Landscape and Urban Planning 97 (2010) 273282

Contents lists available at ScienceDirect

Landscape and Urban Planning

journal homepage: www.elsevier.com/locate/landurbplan

Student performance and high school landscapes: Examining the links

Rodney H. Matsuoka
University of Michigan, School of Natural Resources and Environment, Ann Arbor, MI 48109, USA

a r t i c l e

i n f o

Article history:
Received 25 August 2009
Received in revised form 23 May 2010
Accepted 25 June 2010
Available online 24 July 2010
Keywords:
Nature contact
Academic achievement
Classroom behavior
Mental fatigue
Stress
High school students

a b s t r a c t
High school students today are experiencing unprecedented levels of school-related stress. At the same
time, a growing body of research has linked views of nature with restoration from mental fatigue and
stress reduction. How important are such views for students while they are at school? This study investigated 101 public high schools in southeastern Michigan to examine the role played by the availability of
nearby nature in student academic achievement and behavior. The analyses revealed consistent and systematically positive relationships between nature exposure and student performance. Specically, views
with greater quantities of trees and shrubs from cafeteria as well as classroom windows are positively
associated with standardized test scores, graduation rates, percentages of students planning to attend a
four-year college, and fewer occurrences of criminal behavior. In addition, large expanses of landscape
lacking natural features are negatively related to these same test scores and college plans. These featureless landscapes included large areas of campus lawns, athletic elds, and parking lots. All analyses
accounted for student socio-economic status and racial/ethnic makeup, building age, and size of school
enrollment.
2010 Elsevier B.V. All rights reserved.

1. Introduction
High school students have a great need for restorative and
stress-reducing environments, and this need may be growing.
School work loads and the competition that students face in the
college application process have increased to unprecedented levels in recent years (Mundy, 2005; Ramrez, 2009; U.S. Department
of Education, 2005). Research dealing with life events has cited
school-related issues as the leading sources of stress for this age
group (Ainslie et al., 1996; Kaiser Family Foundation, 2005; Stuart,
2006). In addition, high school dropout rates in major urban areas
throughout the United States are high and student satisfaction
with the high school experience has decreased signicantly (Dillon,
2009; Freeman, 2004).
At the same time, a growing body of research has linked views
of and access to nature with restoration from mental fatigue and
stress reduction. With regard to children and teenagers, this naturebased restoration process has been associated with higher test
scores (Heschong Mahone Group, 2003a), better attention levels
among children suffering from attention-decit hyperactivity disorder (Faber Taylor and Kuo, 2009; Faber Taylor et al., 2001; Kuo
and Faber Taylor, 2004), and greater cognitive functioning (Wells,

Present address: University of Illinois, Department of Landscape Architecture,

101 Temple Buell Hall, MC-620, 611 Taft Drive, Champaign, IL 61820, USA.
Tel.: +1 734 709 0811.
E-mail address: rodney0520@yahoo.com.
0169-2046/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.landurbplan.2010.06.011

2000). Researchers have also hypothesized that such restoration

should be positively linked with better behavior. For example, ndings have associated greater nature exposure with enhanced levels
of self-discipline in children (Faber Taylor et al., 2002). In addition,
both recovery from mental fatigue and stress were postulated to
explain the positive connections found between the presence of
indoor classroom plants and reductions in misbehaviors, feelings
of unfriendliness, and hours of sick leave of junior high school students (Han, 2009). These cognitive, social, and behavior benets
found among children and younger teenagers, then, should translate into better overall high school student performance involving
academic performance, interest in staying in school, and classroom
behaviors.
How important is such contact with outdoor nature for high
students while they are at school? What features of the campus
landscape have the most affect on student academic achievement
and behavior? Surprisingly, there appears to be little information
to answer these questions, particularly with respect to high school
aged students.
1.1. Nature contact benets in diverse settings
1.1.1. School setting
As Owens (1997, p. 158) suggested, there has been limited
interest in improvements to the design of exterior spaces at high
schools. In the context of elementary schools, the Heschong
Mahone Group (2003a) found that ample classroom window views
(i.e., 100 sq. ft. of window area or greater per classroom) that

274

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

included vegetation (i.e., primarily trees or shrubs) or human activity (e.g., playground, lunch area, parking lot), and objects in the
far distance were associated with higher scores on standardized
tests. Other studies in the grade school context have concentrated on playgrounds in preschool, kindergarten, and elementary
school. These studies have found positive connections between natural playscapes and enhanced physical activity (Dyment and Bell,
2007), motor development (Fjrtoft, 2004), creative play behaviors (Dyment and Bell, 2007; Herrington and Studtmann, 1998;
Tranter and Malone, 2004), environmental learning (Tranter and
Malone, 2004), and preference as compared to traditional playgrounds (Ozdemir and Yilmaz, 2008).
1.1.2. Other settings
In the context of the present study, research in workplace settings may be the most pertinent as high school students spend
many hours in school buildings. Studies have provided evidence
that views of nature out of an ofce or factory are associated
with increased employee productivity, enhanced feelings of job
and life satisfaction, greater psychological and physical well-being,
and reduced levels of frustration and stress (Heerwagen and Wise,
1998; Heschong Mahone Group, 2003b; Kaplan, 1993a; Leather et
al., 1998; Shin, 2007).
In addition, the psychological, social, and physical health benets of views of and access to nature for individuals have been shown
in residential settings (De Vries et al., 2003; Faber Taylor et al., 2002;
Gidlf-Gunnarsson and hrstrm, 2007; Jackson, 2003; Kaplan,
2001; Kearney, 2006; Kuo, 2001; Kuo and Sullivan, 2001a,b; Kuo
et al., 1998; Lee et al., 2008; Maas et al., 2009; Sullivan et al., 2004;
Tzoulas et al., 2007; Wells, 2000; Wells and Evans, 2003) including
college dormitories (Tennessen and Cimprich, 1995), prisons (e.g.,
Moore, 1981), and homes for elderly people (Ottosson and Grahn,
2005), and also hospital settings (Curtis et al., 2007; Ulrich, 1984).
A growing body of research, therefore, suggests that views of and
experiences with nearby nature provide many benets for individuals while at work, at home, imprisoned, or hospitalized. In spite of
these studies, however, we know very little about how exposure to
nature affects a tremendously important population high school
students at a time in their development when their academic
performance will set them on a life-course.
1.2. Explanations for these nature benets
Researchers have advanced varied explanations for the benets resulting from contact with nature. Two of the most widely
cited explanations are the attention restoration and the psychoevolutionary theories.
Attention restoration theory proposes that contact with nature
has the potential to restore an individuals directed attention
capabilities. Directed attention fatigue, or mental fatigue, occurs
when the capacity to focus or concentrate is reduced by overuse.
An individual experiencing such fatigue not only may have a
decreased ability to concentrate, but also may become more irritable, distractible, impulsive, antisocial, accident prone, and stressed.
This theory proposes that four sequential stages, which represent
greater levels of restorativeness, are experienced during the process of complete mental restoration. These include clearing the
head of miscellaneous thoughts, resting directed attention abilities, dealing with unresolved concerns, and nally reecting on
priorities, possibilities, values, actions, and goals. Reection represents the nal level of restorativeness, and is the most demanding
of all in terms of both the quality of the environment and the
duration required (Kaplan and Kaplan, 1989, p. 197). Natural environments possess qualities that are supportive of this restoration
process (Kaplan and Kaplan, 1989; Kaplan, 1993b, 1995).

In addition, attention restoration theory proposes that restorative environments possess four important components, namely
being away, extent, fascination, and compatibility (Kaplan and
Kaplan, 1989; Kaplan, 1995). In terms of this study, exposure
to greater levels of nature can provide students with enhanced
senses of both psychological distance from school (being away) and
immersion in conceptual surroundings of sufcient scope to sustain exploration (extent). Such exposure can also provide additional
environmental features that are effortlessly engaging (fascination)
and supportive of a students need for mental restoration (compatibility).
Psycho-evolutionary theory posits that natural settings have a
stress-reducing and calming effect on an individual. Immediate,
subconscious emotional responses play a key role in an individuals initial reaction to the environment. Nature provides a visually
pleasant physical surrounding that reduces stress by producing
positive emotions, sustaining nontaxing attention, and restricting
negative thoughts. Neurophysiological arousal is returned to more
moderate levels, fostering an overall sense of well-being (Hartig et
al., 1991; Ulrich et al., 1991).
In summary, the attention restoration theory concentrates on
cognitive processes while the psycho-evolutionary theory focuses
on emotionally based mechanisms. Nevertheless, both theories
support the idea that nature functions well as a restorative and
stress-reducing environment (Hartig et al., 2003).
1.3. Study overview
In light of the limited prior research, this study is necessarily
exploratory. Many of the school indoor and outdoor characteristics that were utilized to assess student exposure and access to
nature have not been investigated. In addition, this study will not
investigate the possible mechanisms explaining how such nature
contact improves student performance, but will utilize the explanations posited by researchers in contexts largely other than school
settings.
The central proposition of this study is that increased exposure
to nature will be positively associated with student performance,
including both student academic achievement and behavior. This
proposition was tested with the following hypotheses:
1. Higher levels of nature in the views that students have from
the school buildings will be positively associated with student
performance.
2. Higher levels of nature as determined by objectively measured
campus landscape elements will be positively associated with
student performance, in support of the more subjective measures utilized to investigate hypothesis #1.
3. Greater ability of students to view or come into direct contact
with nature, calculated by investigating building features and
school policy, will be positively associated with student performance.
4. A statistical interaction exists between the size of school building windows and the levels of nature in the views afforded
with regard to student performance. This interaction effect will
be positive with the effect of larger windows increasing across
nature levels, and the effect of higher nature levels increasing
across window size.
2. Method
2.1. High schools studied
The high schools studied consisted of 101 public schools
located in southeastern Michigan, USA (Fig. 1). The schools were

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

275

Fig. 1. The high schools studied were located in Lenawee, Livingston, Monroe, Oakland, Washtenaw, and Wayne Counties, Michigan (highlighted in gray).

limited to one region to minimize differences in campus vegetation, layouts, and building designs, and school district policies
and climate. To obtain a more homogenous sample of students,
private high schools, public high schools offering alternative educational or magnet programs, and high schools that were combined
with elementary or middle schools were excluded from the
study.
Of the 137 schools originally contacted, thirty-six schools (26%)
were not included in the nal database. Two schools were undergoing extensive renovations and the school districts of thirty-four
schools either denied permission or required an approval process that was too lengthy to be completed during the time period
allotted for data collection. Fourteen of the excluded schools were
located in inner-cities, eight in other urban settings, eight in rural
areas, and six were urban-fringe schools.
Information about each facility was obtained from the principal,
vice-principal, or other front ofce personnel through interviews
and unscheduled drop-in questioning. Site visits were conducted
to inventory the landscape and building features of each school.
Additional data were obtained from the web sites of the Michigan
Department of Education and Information Technology, Standard

and Poors School Matters, Public School Review, United States

Geological Survey (USGS), and the GIS departments of Wayne and
Livingston counties, Michigan. All of the information collected was
for the 20042005 academic school year.
2.2. Constructs and measures
2.2.1. Student exposure to nature
The investigation into student exposure to nature at each school
involved three groups of measures. First, the views of nature that
that students had from the school buildings were rated. Second,
vegetation levels on the campuses were measured. Third, student
potential access to this vegetation was determined.
2.2.1.1. Views of nature ratings. The degrees of naturalness in the
views from the school cafeteria and the classrooms were separately
rated by the principal researcher, based on site visits to each campus. These two measures represent nature contact during different
student activities, were not signicantly correlated, and were kept
as separate variables:

276

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

Fig. 2. Examples of the level of naturalness in the view from each schools primary cafeteria window.

Cafeteria nature the level of naturalness in the view from each

schools primary cafeteria window was rated on a 5-point scale.
The scale points were dened as follows, based on similar denitions used in a study by Tennessen and Cimprich (1995) (see
Fig. 2):
0- No view consisted of cafeterias without any window to the
outdoors.
1- All built consisted of buildings, roads, and walkways without
any vegetation present.
2- Mostly built indicated that the majority of what could be
seen was built, but with natural elements such as a few trees
and shrubs present, or that the majority of what could be seen
was athletic elds or large expanses of lawn devoid of trees
and shrubs.
3- Mostly natural included evidence of human presence such
as walkways, paved courtyards, and roads along with a mostly
natural setting.
4- All natural consisted of trees, shrubs, and forest remnants
without any evidence of human inuence.
Classroom nature the level of naturalness in the views from
all of the regular classrooms at each school were rated using
the same 5-point scale as dened for the cafeteria nature, and
then averaged. Regular classrooms were dened as follows
classrooms used for teaching such subjects as history, literature, math, social science, or physical science. Classrooms used
for shop (e.g., auto, wood, metal), and other specialized classrooms (e.g., planetarium, greenhouse) were excluded from the
calculations.

2.2.1.2. Vegetation levels on each campus (objective measures). The

amount of vegetation at each school was objectively measured with
the following campus areas and campus vegetation variables:

Campus areas: Each campus was divided into the following three
areas, and the respective acreage of these areas were calculated
from GIS data (see Fig. 3):
Athletic elds football, soccer, and baseball elds, tennis courts,
and other outdoor sport facilities.
Parking lots parking lots and roadways within school boundaries, including all areas of vegetation contained within or
surrounded by these features and the school borders.
Landscaped areas the areas between the school buildings and
athletic elds, parking lots, and school boundaries, and outdoor
courtyards totally surrounded by school buildings.

Fig. 3. Campus boundaries and areas measured.

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

In an attempt to account for enrollment size, for each school

these three areas were divided by this schools total number of students. In this manner, athletic eld, parking lot, and landscaped
areas on a per student basis were calculated.
Campus vegetation in the landscaped area: Site visits were conducted to count the number of trees, measure the area covered
by shrubs and groundcover, and conrm the presence of lawns
depicted in GIS aerial photographs. Lawn area and landscape area
were calculated from GIS data. These three variables measured
unique aspects of the campus landscape and were kept as separate
variables (see Fig. 4).
Tree density the number of trees per acre of landscaped area.
Shrubs per landscaped area the percentage of the landscaped
area made up of shrubs and groundcover.
Lawn per landscaped area the percentage of the landscaped area
made up of mowed grass.
2.2.1.3. Student potential access to nature. The ability of students
to view or come into direct contact with nature on each campus
during the school day was measured with building window sizes
and school policies. The classroom and cafeteria windows were
measured by the principal researcher during site visits. These two
window measures represent potential access to nature during different student activities, were not signicantly correlated, and were
kept as separate variables:
Building window size:
Classroom window area the average total window area of regular classrooms.
Cafeteria window area the total area of the cafeteria windows was categorized by the principal researcher on a 05 scale
reecting the percentage of the primary outward facing wall that
consisted of a window where: 0 = no windows, 1 = 25%, 2 = 50%,
3 = 75%, 4 = 100%, and 5 = windows made up 100% of a greater than
one story tall primary outward facing wall. The total window area
of each schools cafeteria was not used due to the tremendous
diversity of cafeteria room sizes and congurations as compared
to classrooms. The measure chosen was an attempt to avoid more
complicated adjustment calculations (e.g., dividing the total window area by each schools enrollment size).
School policy:
Eat lunch outdoors were the students are allowed to eat lunch
outdoors.
Lunch time lengths amount of time students are given to eat
lunch.
2.2.2. Student performance measures
Student performance was evaluated by looking at ve aggregate
student academic achievement and behavior measures collected
and reported by each school to the Michigan Department of Education. Academic achievement was measured in the following three
ways:
Michigan merit award the percentage of award winners based
on student performance on the Michigan Educational Assessment
Program (MEAP) test. All Michigan public high school students
were required to take this test during the years 19692006
(Michigan Department of Education, 2006).
Graduation rates as reported to the state.
Four-year college plans the percentage of seniors stating that
they planned to attend a four-year college upon graduation.
Seventy-eight of the high schools studied polled their outgo-

277

ing seniors. None of the schools conducted follow-up surveys to

determine how many of their students actually attended college.
Behavior was measured with the following two variables:
Student disorderly conduct the relative frequencies of seven
types of student discipline problems were averaged, namely
student social tensions, bullying, verbal abuse of teachers, insubordination, acts of disrespect for teachers, physical attacks or
ghting, and truancy.
Student criminal activity the occurrences of six types of student
criminal activities were averaged, namely physical violence, illegal possession, vandalism, verbal assault, larceny, and minor in
possession.
2.2.3. Control variables: four important factors related to student
performance
Student socio-economic status, and family and racial/ethnic
background (Coleman et al., 1966; Fowler and Walberg, 1991;
Gottfredson et al., 2005; Hanushek, 1997; Rouse and Barrow, 2006;
Rumberger and Palardy, 2005; Welsh, 2001), school enrollment size
(Cotton, 1996; Ready et al., 2004; Schneider, 2002; Williams, 1990),
and building age (Earthman and Lemasters, 1996; Kumar et al.,
2008; McGuffey, 1982; National Research Council, 2006; Schneider,
2002; Uline, 2000) have repeatedly been shown to be strongly
related to student academic achievement and misbehaviors. Due to
their strong connection with student performance, these following
four aggregate factors for each school were accounted for in the
statistical analyses conducted in this study.
School socio-economic status student participation in free or
reduced lunch programs.
Ethnicity total percentage of all of the minority racial/ethnic
groups.
Enrollment the number of students enrolled.
Building age the age of the primary classroom building.
2.3. Statistical analyses
The hypotheses were evaluated using both linear and nonlinear regression analyses conducted with SPSS version 15.0. Linear
regression was used to determine the predictors of the three academic achievement measures. However, the occurrences of student
disorderly conduct and criminal activity were relatively rare and
did not meet the assumptions of the linear regression models. For
student disorderly conduct, the Poisson regression model was used,
as this model is specically suited for many types of count data
(Gujarati, 2003). The negative binomial regression model was utilized for student criminal activity. The data distribution for this
variable was over-dispersed, which means that the variance (i.e.,
5016.93) was much larger than the mean (i.e., 51.48 occurrences).
In these situations, the negative binomial rather than the Poisson
regression model provides better estimates (Gujarati, 2003).
3. Results
3.1. Relationships between student performance and the four
control variables
Consistent with previous research, three of the four control variables, namely school socio-economic status, ethnic/racial makeup
of the student body, and the age of the main school building, are
strongly related to the ve student performance measures used in
this study (see Table 1). Given these signicant relationships, the
four control variables were introduced into the regression analyses

278

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

Fig. 4. Examples of schools with a high density of trees or high percentage of lawn in the landscaped area.

used to explore the associations between the possible predictors of

student performance and the ve student academic achievement
and behavior outcome measures. No signicant collinearity problems among these variables were observed in any of these analyses.
These four variables were kept in the model regardless of whether
they were signicant predictors of a particular outcome variable.
3.2. Testing the central proposition
A series of regression analyses were conducted with the measures of student exposure to nature as predictor variables and the
ve student performance indices as outcome variables. Predictors
from the three groups of student exposure to nature variables were
found to explain a signicant portion of the variance for one or more
of the outcome variables. These predictors will be discussed individually in terms of the four hypotheses utilized to investigate the
central proposition.

3.2.1. Hypothesis #1
Are higher levels of nature in the views that students have
from the school buildings positively associated with student performance? Yes, as Table 2 shows, cafeteria nature is positively
associated with each of the three measures of student academic
achievement. These ndings are the most signicant discovery
of this study. For Michigan Merit Award recipients, graduation
rates, and four-year college plans, these ratings explain 4.8%,
3.7%, and 12.2% of their respective variance. Additional analysis revealed that cafeteria nature is also signicantly correlated
with cafeteria window area (see Table 3), the latter variable not
being a signicant predictor of student performance. Therefore,
the schools with greater quantities of natural features in their
cafeteria windows also provided their students with larger views
of these outdoor environments. In addition, the average length
of time given for students to eat lunch was 40.2 min (n = 98,
standard deviation = 7.45). Classroom nature, in contrast, was not

Table 1
Correlations among the four control variables and the ve student performance measures.

School SES
Ethnicity
Building age
Enrollment
Michigan merit award
Graduation rates
Four-year college plans
Student disorderly conduct
Student criminal activity
*
**

p < .05.
p < .01.

School SES

Ethnicity

Building age

Enrollment

.73**
.39**
.20*
.83**
.78**
.26*
.43**
.29**

.34**
.05
.63**
.76**
.12
.27**
.11

.02
.35**
.34**
.08
.05
.12

.16
.05
.17
.07
.00

Michigan merit
award

.72**
.50**
.39**
.34**

Graduation rate

.16
.40**
.28**

Four-year college
plans

Student disorderly
conduct

.32**
.37**

.37**

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

279

Table 2
Student performance regressed onto all three groups of nature exposure variables.

Control variables
School SES
Ethnicity
Building age
Enrollment

Michigan merit award

Graduation rates

Four-year college plans

Student disorderly
conduct

Student criminal
activity

p value

p value

p value

p value

<.001

0.46
0.41

<.001
<.001

0.59
0.32
0.31

<.001
<.01
<.01

0.01a

<.001

0.05a
0.02a

0.75
ns
ns
ns

Adjusted R square control variables

Views of nature ratings cafeteria nature

ns
ns

0.68
0.14

ns

0.67
<.001

0.36b

0.17

0.17

ns
ns
ns

<.01

0.25

<.01

ns

ns

ns
ns
ns

0.26
0.31
0.24

<.01
<.05
<.05

ns
ns
ns

0.02a

Student potential access to nature

Classroom window area

ns

ns

0.23

<.01

ns

0.01a

0.73
98

0.69
96

a
b

0.36b
98

0.53
76

<.001
<.01
ns
ns

0.42b

Vegetation levels on each campus (objective measures)

Lawn per landscaped area
0.13
<.05
Parking lot area per student
ns
Athletic elds
ns

Adjusted R square entire model

Observations

p value

<.05
ns
ns

<.05
0.51b
97

Nonstandardized B values are reported for the Poisson and negative binomial generalized linear regression models.
Pseudo R-squared values are reported for the Poisson and negative binomial generalized linear regression models.

signicantly associated with any of the student performance measures.

3.2.2. Hypothesis #2
Did the ndings concerning the associations between the more
objective measures of campus landscape elements and student performance support those of hypothesis #1. Yes, as Table 2 shows,
landscapes made up by mowed grass and parking lots are associated with poorer student performance, than landscapes composed
primarily of trees and shrubs. This agrees with the rating system
used to measure the level of naturalness in the views from the
school building. As discussed in the method section, to receive
the rating of 3, the view had to be mostly natural, in other
words dominated by trees, shrubs, and natural features other than
lawns. Views made up primarily of built elements or large expanses
of lawns without many trees or shrubs were given the lower
rating of 2 signifying a mostly built view. Higher percentages
of lawn are related with fewer students receiving the Michigan
Merit Award and planning to attend four-year colleges, and more
student criminal activity, explaining 1.8%, 2.6%, and 3.4% of the variance, respectively. In addition, greater parking area per student
levels and athletic eld areas are negative predictors of fouryear college plans, accounting for 15.0% and 6.2% of the variance,
respectively. None of the other measures of campus vegetation
used in this study is signicantly associated with student performance.

3.2.3. Hypothesis #3
Is greater ability of students to view or come into contact with
nature, as measured by investigating building features and school
policy, positively related to student performance? Yes, as Table 2
reveals, larger classroom windows are positively associated with
three measures of student performance. Students at schools with
larger average classroom window areas plan to attend four-year
colleges at a higher rate as compared to schools with smaller window areas, and also commit fewer crimes while at school. For
four-year college plans and student criminal activity, this building feature explains 8.4% and 8.5% of the variance, respectively.
There is a signicant correlation between classroom window area
and classroom nature and tree density in the landscaped areas (see
Table 3). Neither of these latter variables are signicant predictors
of student performance. Nonetheless, the views from classrooms
with larger window area are more likely to include greater concentrations of trees and shrubs in the landscapes adjacent to the
classroom buildings.
3.2.4. Hypothesis #4
Do school building window sizes and the levels of nature in
the views afforded affect each other, with regards to student performance, in a statistically signicant manner? No, the results of
this study did not support the expected relationships (e.g., through
mediation) involving classroom and cafeteria windows and their
corresponding views. No theoretical explanation can be provided

Table 3
Correlations among the subjective views of nature ratings, objective measures of campus vegetation in the landscaped area, and building window sizes.
Cafeteria nature
Cafeteria nature
Classroom nature
Tree density
Shrubs per landscaped area
Lawn per landscaped area
Classroom window area
Cafeteria window area
*
**

p < .05.
p < .01.

.21*
.15
.09*
.02
.04
.68**

Classroom nature

.02
.22*
.15
.27**
.13

Tree density

.06
.22*
.27**
.18

Shrubs per landscaped

area

Lawn per landscaped

area

Classroom window
area

.16
.04
.11

.18
.07

.20

280

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

for this nding. Perhaps the ve aggregate student performance

outcome measures utilized in this study were not sensitive enough
to detect these interactions.
3.2.5. Conrmation of the central proposition
Results from the regression analyses, therefore, support the central proposition and indicated that increased exposure to nature
during the school day is positively associated with student performance. In fact, all three student academic achievement measures
and one measure of student behavior are signicantly related to
one or more of the higher nature exposure predictors used (see
Table 2). For the three achievement measures and student criminal activity, the independent contributions of enhanced exposure
to nature variables are between 3.7% and 20.6% of the variance.
Thus, except for student disorderly conduct, the central proposition
that student academic achievement and behavior is associated with
student exposure to nature earns strong support. These effects are
signicant while accounting for the inuences of the four control
variables.
4. Discussion
This study contributes to our understanding of how the physical
environment, particularly natural features, can promote academic
performance in a high school setting. Analyses revealed consistent,
systematically positive relationships between student exposure to
nature during their lunch time and scores on standardized tests,
graduation rates, and plans to attend a four-year college. Furthermore, the study documented that not all forms of vegetation are
positively associated with school achievement and behavior. While
the presence of trees and shrubs has a positive connection, the
prevalence of lawn areas is negatively associated with standardized
test scores, four-year college plans, and criminal behavior. These
ndings held while accounting for a variety of factors known to
inuence academic performance.
4.1. The importance of lunch time
The most consistent results center on student exposure to more
natural landscapes during lunch time. School cafeteria window
views with higher vegetation content are positively associated
with all three measures of student academic achievement. The
importance of these views was unexpected since students spend
substantially more time in the classroom on a given school day.
Nonetheless, lunch time may be especially pertinent as it provides
students with a valuable break from the learning process. Students
can recover from mental fatigue and stress, and reect on events
that occurred during the rst portion of the school day.
While the results suggest that lunch time views play a stronger
role, the positive effects of viewing more vegetation are not limited
just to lunch time. Larger classroom window areas were related
with greater percentages of students planning to attend four-year
colleges and fewer occurrences of student criminal behaviors. In
addition, these larger windows were more likely to view higher
densities of trees and shrubs in the landscapes adjacent to the
school buildings.
These positive associations of contact with nature during lunch
time and from the classrooms with student performance support
the propositions of both the attention restoration and psychoevolutionary theories. The strong impact of the cafeteria context is,
perhaps, more thoroughly explained by the attention restoration
theory. This theory proposes that four sequential stages, representing greater levels of restorativeness, are experienced during
the process of complete mental restoration (Kaplan and Kaplan,
1989; Kaplan, 1995). Although the nature content of classroom
window views is important, students may not have adequate time

during class time. While in class, students are under a teachers

constant supervision and are busy concentrating on the lessons
being presented. Indeed, the reection process may be equated
with daydreaming, a behavior that is denitely frowned upon by
most teachers.
An argument can be made that due to the social interactions
that occur among students during lunch time, mental restoration
may not be possible. For example, researchers have discovered that
people nd solitude more effective than company in regaining the
capacity for directed attention, provided that the safety that company may bring is guaranteed (Staats and Hartig, 2004, p. 209). In
support of this studys ndings, though, one can contend that some
degree of mental restoration may be achieved in a short period of
time. The minimum amount of time needed has yet to be determined. However, studies have revealed that participants exposed
to photographs and videos of natural environments for six to seven
minutes attained some degree of restoration (Berto, 2005; Van den
Berg et al., 2003). In addition, a student who wishes to be alone may
be able to nd a private place during lunch time away from his or
her peers. Such physical privacy would not be possible during class
time. Finally, researchers have found that children playing together
in more natural playgrounds display greater civil and cooperative behaviors (Dyment and Bell, 2008; Herrington and Studtmann,
1998). These researchers did not cite mental restoration per se as
the underlying reason, but these ndings are in agreement with the
tenets of the Kaplans attention restoration theory.
4.2. Landscapes lacking natural features
Large expanses of campus landscape lacking trees and shrubs,
specically lawns and parking lots, are negatively associated with
student performance. In the context of school and work settings,
prior research has shown decreased performance and increased
frustration and stress as well as dissatisfaction with the environment when nature is less available in the immediate view
(Heschong Mahone Group, 2003a,b; Kaplan, 1993a,b; Leather et al.,
1998; Tennessen and Cimprich, 1995). As the ndings reported here
document, however, the presence of a natural feature like mowed
grass is not sufcient. An explanation for these apparently contradictory results is provided by linking the ndings of studies in the
diverse elds of landscape preference, residential neighborhood
satisfaction, and student productivity.
Landscape preference research has consistently shown that
large, at landscapes lacking trees and shrubs are often aesthetically less preferred, as compared to other natural settings (Kaplan
and Kaplan, 1989; Schroeder, 1987; Ulrich, 1986). Studies have also
revealed that views of less preferred landscapes are associated with
lower levels of neighborhood satisfaction and senses of well-being
(Kaplan, 2001; Kearney, 2006). In addition, researchers have determined that student psychological well-being and satisfaction with
academic life are positively related to measures of school performance and productivity (Chambel and Curral, 2005; Chow, 2007;
Cotton et al., 2002). The effects of these variables in school settings
could be expected to be similar to their effects in other contexts.
Further research would help to test the appropriateness of this
explanation for the negative relationships that exist in this study
between views of campus landscapes lacking natural features and
student performance.
4.3. A link to a students future
The ndings of this study revealed that student exposure to
nature measures were signicantly associated with the future fouryear college plans of graduating seniors. Most prior research has
only examined the effects of the physical environment on more
immediate outcomes, such as standardized test scores and behav-

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

iors. This studys results suggest that campus landscape features

can have connections with a students future academic and career
accomplishments.
4.4. Not just an aesthetic amenity
Finally, natural features of the campus landscapes explained
5.2% of the variance in the test scores required to be a recipient
of the Michigan Merit Award. This is comparable to the reported
36% of the variance in test scores explained by school building features in prior research (Earthman, 2004; National Research
Council, 2006). The often overlooked outdoor physical environment
of schools can perhaps have as much inuence on student performance as the more intensely examined indoor features. The campus
landscape should not be considered just an aesthetic amenity, but
as important as the school buildings themselves.
4.5. Limitations
Only the principal researcher was able to visit each of the high
school campuses studied. As a result, the reliability of the building view ratings, building window sizes, and the presence of trees,
shrubs, and areas of lawn could not be analyzed.
In addition, single measures were used in this study to calculate
each of the predictor variables found to be signicantly associated with the student performance outcome measures. The student
exposure to nature variables represented either views to nature
during different student activities or corresponded to distinct elements of the landscape area or school buildings. These variables
were, therefore, not combined. Further studies are needed that
utilize additional methods to measure these variables to enhance
measurement validity.
5. Conclusions
High school dropout rates are a national concern and student
satisfaction with the high school experience has decreased signicantly in recent years. The students who attend experience
a multitude of school-related stresses. Insights provided by the
results of this study can aid school administrators and designers
in building and renovating campus landscapes in ways that may
reduce student mental fatigue and stress levels, increase satisfaction with the school environment, and enhance overall student
academic achievement and behavior.
The overall results are supported by the tenets of both the
psycho-evolutionary and attention restoration theories. The ability to view nature from the cafeteria and classroom was found to
be benecial for student academic achievement and behavior.
Finally, this studys ndings have linked the benets of greater
nature contact not only to the current performance of students, but
also to their future college plans. Regardless of the socio-economic
level, racial/ethnic makeup, enrollment size, and age of the indoor
facilities of a school, high school students seem to benet from
visual access to nearby nature during their school day. Current students attending such schools will benet and the returns on the
funds spent will continue throughout the lifetimes of both these
landscapes and the graduates themselves.
Acknowledgements
This study was supported by funding from the University of
Michigans Rackham Graduate School, administered by the School
of Natural Resources and Environment, the Department of Landscape Architecture at the University of Illinois, Urbana-Champaign,
as well as the USDA Forest Service, Northern Research Station

281

through the efforts of Lynne M. Westphal. The author also wishes

to express deep gratitude for the assistance of many other individuals and institutions. Rachel Kaplan contributed to the study
in many ways: in terms of conceptual ideas, interpretation of the
results, and help in publication revisions. Stephen Kaplan, William
C. Sullivan, and Raymond De Young provided valuable input in
data interpretation and editing. Jason Duvall provided important
insights concerning data analysis. In addition, the author wishes
to thank the front ofce personnel of the 101 Michigan public
high schools analyzed in this study for their cooperation and assistance.

References
Ainslie, R.C., Shafer, A., Reynolds, J., 1996. Mediators of adolescents stress in a college
preparatory environment. Adolescence 31 (124), 913924.
Berto, R., 2005. Exposure to restorative environments helps restore attentional
capacity. Journal of Environmental Psychology 25 (3), 249259.
Chambel, M.J., Curral, L., 2005. Stress in academic life: work characteristics as
predictors of student well-being and performance. Applied Psychology: An
International Review 54 (1), 135147.
Chow, H.P.H., 2007. Psychological well-being and scholastic achievement among
university students in a Canadian prairie city. Social Psychology of Education
10, 483493.
Coleman, J.S., Campbell, E.Q., Hobson, C.J., McPartland, J., Mood, A.M., Weinfeld, F.D.,
et al., 1966. Equality of Educational Opportunity. U.S. Department of Health,
Education, and Welfare, Washington, DC.
Cotton, K., 1996. School Size, School Climate, and Student Performance. Northwest
Regional Educational Laboratory, Portland, OR.
Cotton, S.J., Dollard, M.F., de Jonge, J., 2002. Stress and student job design: satisfaction, well-being, and performance in university students. International Journal
of Stress Management 9 (3), 147162.
Curtis, S., Gesler, W., Fabian, K., Francis, S., Priebe, S., 2007. Therapeutic landscapes in
hospital design: a qualitative assessment by staff and service users of the design
of a new mental health inpatient unit. Environmental Planning C: Government
and Policy 25, 591610.
De Vries, S., Verheij, R.A., Groenewegen, P.P., Spreeuwenberg, P., 2003. Natural
environmentshealthy environments? An exploratory analysis of the relationship between greenspace and health. Environment and Planning A 35 (10),
17171731.
Dillon, S., 2009. Large urbansuburban gap seen in graduation rates. The New York
Times, A14.
Dyment, J.E., Bell, A.C., 2007. Active by design: promoting physical activity through
school ground greening. Childrens Geographies 5 (4), 463477.
Dyment, J.E., Bell, A.C., 2008. Grounds for movement: green school grounds as sites
for promoting physical activity. Health Education Research 23 (6), 952962.
Earthman, G.I., 2004. Prioritization of 31 Criteria for School Building Adequacy.
American Civil Liberties Union Foundation of Maryland, Baltimore, MD.
Earthman, G.I., Lemasters, L., 1996. Review of research on the relationship between
school buildings, student achievement, and student behavior. In: Paper Presented at the Council of Educational Facility Planners, International Annual
Meeting, Tarpon Springs, Florida, October 8.
Faber Taylor, A., Kuo, F.E., 2009. Children with attention decits concentrate better
after walk in the park. Journal of Attention Disorders 12 (5), 402409.
Faber Taylor, A., Kuo, F.E., Sullivan, W.C., 2001. Coping with ADD the surprising
connection to green play settings. Environment and Behavior 33 (1), 5477.
Faber Taylor, A., Kuo, F.E., Sullivan, W.C., 2002. Views of nature and self-discipline:
evidence from inner city children. Journal of Environmental Psychology 22 (12),
4963.
Fjrtoft, I., 2004. Landscape as playscape: the effects of natural environments on
childrens play and motor development. Children, Youth and Environments 14
(2), 2144.
Fowler, W.J.J., Walberg, H.J., 1991. School size, characteristics, and outcomes. Educational Evaluation and Policy Analysis 13 (2), 189202.
Freeman, C.E., 2004. Trends in Educational Equity of Girls & Women: 2004, NCES
2005-016. U.S. Department of Education, National Center for Education Statistics, Washington, DC.
Gidlf-Gunnarsson, A., hrstrm, E., 2007. Noise and well-being in urban residential
environments: the potential role of perceived availability to nearby green areas.
Landscape and Urban Planning 83 (23), 115126.
Gottfredson, G.D., Gottfredson, D.C., Payne, A.A., Gottfredson, N.C., 2005. School climate predictors of school disorder: results from a national study of delinquency
prevention in schools. Journal of Research in Crime and Delinquency 42 (4),
412444.
Gujarati, D.N., 2003. Basic Econometrics, 4th edition. McGraw-Hill Companies, Inc.,
New York.
Han, K.-T., 2009. Inuence of limitedly visible leafy indoor plants on the psychology,
behavior, and health of students at a junior high school in Taiwan. Environment
and Behavior 41 (5), 658692.
Hanushek, E.A., 1997. Assessing the effects of school resources on student performance: an update. Educational Evaluation and Policy Analysis 19 (2), 141164.

282

R.H. Matsuoka / Landscape and Urban Planning 97 (2010) 273282

Hartig, T., Evans, G.W., Jamner, L.D., Davis, D.S., Grling, T., 2003. Tracking restoration
in natural and urban eld settings. Journal of Environmental Psychology 23,
109123.
Hartig, T., Mang, M.M., Evans, G.W., 1991. Restorative effects of natural environment
experiences. Environment and Behavior 23 (1), 326.
Heerwagen, J.H., Wise, J.A., 1998. Green building benets: differences in perceptions
and experiences across manufacturing shifts. Heating/Piping/Air Conditioning
70 (February), 5763.
Herrington, S., Studtmann, K., 1998. Landscape interventions: new directions for the
design of childrens outdoor play environments. Landscape and Urban Planning
42 (24), 191205.
Heschong Mahone Group, 2003a. Windows and Classrooms: A Study of Student
Performance and the Indoor Environment. California Energy Commission, Sacramento, CA.
Heschong Mahone Group, 2003b. Windows and Ofces: A Study of Ofce Worker
Performance and the Indoor Environment, Technical Report P500-03-082-A-9.
California Energy Commission, Sacramento, CA.
Jackson, L.E., 2003. The relationship of urban design to human health and condition.
Landscape and Urban Planning 64 (4), 191200.
Kaiser Family Foundation, 2005. Survey of Teens in the Greater Washington, DC
Area. Henry J. Kaiser Family Foundation, Menlo Park, CA.
Kaplan, R., 1993a. The role of nature in the context of the workplace. Landscape and
Urban Planning 26 (14), 193201.
Kaplan, R., 2001. The nature of the view from homepsychological benets. Environment and Behavior 33 (4), 507542.
Kaplan, R., Kaplan, S., 1989. The Experience of Nature: A Psychological Perspective.
Cambridge University Press, Cambridge, NY.
Kaplan, S., 1993b. The role of natural environment aesthetics in the restorative experience. In: Gobster, P.H. (Ed.), Managing Urban and High-Use Recreation Settings,
General Technical Report NC-163. Forest Service, USDA, St. Paul, MN, pp. 46
49.
Kaplan, S., 1995. The restorative benets of nature toward an integrative framework. Journal of Environmental Psychology 15 (3), 169182.
Kearney, A.R., 2006. Residential development patterns and neighborhood satisfaction impacts of density and nearby nature. Environment and Behavior 38 (1),
112139.
Kumar, R., OMalley, P.M., Johnston, L.D., 2008. Association between physical environment of secondary schools and student problem behavior: a national study,
20002003. Environment and Behavior 40 (4), 455486.
Kuo, F.E., 2001. Coping with poverty impacts of environment and attention in the
inner city. Environment and Behavior 33 (1), 534.
Kuo, F.E., Faber Taylor, A., 2004. A potential natural treatment for attentiondecit/hyperactivity disorder: evidence from a national study. American Journal
of Public Health 94 (9), 15801586.
Kuo, F.E., Sullivan, W.C., 2001a. Environment and crime in the inner city does
vegetation reduce crime? Environment and Behavior 33 (3), 343367.
Kuo, F.E., Sullivan, W.C., 2001b. Aggression and violence in the inner city effects
of environment via mental fatigue. Environment and Behavior 33 (4), 543
571.
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 (6), 823851.
Leather, P., Pyrgas, M., Beale, D., Lawrence, C., 1998. Windows in the workplace: sunlight, view, and occupational stress. Environment and Behavior 30 (6), 739762.
Lee, S.-W., Ellis, C.D., Kweon, B.-S., Hong, S.-K., 2008. Relationship between landscape structure and neighborhood satisfaction in urbanized areas. Landscape
and Urban Planning 85 (1), 6070.
Maas, J., Spreeuwenberg, P., Van Winsum-Westra, M., Verheij, R.A., de Vries, S.,
Groenewegen, P.P., 2009. Is green space in the living environment associated
with peoples feelings of social safety? Environment and Planning A 41 (7),
17631777.
McGuffey, C.W., 1982. Facilities. In: Walberg, H.J. (Ed.), Improving Educational Standards and Productivity: The Research Basis for Policy. McCutchan Publishing
Corporation, Berkeley, CA, pp. 237288.
Michigan Department of Education, 2006. School Assessment and Accountability,
Retrieved May 21, 2006, from: http://www.michigan.gov/mde/0,1607,7-14043092-00.html.
Moore, E.O., 1981. A prison environments effect on health-care service demands.
Journal of Environmental Systems 11 (1), 1734.
Mundy, L., 2005. High anxiety. The Washington Post (October), W20.
National Research Council, 2006. Review and Assessment of the Health and Productivity Benets of Green Schools: An Interim Report (2006). The National
Academies Press, Washington, DC.

Ottosson, J., Grahn, P., 2005. A comparison of leisure time spent in a garden with
leisure time spent indoors: on measures of restoration in residents in geriatric
care. Landscape Research 30 (1), 2555.
Owens, P.E., 1997. Adolescence and the cultural landscape: public policy, design
decisions, and popular press reporting. Landscape and Urban Planning 39,
153166.
Ozdemir, A., Yilmaz, O., 2008. Assessment of outdoor school environments and physical activity in Ankaras primary schools. Journal of Environmental Psychology
28 (3), 287300.
Ramrez, E., 2009. Schools battle student stress with creative strategies.
U.S. News & World Report (March), Retrieved August 22, 2009, from:
http://www.usnews.com/articles/education/2009/03/20/schools-battlestudent-stress-with-creative-strategies.html.
Ready, D.D., Lee, V.E., Welner, K.G., 2004. Educational equity and school structure:
school size, overcrowding, and schools-within-schools. Teaching College Record
106 (10), 19892014.
Rouse, C.E., Barrow, L., 2006. U.S. elementary and secondary schools: equalizing
opportunity or replicating the status quo? Future Children 16 (2), 99123.
Rumberger, R.W., Palardy, G.J., 2005. Test scores, dropout rates, and transfer rates as
alternative indicators of high school performance. American Education Research
Journal 42 (1), 342.
Schneider, M., 2002. Do School Facilities Affect Academic Outcomes? National Clearinghouse for Educational Facilities, Washington, DC.
Schroeder, H.W., 1987. Environment, behavior, and design research on urban forests.
In: Zube, E.H., Moore, G.T. (Eds.), Advances in Environment, Behavior, and Design,
vol. 2. Plenum Press, New York, pp. 87117.
Shin, W.S., 2007. The inuence of forest view through a window on job satisfaction
and job stress. Scandinavian Journal of Forest Research 22, 248253.
Staats, H., Hartig, T., 2004. Alone or with a friend: a social contect for psychological
restoration and environmental preferences. Journal of Environmental Psychology 24, 199211.
Stuart, H., 2006. Psychosocial risk clustering in high school students. Social Psychiatry and Psychiatric Epidemiology 41 (6), 498507.
Sullivan, W.C., Kuo, F.E., DePooter, S.F., 2004. The fruit of urban nature: vital neighborhood spaces. Environment and Behavior 36 (5), 678700.
Tennessen, C.M., Cimprich, B., 1995. Views to nature: effects on attention. Journal of
Environmental Psychology 15 (1), 7785.
Tranter, P.J., Malone, K., 2004. Geographies of environmental learning: an exploration of childrens use of school grounds. Childrens Geographies 2 (1), 131155.

Tzoulas, K., Korpela, K., Venn, S., Yli-Pelkonen, V., Kazmierczak,

A., Niemela, J., et al.,
2007. Promoting ecosystem and human health in urban areas using Green Infrastructure: a literature review. Landscape and Urban Planning 81 (3), 167178.
U.S. Department of Education, 2005. Americas High School Graduates: Results from
the 2005 NAEP High School Transcript Study. U.S. Department of Education,
Washington, DC.
Uline, C.L., 2000. Decent facilities and learning: Thirman A. Milner Elementary School
and beyond. Teaching College Record 102 (2), 442460.
Ulrich, R.S., 1984. View through a window may inuence recovery from surgery.
Science 224 (4647), 420421.
Ulrich, R.S., 1986. Human responses to vegetation and landscapes. Landscape and
Urban Planning 13, 2944.
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 Psychology 11, 201230.
Van den Berg, A.E., Koole, S.L.K., van der Wulp, N.Y., 2003. Environmental preference
and restoration: (how) are they related? Journal of Environmental Psychology
23, 135146.
Wells, N.M., 2000. At home with nature effects of greenness on childrens cognitive functioning. Environment and Behavior 32 (6), 775795.
Wells, N.M., Evans, G.W., 2003. Nearby nature a buffer of life stress among rural
children. Environment and Behavior 35 (3), 311330.
Welsh, W.N., 2001. Effects of student and school factors on ve measures of school
disorder. Justice Quarterly 18 (4), 911947.
Williams, D.T., 1990. The Dimensions of Education: Recent Research on School
Size. Clemson University, Strom Thurmond Institute of Government and Public
Affairs, Clemson, SC.
Rodney H. Matsuoka is currently a post doctoral research scientist in the Department of Landscape Architecture at the University of Illinois, Urbana-Champaign.
His research explores the human needs and preferences that underlie successful
landscape designs. His current efforts focus on the psychological, social, and health
benets provided by contact with nature.