Content uploaded by Nancy M Wells
Author content
All content in this area was uploaded by Nancy M Wells on Aug 06, 2014
Content may be subject to copyright.
http://eab.sagepub.com
Environment and Behavior
DOI: 10.1177/0013916503035003001
2003; 35; 311 Environment and Behavior
Nancy M. Wells and Gary W. Evans
Nearby Nature: A Buffer of Life Stress among Rural Children
http://eab.sagepub.com/cgi/content/abstract/35/3/311
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
On behalf of:
Environmental Design Research Association
can be found at:Environment and Behavior Additional services and information for
http://eab.sagepub.com/cgi/alerts Email Alerts:
http://eab.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://eab.sagepub.com/cgi/content/refs/35/3/311 Citations
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
10.1177/0013916503251445ARTICLEENVIRONMENT AND BEHAVIOR / May 2003Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN
NEARBY NATURE
A Buffer of Life Stress Among Rural Children
NANCY M. WELLS is an environmental psychologist interested in the impact of the
built and natural environment on human well-being through the life course. Her work
includes the study of housing transitions among impoverished urban families, the im-
pact of nearby nature on individual and community well-being, and environments for
elderly.
GARY W. EVANS is an environmental and developmental psychologist interested in
the effects of the physical environment on children and their families. His major
research interests include environmental stressors, housing, and poverty.
ABSTRACT: Identifying mechanisms that buffer children from life’s stress and
adversity is an important empirical and practical concern. This study focuses on
nature as a buffer of life stress among rural children. To examine whether vegetation
near the residential environment might buffer or moderate the impact of stressful life
events on children’s psychological well-being, data were collected from 337 rural
children in Grades 3 through 5 (mean age = 9.2 years). Dependent variables include a
standard parent-reported measure of children’s psychological distress and children’s
own ratings of global self-worth. In a rural setting, levels of nearby nature moderate
the impact of stressful life events on the psychological well-being of children. Spe-
cifically, the impact of life stress was lower among children with high levels of nearby
nature than among those with little nearby nature. Implications of these finding are
discussed with respect to our understanding of resilience and protective mechanisms.
Keywords: nature; restoration; children; stress; housing
During the past 30 years, a substantial body of literature has illustrated that
the natural environment has profound effects on the well-being of adult
311
AUTHORS’ NOTE: We are grateful to the many families who participated in our
research. We also thank Jana Cooperman, Kim English, Missy Globerman, and Amy
Schreier for their assistance with data collection. This research was partially sup-
ported by the John D. and Catherine T. McArthur Foundation Network on Socio-
ENVIRONMENT AND BEHAVIOR, Vol. 35 No. 3, May 2003 311-330
DOI: 10.1177/0013916503251445
© 2003 Sage Publications
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
humans. Time spent in contact with the natural environment has been associ-
ated with better psychological well-being (R. Kaplan, 1973), superior cogni-
tive functioning (Cimprich, 1990; Hartig, Mang, & Evans, 1991; Kuo, 2001;
Tennessen & Cimprich, 1995), fewer physical ailments (E. O. Moore, 1981;
West, 1986), and speedier recovery from illness (Ulrich, 1984; Verderber,
1986; Verderber & Reuman, 1987). In addition to the studies of adults,
research has more recently begun to explore children’s relationships to the
natural environment (e.g., Bartlett, 1997, 1998; Coley, Kuo, & Sullivan,
1997; Kahn & Kellert, 2002; R. C. Moore, 1986; Sebba 1991; Sobel, 1993).
And although some research has examined the direct effects of nature on chil-
dren’s functioning or well-being (Faber Taylor, Kuo, & Sullivan, 2001, 2002;
Faber Taylor, Wiley, Kuo, & Sullivan, 1998; Grahn, Mårtensson, Lindblad,
Nilsson, & Ekman, 1997; Wells, 2000), the notion that nature might buffer or
moderate the effects of stress or adversity has not been studied. It is widely
accepted that the environment is likely to have a more profound effect on chil-
dren due to their greater plasticity or vulnerability, and yet there is need for
research, especially regarding the potential buffering effects of the natural
environment on children’s well-being.
This article first contains a brief review of the literature regarding chil-
dren’s affinity for the natural environment and the research, largely of the past
decade, exploring the beneficial effects of nature on children. The article then
explores the notion that nature may act as a buffer or moderator, mitigating
the impact of stress or adversity on children. Finally, an empirical study will
be presented examining whether access to nature may act as a buffer.
CHILDREN’S PREFERENCE FOR THE OUTDOORS
Numerous studies have illustrated children’s predilection for outdoor set-
tings. The documentation of children’s preference for green natural spaces is
neither frivolous nor insignificant. From an evolutionary perspective, it is
reasonable to expect that humans will have an affinity for settings that are
beneficial, therapeutic, or healthful. As S. Kaplan and R. Kaplan (1982)
pointed out, “An organism must prefer those environments in which it is
likely to thrive” (p. 147). Thus, preference in this context is an expression of
human needs. Preferred environments are likely to afford long-term surviv-
ability and are likely to be the settings in which humans are more likely to
312 ENVIRONMENT AND BEHAVIOR / May 2003
economic Status and Health; the W. T. Grant Foundation; the College of Human Ecol-
ogy at Cornell University; the National Institute for Child Health and Human Devel-
opment, 1F33 HD08473-01; the USDA New York State; Hatch Grants 327404,
327407, 327416; and the National Institute of Mental Health, T32MH19958-05.
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
function effectively (R. Kaplan & S. Kaplan, 1989; S. Kaplan & R. Kaplan,
1982; Parsons, Tassinary, Ulrich, Hebl, & Grossman-Alexander, 1998;
Ulrich, 1983; Zube, Pitt, & Evans, 1983).
Numerous studies have documented that children’s preferred environ-
ments include a predominance of natural elements (Korpela, 2002). For
example, R. C. Moore (1986) reported that when urban children aged 9 to 12
were asked to make a map or drawing of all their favorite places, 96% of the
illustrations were of outdoor places. In fact, only four drawings mentioned
interior spaces and just one drawing was devoted exclusively to building inte-
riors. The children’s most frequently drawn favorite places were lawns, play-
grounds and schoolyards, their own home, local parks, and single trees.
Similarly, a British study found that when 237 children living both on and off
public housing estates were asked what they like most about where they lived,
75% of the estate children and 45% of the nonestate children answered in
terms of places to play outdoors (Department of the Environment, 1973).
Sobel (1993) documented the preference for natural play spaces among both
British and Caribbean children. A retrospective study by Sebba (1991)
reported that when asked to describe the most significant or favorite place of
their childhood, 97% of adults indicated outdoor places. And in his 1977
international exploration of the experience of growing up in cities, Lynch dis-
covered a universal appreciation for vegetation. Youth participating in the
study frequently suggested that more trees be planted in their city. “The hun-
ger for trees is outspoken and seemingly universal....Landscaping should be
as essential a part of the basic infrastructure of a settlement as electricity,
water, sewer, and paving” (Lynch, 1977, pp. 56-57). Given the evidence, it is
reasonable to expect that green natural settings, preferred by children, would
also have a beneficial effect on children’s well-being.
DOES NATURE MATTER? DIRECT EFFECTS OF
THE NATURAL ENVIRONMENT ON CHILDREN
Studies of the direct effects of the natural environment on children’s well-
being include those examining the absence or inaccessibility of nearby nature
and those comparing the presence of nearby trees and vegetation with cir-
cumstances where vegetation is absent or minimal. These studies employ a
range of dependent variables, including cognitive, behavioral, and social
outcomes.
In exploring the notion that natural environments may be beneficial to
children’s well-being, insight can be gained from situations when children
are deprived of contact with nature, when nature is “out of reach.” In two
qualitative studies, Bartlett (1997, 1998) described the circumstances of
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 313
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
young children who, due to a confluence of poverty, lack of space, and over-
taxed parents, are unable to play outside. One of the children is described as
follows:
Lee often throws tantrums in the doorway when his older brothers go outdoors
and tries to fight his way out. Windows are kept shut, and outside doors are
locked....Leeisafrustrated child. What he most wants, active play outdoors
with other children, is off limits, literally out of reach. ...Inhisattempts to
escape...Lee[frequently] hurts himself. ...Heexperiences himself as wild,
uncontrollable, and potentially dangerous; but also as thwarted and help-
less. . . . Lee has long since exhausted the limited possibilities in [the] two
rooms, and his frustration alternates with listless withdrawal. (Bartlett, 1997,
pp. 43-44)
Along this same vein, in his study of families in Zurich, Hüttenmoser (1995)
found that 5-year-old children who, due to dangerous traffic conditions,
could not easily play outdoors unsupervised exhibited poorer social, behav-
ior, and motor skills and had fewer playmates than children with easy access
to the outdoors. These data make a strong case for the importance of nature to
well-being.
The majority of research exploring the beneficial effects of nature on chil-
dren has been conducted since the mid-1990s, and nearly all of this work has
focused on children living in urban contexts. Several studies have examined
the impact of nature on children’s cognitive functioning. For example, a
Swedish study compared the effects of the natural environment on children’s
physical and cognitive abilities within two different day care settings. Grahn
et al. (1997) found that children attending an “outdoors in all weather” day
care facility with surrounding orchards, pastures, and woodlands had better
motor coordination and greater attention capacity than did children who
attended an urban day care center surrounded by tall buildings. Wells (2000),
in one of the few studies to employ a longitudinal design, examined the
effects of green natural residential settings on children’s cognitive function-
ing. Children who moved to housing with more nature nearby tended to have
higher levels of cognitive functioning than children who experienced less
increase in the amount of nearby nature from premove to postmove. More
recently, Faber Taylor and her colleagues Kuo and Sullivan (2001) docu-
mented that activities in green settings tend to lower the symptoms of chil-
dren who struggle with a chronic attention deficit due to Attention Deficit
Disorder (ADD). And in a second study, Faber Taylor et al. (2002) examined
the effect of nearby nature on self-discipline among girls living in Chicago
public housing. Findings indicated that mothers’ratings of the naturalness of
314 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
window views was predictive of the girls’ ability to concentrate, to inhibit
impulses, and to delay gratification. Collectively, these studies offer substan-
tial evidence of the effects of nature on children’s functioning. Other work by
Kuo and Sullivan and their colleagues has focused on the social effects of the
natural environment on families living in public housing complexes in Chi-
cago. These researchers compared the behavior and activities of public hous-
ing residents in two outdoor spaces—one with many trees and the other
relatively barren with few trees. Coley et al. (1997) found that the presence of
trees and vegetation in outdoor public spaces was associated with greater use
of these spaces by both youth and adult residents. Faber Taylor et al. (1998)
found that the green spaces were more supportive of children’s play and that
children had more access to adults in greener outdoor spaces than in the rela-
tively barren spaces. Areas with natural landscaping appear to promote
opportunities for social interaction as well as better adult supervision of chil-
dren in poor urban neighborhoods. These public housing studies are also
noteworthy because of the quasi-random assignment of residents to apart-
ments. Long waiting lists essentially ensure that residents take whatever
apartment becomes available—in other words, they rarely choose a particu-
lar residence.
Together, the collection of studies described here suggests that not only do
children prefer to spend time in natural settings, but disconnection from the
natural environment negatively affects the well-being of children. Further-
more, the availability and use of green, outdoor spaces contributes to cogni-
tive function and well-being as well as to social interaction and social
connectedness within impoverished, urban communities.
NATURE AS A BUFFER
Considerable empirical research has addressed the impact of negative
environmental factors on the health and well-being of children. We know, for
instance, that residential crowding and chronic noise (Evans, 2001), pollut-
ants such as lead (Needleman, 1994; Spreen, Tupper, Risser, Tuokko, &
Edgell, 1984), and poor housing quality (Bartlett, 1987, 1988; Evans, Wells,
& Moch, in press; Gifford, in press) all detrimentally affect children psycho-
logically, cognitively, or physiologically. Increasingly, researchers are also
exploring the idea that environmental characteristics may function as buffers
or moderators of adverse conditions, serving as protective factors and con-
tributing to resilience among children. Physical characteristics of the home
environment, such as the availability of toys and materials, a variety of stimu-
lation, and adequate space for privacy and exploration have been identified as
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 315
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
potentially protective factors (Bradley, Caldwell, & Rock, 1990; Wachs &
Gruen, 1982). Safe play space, greater stimulation, and low residential den-
sity have been empirically linked to greater resilience among low birth
weight children living in poverty (Bradley et al., 1994). Borrowing from
Garbarino’s (1982) risk-opportunity framework, Dunst and Trivette (1994)
have proposed that factors that promote as well as those that impede chil-
dren’s development and well-being ought to be incorporated in theoretical
and empirical work. They suggested and provided empirical evidence that
“the absence of problems (risk factors) does not adequately consider the envi-
ronmental conditions related to optimal performance” (p. 302). There is a
need to identify environmental factors that bolster functioning and well-
being of children. Wells (2000) called for further exploration of the notion
that the natural environment, in particular, might buffer the potential effects
of stress and adversity, contributing to the resilience of children, adults, fami-
lies, and communities.
Buffer is another term for an interaction effect or a moderator (Baron &
Kenny, 1986; Evans & Lepore, 1997). A moderating variable changes the
impact of another independent variable, explaining how or under what cir-
cumstances the independent variable affects the outcome variable. Although
a moderator or buffer may occur in the presence of a direct or main effect, this
is not necessarily the case. A main effect of nature means such exposure
directly affects health or well-being. A buffering effect means that nature
attenuates the adverse effects of stressors or other adverse main effects on
health or well-being. Herein, we examine both the main (direct) effects of
nature and stressful life events alone and the moderator (interactive) effects
of nature and life stress on children’s well-being.
Perhaps access to vegetation and natural areas can help to bolster the resil-
ience of children and others encountering stress, challenge, or adversity.
Although research has not explicitly examined the buffering hypothesis, a
few laboratory studies with adults suggest that the effects of stressors on
well-being may be moderated by exposure to nature. With respect to the
amount of nature in commuters’ roadside views, Parsons and his colleagues
(1998) found that participants who viewed videotapes of nature-dominated
drives experienced quicker and more complete recovery from induced stress
than did participants who viewed artifact-dominated drives. And in another
study, participants presented with videotapes of natural settings recovered
faster and more completely from their exposure to a stressful movie than did
those presented with a videotape of an urban setting (Ulrich et al., 1991).
Recovery from illness also shows parallel trends (Ulrich, 1984). And
although no empirical research with children has addressed the role of nature
316 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
as a potential buffer, much of Kuo, Sullivan, Faber Taylor, and colleagues’
work (Coley et al., 1997; Faber Taylor et al., 1998, 2002; Kuo, Sullivan,
Coley, & Brunson, 1998) is suggestive of such a possibility, as they explore
the benefits of nearby nature on children living within an impoverished, often
treacherous, setting. Furthermore, Kuo (2001), although not statistically
examining nature as a moderator, found that nearby nature, by bolstering the
attention resources of adult inner-city public housing residents, enhanced
their ability to function effectively and make life decisions.
We examine nearby nature as a buffer of the impact of stressful life events
on the well-being of children, with two standardized measures of psychologi-
cal distress (maternal report) and global self-worth (child self-report). We
hypothesize that there will be an interaction effect—that the adverse impact
of stressful life events will be less severe in the presence of nearby nature. In
addition, we predict direct effects of both stressful life events and nature on
children’s psychological distress and global self-worth. We examine these
buffering and main effect hypotheses among a sample of rural children.
METHOD
PARTICIPANTS
A total of 337 children participated in this study. The children lived in
small towns in five rural upstate New York communities. They were in
Grades 3 through 5 (mean age = 9.2 years). Of these, 49% of the participants
were female. Regarding the children’s parents, 56% were married and 44%
were single, divorced, or widowed. A majority (95%) were White; 3% were
Black. Of the children’s mothers, 63% had completed at least some college;
33% had 2-year, 4-year, or graduate degrees. The mean income-to-needs
ratio for these families was 1.79 (standard deviation 1.66), where a ratio of
1.0 or below represents the U.S. government’s definition of poverty. The
income-to-needs ratio is an annually adjusted, per capita index comparing
household income to federal estimates of minimally required expenditures
for food and shelter.
CONSTRUCTS AND MEASURES
Naturalness. A naturalness scale of the residential environment was
developed as part of a detailed housing scale instrument (Evans, Wells, Chan,
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 317
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
& Saltzman, 2000) for a larger study of housing quality and mental health.
The Naturalness scale consists of four items regarding the amount of nature
in the window view, the number of live plants indoors, and the material of the
outdoor yard. The Naturalness scale items are summarized in Table 1.
Children’s stressful life events. The Lewis Stressful Life Events Scale
(Lewis, Seigel, & Lewis, 1984) was used to assess the frequency of stressful
events in the children’s lives. This 20-item scale was originally developed
and tested on a large sample of 5th- and 6th-grade children. Sample items
include “In the last three months, did your family move from one home to
another?” “How often were you picked on or made fun of by other kids?”
“How often did you fight or argue with your parents?” “How often did some-
one make you try something new like a cigarette that you really didn’t want to
try?” and “How often have you been punished or gotten in trouble at school?”
For this study, only the frequency of stressors was employed, not the severity
ratings. Response options used were a lot, sometimes, and never.
Children’s psychological distress. Two measures were used as indicators
of children’s distress. First, the Rutter Child Behavior Questionnaire was
used to assess the children’s psychological distress. The Rutter scale is a stan-
dardized instrument designed to measure psychological health in nonclinical
samples of children. The Rutter scale is a widely used instrument and has
well-documented psychometric properties (Boyle & Jones, 1985; Rutter,
Tizard, & Whitmore, 1970). The scale consists of 26 items concerning com-
mon childhood symptoms indicative of behavioral conduct disorders (e.g.,
bullies other children) as well as symptoms of anxiety and depression (e.g.,
often appears miserable, unhappy, tearful, or depressed). These items are
rated by the mother on a 3-point continuum (0 = does not apply,1=applies
somewhat, 2 = certainly applies).
318 ENVIRONMENT AND BEHAVIOR / May 2003
TABLE 1
The Four-Item Naturalness Scale
Naturalness Scale Item Area of House
What is the view? 3 = > 1/2 natural,
2 = < 1/2 natural, 1 = no natural, 0 = none Living room, kitchen
How many live plants are in the room?
2 = > 3 plants, 1 = 1 to 3 plants, 0 = none Living room
Yard—What material is it? 3 = grass,
2 = dirt, 1 = concrete, 0 = other Outdoors
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
The Global Self-Worth subscale of the Harter Competency Scale (Harter,
1982) was used to index children’s self-perception of psychological well-
being. This instrument requires that children respond in a forced-choice for-
mat, choosing which of two bipolar behavioral descriptions are “really true or
sort of true of you.” The 6-item subscale includes items such as “Some kids
like the kind of person they are but other kids often wish they were someone
else” and “Some kids are often unhappy with themselves but other kids are
pretty pleased with themselves.”
PROCEDURE
All data were collected using a uniform protocol within the home. One
researcher conducted a series of survey and interview questions with the
mother. These items included the Rutter Child Behavior Questionnaire. The
second researcher conducted a detailed housing characteristics checklist,
which included the Naturalness items, and also administered the Harter
Competency Scale and the Lewis Stressful Life Events Scale to the child.
RESULTS
The focus of our analytic strategy was to assess whether nearby natural
elements would moderate the effects of stressful life events on children’s
well-being. In addition, we examined whether nearby nature and life stress
each had direct effects on children’s well-being. Identical analytic proce-
dures were followed for both of the dependent measures—children’s psycho-
logical distress (Rutter) and global self-worth (Harter). Using hierarchical
regression analyses, the families’ socioeconomic status as measured by the
income-to-needs ratio was entered first to control for its effects. Next, the
main effects of first nearby nature and then stressful life events (Lewis) were
each examined while controlling for income. Last, the interaction of stressful
life events and nearby nature was examined following the inclusion of the
covariate (income) and both main effects in the model. Each of the four steps
is represented by one line in Tables 2 and 3.
First we examined children’s psychological distress as the dependent vari-
able. The initial step of the analysis (line 1 in Table 2) controlled for the
explanatory power of socioeconomic status, F(1, 336) = 49.57, p < .001. This
shows that children of higher socioeconomic status tend to have significantly
lower levels of psychological distress. Next, the direct effect of nearby nature is
taken into account, F(2, 335) = 6.27, p = .05. This is also significant, indicating
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 319
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
that children with more nature near the home exhibit less psychological dis-
tress. The third line examines the direct effect of children’s stressful life
events on levels of psychological distress, F(2, 336) = 53.14, p < .001. This is
highly significant and indicates that children exposed to greater stressful life
events experience greater psychological distress. The last line incorporates
the interaction of nearby nature and stressful life events. This is also statisti-
cally significant. Nearby nature was found to buffer the effects of stressful
life events on children’s psychological distress, F(4, 333) = 4.73, p < .05. The
effect of stressful life events on children’s psychological distress varies as a
function of the amount of nearby nature to which the children are exposed.
Specifically, the impact of stressful life events on psychological distress is
weaker under conditions of high nature than under low nature conditions.
The interaction effect is illustrated in Figure 1.
1
The difference in levels of
psychological distress between low nature exposure and high nature expo-
sure was particularly pronounced among children who experienced the high-
est levels of stressful life events.
320 ENVIRONMENT AND BEHAVIOR / May 2003
TABLE 2
Regression of Children’s Psychological Distress (Rutter) Onto Nature,
Life Stress, and the Interaction of Stress
Nature, Controlling for Income
Model Predictor Total
R
2
∆R
2
F
(
∆R
2
)
df b SE b p
Control Income .129 .129 49.57** 1, 336 –1.99 .283 .000
Main effect Nature .145 .016 6.27* 2, 335 2.36 .941 .013
Main effect Life Stress .240 .113 53.14** 2, 336 –.352 .050 .000
Interaction Nature × Life
Stress .259 .011 4.73* 4, 333 –.313 .144 .030
NOTE:Differences in degrees of freedom for main effect analyses reflect variations in sample size.
*
p
< .05. **
p
< .001.
TABLE 3
Regression of Children’s Global Self-Worth Onto Nature, Stressful Life
Events, and Interaction of Stress
Nature, Controlling for Income of Family
Model Predictor Total
R
2
∆R
2
F
(
∆R
2
)
df b SE b p
Control Income .050 .050 15.86** 1, 299 .118 .030 .000
Main effect Nature .066 .016 5.05* 2, 298 –.228 .102 .025
Main effect Life Stress .164 .114 40.54** 2, 298 .034 .005 .000
Interaction Nature ×
Life Stress .200 .029 10.64* 4, 296 .052 .016 .001
NOTE:Differences in degrees of freedom for main effect analyses reflect variations in sample size.
*
p
< .05. **
p
< .001.
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
Identical analyses were conducted with respect to the second dependent
measure, children’s global self-worth (Table 3). The first line reveals that
socioeconomic status, the control or covariate, is positively related to chil-
dren’s global self-worth, F(1, 299) = 15.86, p < .001. Degrees of freedom
vary because of missing data. Next, there was a main effect of nature, F(2,
298) = 5.05, p < .05, such that higher nature corresponds to higher global self-
worth. In addition, there was a highly significant main effect of stressful life
events, F(2, 298) = 40.54, p < .001. Analyses also revealed an interaction
effect of nature and stressful life events on global self-worth, F(4, 296) =
10.64, p < .001. As shown in Figure 2, nature moderates the impact of stress-
ful life events on the self-worth of rural children. The link between stressors
and self-worth is weaker for those children with greater exposure to nature.
DISCUSSION
Consistent with our hypothesis, the results of this study suggest that the
presence of nearby nature moderates or buffers the impact of life stress on
children. The psychological effects of stressful life events such as family
relocation, being picked on or punished at school, or being subject to peer
pressure varied depending on the amount of nearby nature to which the chil-
dren, aged 6 through 12, had access. This moderating effect occurred with
both dependent measures—parent-reported psychological distress and chil-
dren’s own reports of global self-worth. With respect to psychological
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 321
10.22
6.31
7.86
13.23
9.42
0
2
4
6
8
10
12
14
Low Medium High
Stressful Life Events
Psychological Distress
low nature
high nature
7.22
Figure 1: Nature Moderates Effects of Stressful Life Events on Psychological
Distress
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
distress, the expected elevations in response to greater life stress are attenu-
ated in the presence of greater exposure to nearby nature. As shown in Figure
1, stressful life events have less impact on psychological distress under high
nature conditions than under low nature conditions. And in the case of global
self-worth, nature also buffers the effect of stressful life events. Children with
a high degree of exposure to nature seem to be protected from the impact of
life stress. This buffering appears to be greatest for those at most risk—those
experiencing the highest levels of life stress. Nature’s role as a moderator in
these analyses is illustrated by comparing the slopes of low nature versus
high nature in Figures 1 and 2. The slope of the high nature line is less steep—
representing a dampening of the effects of life stress.
Together, these results offer evidence of the potential for nearby nature to
moderate the impact of life stress on rural children. The findings, for both
dependent measures, are consistent with Rutter’s (1987) criteria for a protec-
tive effect—that the moderator “either has no effect in low risk populations or
its effect is magnified in the presence of the risk variable” (p. 317). One sees
here (Figures 1 and 2) a greater buffering effect for those with exposure to
greater life stress. The protective or buffering effects of nearby nature are
strongest for the most vulnerable children—those experiencing the highest
levels of stressful life events. This illustrates an attenuating interactive effect
(Evans & Lepore, 1997).
The data also provide support for the hypothesized direct effects of life
stress and nature. The direct effects of both nature and life stress were signifi-
cant for both psychological distress and global self-worth. Children
322 ENVIRONMENT AND BEHAVIOR / May 2003
3.53
3.2
2.87
3.4
3.22
3.31
0
0.5
1
1.5
2
2.5
3
3.5
4
Low Medium High
Stressful Life Events
Global Self-Worth
low nature
high nature
Figure 2: Nature Moderates the Effects of Stressful Life Events on Global Self-
Worth
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
experiencing more nature were rated by their mothers as being lower in
symptoms of psychological distress. These same children also perceive
themselves as being higher in self-worth. Analogously, children experienc-
ing more stressors in their lives rated themselves as lower in self-worth and
were judged by their mother to have more symptoms of psychological dis-
tress. These patterns of results are noteworthy from a methodological per-
spective given the cross-rater stability.
This study makes several contributions. First, the findings are significant
because, unlike those of prior research, they provide evidence of the natural
environment’s buffering effect—indicating that the impact of life stress var-
ies depending on levels of nature exposure. In addition, this work is relatively
unique in its focus on children. Most prior work examining the effects of
nature has focused on adults, whereas relatively little prior research has
examined the impact of the natural environment on children (Grahn et al.,
1997; Faber Taylor et al., 1998, 2001, 2002; Wells, 2000). Third, the rela-
tively green rural context of this study means that our tests are conservative.
Findings of a similar study in an urban setting, where the amount of nearby
nature is likely to be more variable, would likely be stronger. Moreover, the
use of two dependent variable measures further enhances the validity of
these findings. Last, this work also differs from prior research in its measure-
ment of nature as a continuous variable. The measurement of nature as a con-
tinuous, rather than a dichotomous, variable allows us to more confidently
rule out possible confounding variables, which might be collinear with
nature in a circumstance where nature is measured dichotomously. For exam-
ple, in some situations the absence or presence of nearby nature is likely to
covary with other factors, such as the overall quality of the environment.
ALTERNATIVE INTERPRETATIONS
Our findings suggest that nature buffers the adverse impact of stressful life
events on psychological distress and global self-worth. It is important how-
ever, in field studies such as this, to consider the possibility of a self-selection
bias—the notion that a person may “select into” a housing situation or other
circumstance and that this accounts for the reported effects. In other words,
perhaps it is not the case that nature buffers the impact of stressful life events
but, rather, that some parental characteristics such as personality or socioeco-
nomic status influence the choice of a home with natural surroundings and/or
the level of stressful life events experienced by the child. We believe that the
selection bias explanation is unlikely for four reasons. First, because we con-
trol for income in the first step of our analyses, we eliminate the possibility
that family socioeconomic status underlies our results. This is critical
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 323
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
because many housing selection biases are associated with income. Second,
generally there is evidence in the literature against the self-selection explana-
tion of relationships between residence characteristics and well-being. For
example, Evans et al., (2000) documented that changes in housing quality
explained changes in psychological well-being when people relocated from
inadequate housing to better housing. Furthermore, in another housing study
where families were randomly assigned to residences, housing characteris-
tics were associated with psychological health (Fanning, 1967). Third,
because we focus on children, who do not choose where they live, effects of
parent selection biases would have to operate cross-generationally. Fourth,
any alternative explanation for our results, including selection bias, has to
account for the interaction of nature and life stress on the outcomes. Recall
that our principal focus herein is on the buffering or interactive effects, not the
main effects, of nature. The argument that personality or some other con-
struct is a viable alternative explanation is much more difficult to make for
interaction outcomes.
EXPLANATORY MECHANISMS
This study illustrates that access to nearby nature can buffer the effects of
stressful life events on psychological distress among children. The result
makes the powerful suggestion that vegetation and natural elements in or
near a residential setting may be among a variety of potential protective fac-
tors that can partially shield children from the impact of stress and adver-
sity—contributing to their resilience. What this work does not tell us is
explicitly how or why this occurs. Identifying protective factors is not
enough; it is essential to also delineate the specific mechanisms through
which such protection occurs (Evans & Lepore, 1997; Rutter, 1987).
Although the data in the current study are insufficient to allow an empirical
exploration of this question, prior theory and empirical research provide a
framework for further speculative interpretation. Two potential explanatory
mechanisms will be briefly explored: (a) social support and (b) attention
restoration.
Social support is one possible candidate for a mechanism to explain the
moderating function of nearby nature. Perhaps natural areas, even in rela-
tively rural settings, draw children together, providing a context for making
friends. The social support provided by these rural children’s friendships
may in turn help to buffer the impact of life stress. Consistent with the notion
that green, outdoor spaces foster social interaction and community networks
is Hüttenmoser’s (1995) finding that both children and parents who lived in
places that allowed for outdoor access had more than twice as many
324 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
playmates or friends than did those living in places with restricted outdoor
access due to traffic. Additional support for this idea is provided by Coley
et al. (1997), Faber Taylor et al. (1998), and Kuo (2001), who showed that
landscaped areas support social interaction, children’s play, and neighbor-
hood social ties in an urban setting. Numerous studies indicate that social
support buffers the effects of stress in both adults (Cohen, Underwood, &
Gottlieb, 2000; McNaughton, Patterson, Irwin, & Grant, 1992; Roy &
Steptoe, 1994; Vaux, 1988) and in children (Cauce, Reid, Landesman, &
Gonzales, 1992; Wills, Blechman, & McNamara, 1996). The idea that
nature, by fostering social support, may buffer the impact of life stress merits
further exploration, particularly with respect to children.
Another possible explanatory mechanism involves the theory that expo-
sure to natural elements help to restore the ability to focus one’s attention—in
other words, “directed attention capacity.” According to this theoretical per-
spective (R. Kaplan & S. Kaplan, 1989; S. Kaplan, 1995; S. Kaplan & R.
Kaplan, 1983), exposure to nature bolsters one’s cognitive resources by
allowing neural inhibitory mechanisms to rest and recover from use. This
recovery occurs due to four characteristics, most often found in natural envi-
ronments. Nature’s tendency to draw one’s attention effortlessly (fascina-
tion) allows inhibitory mechanisms that underlie attention to rest, while at the
same time, the sense of being away from one’s daily concerns permits a men-
tal vacation, and the extent of the environment provides a scope or depth in
which one can become immersed. Last, a setting that is compatible or well-
matched with one’s inclinations allows attention to rest. A person whose
attention resources have been restored will be able to inhibit the urge to
respond to potentially distracting stimuli, able to focus attention, and able to
more effectively manage the challenges of daily life. Support for attention
restoration theory has been found among breast cancer patients (Cimprich,
1990), backpackers (Hartig et al., 1991), and college students (Tennessen &
Cimprich, 1995), among others. Nature’s bolstering of attention resources
may enable children to think more clearly and cope more effectively with life
stress. Greater cognitive clarity may enable children to seek out activities or
resources to fortify themselves against life stress as well as enable them to
resist the inclination to react to certain stressors or potential distractions.
Related to this are the findings of Faber Taylor et al. (2002) that nature fosters
self-discipline among urban girls and Kuo (2001) that attention function
mediates the relationship between nature and life functioning (i.e., coping
and decision making) among urban women. The explanatory linkage or
mechanism underlying the buffering effect of nature needs to be explored in
future research.
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 325
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
FUTURE RESEARCH
In addition to further exploration of the mechanisms underlying nature’s
capacity to bolster resilience, research ought to explore the generalizability
of the present findings. In an urban setting, where barren treeless spaces are
more common and the amount of nature is more variable, it is possible that
nature might play a stronger role as a moderator of stress or adversity. Future
work might explore whether nearby nature has a similarly buffering effect in
more urban settings and with other populations and ages.
Longitudinal work would be a valuable adjunct to this study. First, it
would enable a more rigorous causal evaluation. Second, it would provide
insight into the impact of nearby nature on human well-being through the life
course. Research from a life course perspective might explore questions such
as, How do people’s relationships with the natural environment change
throughout a lifetime? How might frequent exposure to nature in infancy or
childhood affect reaction to stressors and general functioning later in life?
Longitudinal research exploring people’s relationships to nature over time
and during various periods of life would be a valuable contribution to our
understanding of human/natural environment relationships.
IMPLICATIONS
The implications of this work may be particularly relevant with respect to
impoverished children. If access to nearby nature is indeed a protective fac-
tor, contributing to the resilience of children and youth, then if nearby nature
is lacking, it is one more strike against poor children who already face tre-
mendous disadvantage (Evans & Kantrowitz, 2002; McLoyd, 1990). More-
over, it is the case that, in urban areas, children from low-income families
have less access to natural areas. In New York City in 1989, for example, poor
children lived in neighborhoods with less than half the park space as nonpoor
children. The average poor child in New York City had 17 square yards of
park space per child in his or her community district; a nonpoor child lived in
a community district with 40 square yards of park per child (Sherman, 1994).
Moreover, there is parallel evidence with respect to adults. In the United
Kingdom, manual laborers in comparison to professionals are twice as likely
to have no access to a private garden area and 4 times more likely to have an
outdoor space that is too small to sit outside in the sun (Townsend, 1979). The
current data indicate that the link between income or socioeconomic status
and access to the natural environment also exists in a rural context, F(1, 338)
= 4.72, p < .05. Although not a focus of the analysis in this article, it is striking
that access to the natural environment appears to be related to socioeconomic
326 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
status even in rural areas. Although we might expect such a relationship in an
urban context, where wealthier families may live close to parks and have
yards surrounding their own dwellings, this is surprising in a rural area.
Remarkably, the ubiquitous environmental justice issue documented in met-
ropolitan areas (Evans & Kantrowitz, 2002) may be mirrored in rural settings
as well.
The notion that nature plays a buffering or moderating role in protecting
children from the brunt of life stress has powerful implications for policy and
design. Natural areas proximate to housing and schools are essential features
in an effort to foster the resilience of children and perhaps to promote their
healthy development. Moreover, the fact that this result occurs within a rural
context is particularly noteworthy. Although a recent body of empirical
research has indicated that nature is beneficial to urban youth, little work has
explored the effects of nature on rural children. The data herein suggest that
there is little “ceiling effect” with respect to the benefits of exposure to the
natural environment. Even in a setting with a relative abundance of green
landscape, more appears to be better when it comes to bolstering children’s
resilience against stress or adversity.
NOTE
1. The plots presented in Figures 1 and 2 are created by centering both independent variables
(i.v.) (subtracting the mean from each value). Regressions are then rerun using centered i.v.s with
normal dependent variables (d.v.). Unstandardized beta coefficients from the regression analysis
are then used to determine the equation describing the d.v. Dichotomized or trichotomized inde-
pendent variable values (derived by adding and subtracting the standard deviation to zero—the
mean of the centered variable) are then entered into the equation to represent the results in cate-
gorical terms. These values are then plotted (Aiken & West, 1991).
REFERENCES
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.
Newbury Park, CA: Sage.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psy-
chological research: Conceptual, strategic, and statistical consideration. Journal of Person-
ality and Social Psychology, 51, 1173-1182.
Bartlett, S. N. (1997). No place to play: Implications for the interaction of parents and children.
Journal of Children and Poverty, 3, 37-48.
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 327
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
Bartlett, S. N. (1998). Does inadequate housing perpetuate children’s poverty? Childhood: A
Global Journal of Child Research, 5, 403-420.
Boyle, M., & Jones, S. (1985). Selecting measures of emotional and behavioral disorders of
childhood for use in general populations. Journal of Clinical Psychology and Psychiatry, 26,
137-159.
Bradley, R. H., Caldwell, B. M., & Rock, S. L. (1990). Home environment classification system:
A model for assessing the home environments of developing children. Early Education and
Development, 1, 238-265.
Bradley, R. H., Whiteside, L., Mundfrom, D. J., Casey, P. H., Kelleher, K. J., & Pope, S. K.
(1994). Early indications of resilience and their relation to experiences in the home environ-
ments of low birthweight, premature children living in poverty. Child Development, 65, 346-
360.
Cauce, A., Reid, M., Landesman, S., & Gonzales, N. (1992). Social support in young children. In
B. Sarason, I. Sarason, & G. Pierce (Eds.), Social support: An interactional view (pp. 64-94).
New York: John Wiley.
Cimprich, B. E. (1990). Attentional fatigue and restoration in individuals with cancer. Unpub-
lished doctoral dissertation, University of Michigan, Ann Arbor.
Cohen, S., Underwood, L., & Gottlieb, B. (2000). Social support measurement and intervention.
New York: Oxford University Press.
Coley, R. L., Kuo, F. E., & Sullivan, W. C. (1997). Where does community grow? The social con-
text created by nature in urban public housing. Environment & Behavior, 29, 468-494.
Department of the Environment. (1973). Children at play. London: Her Majesty’s Stationery
Office.
Dunst, C. J., & Trivette, C. M. (1994). Methodological considerations and strategies for studying
the long-term effects of early intervention. In S. L. Friedman & H. C. Haywood (Eds.),
Developmental follow-up: Concepts, domains, and methods (pp. 277-313). San Diego, CA:
Academic Press.
Evans, G. W. (2001). Environmental stress and health. In A. Baum, T. Revenson, & J. E. Singer
(Eds.), Handbook of health psychology (pp. 365-385). Mahwah, NJ: Lawrence Erlbaum.
Evans, G. W., & Kantrowitz, E. (2002). Socioeconomic status and health: The potential role of
environmental risk exposure. Annual Review of Public Health, 23, 202-231.
Evans, G. W., & Lepore, S. J. (1997). Moderating and mediating processes in environment—
Behavior research. In G. T. Moore & R. W. Marans (Eds.), Advances in environment, behav-
ior and design (Vol. 4, pp. 255-285). New York: Plenum.
Evans, G. W., Wells, N. M., Chan, H.-Y. E., & Saltzman, H. (2000). Housing quality and mental
health. Journal of Consulting and Clinical Psychology, 68, 526-530.
Evans, G. W., Wells, N. M., & Moch, A. (in press). Housing and mental health: A review of the
literature and a conceptual and methodological critique. Journal of Social Issues.
Faber Taylor, A., Kuo, F. E., & Sullivan, W. C. (2001). Coping with ADD: The surprising con-
nection to green play settings. Environment & Behavior, 33, 54-77.
Faber Taylor, A., Kuo, F. E., & Sullivan, W. C. (2002). Views of nature and self-discipline: Evi-
dence from inner city children. Journal of Environmental Psychology, 22, 49-63.
Faber Taylor, A., Wiley, A., Kuo, F. E., & Sullivan, W. C. (1998). Growing up in the inner city:
Green spaces as places to grow. Environment & Behavior, 30, 3-27.
Fanning, D. M. (1967). Families in flats. British Medical Journal, 4, 382-386.
Garbarino, J. (1982). Children and families in the social environment. New York: Aldine.
Gifford, R. (in press). Satisfaction, health, security and social relations in high rise buildings. In
A. Seidel & T. Heath (Eds.), Social effects of the building environment. London:E&FN
Spon.
328 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
Grahn, P., Mårtensson, F., Lindblad, B., Nilsson, P., & Ekman, A. (1997). Ute på dagis. Stad and
Land, Nr. 145 [Outdoor daycare. City and country]. Hässleholm, Sverige: Norra Skåne
Offset.
Harter, S., (1982) The perceived competence scale for children. Child Development, 53, 87-97.
Hartig, T., Mann, M., & Evans, G. W. (1991). Restorative effects of natural environment experi-
ences. Environment & Behavior, 23, 3-26.
Hüttenmoser, M. (1995). Children and their living surroundings: Empirical investigations into
the significance of living surroundings for the everyday life and development of children.
Children’s Environments, 12, 403-413.
Kahn, P. H., & Kellert, S. R. (Eds.). (2002). Children and nature: Psychological, sociocultural,
and evolutionary investigations. Cambridge, MA: MIT Press.
Kaplan, R. (1973). Some psychological benefits of gardening. Environment & Behavior, 5, 145-
152.
Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological perspective.New
York: Cambridge University Press.
Kaplan, S. (1995). The restorative benefits of nature: Toward an integrative framework. Journal
of Environmental Psychology, 15, 169-182.
Kaplan, S., & Kaplan, R. (1982). Humanscape: Environments for people. Ann Arbor, MI:
Ulrich’s.
Kaplan, S., & Kaplan, R. (1983). Cognition and environment: Functioning in an uncertain
world. Ann Arbor, MI: Ulrich’s.
Korpela, K. (2002). Children’s environments. In R. B. Bechtel & A. Churchman (Eds.), Hand-
book of environmental psychology (pp. 363-373). New York: John Wiley.
Kuo, F. E. (2001). Coping with poverty: Impacts of environment and attention in the inner city.
Environment & Behavior, 33, 5-34.
Kuo, F. E., Sullivan, W. C., Coley, R. L., & Brunson, L. (1998). Fertile ground for community:
Inner-city neighborhood common spaces. American Journal of Community Psychology, 26,
823-851.
Lewis, C. E., Siegel, J. M., & Lewis, M. A. (1984). Feeling bad: Exploring sources of distress
among pre-adolescent children. American Journal of Public Health, 74, 117-122.
Lynch, K. (Ed.). (1977). Growing up in cities: Studies of the spatial environments of adolescence
in Krakow, Melbourne, Mexico City, Salta, Toluca, and Warszawa. Cambridge, MA: MIT
Press.
McLoyd, V. (1990). The impact of economic hardship on Black families and children: Psycho-
logical distress, parenting, and socioemotional development. Child Development, 61, 311-
346.
McNaughton, M. E., Patterson, T. L., Irwin, M. R., & Grant, I. (1992). The relationship of life
adversity, social support, and coping to hospitalization with major depression. Journal of
Nervous and Mental Disease, 180, 491-497.
Moore, E. O. (1981). A prison environment’s effect on health care service demands. Journal of
Environmental Systems, 11, 17-34.
Moore, R. C. (1986). Childhood’s domain. London: Croom Helm.
Needleman, H. L. (1994). Preventing childhood lead poisoning. Preventive Medicine, 23, 634-
637.
Parsons, R., Tassinary, L. G., Ulrich, R. S., Hebl, M. R., & Grossman-Alexander, M. (1998). The
view from the road: Implications for stress recovery and immunization. Journal of Environ-
mental Psychology, 18, 113-139.
Roy, M. P., & Steptoe, A. (1994). Daily stressors and social support availability as predictors of
depressed mood in male firefighters. Work Stress, 8, 210-219.
Wells, Evans / LIFE STRESS AMONG RURAL CHILDREN 329
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American Journal of
Orthopsychiatry, 57, 316-331.
Rutter, M., Tizard, J., & Whitmore, K. (1970). Education, health, and behavior. London:
Longmans.
Sebba, R. (1991). The landscapes of childhood: The reflections of childhood’s environment in
adult memories and in children’s attitudes. Environment & Behavior, 23, 395-422.
Sherman, A. (1994). Wasting America’s future: The Children’s Defense Fund report on the cost
of child poverty. Boston: Beacon.
Sobel, D. (1993). Children’s special places: Exploring the role of forts, dens, and bush houses in
middle childhood. Tucson, AZ: Zephyr.
Spreen, O., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. (1984). Human developmental
neuropsychology. New York: Oxford University Press.
Tennessen, C. M., & Cimprich, B. (1995). Views to nature: Effects on attention. Journal of Envi-
ronmental Psychology, 15, 77-85.
Townsend, P. (1979). Poverty in the United Kingdom. Berkeley: University of California Press.
Ulrich, R. S. (1983). Aesthetic and affective responses to natural environments. In I. Altman & J.
F. Wohlwill (Eds.), Human behavior and environment: Vol. 6. Behavior and the natural envi-
ronment (pp. 85-125). New York: Plenum.
Ulrich, R. S. (1984). View through a window may influence recovery from surgery. Science, 224,
420-421.
Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E., Miles, M. A., & Zelson, M. (1991). Stress
recovery during exposure to natural and urban environments. Journal of Environmental Psy-
chology, 11, 201-230.
Vaux, A. (1988). Social support. New York: Praeger.
Verderber, S. (1986). Dimensions of person-window transactions in the hospital environment.
Environment & Behavior, 18, 450-466.
Verderber, S., & Reuman, (1987). Windows, views, and health status in hospital therapeutic
environments. Journal of Architectural and Planning Research, 4, 120-133.
Wachs, T. D., & Gruen, G. E. (1982). Early experience and human development. New York:
Plenum.
Wells, N. M. (2000). At home with nature: The effects of nearby nature on children’s cognitive
functioning. Environment & Behavior, 32, 775-795.
West, M. J. (1986). Landscape views and stress responses in the prison environment. Unpub-
lished master’s thesis, University of Washington, Seattle.
Wills, T. A., Blechman, E., & McNamara, G. (1996). Family support, coping, and competence.
In E. M. Hetherington & E. Blechman (Eds.), Stress, coping and resilience in children and
the family (pp. 107-133). Hillsdale, NJ: Lawrence Erlbaum.
Zube, E., Pitt, D., & Evans, G. W. (1983). A lifespan developmental study of landscape assess-
ment. Journal of Environmental Psychology, 3, 115-128.
330 ENVIRONMENT AND BEHAVIOR / May 2003
at OhioLink on April 5, 2009 http://eab.sagepub.comDownloaded from
