplaces, other settings in the activitycycle, and, possibly, settings in the activity cycles of others, such as a spouse and/orchildren. In view of the potentially greater salience and/or potency of stressors en-countered in the residence and the relative importance of the home in restoration,we also expect that some people will resist the entry of paid work into the home.Furthermore, we recognize that those who do choose to telework risk not onlynew demands at home, but also quantitatively and qualitatively diminished oppor-tunities for restoration. As an adaptive strategy, then, telework itself may requirefurther adaptations, and not only on the individual and household levels. We takeeach of these points in turn in the following.

Deciding to Telework

A worker’s decision to telecommute has been framed by Mokhtarian andSalomon (1994) as an outcome of the interplay of facilitators, constraints, anddrives. Facilitators include employer support and independence of job tasks fromlocation. Conversely, external and internal constraints in their model include a lackof employer support, the location dependency of job tasks, and needs for socialand/or professional interactions at the workplace. For a variety of reasons, then,not all workers have the conditions in place to make telework a viable option.

Among drives Mokhtarian and Salomon (1994) include the desire to get morework done, to reduce stress associated with the social environment at the mainworkplace, to gain greater control over the physical work environment, to satisfypersonal orientations toward independence or withdrawal from other people, to


have more time with the family or more flexibility in managing family duties(and so reduce role conflicts), to have more flexibility in arranging time so thatone can more easily accommodate preferred leisure activities, and to avoid a longor stressful or expensive commute. Clearly, then, reducing stress and facilitatingrestoration figure in their model (see also Salomon & Mokhtarian, 1997).

As the reasons for teleworking vary across people, so also does the representa-tion of the home and household and issues of stress and restoration in the decisionprocess. For example, Mokhtarian, Bagley, and Salomon (1998) report variationsin drives across gender and household type (i.e., presence of children). Amongthe San Diego municipal employees in their sample, women with children gavesubstantially higher importance ratings to family benefits (e.g., taking care of adependent, spending time with family) as motivations for telecommuting than didmen with children, whereas women and men alike gave relatively low importanceratings to family benefits when no children were present in the household. Theseresults speak to how working mothers use telework to resolve role conflicts likethose discussed earlier.

Travel diary data provide evidence of just how consideration of the relativelocations of the residence, workplace, and other settings in activity cycles mightfigure in the decision to telework. The State of California Telecommuting PilotProject involved waves of data collection before and after a group of state em-ployees began to telecommute (JALA Associates, cited in Saxena & Mokhtarian,1997). In each data collection wave the telecommuters completed a travel diary onthree successive days, at least one of which was a telecommuting day. Using dataon all trips aside from those between residence and work, Saxena and Mokhtarian(1997) studied the impact of telecommuting on the telecommuters’ activity spaces,or the set of locations traveled to in day-to-day activities. They found that tripsclustered more tightly around the residence on telecommuting days; trip ends werecloser to the residence and more evenly distributed around it (rather than being dis-proportionately oriented toward the work location, indicating activities performedbetween home and work). These results inform previously published project re-sults (summarized by Saxena & Mokhtarian, 1997) showing a reduction in tripsdue to telecommuting. They illustrate how a decision to telework might aim toreduce everyday demands not only by reducing commuting, but also by reducingdistances traversed on other household trips.

Resistance to Telework

Again, though, not all people have the option to telework. And of those who dohave the option, not all decide to do so. We can now look back on enthusiastic butunfulfilled projections of widespread teleworking as a substitute for commuting(Salomon, 1998). In Sweden, relatively few gainfully employed people regularlywork at a location other than the ordinary workplace during regularly scheduled


working hours; surveys give an estimate of 8% or less for the years 1997, 1998, and1999 (Vilhelmson & Thulin, 2001). Of the 8% who reported working remotely, notall made use of computers and Internet links in their work, despite a high degreeof access to and experience with computers in the population.

Why the gap between projections and reality? Among other factors, Salomon(1998) points to a lack of sensitivity to the social complexities surrounding theadoption of telework. Modeling telework adoption as an alternative to commutingrequires specifying the person or people whose behavior is being modeled. Adop-tion of telework is, as we alluded to earlier, the joint decision of at least two actors,the employer and employee. We have already mentioned lack of employer supportas a constraint on telework adoption. Other constraints may be imposed by theemployee and/or other household members, with a view to resisting the entry ofpaid work practices into the home and in doing so preserving coping and restora-tion functions. Among constraints on teleworking in their model, Mokhtarianand Salomon (1994) list some related to the unsuitability of the residence as aworkplace (e.g., lack of appropriate space, conflicts with household members),as well as a loss of benefits from the commute. For example, the commute mayhelp some people accomplish the transition between work and family roles (seeEllison, 1999).

Telework as a Stimulus to Further Adaptations on Multiple Levels

Although telework may afford individuals and households new means to ad-dress everyday demands, in practice it imposes demands of its own. Certainly, somepeople appreciate the added scheduling flexibility and closer integration of workand home life, as seen, for example, in the use of household tasks as restorative in-terludes in concentration-intensive work (Hytter, 1995). Yet others are troubled byblurred work–home boundaries (e.g., Hill, Hawkins, & Miller, 1996). With workclose at hand, the home may lose value as a place of refuge (Ahrentzen, 1989).

This blurring of boundaries between work and family can be seen in theamount of time spent working and its disposition over the day (Allvin et al.,1998). Analyses of time-use data from Sweden and Canada (Michelson, 2000)are informative here, even though they focus on home-based work in general, andnot only telework. A key finding is that those who worked from home intensively(≥ 4 hr/day) spent more time on paid work than those who did not work fromhome so intensively, a result that contrasts with the notion that teleworking freesup time by eliminating commutes. Moreover, the home-based workers had a greaternumber of distinct work episodes than those not working at home on the day ofthe survey. An evening work episode also distinguished the home-based workers.

Such a work pattern may engender chronic stress arousal. In a recent studyreported by Lundberg and Lindfors (2002), the psychophysiological arousal ofteleworkers was studied during office work, during telework at home, and on a


work-free day at home. Physiological arousal was higher during telework andoffice work than during relaxation in the home. Yet telework was associated withlower cardiovascular arousal than work at the office, perhaps because the workerscould choose to perform fewer and different work tasks than they would in theoffice. However, an elevated level of epinephrine secretion was observed amongmales in the evening after teleworking as compared to the evening after work atthe office. Daily diary entries suggest that the elevation resulted from continuedwork after normal working hours.

Individuals and households make various temporal, spatial, and interpersonaladaptations to address the problem of blurred work–family boundaries and other-wise improve the viability of telework. Needs for a suitable place and uninterruptedtime to carry out the work underlie the adaptations that households make. Theseinvolve, for example, physical changes or arrangements in the residence, schedul-ing, the restriction of the teleworker’s accessibility to other household members,and modification of interpersonal behavior norms (e.g., Ahrentzen, 1989, 1990;see also Gurstein, 1991). A separate room appears to have particular value. Forexample, in a study of Swedish teleworkers, those who could use a separate roomfor work at home perceived less spatial overlap between work and nonwork lifeand evaluated that overlap more positively than those who used only part of a roomfor their telework (Hartig, Kylin, & Johansson, 2002).

The adaptations required by telework extend beyond the household level. Asan example of feedback from households to employers regarding difficulties withtelework, another alternative has emerged: telecommuting centers located closerto the residence than to the employer’s main facilities. Preference for home-basedversus center-based commuting appears to reflect on how telework is combinedwith family life; some evidence suggests that teleworkers with very young children(less than 2 years old) will prefer working at home, while those with somewhatolder children will prefer the opportunity afforded by a center for ready escapefrom (and return to) household distractions (Stanek & Mokhtarian, 1998).

We might project other ways in which increasing rates of home-based tele-working would feed back into high-order processes. Consider the impacts on hous-ing markets and the provision of housing (e.g., Handy & Mokhtarian, 1995). Forexample, people who currently telework or who intend to telework may, in plan-ning a residential relocation or remodel, prioritize spatial and technological re-quirements for work at home. The marketing of new and remodeled homes inSweden already calls attention to provisions made for high-speed Internet access,though not necessarily for home-based telework in particular.

To take another example, increasing rates of home-based teleworking couldtranslate into greater spatial dispersion of residences (e.g., Helling & Mokhtarian,2001). Swedish families with young children have expressed a greater preferencefor housing at a greater distance from the city center than other life cycle groups(Lindberg et al., 1992), and telework may support such a preference. Although


such dispersion might satisfy people’s desires to escape urban stressors or threatsto the safety of their children, the commutes to work they would then make, thoughfewer, would involve longer distances, as would other household trips.

In closing this illustration of the interplay between household and higher-order processes, we note that telework adoption has a paradoxical connectionwith the cultural lag that generates role conflicts for working mothers in dual-income families. Those women may telework to resolve work–family conflicts,as suggested by Mokhtarian et al.’s (1998) results. Yet, in doing so, they mayhelp perpetuate a gendered division of labor in which women continue to shoulderprimary responsibility for child care and domestic work (Ellison, 1999; Michelson,2000).

Practical Implications

At the outset of this article, we described the facts and acts of residence asintegral to the arrangements that people make to manage change in themselves andtheir environments. As our discussion of telework illustrates, those adaptive ar-rangements can undergo revision with new opportunities for ameliorating existingdemands and reducing their negative consequences, not least chronic stress. Suchopportunities can be created or capitalized upon by individuals and households, orby others acting on their behalf, such as employers, planners, and politicians.

We see at least four ways in which our social ecological model can help guidea search for opportunities for ameliorating chronic stress. First, it encourages asearch that goes beyond new means for reducing particular demands themselvesto include new means for increasing access to coping resources or better promot-ing restoration (which in turn means increased availability of coping resources;Hartig, 2001). Second, it directs attention to settings other than that in which theperson encounters the demands of focal concern. So, for example, a person mayseek to offset the impact of intractable demands in one setting by enhancing therestorativeness of another setting. Third, the model encourages a search for waysto intervene on activity cycles considered as a whole. For example, this could in-volve changing the constellation of settings and/or the relative spatial and temporallocations of those settings, as we saw in the case of telework. It could also involvesuspending some subset of demanding daily activities, as with an annual vacationtaken away from home (cf. Westman & Eden, 1997). Fourth, the model makes clearthat opportunities for ameliorating chronic stress can be sought in the ways justmentioned not only on the individual or household level, but also on higher spa-tial and social or organizational levels (cf. Stokols, 1996). As a historic example,Frederick Law Olmsted (1870) viewed parks and tree-lined promenades as plan-ning provisions that placed opportunities for restorative interludes in the paths ofharried city dwellers. Similarly, vacation legislation in Sweden and other West-ern European countries exemplifies a population intervention that allows many


people an extended respite from demanding daily routines (Statens OffentligaUtredningar, 1988; Westman & Eden, 1997).

In conjunction with its utility for identifying intervention alternatives, oursocial ecological model can aid in evaluating stress interventions that involve theresidence. For one, it encourages comparisons of the effectiveness of measuresoutside the residence (e.g., to reduce demands in the workplace) and measureswithin the residence (e.g., to enhance its restorative quality). For another, it en-courages questions about potential negative consequences of interventions. In thecase of telework, innovations in ICT opened up new possibilities for people tocoordinate demands, resources, and restoration possibilities across settings withintheir activity cycles. Yet, for some people, telework has had the unanticipated ef-fect of diminishing the value of the residence as a place for restoration. Such aneffect may translate into greater mobility during leisure time (cf. Fuhrer, Kaiser,& Hartig, 1993). Anticipating negative effects on home life, other people havechosen to not telework, thereby also weakening the hopes of governmental entitieswho saw in telework a way to ease the load borne by transportation systems.

References

Ahrentzen, S. (1989). A place of peace, prospect, and . . . a P.C.: The home as office. Journal ofArchitectural and Planning Research, 6, 271–288.

Ahrentzen, S. (1990). Managing conflict by managing boundaries: How professional homeworkerscope with multiple roles at home. Environment and Behavior, 22, 723–752.

Ahrentzen, S., Levine, D. W., & Michelson, W. (1989). Space, time, and activity in the home: A genderanalysis. Journal of Environmental Psychology, 9, 89–101.

Aldwin, C. (1994). Stress, coping, and development. New York: Guilford.Allvin, M., Aronsson, G., Hagstrom, T., Johansson, G., Lundberg, U., & Skarstrom, E. (1998). Granslost

arbete eller arbetets nya granser [Boundaryless work or work’s new boundaries] (Rapport1998:21, Arbete och Halsa Vetenskaplig Skriftserie). Solna, Sweden: Arbetslivsinstitutet.

Antonovsky, A. (1979). Health, stress, and coping. San Francisco: Jossey-Bass.Appley, M. H., & Trumbull, R. (1986). The dynamics of stress: Physiological, psychological and social

perspectives. New York: Plenum.Axhausen, K. W., & Garling, T. (1992). Activity-based approaches to travel analysis: Conceptual

frameworks, models, and research problems. Transport Reviews, 12, 323–341.Barnett, R. C., & Hyde, J. S. (2001). Women, men, work, and family. An expansionist theory. American

Psychologist, 56, 781–796.Belanger, F. (1999). Workers’ propensity to telecommute: An empirical study. Information and Man-

agement, 35, 139–153.Bolger, N., DeLongis, A., Kessler, R. C., & Wethington, E. (1989). The contagion of stress across

multiple roles. Journal of Marriage and the Family, 51, 175–183.Brown, B. B., & Perkins, D. D. (1992). Disruptions in place attachment. In I. Altman & S. M. Low

(Eds.), Place attachment (pp. 279–304). New York: Plenum.Brown, P., & Masterson-Allen, S. (1994). The toxic waste movement: A new type of activism. Society

and Natural Resources, 7, 269–287.Campbell, J. (1983). Ambient stressors. Environment and Behavior, 15, 355–380.Cannon, W. B. (1932). The wisdom of the body. New York: Norton.Carter, N. L. (1996). Transportation noise, sleep, and possible after-effects. Environment International,

22, 105–116.


Catalano, R. (1979). Health, behavior and the community: An ecological perspective. New York:Pergamon.

Catalano, R., Hansen, H.-T., & Hartig, T. (1999). The ecological effect of unemployment on the inci-dence of very low birthweight in Norway and Sweden. Journal of Health and Social Behavior,40, 422–428.

Chapin, F. S. (1974). Human activity patterns in the city. London: Wiley.Cohen, S., Doyle, W. J., & Skoner, D. P. (1999). Psychological stress, cytokine production, and severity

of upper respiratory illness. Psychosomatic Medicine, 61, 175–180.Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological

Bulletin, 98, 310–357.Cooper Marcus, C., & Sarkissian, W. (1986). Housing as if people mattered: Site design guidelines for

medium-density family housing. Berkeley, CA: University of California Press.Despres, C. (1991). The meaning of home: Literature review and directions for future research

and theoretical development. Journal of Architectural and Planning Research, 8, 96–115.

Driver, B. L., & Knopf, R. C. (1976). Temporary escape: One product of sport fisheries management.Fisheries, 1, 24–29.

Eckenrode, J., & Gore, S. (1990). Stress and coping at the boundary of work and family. In J. Eckenrode& S. Gore (Eds.), Stress between work and family (pp. 1–16). New York: Plenum.

Ellison, N. B. (1999). Social impacts: New perspectives on telework. Social Science Computer Review,17, 338–356.

Espie, C. A. (2002). Insomnia: Conceptual issues in the development, persistence, and treatment ofsleep disorders in adults. Annual Review of Psychology, 53, 215–243.

Evans, G. W., & Cohen, S. (1987). Environmental stress. In D. Stokols & I. Altman (Eds.), Handbookof environmental psychology: Vol. 1 (pp. 571–610). New York: Wiley.

Evans, G. W., Jacobs, S. V., Dooley, D., & Catalano, R. (1987). The interaction of stressful life eventsand chronic strains on community mental health. American Journal of Community Psychology,15, 23–34.

Evans, G. W., & Kantrowitz, E. (2002). Socioeconomic status and health: The potential role of envi-ronmental risk exposure. Annual Review of Public Health, 23, 303–331.

Frankenhaeuser, M., & Johansson, G. (1986). Stress at work: Psychobiological and psychosocialaspects. Applied Psychology: An International Review, 35, 287–299.

Frankenhaeuser, M., Lundberg, U., Fredrikson, M., Melin, B., Tuomisto, M., Myrsten, A.-L., Hedman,M., Bergman-Losman, B., & Wallin, L. (1989). Stress on and off the job as related to sex andoccupational status in white-collar workers. Journal of Organizational Behavior, 10, 321–346.

Fuhrer, U., Kaiser, F. G., & Hartig, T. (1993). Place attachment and mobility during leisure time.Journal of Environmental Psychology, 13, 309–321.

Garling, T., & Friman, M. (2002). A psychological conceptualization of residential choice and satisfac-tion. In J. I. Aragones, G. Francescato, & T. Garling (Eds.), Residential environments: Choice,satisfaction, and behavior (pp. 55–80). London: Bergin & Garvey.

Garling, T., Kalen, T., Romanus, J., Selart, M., & Vilhelmson, B. (1998). Computer simulation ofhousehold activity scheduling. Environment and Planning A, 30, 665–679.

Glyptis, S. A., & Chambers, D. A. (1982). No place like home. Leisure Studies, 1, 247–262.Golledge, R. G., & Stimson, R. J. (1997). Spatial behavior: A geographic perspective. London:

Guilford.Gurstein, P. (1991). Working at home and living at home: Emerging scenarios. Journal of Architectural

and Planning Research, 8, 164–180.Hagerstrand, T. (1970). What about people in regional science? Papers of the Regional Science Asso-

ciation, 24, 7–21.Hammitt, W. E. (1980). Outdoor recreation: Is it a multi-phase experience? Journal of Leisure Research,

12, 107–115.Handy, S., & Mokhtarian, P. (1995). Planning for telecommuting: Measurement and policy issues.

Journal of the American Planning Association, 61, 99–111.Hartig, T. (1993). Nature experience in transactional perspective. Landscape and Urban Planning, 25,

17–36.


Hartig, T. (2001). Guest editor’s introduction [Special issue on restorative environments]. Environmentand Behavior, 33, 475–479.

Hartig, T., Kylin, C., & Johansson, G. (2002). Stress and restoration in the home: A study of tele-work(ers). Manuscript in preparation.

Haynes, S. N., Gannon, L. R., Orimoto, L., O’Brien, W. H., & Brandt, M. (1991). Psychophysiologicalassessment of poststress recovery. Psychological Assessment, 3, 356–365.

Helling, A., & Mokhtarian, P. L. (2001). Worker telecommunication and mobility in transition: Con-sequences for planning. Journal of Planning Literature, 15, 511–525.

Heslop, P., Davey Smith, G., Metcalfe, C., Macleod, J., & Hart, C. (2002). Sleep duration and mortality:The effect of short or long sleep duration on cardiovascular and all-cause mortality in workingmen and women. Sleep Medicine, 3, 305–314.

Hill, E. J., Hawkins, A. J., & Miller, B. C. (1996). Work and family in the virtual office—Perceivedinfluences of mobile telework. Family Relations, 45, 293–301.

Hochschild, A. R. (1989). The second shift. New York: Avon Books.Holahan, C. J., Moos, R. H., & Schaefer, J. A. (1996). Coping, stress resistance, and growth: Con-

ceptualizing adaptive functioning. In M. Zeidner & N. S. Endler (Eds.), Handbook of coping:Theory, research, applications (pp. 24–43). New York: Wiley.

Hytter, K. (1995). Datorn pa koksbordet. En studie av datorstott distansarbete och dess konsekvenserfor individ, foretag och samhalle [The computer at the kitchen table: A study of computer-supported remote work and its consequences for the individual, company, and society]. Lund,Sweden: Sociologiska Institutionen, Lunds Universitet.

Jenkins, V. S. (1994). The lawn: A history of an American obsession. Washington, DC: SmithsonianInstitution Press.

Johansson, G., Aronsson, G., & Lindstrom, B. O. (1978). Social psychological and neuroendocrinereactions in highly mechanised work. Ergonomics, 21, 583–599.

Kanner, A. D., Coyne, J. C., Schaefer, C., & Lazarus, R. S. (1981). Comparison of two modes ofstress measurement: Daily hassles and uplifts versus major life events. Journal of BehavioralMedicine, 4, 1–39.

Kaplan, S. (1995). The restorative benefits of nature: Toward an integrated framework. Journal ofEnvironmental Psychology, 15, 169–182.

Karasek, R., & Theorell, T. (1990). Healthy work: Stress, productivity, and the reconstruction of workinglife. New York: Basic Books.

Keane, C. (1998). Evaluating the influence of fear of crime as an environmental mobility restrictor onwomen’s routine activities. Environment and Behavior, 30, 60–74.

Kelly, J. R., & Kelly, J. R. (1994). Multiple dimensions of meaning in the domains of work, family,and leisure. Journal of Leisure Research, 26, 250–274.

Kemeny, J. (1992). Housing and social theory. London: Routledge.Kluger, A. N. (1998). Commute variability and strain. Journal of Organizational Behavior, 19, 147–

165.Lavie, P. (1996). The enchanted world of sleep. New Haven, CT: Yale University Press.Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.Lee, I. M., Paffenberger, R. S., & Hennekens, C. H. (1997). Physical activity, physical fitness, and

longevity. Aging—Clinical and Experimental Research, 9, 2–11.Lepore, S. J. (1997). Measurement of chronic stressors. In S. Cohen, R. C. Kessler, & L. U.

Gordon (Eds.), Measuring stress: A guide for health and social scientists (pp. 102–120). Oxford,England: Oxford University Press.

Lepore, S. J., & Evans, G. W. (1996). Coping with multiple stressors in the environment. In M. Zeidner& N. S. Endler (Eds.), Handbook of coping: Theory, research, applications (pp. 350–377). NewYork: Wiley.

Lewin, K. (1951). Field theory in social science. New York: Harper and Row.Lindberg, E., Hartig, T., Garvill, J., & Garling, T. (1992). Residential location preferences across the

lifespan. Journal of Environmental Psychology, 12, 187–198.Linden, W., Earle, T. L., Gerin, W., & Christenfeld, N. (1997). Physiological stress reactivity and

recovery: Conceptual siblings separated at birth? Journal of Psychosomatic Research, 42, 117–135.


Lundberg, U. (1996). Influence of paid and unpaid work on psychophysiological stress responses ofmen and women. Journal of Occupational Health Psychology, 1, 117–130.

Lundberg, U., Dohns, I. E., Melin, B., Sandsjo, L., Palmerud, G., Kadefors, R., Ekstrom, M., & Parr,D. (1999). Psychophysiological stress responses, muscle tension, and neck and shoulder painamong supermarket cashiers. Journal of Occupational Health Psychology, 4, 245–255.

Lundberg, U., & Frankenhaeuser, M. (1999). Stress and workload of men and women in high-rankingpositions. Journal of Occupational Health Psychology, 4, 142–151.

Lundberg, U., & Lindfors, P. (2002). Psychophysiological reactions to telework in female and malewhite-collar workers. Journal of Occupational Health Psychology, 7, 354–364.

Lundberg, U., Mardberg, B., & Frankenhaeuser, M. (1994). The total workload of male and femalewhite collar workers as related to age, occupational level, and number of children. ScandinavianJournal of Psychology, 35, 315–327.

Margai, F. L. (2001). Health risks and environmental equity: A geographical analysis of accidentalreleases of hazardous materials. The Professional Geographer, 53, 422–434.

Matte, T. D., & Jacobs, D. E. (2000). Housing and health: Current issues and implications for researchand programs. Journal of Urban Health, 77, 7–25.

Matthews, K. A., & Rodin, J. (1989). Women’s changing work roles: Impact on health, family, andpublic policy. American Psychologist, 44, 1389–1393.

McEwen, B. (1998). Protective and damaging effects of stress mediators. New England Journal ofMedicine, 338, 171–179.

Melin, B., & Lundberg, U. (1997). A biopsychosocial approach to work-stress and musculoskeletaldisorders. Journal of Psychophysiology, 11, 238–247.

Michelson, W. (1988). Divergent convergence: The daily routine of employed spouses. Ekistics,331/332, 193–199.

Michelson, W. (2000). Home-based employment and quality of life: A time-use analysis. In E. Diener& D. R. Rahtz (Eds.), Advances in quality of life theory and research (pp. 183–203). New York:Kluwer.

Mokhtarian, P. L., Bagley, M. N., & Salomon, I. (1998). The impact of gender, occupation, and presenceof children on telecommunicating motivations and constraints. Journal of the American Societyfor Information Science, 49, 1115–1134.

Mokhtarian, P. L., & Salomon, I. (1994). Modeling the choice of telecommuting: Setting the context.Environment and Planning A, 26, 749–766.

Olmsted, F. L. (1870). Public parks and the enlargement of towns. Cambridge, MA: Riverside Press(Reissued by Arno Press, New York, 1970).

Parrott, A. C. (1999). Does cigarette smoking cause stress? American Psychologist, 54, 817–820.Pearlin, L. I. (1989). The sociological study of stress. Journal of Health and Social Behavior, 30,

241–256.Repetti, R. L. (1989). Effects of daily workload on subsequent behavior during marital interaction.

Journal of Personality and Social Psychology, 57, 651–659.Repetti, R. L. (1992). Social withdrawal as a short-term coping response to daily stressors. In H.

S. Friedman (Ed.), Hostility, coping, and health (pp. 151–165). Washington, DC: AmericanPsychological Association.

Repetti, R. L. (1994). Short-term and long-term processes linking job stressors to father-child interac-tion. Social Development, 3, 1–15.

Robinson, J. P., & Godbey, G. (1997). Time for life: The surprising ways Americans use their time.University Park, PA: Pennsylvania State University Press.

Rydenstam, K. (1992). I tid och otid: En undersokning om kvinnors och mans tidsanvandning 1990/91[At all times: How women and men use their time 1990/91] (SCB Report No. 79). Stockholm,Sweden: Statistics Sweden.

Saegert, S. (1985). The role of housing in the experience of dwelling. In I. Altman & C. M. Werner(Eds.), Home environments (pp. 287–309). New York: Plenum.

Salomon, I. (1998). Technological change and social forecasting: The case of telecommuting as a travelsubstitute. Transportation Research C, 6, 17–45.

Salomon, I., & Mokhtarian, P. L. (1997). Coping with congestion: Understanding the gap betweenpolicy assumptions and behavior. Transportation Research D, 2, 107–123.


Saxena, S., & Mokhtarian, P. L. (1997). The impact of telecommuting on the activity spaces of partic-ipants. Geographical Analysis, 29, 124–144.

Sayette, M. A. (1999). Does drinking reduce stress? Alcohol Research and Health, 23, 250–255.Schoggen, P. (1989). Behavior settings. Stanford, CA: Stanford University Press.Scully, D., Kremer, J., Meade, M. M., Graham, R., & Dudgeon, K. (1998). Physical exercise and

psychological well-being: A critical review. British Journal of Sports Medicine, 32, 111–120.

Shapiro, D. H., Jr., Schwartz, C. E., & Astin, J. A. (1996). Controlling ourselves, controlling our world:Psychology’s role in understanding positive and negative consequences of seeking and gainingcontrol. American Psychologist, 51, 1213–1230.

Shin, B., Sheng, O. R. L., & Higa, K. (2000). Telework: Existing research and future directions. Journalof Organizational Computing and Electronic Commerce, 10(2), 85–101.

Somerville, P. (1997). The social construction of home. Journal of Architectural and Planning Research,14, 226–245.

Stanek, D. M., & Mokhtarian, P. L. (1998). Developing models of preference for home-based andcenter-based telecommuting: Findings and forecasts. Technological Forecasting and SocialChange, 57, 53–74.

Statens Offentliga Utredningar. (1988). Om semester [About vacation] (SOU 1988:54). Stockholm,Sweden: Arbetsmarknadsdepartementet.

Steptoe, A. (1997). Stress and disease. In A. Baum, S. Newman, J. Weinman, R. West, & C. McManus(Eds.), Cambridge handbook of psychology, health, and medicine (pp. 174–177). Cambridge,MA: Cambridge University Press.

Steptoe, A., Kimbell, J., & Basford, P. (1998). Exercise and the experience and appraisal of dailystressors: A naturalistic study. Journal of Behavioral Medicine, 21, 363–374.

Sterling, P., & Eyer, J. (1988). Allostasis: A new paradigm to explain arousal pathology. In S. Fisher& J. Reason (Eds.), Handbook of life stress, cognition, and health (pp. 629–649). London:Wiley.

Stokols, D. (1976). The experience of crowding in primary and secondary environments. Environmentand Behavior, 8, 49–86.

Stokols, D. (1992). Establishing and maintaining healthy environments: Toward a social ecology ofhealth promotion. American Psychologist, 47, 6–22.

Stokols, D. (1996). Translating social ecological theory into guidelines for community health promo-tion. American Journal of Health Promotion, 10, 282–298.

Stokols, D., & Novaco, R. W. (1981). Transportation and well-being. In I. Altman, J. F. Wohlwill, &P. Everett (Eds.), Transportation and behavior (pp. 85–130). New York: Plenum.

Stokols, D., & Shumaker, S. A. (1981). People in places: A transactional view of settings. In J. H.Harvey (Ed.), Cognition, social behavior, and the environment (pp. 441–488). Hillsdale, NJ:Lawrence Erlbaum.

Stokols, D., & Shumaker, S. A. (1982). The psychological context of residential mobility and well-being. Journal of Social Issues, 38, 149–171.

Syme, S. L. (1990). Control and health: An epidemiological perspective. In J. Rodin, C. Schooler, & K.W. Schaie (Eds.), Self-directedness: Cause and effects throughout the life course(pp. 213–229). Hillsdale, NJ: Lawrence Erlbaum.

Tognoli, J. (1987). Residential environments. In D. Stokols & I. Altman (Eds.), Handbook of environ-mental psychology: Vol. 1 (pp. 655–690). New York: Wiley.

Ulrich, R. S. (1983). Aesthetic and affective response to natural environment. In I. Altman & J. F.Wohlwill (Eds.), Behavior and the natural environment (pp. 85–125). New York: Plenum.

U.S. Census Bureau. (2002). Facts about Census 2000 residence rules. Retrieved May 13, 2003, fromhttp://www.census.gov/population/www/censusdata/resid rules.html

Vilhelmson, B., & Thulin, E. (2001). Is regular work at fixed places fading away? The developmentof ICT-based and travel-based modes of work in Sweden. Environment and Planning A, 33,1015–1029.

Weinberger, D. A. (1990). The construct validity of the repressive coping style. In J. L. Singer (Ed.),Repression and dissociation: Implications for personality theory, psychopathology, and health(pp. 337–386). Chicago: University of Chicago Press.


Westman, M., & Eden, D. (1997). Effects of a respite from work on burnout: Vacation relief andfade-out. Journal of Applied Psychology, 82, 516–527.

Wolpert, J. (1966). Migration as an adjustment to environmental stress. Journal of Social Issues, 22,92–102.

World Health Organization. (1946). Constitution. Geneva, Switzerland: Author.

TERRY HARTIG is an Associate Professor of Psychology with the Institute forHousing and Urban Research and the Department of Psychology of Uppsala Uni-versity. He completed graduate training in social ecology at the University ofCalifornia at Irvine and postdoctoral training in social epidemiology at the Uni-versity of California at Berkeley. His research focuses on health values of natureexperience, restorative environments, health in relation to residence, and the socialecology of stress and restoration.

GUNN JOHANSSON is Professor of Work and Organizational Psychology atStockholm University. In her research she analyzes demands and resources inworking life that affect psychobiological stress reactions, health, and the individ-ual’s occupational and personal development. She applies a gender perspectivethroughout her work, and places special emphasis on workloads such as time pres-sure, limited control and/or decision latitude, and a high rate of organizationaland technological change. She is one of the principal investigators in the “NewBoundaries of Work” project, a study of Swedish teleworkers.

CAMILLA KYLIN is a doctoral student within the Work and Organizational Psy-chology Division of the Department of Psychology at Stockholm University. Herstudies focus on the balance/interaction between work and other life domains. Asthe working life of today has assumed a boundaryless character through flexibi-lization in time and space, the aim of her research is to understand the effectsof flexible work arrangements on work/nonwork relations, individuals’ copingstrategies, health, and well-being.

TERRY HARTIG

Documents

Transcript of TERRY HARTIG