ArticlePDF Available

Abstract and Figures

At first glance, the concept of density is wonderfully appealing to planners. It is an objective, quantitative, and, by itself, neutral term. However, a second and third glance reveals that it is a very complex concept. Some of the complexity is inherent to the nature of the phenomena associated with density, but part of the complexity stems from the different ways in which density is defined and used in different countries and different disciplines. This review of the literature presents this complexity in an attempt to contribute to a better understanding of the concept and a more careful approach to its use. The review includes both academic and practice literature from the planning, urban studies, and environment-behavior disciplines and selected planning documents from countries around the world.
Content may be subject to copyright.
Journal of Planning Literature
DOI: 10.1177/08854129922092478
1999; 13; 389 Journal of Planning Literature
Arza Churchman Disentangling the Concept of Density
The online version of this article can be found at:
Published by:
can be found at:Journal of Planning Literature Additional services and information for Email Alerts: Subscriptions: Citations
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
Journal of Planning Literature
The Concept of Density
Disentangling the
Concept of Density
Arza Churchman
At first glance, the concept of density is wonderfully appeal-
itis a very complex concept. Some of the complexity is inher-
ent to the nature of the phenomena associated with density,
but part of the complexity stems from the different ways in
which density is defined and used in different countries and
different disciplines. This review of the literature presents
thiscomplexityin an attempt to contribute to a better under-
The review includes both academic and practice literature
fromthe planning, urbanstudies, and environment-behavior
disciplines and selected planning documents from countries
around the world.
At first glance, the concept of density is wonderfully
appealing to planners because it is objective, quantita-
tive, and neutral. What more could one ask for? How-
ever, on second and third glance, it becomes clear that
density is a very complex concept. Some of the com-
plexityis inherent tothenature of thephenomenaasso-
ciated with density, but some of the complexity stems
fromthe different waysinwhich density is definedand
used in different countries and different disciplines.
Thetopic of density is cross disciplinary, as reflected in
the fact that the authors and publications surveyed for
thisarticle come from planning,urbandesign, architec-
ture, environment-behavior studies, transportation,
economics, sociology, psychology, anthropology, and
The purpose of this review of the literature is to pre-
sent this complexity, thereby contributing to a better
understanding of the concept and a more careful ap-
proach to its use. The article includes academic litera-
ture that deals with density and planning documents
gatheredfrom bothreadilyavailable published sources
and from colleagues in various parts of the world in
whichdensity is an issue thatisspecifically addressed.1
These planning documents demonstrate how the con-
cept of density is addressed in practice. Regrettably,
this review focuses almost completely on the literature
available in the English language. This is particularly
unfortunate with regard to the planning documents
used as examples. As a result, these examples are in no
way a representative sample. Nevertheless, something
canbe learnedfrom these documentsby seeingwhatis-
sues are addressed and what approaches are adopted
as long as we are careful to make no generalizations
otherthan thatthere arebothcommonalities anddiffer-
ences within and between countries.
Thereis no consensus on most of the issues raised in
this article; thus, the article presents some of the rele-
vant pros and cons on each issue. Some of the differ-
“locations”, whether those locations are physical, cul-
tural, or professional. Other differences are a function
ARZA CHURCHMAN is an associate professor on the Faculty of
Architecture and Town Planning at Technion-Israel Institute of
Technology, Technion City, Haifa, Israel.
Journal of Planning Literature, Vol. 13, No. 4 (May 1999).
Copyright © 1999 by Sage Publications, Inc.
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
ofthe nature and complexity of density-related phe-
nomena,whichnoone approachor singlestudy canad-
dress or encompass.
The topic of residential density intersects with so
many other issues that to do it justice one needs an
Internet site. Virtually every concept or issue requires
or justifies a detour that involves an in-depth elabora-
tion and expansion into its own literature and into re-
lated literatures. However, this is not possible within
the pages of a journal article, so this article highlights
the relevant issues and the debates surrounding those
issues. It cannot be my role to resolve these issues and
debates; rather, I make their complexity explicit and
visible. These density-related concepts or topics in-
cludeurban form, citysize, preferred buildingorsettle-
ment type, site design, economic issues and policy,
zoning and other land use issues, social issues and val-
ues,women’sissues, children’sdevelopment,cognitive
and perceptual processes, stress, sustainable develop-
ment,compactcities,street andtransportation systems,
conflictsbetweenpublictransportation and the private
car, urban sprawl, environmental quality policies, and
the role of professional planning and government in
setting density standards.
Three concepts are used to address the issue of den-
sityandhowdensity affectspeople’s lives:density, per-
ceived density, and crowding (Alexander 1993).
Density is a term that represents the relationship be-
tween a given physical area and the number of people
whoinhabitor use that area. It is expressed as a ratio of
population size or number of dwelling units (the nu-
merator) to area units (the denominator). Density is an
given density level is positive or negative. Psycholo-
gists distinguish between spatial and social density.
Spatial density is created by a given number of people
withindifferent size spaces. Social density iscreatedby
differentnumbers of peoplewithin the samespace.The
argument is that these two types of density are experi-
enced differently (Baum and Paulus 1987; Russell and
Snodgrass 1987; Altman 1975). This distinction is simi-
lar to Hitchcock’s (1994) analysis of the difference be-
tween increasing density by reducing residential land
areafor the samenumberof people or byincreasingthe
number of people in the same residential land area.
Perceived density and crowding are based on the
principle that the same density can be perceived and
evaluated in very different ways, by different people,
underdifferent circumstances,indifferent culturesand
countries. Thus, even though planners operate on the
level of density, they must be cognizant of the fact that
peopleexperience and liveina multilevel situationthat
manifestsitself in interactions between density and the
perception and evaluation of density.
Perceiveddensity isdefinedas anindividual’spercep-
tion and estimate of the number of people present in a
given area, the space available, and the organization of
thatspace. Cues in the environment thatrepresentpeo-
ple and their activities play critical roles in this percep-
tion of density (Rapoport 1975). Perception is, by
definition,subjective becauseit isdeterminedby thein-
dividual and neutral because it does not include an
evaluative component.
Crowding is defined as the subjective evaluation by
an individual that a given density and perceived den-
sity is negative. Crowding is also defined as a state of
psychological stress that accompanies density that is
evaluated as too high (Evans and Cohen 1987; Sund-
strom1978).Itis a psychological state, the outcome of a
subjectiveandexperientialprocess that includes an ap-
praisal of physical conditions, situational variables,
personal characteristics, and coping assets (Baum and
Paulus 1987; Altman 1975; Stokols 1972). Thus, crowd-
ing represents a subjective, qualitative, and affective
(emotion laden) experience. There is a need for a more
general term than crowding for the subjective evalua-
tionofdensity. Rapoport(1975)pointedoutmanyyears
ago that research addresses the negative subjective as-
pects of density (i.e., crowding) but virtually ignores
thepositive subjective aspects. There has been some re-
centattention paid to this lacuna that hopefully will re-
sult in a term for positive evaluations of density.
One might assume that density, an objective and
case (Hitchcock 1994; Alexander 1993). There is no one
accepted measure of density between or within coun-
triesor even within metropolitan regions. For example,
a report by Lehman and Associates (1995) points out
that even between municipalities in the greater To-
ronto, Canada, area, there is no consistency in the defi-
nition of density. In general, density measures vary in
several ways. First, different numerators and denomi-
nators are used in different countries. Some countries
define density according to the number of people per
givenarea (population density),butother countries de-
fine density according to the number of dwelling units
per given area (residential density).2Second, a variety
of land units, including acre, hectare, square mile, and
square kilometer, are used as the denominator. Third,
although it is common to distinguish between net and
grossdensity, thedefinition ofnetand grossdensityvar-
ies from place to place (see Table 1). For these reasons, it
is very difficult to compare the densities of different
countries (Alexander 1993). Hitchcock (1994) suggests
that every document should include a set of definitions
forthedensityinquestion.To mitigateatleastpartofthis
problem, I convert all denominators into a common
390 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
term, hectare (1 hectare equals 2.47 acres, 1 square kilo-
meter equals 10 hectares, and 1 square mile equals 2.58
square kilometers). Where it is possible, I also specify
whether the reference is to net or gross density.
Thetranslationof populationdensityintoresidential
density requires knowledge about the relevant house-
hold size in a particular context (Hitchcock 1994; Alex-
ander 1993). Given the broad range of household sizes
that characterize different countries, regions, and
populationgroups, this is neither a simplenoraninsig-
nificant task. For example, the average household size
Greenberg,Ltd. 1991b).Inthe Netherlands,theaverage
householdsizeranges from 2.2 to 2.5 persons (Ministry
of Housing 1996). In Israel, the average household size
for the secular Jewish population in 1994 was 3.3 per-
sons,for ultra orthodox Jewish households 5.0persons,
for Moslem households 5.6 persons, and for Christian
households3.9 persons (Churchman et al. 1996).InSin-
gapore,theaverage household size in 1984 was 4.4 per-
sons (Wong and Yeh 1985).
Populationdensitygivesabetteridea ofthenumbers
of people likely to make use of the area in question.
Residential density provides an indication of the
number of dwelling units in that area. However, resi-
dential density and population density both represent
averages; therefore, any discussion of density must be-
ware the pitfalls of averages, especially when the area
of concern is large (Hitchcock 1994). As an example,
Loo and Ong (1984) reported that the average popula-
tion density in San Francisco was 183 persons per hec-
tare. Yet, the variance between neighborhoods was
tremendous, with part of the range in density from at
least86to 1,838 persons per hectare. The same problem
The Concept of Density 391
TABLE 1. Density Measures in Residential Areas
Parcel density (net-net density, Parcel density is measured in areas designated for residences. The two main ways to
net site density, net density, express this density are dwelling units per area and floor area per area. In some cases
lot density) (Toronto, Israel, some regions of the United States), the measure consists only of the
number of dwelling units built on parcels allocated for residence—it excludes roads,
parks, and other public lands (Alterman and Churchman 1998; Berridge Lewinberg
Greenberg, Ltd. 1991b; Wentling 1991). In the Netherlands, net density includes
neighborhood-related spaces such as the land of the houses, schools, local streets, and
local parks (van Andel 1998). The measure of floor area per area density is expressed in
the ratio between the floor area and the lot area, both expressed in square meters. This
measure is especially useful when the same parcel consists of land for residential and
nonresidential purposes or in areas of high density and large buildings. Since the parcel
density denominator is precisely defined, in contrast to other density measures, it is the
most unambiguous measure.
Street density (net density) This measure includes the area of the public street rights-of-way that provide access to
the residential parcels. The prevalent numerator is the number of dwelling units,
whereas the denominator is typically the parcel area plus half of the public rights-of
way adjacent to the residential parcels.
Gross residential area density This term expresses the living space of the population in the residential area, including
(gross site density, residential both private and public space. This measure is useful because many residential areas
density, residential area density, include a limited variety of nonresidential uses meant to serve the local residents, such
gross density, gross living area as parks, schools, community centers and so forth. It takes into account the space
density, neighborhood density) needed by a given residential population, when all the residentially related uses are
taken into account, in addition to public streets and the residential parcels. Gross
residential density is the most ambiguous measure, because some neighborhoods may
include land for purposes that serve a wider population than that of the specific area,
for example, zoos, theaters, and so forth. Wentling (1991), for example, defines gross
density in parts of the United States as the number of units per acre of initial
undeveloped site.
Density measures beyond Since the denominator in this measure includes the entire municipal area, it will reflect
residential areas ( population the lowest density mentioned so far, because the municipal region includes land that
density, community density) has other than residential uses, as well as undeveloped land. Hitchcock (1994) points
out that as the amount of undeveloped land differs from city to city, it is difficult to
compare the density of different cities whose density measure is based on dwelling
units. On the other hand, Berridge Lewinberg Greenberg, Ltd. (1991a) claims that two
measures—gross population density and gross urban density—do not include, in most
cases, undeveloped areas and that these measures are therefore useful for a comparative
study of cities.
SOURCE: Hitchcock (1994).
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
exists with average residential density, even at the
neighborhood scale, because there may be a very un-
equal distribution of units over geographic space
(EvansandCohen 1987). This is especially true in areas
with different building types. In a recent research proj-
ect, Alterman and Churchman (1998) partially resolve
this problem by proposing that separate residential
densities be calculated for each building type (e.g., sin-
glefamily, multifamily, orhighrise) so thattheaverage
would not mask the variability in densities between
building types. Another potential problem with resi-
dential density is that local public open space may be
scattered or concentrated. Although this does not
change the density figure, it may have major implica-
tions,particularly for children’s outdoor play opportu-
nities (van Andel 1998; Calthorpe 1993).
Complicating the matter even further is the fact that
density is often confounded with other variables such
aspopulation size and attendantlevelsof resources, es-
pecially in developing countries (Jain 1987). Other con-
founding variables include building type (Alexander
1993;BaumandPaulus 1987; van Vliet 1985), suburban
orinner-city locations (Shlay 1985; Michelson1977),so-
cioeconomic status (Shannon and Cromley 1985; Shlay
1985),and otherstressors suchasnoise,heat,and pollu-
tion.Asaresult,manyofthe statementsmade aboutthe
problems associated with high density may actually be
a function of these other variables or of the interaction
between these variables and density.
housing is particularly instructive. Although high-rise
buildings are intuitively associated with high residen-
tial density, there is no necessary relationship between
the two (Michelson 1977). For example, a study con-
ducted by the Tel Aviv Town Planning Department
showed that high-rise buildings do not necessarily
equal high densities. Within the city, there are areas of
low-risebuildings of up to four stories varying from 80
to 240 net dwelling units per hectare; areas of eight- to
nine-story buildings with net densities of 100 and 290
dwelling units per hectare; and a net density of 250
dwelling units per hectare in an area of sixteen-story
buildings. A report prepared by Lehman and Associ-
ates(1995) gives examples of areas in the Toronto, Can-
ada, metropolitan area in which net densities of
between 120 to 230 dwelling units per hectare were
achieved with low-rise buildings of up to five stories.
The research dealing with the subject of residential
densitycanbedivided intohistorical periodsaccording
to the focus of the research. Studies from the 1960s and
1970s were interested in the social and psychological
ramifications of high density of mostly undefined lev-
els. In the 1980s, density studies focused on how den-
sity affected the physical layout of a building, a
neighborhood, or a settlement. Toward the end of the
1980s, ecology emerged as the new focus of interest in
densityresearch, marking atendencythat became even
stronger in the 1990s.
Most studies deal with high density and its effects,
but different disciplines ask different questions, con-
sider the subject from different angles, use different
concepts,andapplydifferent methods. The attitude to-
wardhigh density dependsto some extentonthe scien-
tific discipline of the researcher conducting the study.
Psychologistsand sociologists concentrate on thedetri-
mental effects of density. Economists, transportation
experts, and environmentalists assert both the advan-
tagesand disadvantages ofhigh density. Multidisciplin-
ary researchers in the field of environment-behavior
studies try to examine density in all its complexity and
to uncover the negative and positive aspects of density
in different contexts and under different conditions.
Unfortunately, researchers often refer to density in
relativeterms,suchashighor mediumdensity,without
specifying numbers. The researchers in question are
from various countries; thus, high, medium, or low
densities may have very different numerical values.
Thisvariation makes it impossible to relateconclusions
to the objective facts of the density in question (see Ta-
bles2and3).Forexample, inthe Netherlands,10 dwell-
100 units is considered high density (van Andel 1998).
In Israel, 20 to 40 dwelling units per net hectare is con-
sideredlowdensity,and 290units perhectare isconsid-
ered high density (Alterman and Churchman 1998).
Even given the differences in the definition of net den-
sity between the two countries, this is still a significant
difference. It is no wonder that a comprehensive out-
look has not crystallized on the subject and the signifi-
canceof density.Itis impossibletocome upwitha clear
and generally accepted picture.
Much of the concern with density in planning and
other related fields has been over high urban density
and its assumed negative effect on the quality of life of
urban residents. The city has historically been
perceived to be a place of overcrowding, noise, dirt,
crime, poverty, disease, and so forth (Radberg 1998;
Lehman and Associates 1995; Gowling and Penny
1988). The high density existing in cities during the
early period of the Industrial Revolution was seen as
sult,planningcontrols(inCanada andGreat Britain,for
example) usually specified maximum densities. The
planning reaction was a strong movement toward
States and Canada, this took the form of a move to the
suburbs,butinGreatBritainand Sweden, it resulted in
garden cities (Madanipour 1996; Gowling and Penny
392 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
1988). The garden city movement is described by Rad-
berg (1998) as representing decentralized urban
growth. The assumption was that these relatively low-
density residential areas would not suffer from the ills
found in high-density cities and would offer a higher
quality of life to residents. More recently, there have
been many second thoughts on, and strong criticisms
of, these trends. Environmentalists express concern
about the environmental implications of low density
(Van der Ryn 1986), and urbanists are concerned about
thedeclineofthe city (Lehman and Associates 1995; Ja-
cobs 1961) or of the community (Scully 1994; Smyth
1992). Questions about low densities also have been
posed by those who are concerned about the efficient
use of land and public services (Lehman and Associ-
ates 1995); by feminists and researchers who argue
that low-density suburbs are inimical to women’s
lives, especially employed women with children and
single parents (Churchman 1993); and by sociologists
who criticize the social homogeneity and the social
segregation in these low-density areas (Smyth 1992;
Shannonand Cromley 1985).Thereare some, ofcourse,
who mention all of these problems (e.g., Calthorpe
Although the arguments against low density are
widelyaccepted andhavebeen veryinfluentialin plan-
ning in the past two decades, more and more dissident
voices can be heard. There are those who question the
basic assumptions of many of the arguments against
low density and those who contend that the analysis of
theproblem is wrong orthat the picture isnotthat clear
as to how to achieve desired quality of life or equity
goals. Others argue that some of these goals are wrong
or misguided (see, for example, Breheny 1996; Troy
1995b, 1992; Gordon and Richardson 1997).
The current planning discourse on these issues is fo-
cusedonquestions ofthe advisabilityof increasingresi-
dential densities or the necessity for it. Terms used in
the discourse clearly reflect a point of view. For exam-
ple, some speak of urban sprawl (Ewing 1997) versus
urbanconsolidation(Troy 1996;Orchard1995)orurban
intensification (Jenks et al. 1996). Others discuss reur-
banization(Berridge Lewinberg Greenberg, Ltd.1991a,
1991b), urban compaction (Breheny 1996; Troy 1996),
compactcities (Jenkset al.1996b;Van derRyn andCalt-
horpe 1986b), and sustainable cities (Haughton and
Hunter 1994; Walter et al. 1992). And still others talk of
town cramming (City of Newcastle upon Tyne 1993;
Barton 1992; Breheny 1992a).
There are a number of planning and design ap-
proaches coexisting at the moment, each of which has
its own proponents and opponents. When density is a
central issue, it is perceived as a means to some end.
not even in terms of density. However, I integrate the
relevantprinciples and claims of these approaches into
a discussion of the advantages and disadvantages of
higher densities.
Two of the approaches, new urbanism and transit-
orienteddevelopment,are concerned with many of the
issues raised by the other approaches. However, the
motivation of the proponents of new urbanism and
transit-oriented development is not that of increasing
densities—anyincreasein densitythatis achievedisba-
sicallyaby-product of a minimal nature. The emphasis
ofthe newurbanism movementison smalltowns. New
urbanists envision towns or neighborhoods that are
compact, mixed use, and pedestrian friendly (Madani-
pour1996). Newurbanismhas comeinfor criticism,for
example, for its emphasis on shape and form rather
thansubstance (Peel1995) andforinvokingthename of
Jane Jacobs when new urbanist ideas are actually very
different (Montgomery 1998). In terms of the issue of
density, the aesthetic, spatial, and programmatic prin-
ciples of this approach are considered to be applicable
in urban and suburban conditions at any density. Den-
The Concept of Density 393
TABLE 2. Single-Family Net Dwelling Unit Density
Country Hectare Source
Australia 10 Haughton and Hunter 1994
France 27 Alterman 1997
Israel 20 to 40 Alterman and Churchman 1998
Netherlands 10 van Andel 1998
Sweden 20 Radberg 1998
United States 12aBannister 1992
a.Thisis theaverage rate.Thelow rangesare 2.5to10, andthe
high ranges are 17 to 29 dwelling units per hectare.
TABLE 3. Multifamily Net Dwelling Unit Density
Country Hectare Source
Toronto, Canada 53 to 318 Lehman and Associates
Vancouver, Canada 47 to 89 Vancouver City
Planning Department
Great Britain 28 to 48 Goodchild 1994
Israel 290 Alterman and
Churchman 1998
Netherlands 100 van Andel 1998
Singapore 200 Wong and Yeh 1985
Sweden 250 to 300 Radberg 1998
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
sity is mentioned as a way of making walking, cycling,
andpublic transit use viable alternatives to automobile
use. Yet, increased density is not listed as one of the
goals of new urbanism, nor do proponents of new ur-
banism mention the densities achieved when they de-
scribe their projects (Katz 1994).
The emphasis of transit-oriented development,
whose principal proponent is Peter Calthorpe (1993,
1992), is to plan balanced, mixed-use areas with a sim-
ple cluster of housing, retail space, and offices within a
one-quarter mile walking radius of a light rail system.
The motivation for transit-oriented development is to
improve the ills brought about by dependence on the
automobile and the mismatch that exists between old
suburban patterns and the postindustrial culture. The
goal is to preserve open space and reduce automobile
trafficwithout necessarilyincreasing density.Calthorpe
(1993)defines averagenet residentialdensities ofurban
transit-oriented developments as 44 dwelling units per
hectare, with densities of 62 to 123 units per hectare for
up to three-story apartment buildings.
The vision of the compact city, embraced by the
European community, has had a significant effect in
manycountries.Its basic principles include the intensi-
fication of the use of space within the city through
higher residential densities and centralization, mixed
land uses, and limits on development beyond the pe-
riphery of the city. Proponents of the compact city ap-
proach assume that concentrated development will
reducethe need to travel by car, thereby reducing vehi-
cle emissions and leading to energy savings (Thomas
and Cousins 1996). The approach combines concerns
about how urban growth affects the environment and
concerns about the future quality of life in urban areas
(Breheny 1992a).
Thereis no consensus as to whetherthenotion of the
compactcity isdesirable, achievable, oreven whetherit
is a sustainable urban form. (See Breheny 1992b and
Jenkset al. 1996a for explications of many of the princi-
ples of the compact city and the questions raised re-
garding them.) Troy (1992, 1995b, 1996), for example,
arguesthat the compact city approach adopted as a na-
tional policy in Australia in the 1980s (Orchard 1995) is
notthe best answer for the issues facing the country to-
day.Troy’sarguments are that urban consolidation is a
smoke screen for other problems and that compact cit-
ies will not achieve promised land and infrastructure
savings. Gordon and Richardson (1997) claim that the
case for the compact city has not been proven in envi-
ronmental, economic, or transportation terms, and
compact cities should not be adopted as a goal because
they contradict the overwhelming preference for low-
density development. Ewing (1997) responds by inter-
pretingcompactdevelopmentas someconcentration of
housing and some mixing of land use but not high-
density development. To complicate matters even fur-
ther,Preiser (1989) argues that thesprawlpattern of ur-
ban growth, which is characterized by discontinuous
development, can lead to higher densities in areas that
are initially skipped over (depending on other factors);
therefore, sprawl may not be such a detrimental devel-
opment pattern. Goodchild (1994), from a different
point of view, questions the ability of the compact city
to achieve its promised benefits. For example, he says
that one cannot take for granted that the people who
live in compact cities will travel more by public trans-
portation or on foot because much depends on resi-
dents’ attitudes and behaviors.
The sustainable city approach is confronted by the
problem of the definition of the term sustainability,
which is interpreted both broadly and narrowly. Sus-
tainabledevelopmentinabroadsense includesecologi-
cal sustainability (the continued productivity of
ecosystems), economic sustainability (economic
growth), and social sustainability (the maintenance of
socialvalues, traditions,and institutions)(Ewing 1997).
Breheny (1992a) reduces sustainability to the adoption
of policies that minimize both local resource consump-
tion and pollutants. Often, the terms compact city and
sustainable city are used interchangeably because their
basic principles in terms of urban issues are similar.
Oneof these principlesinvolvesincreasing densities on
the assumption that increased densities will have the
desired environmental results. However, Goodchild
(1994) argues that “very high” densities have undesir-
able consequences in terms of sustainable develop-
ment. He refers to the argument that sustainable
development requires a general reduction in net resi-
dential densities and quotes Rydin (1992), who argues
thatthe optimum for sustainable development is thirty
to thirty-five dwelling units per hectare.
Hitchcock(1994) and Orchard(1995)direct attention
to the fact that, on the whole, the discussion about in-
creasingdensityandreducingurban landconsumption
concentrates almost totally on residential densities. It
neglects all of the other land uses that make up a city,
eventhough theseland usesrepresent asignificantpro-
portionof acity’stotal landarea.If thesenonresidential
land uses are not taken into account, the reduction in
land consumption achieved by increasing residential
density will not be as great as initially conceived be-
causeservicesandamenities willhave tobe augmented
to accommodate the increased population (see also
Goodchild1994). Hitchcockalso cautionsthatincreases
in population in individual areas of the city affect land
consumption in the urban region as a whole. Thus, he
argues that achieving the goals associated with higher
densitiesmust take into accountnumerous interactions
withthelargergeographic areaand mustexamine spot,
rather than average, densities. Handy (1996) in her dis-
394 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
cussion of travel behavior and its relationship to den-
sity makes a similar point about not using averages.
This section provides a review of the ways in which
density issues are incorporated into plans in different
countriesand atdifferentscales (national,regional,and
metropolitan). Approaches to residential density vary
withinand betweencountries. Thisisan obviousneces-
sity because of differing historical, political, economic,
geographic, physical, social, cultural, demographic,
technological, and ecological contexts. However, plan-
ners and decision makers all over the world are also
cognizantof the zeitgeist at any givenpointintime and
ofwhat is being done in other countries. Thus, many of
the same ideas and approaches are present in some
form or another in most of these plans. Contextual dif-
ferences lead to somewhat different policy goals and
measures taken to achieve those goals.
Contextual factors also play a major role in motiva-
tionsfor focusing on density. For example, in countries
such as Israel, where land is scarce or perceived to be
scarce,the primary goal is to makemoreefficientuse of
land to preserve agricultural land or natural open
spaces (Alterman and Churchman 1998). Other coun-
tries that do not have a scarcity of land may be more
concernedaboutenvironmental factorsandsustainable
development (e.g., Norway) or about economic devel-
opment (e.g., Australia). Areas with little population
growth, such as the City of Newcastle upon Tyne
(1993), adopt one approach. Areas that anticipate a
large population growth, such as the metropolitan To-
rontoarea(BerridgeLewinbergGreenberg,Ltd. 1991a),
adopt another.
Areviewof the various goals that national, regional,
andlocalplanning authoritiesare tryingtoachieve,and
the implications those goals have for residential den-
sity, is thus instructive. A few planning documents are
presented here as somewhat detailed case studies to il-
lustrate some of the various combinations of goals that
exist. These are followed by a summary of different
categories of goals that can be identified in these and
other planning documents.
Randstad, the Netherlands
This case study deals with national strategic plan-
ning,a type of planning in which the Dutch are consid-
ered to excel. The following description is based on
researchby Manshaden and deSchmidt(1992),van der
and van der Cammen et al. (1988). The Netherlands, a
smallcountry,isthe mostdensely populatedcountry in
Europe, with a population density of 439 people per
squarekilometer.The country’sland policyis alsoquite
different from most other countries, and it has mani-
festedininteresting developmentsover timeintermsof
density. The urban agglomeration of Randstad is the
most highly urbanized area in the Netherlands. Forty-
five percent of the population lives in one-quarter of
in an average density of more than one thousand per-
sons per square kilometer (one hundred persons per
square hectare).
Dutch national planning since the 1950s has focused
on the “Randstad” and “Green Heart” concepts. The
initialapproach to these areas represented a general re-
jection of unbridled growth, particularly suburban
growth,andanemphasison newtowns. Thechief inno-
vationofthe Second Report is the notion of concentrated
deconcentration. The Randstad was designed to be a
horseshoe-shapedpatternofurban settlementsarrayed
aroundthe GreenHeart. The Randstadwas subdivided
into conurbations, city regions, and agglomerations,
which were all separated by green corridors. In the
1970s, the Third National Physical Planning Report took
theprotection of the Green Heart region as its most im-
portant mission.
In the mid-1980s, the Dutch national government
embracedapolicy thatfavoredacompactcity approach
witha focus on thedevelopmentand redevelopment of
existing cities at greater densities. In the Fourth Report
Extra, emphasis on obstructing continuous urban
sprawl, intensifying land uses in urban areas, and pre-
serving nonurbanized space was even greater. All new
developmentwasto occurwithin theRandstad andto be
rating environmental problems was also emphasized in
the report. The compact city (representing high density
terion on which the plans of local authorities would be
assessed by provincial and state planning agencies.
Singapore represents a very unique case in many
ways, including its large population and increasing
need for land for both residential and nonresidential
uses.Becauseof these problems, the Singapore govern-
ment committed to the provision of public housing for
the majority of the country’s population (85 percent in
teresting. Since Singapore obtained its independence,
decisionmakers,town planners, andarchitectshave ex-
pressed concern over the country’s limited land re-
sources. Net residential density (defined as dwelling
units in a site or area including access roads and car
parks) in public housing projects constructed during
the 1960s (mainly composed of one- to three-room
apartments) was between 200 and 500 dwelling units
per hectare. In developments from the 1970s, which
The Concept of Density 395
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
are primarily composed of three- or four-room apart-
ments, net residential density was reduced to between
170 and 250 dwelling units per hectare. In recent years,
thereis agrowing tendencyinSingapore tobuild larger
apartments, with the net density goal set at 200 dwell-
ingunits per hectare. This density is seen to becounter-
balanced by the decrease in average household size,
from 6.2 persons in 1968 to 4.4 persons in 1984. Net
population density, which decreased from 1,000 per-
sons per hectare in 1981 to 880 persons per hectare in
in 1990. Residential density is the foremost criterion of
site planning for housing development in Singapore.
Housing development primarily takes the form of
high-rise (twenty-five-story point block and ten- to
twelve-story slab block buildings) and high-density
buildings (Wong and Yeh 1985). The main goal is to
providea favorable physical environment within these
Israelisa small country with an effective population
density (effective population density deducts the 50
percent of the country’s area that is mountainous des-
ert)of more than five hundred persons per square kilo-
meter (fifty persons per hectare). Since Israel has the
highest birthrate among industrialized countries and
acceptsmass immigration, the nation will become even
more densely inhabited in the future. Israel’s use of ur-
banland is relatively intense, and most urban residents
live in apartment houses. Residential net densities
rangefrom 20 dwelling units per hectare to 290 units or
even higher (Alterman and Churchman 1998).
There is a strong consensus among planners and
policymakers in Israel that the rapid depletion of land
resources necessitates even higher densities. A review
of recent metropolitan, regional, and national master
plans reveals that this consensus has been expressed
simply as a call for increased densities, without any
specification of what kinds of increases may be re-
quiredor appropriate.Forexample, the goalof the 1992
National Master Plan (Israel Ministry of the Interior
1992) was to cope with a mass immigration from the
former Soviet Union and from Ethiopia. The antici-
pated need to absorb 1.6 million new immigrants
within five years (in a country with about 4.5 million
residents) set the principles that no new small settle-
mentsshould beapproved,that the populationof exist-
ing settlements should be expanded, and that the
density of existing cities should be increased within
their present boundaries. However, the plan does not
definethe densitiesrequired to implementthese princi-
ples. A regional master plan, the Master Plan and Devel-
opment Plan for the Northern Region (Shefer et al. 1997),
alsodoesnotquantitativelydefinedensities. However,
at present there is a proposal for an amendment to the
plan that will specify minimum average gross popula-
the region. For example, a city of more than fifty thou-
sandpersonsina central area of the region would have
toattain aminimum averagedensityof seventhousand
persons per square kilometer (seven hundred per hec-
tare). A similar city in a more peripheral area of the re-
gion would have to attain a minimum average density
of six thousand persons per square kilometer or six
hundred persons per hectare (Shefer et al. 1997).
Arecentpolicy wasproposedtoincreasetheefficient
useof urban land for residential development.Thepol-
icy included safeguards that were designed to mini-
mize possible negative effects on residents’ quality of
life or limits on freedom of choice between different
housing types (Alterman and Churchman 1998). The
policyproposes thateachvillage or townbe required to
gradually decrease its proportion of single-family,
low-density dwellings to intensify residential land use
before officials can request an expansion of town bor-
ders at the expense of agricultural land or open space.
The intensification requirement also ensures that the
higherdensitiesprovide forthepublicservice,amenity,
privacy, and open-space needs of residents. The pro-
posal defines low net densities as a minimum of forty-
five dwelling units per hectare and medium net densi-
ties as one hundred to two hundred dwelling units per
hectare. It regards net densities higher than these as
negative and not recommended.
Newcastle upon Tyne, Great Britain
The development plan for the City of Newcastle
uponTyne (1993)addresses theissue ofexistingand fu-
ofNewcastle uponTyne suffered forseveral yearsfrom
population declines that resulted from young adults
and young families with children moving out of the
city. The city was left with a high proportion of depen-
dent groups, such as elderly people and single-parent
families, which require many support services but are
the number of small households brought about an in-
creaseddemand for dwellings, despitetheoverall drop
inpopulation.Most oftheexistingstockof high-density
housesalreadyhadbeenrehabilitatedand renovatedin
the1970s. As land became scarce, there was a tendency
toward town cramming that was manifested in higher
densitiesand pressure for thedevelopmentofland that
was allocated for open space or economic activities.
The city’s plan allocated areas for development in
terms of the environmental capacity of the built-up
area.Development thatoverstepped theboundariesset
by the plan was seen as unacceptably damaging to the
environment. The residential goal set by the plan was
396 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
“tostabilize the city’s population atthelevelprevailing
at the beginning of the Plan period, by providing suffi-
cient land for a wide range of types and qualities of
housing development” (City of Newcastle upon Tyne
1993, 9). Increased density was not one of the stated
goals of the plan, but it was included as one of the
means to achieve the goal of providing varied and af-
fordablehousingthat wouldmeet theneedsofdifferent
groupswithinthe city’spopulation. Theplan presentsa
list of sites within the city in which the minimum den-
sity was set at a nominal thirty dwelling units per hec-
tare.3The actual density was to be determined in the
masterplans of the specific sites that were allocated for
new housing development. The goal for some existing
local authority housing sites was to reduce density
through redevelopment, yet the plan does not mention
existing and planned density figures.
Othergoals set by the Newcastle planwere“towork
towardsthe longterm goalofsustainability byprogres-
sively reducing both the need for energy and other
natural resources, and the environmental impact
caused by their use;” to improve public transportation;
and to increase pedestrian and bicycling opportunities
(CityofNewcastleupon Tyne1993, 13).However, none
of the plan’s goals is specifically linked to the issue of
residential density, perhaps because officials believe
that existing density levels are high enough to support
these goals.
Toronto, Ontario, Canada
The 1991 proposed Reurbanization Plan for Metro-
politan Toronto, Canada, consists of two parts: a study
of the reurbanization process (Berridge Lewinberg
Greenberg, Ltd. 1991b) and planning guidelines (Ber-
ridgeLewinbergGreenberg, Ltd.1991a).Thegoalofthe
reurbanization plan was to accommodate three hun-
dred thousand new residents and three hundred thou-
sand new jobs in twenty to thirty years. Toronto has a
number of advantages that facilitate this growth.
Among these advantages are a vast supply of under-
used sites that are distributed throughout the metro-
politan area and a sophisticated public transit network
of buses, streetcars, and subways that reaches into
everycornerof themetropolitanarea.Oneofthecentral
goals set forth in the reurbanization plan was the ac-
commodationof as muchgrowthas possible within the
city’s already built-up urban areas. Urban density was
specifically linked to several planning goals: (1) the re-
duction of automobile dependency through the crea-
and transit use; (2) the preservation of open space
within the city by not allowing increases in density to
occur at the expense of land for the “public realm,”
which consists of parks, streets, sidewalks, and other
openspaces; (3) the promotion ofdiversityinbuildings
and living and working environments; and (4) the pro-
visionof a range of housing types at afullrangeof den-
sities within the City of Toronto to accommodate the
city’s increasingly diverse population.
The Berridge, Lewinberg Greenberg, Ltd. (1991b)
studyof theviability ofthesegoals formetropolitan To-
rontoreached the followingconclusions. First, theneed
topreservefarmland andopenspaceattheperipheryof
the city implied that development densities within the
metropolitan area should always be maximized,
withinspecifiedparameters. Thereportnoted thateven
ifdensities were to double, there was no need to build
high-rise buildings because maximum densities can
be attained with single-family houses and low-rise
apartment buildings. Second, to accommodate the ex-
pectedincrease of three hundred thousand residents at
gross residential densities of one hundred persons per
would require development. Yet, the report notes that
such a density already exists in at least one of the city’s
neighborhoods.Third, high urban density is associated
witha higher degree ofwalkingand bicycling, butden-
sity is not the only factor influencing the decision to
walkorcycle.The reportnotes thatfactors relatedtour-
banstructure,especially the mix of land uses in a given
area, affect these behavior patterns.
Finally,asin othercities,theuseofpublictransporta-
tion in Toronto increases with an increase in density.
Toronto’s overall density exceeds the minimum gross
urban density benchmark (30 persons per hectare)
above which higher densities will increase the use of
public transportation. However, densities vary within
Toronto,withhigher densitiesconcentratedintheinner
city. Some areas on the outskirts of the city, especially
single-familyareas in suburban neighborhoods, do not
reachtherequired densitybenchmark. Increaseddensi-
ties in these areas would support greater use of transit.
Thegrossdensities ofotherresidentialareasrange from
150to250persons per hectare, which is within the high
transit use range. The fact that these areas consist of
four-story low-rise apartment buildings and high-rise
buildings indicates that density ranges that support
transit use can be achieved with low-rise buildings.
A basic premise of Toronto’s redevelopment plan is
that when the nature of the changed urban form is de-
termined, residential uses cannot be treated separately
fromemploymentuses. Thus,a newmeasure ofdensity
was proposed: gross reurbanization density. Gross
reurbanizationdensity is defined asthenumber of resi-
dents and employed persons per hectare, regardless of
the relative predominance of members of each group
(BerridgeLewinbergGreenberg, Ltd.1991a, 1991b).For
example, a gross reurbanization density of one hun-
dredresidentsandemployedpersonsperhectare could
include seventy-five employed persons and twenty-
The Concept of Density 397
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
five residents or any other combination of these two
groups. This density measure supports the implemen-
tation of a balanced mix of land uses in a given area.
The draft guidelines for the reurbanization of metro-
politan Toronto adopt a hierarchical, multicentered ur-
ban structure, which distinguishes between three levels
of centers. The gross density ranges that are recom-
mendedfor each center are mainly basedonlevelsof ex-
isting or future transit availability. Low-density centers
are to have between 125 and 175 residents and workers
per hectare, medium-density centers between 250 and
centers between 400 and 500 residents and workers per
hectare.Regardless of density, each center istobedevel-
oped in a compact manner with densities maximized
within these compact parameters. These guidelines re-
flectthenotion thatareas withmixedusesareabletosus-
tain higher densities than single-use residential areas
(Berridge Lewinberg Greenberg, Ltd. 1991a).
Types of Goals Related to Density
A summary of the types of goals that are linked to
density in these and other planning documents indi-
cates that they can be divided into goals that relate to
environmental quality, transportation systems, physi-
cal infrastructure and urban form, social factors, and
economic factors. Higher density goals that fall under
each of these categories are listed below.
Higher density goals that are related to environ-
mental quality include the following:
1. reducing the need for energy and other natural re-
sources and associated environmental effects (Re-
gionalMunicipalityof York1994;City ofNewcastle
upon Tyne 1993),
2. improvingair quality throughincreased transit use
and reduced car trips (Regional Municipality of
York 1994),
3. saving energy by planning high-density mixed
land uses (Stenhouse 1992),
4. protecting farmland and natural resources (Alter-
manand Churchman 1998;Faludi andvan derValk
1994; Regional Municipality of York 1994; Berridge
Lewinberg Greenberg, Ltd. 1991a), and
5. preservinggreen openspaces and air,water, fauna,
and flora systems within the plan’s boundaries
(Martin County 1994; Regional Municipality of
York 1994; New York City Planning Commission
1993; Berridge Lewinberg Greenberg, Ltd. 1991a).
Higher density goals that are related to transporta-
tion systems include the following:
1. reducing the frequency of use of private vehicles
and shortening routes to various land uses (Wood-
hull 1992),
2. encouragingthe useof publictransportation by im-
provingthe qualityof publictransit systemsand by
providing easy access to mass transportation sys-
tems through high-density development (New
York City Planning Commission 1993; Berridge
Lewinberg Greenberg, Ltd. 1991a), and
3. increasing the incidence of walking and cycling
(Berridge Lewinberg Greenberg, Ltd. 1991a).
Higher density goals related to physical infrastruc-
ture and urban form include the following:
1. meeting the need for more dwelling units that re-
sults from an increase in the number of households
(City of Newcastle upon Tyne 1993; Israel Ministry
oftheInterior1992; Berridge LewinbergGreenberg,
Ltd. 1991a),
2. obviating the problem of the gradual depletion of
land reserves in specific urban areas (City of New-
castle upon Tyne 1993),
3. intensifying the use of urban areas (Martin County
1994; Manshaden and de Schmidt 1992),
4. creating a hierarchical multicentered urban struc-
turethatenables gradations ofdensityand avariety
of residential choices (Berridge Lewinberg Green-
berg, Ltd. 1991a), and
5. providing a favorable physical environment in
terms of maximum building heights with at least
minimal spacing between buildings (Wong and
Yeh 1985).
Higherdensity goals that are related to social factors
include the following:
1. providing a wide range of housing types and den-
sity levels to provide choice and meet the needs of
an increasingly diverse population (Regional Mu-
nicipality of York 1994; City of Newcastle upon
Tyne 1993; New York City Planning Commission
1993; Berridge Lewinberg Greenberg, Ltd. 1991a);
2. ensuringa sufficient supplyof apartmentsin future
years (Regional Municipality of York 1994; Wong
and Yeh 1985);
3. creating a livable urban environment (Berridge
Lewinberg Greenberg, Ltd. 1991a), possibly as Ja-
cobs and Appleyard (1987) define livability as a
placein whicheveryone can livein relativecomfort
in a well-managed environment that is relatively
devoid of nuisance, overcrowding, noise, danger,
air pollution, dirt, trash, and other unwelcome in-
4. meeting the needs of particular groups in society,
including single-parent families, the elderly (Ber-
ridge Lewinberg Greenberg, 1991b), and low- and
moderate-income households (Martin County
5. redeveloping at densities that are high enough to
recapture a neighborhood’s former vitality (New
York City Planning Commission 1993); and
6. bringing buildings closer to the street to provide
“eyes on the street” for safety purposes (New York
City Planning Commission 1993).
398 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
Higher density goals that are related to economic
factors include the following:
1. promotingthecritical massnecessary tosupportlo-
calretailandservice areas (NewYorkCityPlanning
Commission 1993);
2. attractingbusinesses,hotels,shopping, andupscale
residentialdevelopment to urban areas (Faludiand
van der Valk 1994);
3. improving a city’s economic efficiency (Frost and
Dingle 1995; Troy 1995b);
4. enabling the construction of low cost, middle-
density housing (two- and three-story row houses
withunenclosed parking andno elevator) in neigh-
borhoods in which this is appropriate (New York
City Planning Commission 1993); and
5. enabling the use and extension of necessary urban
services in an efficient and economical manner
(New York City Planning Commission 1993).
Planningprofessionalsdifferin their opinions about
the importance of these goals and whether increased
density is the right way to achieve them. Furthermore,
many researchers contend that high densities have nu-
merous negative effects. These issues are addressed in
the next section.
Advantages and Disadvantages of (Relatively)
High Densities
Theadvantagesanddisadvantages of high densities
that are presented here are discussed in the literature
includedin thisreview.To reiterate,thelevel of density
thatisconsidered high varies between and even within
countries, cultures, socioeconomic classes, contexts,
and at different stages of development. Every aspect of
high density has both advantages and disadvantages,
but whether an advantage or disadvantage applies in a
givensituation depends oncontextin its mostinclusive
sense. Furthermore, all of these advantages and disad-
vantages are on some level theoretical—they repre-
sentpossibilitiesorpotentials,notcertainties or inevi-
tabilities. Whether they actually exist in a particular
situation depends on the characteristics of that place
and time. For example, it is quite commonly accepted
portunityfor a high-quality public transit system. Yet,
theopportunity does not guaranteethatsuch a system
indeed exists and thus that residents benefit from this
One of the problematic aspects of attempting to link
density and positive or negative consequences is the
tendency to make assumptions about how various lev-
els of density affect people’s lives. There is no system-
atic evidence as to whether higher densities affect
everyoneor most people the same, who is affected, and
to what extent they are affected. Fischer (1976) argued
that we do not even know how much of an “average”
urban resident’s life actually is spent in high-density
situations. For example, the assumptions are often
made that a person living in a high-density area en-
counters a large number of people in various parts of
that environment, that many of these encounters are
withstrangers, and thatthesemultiple encounters with
strangers have negative behavioral, cognitive, percep-
tual, and emotional consequences. These assumptions
should be taken as hypotheses that require testing
rather than as facts (Churchman and Ginsberg 1984b;
McCarthy and Saegert 1978).
The pros and cons of low density are not presented
here because they are the implicit opposite of the pros
and cons of high density (see Haughton and Hunter
1994 for a discussion of the pros and cons of high and
low density). The order of presentation of the advan-
tages and disadvantages of relatively higher densities
parallelsthe order ofpresentationof the planninggoals
in the previous section. There is no value statement re-
flected in the order. However, it should be noted that
this list also includes personal advantages and disad-
vantagesthat couldnotbe separated fromsocial factors
in the list of density-related goals that are specified in
1. High density can help protect agricultural land
from urbanization (Alterman 1997; Burton and
Matson 1996).
2. High density results in less depletion of the natural
resources needed for construction purposes (Bre-
heny 1992a).
3. Built forms that facilitate higher net densities may
result in significant reductions in energy demands
(Owens 1992; Stenhouse 1992). Energy use within
buildings can be reduced by passive solar architec-
ture, superior insulation, and energy-saving tech-
nology (Stenhouse 1992) or by built forms with
low-surface areas and combined heat and power
systems(Rydin 1992). Owens (1992) notesthat very
differentdensities (ranging from 37 to250 dwelling
units per hectare) are attainable using combined
heat and power systems, depending on discount
rates and fuel prices.
4. Decreased pollution from vehicle exhausts can be
achievedasaresultof adeclineinthe use ofcars,the
mixing of land uses, the provision of efficient and
accessiblepublictransportation,andwalking (Sten-
house1992; Owens 1992).High densities havebeen
found to be associated with lower gasoline con-
sumption per capita (Breheny 1996; Newman and
Kenworthy 1989); however, this is a controversial
issue(e.g.,Gordon andRichardson1997; Jenksetal.
1996a; Orchard 1995).
5. Decreased emission of pollutants may result from
energy-saving land use plans and from energy-
efficient buildings (Breheny 1992a).
The Concept of Density 399
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
1. High density may result in a decrease in the total
number of car trips (Breheny 1992a). Nasar (1997)
found lower automobile dependency scores in
high- versus low-density neighborhoods. These
differences were greater for older people, women,
and households with no children.
2. Highdensitymayresult inadecreasein thenumber
of kilometers per trip (Bannister 1992; Stenhouse
1992; Woodhull 1992; Berridge Lewinberg Green-
berg, Ltd. 1991b).
3. High density has been found to be related to a
higher proportion of travel on public transit, to
greater public transit service provision per person,
andto transit useby ahigher proportionof workers
(Breheny 1996; Newman and Kenworthy 1989). In-
creased public transit use, in turn, may reduce pol-
lution emissions (an environmental advantage).
4. High density enhances the opportunity to use pub-
lic transportation, since high density brings the de-
velopment of public transportation systems to the
thresholdsofprofitabilityandefficiency. The report
prepared by Berridge Lewinberg Greenberg, Ltd.
(1991b)adopts several benchmarksfor the relation-
ship between residential density and transit use. It
suggeststhat 17 to 75 dwellingunits per nethectare
are necessary to sustain significant transit use, and
150 dwelling units result in a modal split of differ-
enttransportation typesin which morethan 50 per-
cent is public transit.
5. As a result of an increase in transit use, traffic con-
gestion in residential, work, and commercial cen-
ters may decrease (Berridge Lewinberg Greenberg,
Ltd. 1991b).
6. Public transit can be more energy efficient (Reid
1986). Handy (1996) reminds us that it is the set of
choices correlated with density, not density itself,
that shapes travel behavior. Bannister (1992) dis-
cusses the interaction between socioeconomic cir-
cumstances and people’s propensity to travel with
different frequencies, trip lengths, and transporta-
tionmodes. Gender should beadded to these inter-
vening variables (Pickup 1984). Self (1997)
questions the effect that a change in density would
make. He argues, for example, that a 50 percent in-
crease in the density of Canberra, Australia, would
produce only a modest increase in public transit
7. High density offers more opportunities to walk or
ride a bicycle to work, service, and entertainment
facilities (Bannister 1992; Woodhull 1992).
1. Highdensities may resultin economiesof scalethat
facilitate the use of better quality and more attrac-
tive building materials (Hitchcock 1994).
2. Highdensity enables the use of a building complex
as an element of the urban composition (Hitchcock
3. High density allows for a variety of densities and
typesof construction in a given region. Variationin
density and construction, in turn, makes the envi-
ronment more interesting (Hitchcock 1994).
4. High-densitydevelopmentin theproximityof pub-
lictransportation linescan decrease the demand for
land located further from these lines (Shireman
5. High-densitydevelopmentas infillinexisting areas
canrevitalizethoseareas andreducethepressure to
develop open spaces (Berridge Lewinberg Green-
berg, Ltd. 1991a).
1. High density facilitates the supply of a variety of
relatively high-quality resources and housing,
health,education, culture, recreation, and munici-
pal service opportunities (Churchman et al. 1996;
Jenkset al. 1996; Audiracand Smith 1992) andem-
ployment opportunities (Berridge Lewinberg
Greenberg,Ltd.1991b) ofa qualitynotprovided in
low-density housing. This is particularly impor-
tant to meet the needs of various population
groups and household types and of people with
different interests and lifestyles (Forsyth 1997;
Lehman and Associates 1995; Churchman 1993;
Wohlwill 1985).
2. As a result of high density, a large number of ser-
vices may be located within walking distance from
dwellingsand inclose proximityto eachother. This
encourages walking and bicycling, thereby ena-
bling individuals (such as children or the elderly)
who cannot drive a car, people of limited means
whodo notown a private car, orpeople whowould
rather not use their private car to move around in-
dependently(Churchman 1993; vanVliet 1985; Loo
and Ong 1984). However, Christoforidis (1993) ar-
guesthat peoplethink interms of time not space,so
a three-minute drive may seem closer than a five-
minute walk.
3. Highdensitymakesitpossible to maintainanacces-
sible, comfortable, and frequent system of public
transportation that contributes to the independent
ability of various kinds of population groups (chil-
dren, teenagers, the elderly, the handicapped, and
thosewithout cars) toavail themselves ofresources
that are out of their immediate reach (Churchman
et al. 1996;Hillman 1996). Theimportance of such a
publictransportation systemis thatit isan essential
linkbetween homes,services, employment,and so-
cial networks. Where such a system does not exist,
orwhere it onlyfills minimum needs,the quality of
lifefor those whodepend on itis seriously affected.
Womenareespeciallyaffectedbecause theyusually
shoulder the burden of managing the demands of
family and work, among others, in the absence of a
proper public transportation system (Churchman
et al. 1996; Haughton and Hunter 1994).
4. High density frees land for recreation and open
space(VanderRyn and Calthorpe1986a).Aconsid-
erable amount of land for open space may also be
released if higher densities result in a reduction in
automobile dependency (Owens 1992).
5. Highdensity facilitates more activities in the center
of a city, thus contributing to the city’s vitality
400 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
(Jenks et al. 1996; Haughton and Hunter 1994; Rob-
erts 1978).
6. High density may offer possibilities for social sup-
port and attachment on one hand and the potential
for obtaining desired kinds of privacy (e.g., ano-
nymity)on the other (Lehmanand Associates 1995;
Jain1987; Churchman and Ginsberg1984a; Roberts
7. Highdensityenablesdiversity inthechoiceof peers
and associates, not only because more people are
present but also because there is greater variety
among those present (Churchman 1993; van Vliet
1985; Wohlwill 1985).
1. Highdensity makes it possibleto economize on the
construction costs of housing units. Higher density
housingclusters,in whichdwellingunits aggregate
more closely to leave larger common open spaces,
can be much more economical (Studies quoted by
Alexander 1993). High-density development re-
duces land, infrastructure, and building costs
(Christoforidis 1993; Preiser 1992).
2. High density is economically efficient because it is
based on dense construction on high-priced land
(Ottensmann1977 and Buttler 1981quoted in Alex-
ander 1993).
3. High density affords economies of scale in relation
to the public and private provision of urban infra-
structure, services, and amenities (Haughton and
Hunter 1994; Hitchcock 1994).
4. Highdensity allowsfor theefficient useof landand
publicservices yet maintainsa high“quality oflife”
(Lehman and Associates 1995).
5. Highdensity is more spatially and energyefficient.
It also requires less land for urban development
(Lehman and Associates 1995). (See Gordon and
Richardson 1997 for a challenge to this claim.)
6. High density makes public transportation more vi-
able (Rydin 1992; Reid 1986).
7. Smyth (1992) argues that 40 percent of the initial
costof development in conventional land use plan-
ning is automobile related (e.g., freeways, streets,
stoplights, parking lots, driveways, garages, park-
ing structures, and associated land). A different
kindoflanduseplanning, based onhigherdensities
and public transportation, could reduce some por-
tion of this cost.
8. High density (along with mixed land uses) allows
for the technical and economic viability of certain
energy technologies and transportation systems
(Owens 1992).
1. Highdensity may resultin the loss of open and rec-
reational space (Jenks et al. 1996; Breheny 1992a)
and thereby reduce an area’s capacity to absorb
rainfall (Troy 1996).
2. High-densityconstructionmayrequirehigh energy
use (Rydin 1992).
3. High density limits the use of some forms of ambi-
ent energy systems, such as passive solar power
(Owens 1992; Rydin 1992).
4. A high-density area may be subject to congestion
andpollution(Breheny 1992a).Higherlevels ofpol-
lution also may occur because of reduced space for
treesandshrubs thatpurify theairand coolthe area
(Troy 1996).
5. High density reduces the capacity to cope with do-
mesticwastes anddecreases opportunitiesfor recy-
cling (Troy 1996).
1. Highdensitymayleadto traffic congestion(Jenkset
al.1996; Rydin 1992)and to an increased number of
traffic accidents (Troy 1996).
2. Highdensity maycreate pedestriancongestion and
congestion in public transportation facilities if pe-
destrian and public transportation systems have
notbeen developed toaccommodate high densities
(Ruback and Pandey 1992; Roberts 1978).
3. High-rise, high-density construction may cause a
major point source of congestion at the street level
(Troy 1996).
1. Thereis less choiceas tothe placementof abuilding
on a lot when net density increases (Hitchcock
2. High-rise, high-density construction may obstruct
views, cause shadowing, and give a visual sense of
lack of proportion (Hitchcock 1994).
1. High density may cause psychological stress and
violationsofpersonalspace(Jain 1987;LooandOng
2. High density may lead to physiological overstimu-
lation, negative health effects (Evans and Cohen
1987), or cognitive overload because of the de-
mands of functioning in a very dense environment
(Baum and Paulus 1987).
3. High density may lead to constraints on an indi-
vidual’sbehaviorand freedomofchoice (Baum and
Paulus 1987).
4. Negative personal consequences associated with
higher densities may be manifested in anxiety, so-
cial withdrawal, and a feeling of loss of control
(Baum and Paulus 1987; Fleming et al. 1987; Jain
5. High density may invoke a feeling of reduced pri-
vacy and personal security (Troy 1996; McCarthy
and Saegert 1978).
6. High density may lead to difficulty in supervising
the location of children’s outdoor play spaces and
choice of friends (Aiello et al. 1985).
1. High density may lead to competition between
groups for space and to other social conflicts (Jain
1987; Loo and Ong 1984).
2. High density may be associated with a severing of
social ties (McCarthy and Saegert 1978), perhaps
becausepeopleare less abletoregulate theircontact
The Concept of Density 401
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
with others (Fleming et al. 1987). Higher density
mayalso be related to alower overall sense of com-
munity (Wilson and Baldassare 1996).
3. High density (with high-rise buildings) may pro-
mote social segregation (Radberg 1996).
4. Highdensitymaybeassociated withrelativelylittle
public open space (Breheny 1992a; Simon and
Wekerle 1987). Yet, van Andel (1998) did not find a
clear relationship between neighborhood open
space and the amount of open space per child. This
differencein results clearly illustratesthat there are
otherinterveningfactorsthat affectwhetherthereis
publicopenspaceina high-density neighborhood.
5. High density may make it difficult to maintain an
automobile and thus may result in the loss of a
status symbol (Mullins 1995; Rapoport 1977).
1. Veryhigh-density construction may bemore costly
than medium- or low-density construction (Alex-
ander 1993).
2. The operational energy costs of buildings increase
for taller high-density construction (Troy 1996).4
High-rise, high-density buildings frequently cost
moretobuild andmaintain(Ewing 1997;Haughton
and Hunter 1994).
3. The value of land in the city center may soar as the
result of high-density development (Alexander
4. High density in the city may have a detrimental ef-
fectoneconomicdevelopmentin surroundingrural
regions (Breheny 1992a).
5. Land absorption for high-density projects takes
longer than for low-density projects because more
unitsmust be sold toabsorb each acre of land (Prei-
ser 1992).
6. Higher density development in inner-city areas
may require the very costly upgrading of existing
infrastructure (Troy 1992).
Considering all of these potential advantages and
disadvantages, in very simplistic terms, one could rec-
ommend maximizing the advantages of high density
and minimizing the disadvantages. However, this is
not as simple as it sounds because (1) there is no clear-
cut agreement among professionals and researchers as
to what is an advantage and what is a disadvantage of
high density; (2) for the residents and users of an envi-
ronment, one person’s advantage may be another per-
son’s disadvantage; (3) at least some of the factors are
not under the control of planners or politicians, includ-
ing subjective interpretations by residents and users
(ChurchmanandGinsberg1984b);and(4) subjectivein-
tervening variables that relate to the concepts of per-
ceived density and crowding. These concepts will be
described in the next section.
Furthermore, many of the statements made by plan-
nersandresearchersareassertionsand assumptions,not
substantiatedfacts.This would be acceptable if these as-
sertions were stated as hypotheses that need to be ex-
amined and tested (see, for example, Churchman
forthcoming), but they usually are not. Perhaps even
more problematic, many of these assertions are stated in
deterministiclanguage that says thatifonedoes such and
such in the physical environment, people will behave or
feel in a particular way. Audirac and Shermyen (1994)
pointto an exampleofdeterministic language in theTND
ordinances:“by walkingin definedpublic spaces,citizens
come to know each other, and to watch over their collec-
tivesecurity.”Yet,thereisnoway toknow whetherciti-
zenswillindeed walkinthesepublicspaces orwhether,
is also no way to know whether citizens will perceive
these spaces, in terms of collective security, as places
that they are responsible for watching over.
It is important, therefore, that we begin by not using
deterministic language (Haughton and Hunter 1994).
Wemust also recognize the contingent nature of all de-
sign and planning actions and that these actions inter-
act with many other variables, especially the choices
made by people.
Theconcepts of perceived density and crowding are
grounded in a contextual approach to the study of the
person-environment unit. This means that the theory
and research on these concepts attempts to identify the
conditions under which high density does or does not
lead to negative effects on health, behavior, feelings, or
attitudes. The difficulty in defining high density illus-
trates the importance of a contextual approach that in-
corporates all relevant physical, social, cultural,
economic, geographic, ecological, technological, and
personal aspects of a situation. There is great variation
in terms of which aspect is relevant in a given situation
andinwhatwaysthataspect isrelevant(Stokols 1987).
Studies that have attempted to investigate the sub-
jective aspects of density have a number of methodo-
logicallimitationsthatmustbetaken intoaccount.First,
resultsfromone environmentalscaleareoftenassumed
to be relevant to other environmental scales. For exam-
inside the home), which is clearly very different from
residentialdensity (densityoutside thehome),whether
at the building, street, or city level. Second, some of the
research has been short term and conducted in
laboratory-type settings in which “extraneous” vari-
ables are controlled. These studies suffer from serious
problems of ecological validity (Stokols 1987). Ecologi-
cal validity suggests that the assumption that a phe-
nomenon studied in a laboratory setting will be the
same in a real-life situation is unjustified. A similar ca-
402 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
veatis legitimateforpast attemptsto argue thatone can
learn about human behavior from animal studies.
Third,itisnotasimpletask todefine therelevant physi-
cal area for examination, particularly at the neighbor-
hood scale. There is much evidence that indicates that
residents’subjective definitionof theirneighborhoodis
oftenverydifferentfrom the official definitions of their
neighborhood’s boundaries (Chaskin 1998; Rapoport
1997).(See Churchman and Ginosarforthcoming for an
attempt to deal with this problem.) It is also not clear
that the neighborhood is a relevant and important set-
tingfor everyone inthesame manner or tothesame de-
gree. The amount of time adults who work outside the
home spend in a neighborhood is likely to be very dif-
ferent from the time spent by children, the elderly, and
people without access to private or public transporta-
tion.Furthermore,eachofthesegroupsmay experience
different environments and different numbers of envi-
ronmentsintheirdailylives thathave differentlevels of
effectivedensity. Thus, even if people nominally livein
the same neighborhood, they may experience very dif-
ferent levels of density and combinations of densities
overtime. Whatdoesit meanifa person livesin a dense
neighborhoodbutworks inaverylow-densityenviron-
ment? Or what does it mean if a person lives in a high-
density neighborhood and attends a high-density
school, plays on a high-density playground, and walks
in high-density streets?
Perceived Density
Perceived density is defined as an individual’s per-
ception and estimate of the number of cues in the envi-
ronment that represent people and their activities.
Factors contributing to perceived density are hypothe-
sizedto include the perceptual,associational-symbolic,
and physical aspects of an environment; the temporal
aspects of activities; and the sociocultural characteris-
ticsofactors andsettings(Rapoport1975).Thenotionof
perceiveddensityis based on the fact that any environ-
ment’s nature, the potential for action that an
environment offers, and the behavior appropriate for
that environment. Certain physical and social cues can
beread and interpretedas indicating ahigh-densityen-
vironment. Other cues can be read as indicating a less
denseenvironment. Inbothcases, thesecues are atleast
partly independent of the actual number of people per
unit area (Rapoport 1977). The degree to which a per-
ceived environment makes demands on our attention
andthelevelofinformationprocessingan environment
requires are related to the degree of density that is per-
ceived. These factors interact with a person’s percep-
tualabilitiesbecauseindividual thresholdsfor visualor
auditory stimuli may be very different. For example, a
person who is blind will judge the level of information
provided by an environment differently than a person
whois not blind. However, such ajudgmentis not in it-
self an evaluation. Evaluation is the step during which
the appropriateness of this level of density is assessed.
Evaluation is the essence of crowding.
Physical variables are hypothesized to be related to
perceived density by affecting the number of physical
sensory stimuli in an environment that indicate the ac-
tual or potential presence of people. These physical
variables include tight or open spaces; intricate or sim-
ple spaces; large or small building height to space ra-
tios;numerousorfew signs, lights, cars, and people (or
theirtraces); thepredominanceof artificialversusnatu-
Marcus and Sarkissian 1986; Rapoport 1975); and the
presence or absence of nonresidential or mixed land
uses (Vancouver City Planning Department 1978).
Cooper Marcus and Sarkissian (1986) recommend a
list of design variables that may serve to reduce per-
ceived density. These design variables include a rela-
tively small neighborhood size, greater spacing
between buildings, visual and functional accessibility
fromadwelling unittoopenspaces,respect forprivacy,
division into small clusters, diverse elevation designs,
fewer dwelling units that use the same building en-
trance,minimal noiseinfiltration, well-located commu-
nity services, and convenient parking. The importance
oftheconcept of perceived density is that it shows how
physicalphenomena can be manipulated in an attempt
toincrease the probabilityofgreater or lesserperceived
densities (Jacobs and Appleyard 1987).
The theoretical approaches and proposed hypothe-
ses for the psychological and social significance of
crowding and the research available that deals with
crowding in the context of residential density are pre-
sented in this section. Crowding is the concept associ-
ated with the negative psychological and social
ing the crowding experience stems from the fact that
feeling of crowding may be occasioned by intraper-
sonal,interpersonal, orphysical conditionsorby thein-
teraction between all of these conditions in a given
situation. In high-density conditions, for example,
stressmay beexperiencedas aresult of heator noise, an
overflow of social stimuli, an excess of interpersonal
physicalproximity, toomany partnerssharingan inter-
action, or some combination of all of these factors. The
difficulty in identifying crowding lies in the complex
conditions that give rise to stress at different levels of
The Concept of Density 403
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
The stress occasioned by crowding experiences may
result in coping responses and in short-term and long-
term psychological, physical, and physiological conse-
quences. The individual effects of the feeling of crowd-
ing can consist of psychological, behavioral, and
delayed cumulative responses. Psychological re-
sponses to distress include perceptual, cognitive, or
emotional processes of adjustment or changes in atti-
tudes toward other people. Behavioral responses may
include assertiveness or psychological withdrawal,
physical withdrawal (or the intent to withdraw) from
the scene (Ginsberg and Churchman 1984), attempts to
changethe environment or adapttothe demands of the
situation, or changes in task performance or coping re-
1997). Delayed cumulative responses include changes
in health and in performance that take place after the
emergence of the feeling of crowding (Sundstrom
1978). Cumulative effects may result from the stress it-
self,the effort expended during coping, the effort spent
in adaptation, or some combination of these responses
(Baumand Paulus 1987). A cautionary noteiscalled for
because virtually all available data on these conse-
quences are related to living density, not residential
density(Altman and Rogoff 1987). These consequences
must, therefore, still be considered hypothetical.
Four theoretical models have been proposed to ex-
plain the conditions under which crowding will be ex-
perienced, the underlying processes of the crowding
experience, and the mechanisms that account for the
negative effects of high density. The behavioral con-
straint model holds that density that interferes with
goal attainment, restricts or inhibits movement, and is
generally noxious because of reduced freedom may be
evaluatedas crowding. Thecontrol-densitymodel sug-
gests that density that makes environments more un-
predictableandallowsless control over a situation and
over privacy may be evaluated as crowding. The over-
load/arousalmodel suggeststhat a feelingof crowding
results when density generates excessive stimulation
that overwhelms the sensory systems. This causes an
overload or overarousal of the nervous system. The
density/intensity model indicates that high density
may intensify existing life stresses and problems, such
asinterpersonalrelations (Evans and Lepore 1992). It is
very probable that some combination of the four mod-
els operates simultaneously in any given situation.
Based on these models and on the results of empiri-
cal studies, researchers have suggested variables that
may interact with density in such a way as to lead to a
feelingofcrowding. Thesevariables canbe dividedinto
variables that pertain to the physical environment, to
the social environment, to situational characteristics,
and to the characteristics of the individual(s) involved
in the situation. However, it is critical to recognize that
these variables are interrelated and none operate inde-
pendently, and many of them are culturally defined.
Density. This is, of course, the necessary first vari-
able. There is no existing research that addresses the
questionof how important net or gross residential den-
sity is to the experience of crowding. However, there is
no direct relationship between density and crowding
(Gifford 1997)—this has been an accepted basic tenet
sinceStokols (1972)firstdistinguished betweenthetwo
andstress generated byhigh density diminishesinpro-
portionto increases in the number and quality of avail-
able physical resources and services, such as social
services and public parks (Gifford 1997; Jain 1987).
Community size. The possibility exists that the same
density is experienced differently in large and small
neighborhoods(Gifford1997). Theremayalsobediffer-
encesin reactions todifferentsize projects thathavethe
same density in terms of the number of dwelling units
(Bonnes et al. 1991; Vancouver City Planning Depart-
ment 1978).
Housing type homogeneity. This variable may have an
larhousing may assume that they have similar lifestyles
and norms of behavior (Gifford 1997; Merry 1987).
Greater space between houses. Because this affords
moreprivacy,it may enable residents to avoid conflicts
between neighbors (Merry 1987).
Balance between built and open spaces (public and pri-
vate). Open space is important to people for many rea-
sons. These reasons include the fact that open space
provides opportunities for specific kinds of behavior
andrecreation,forprivacy, forreducing perceivedden-
sity,and,insomeplaces,for indicatingsocial statusand
quality of life (Bonnes et al. 1991; Merry 1987).
Weather.There issome indicationthatheat caninten-
sify the negative evaluation of a given density (Ruback
and Pandey 1992).
Mixed land use. Proximity to commercial and indus-
trialactivitiesis foundtoberelatedto feelingsofcrowd-
ing (Fleming et al. 1987; Schmidt et al. 1979).
Defensemeasures. Defense measures, such as erecting
atall fence or locking doors,areefforts to prevent inter-
ference that is caused by the presence of others. Re-
course to a greater number of defense measures may
mitigate the feeling of crowding (Jain 1987).
Sensory overload. Sensory overload reduces an indi-
vidual’scapacity to attend to the stimuli important to a
404 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
specificsituation (Gifford1997;Holohan andWanders-
man 1987).
Control.Theability toregulatesocialinteractionsand
choose the frequency and duration of interpersonal
contacts diminishes the feeling of crowding (Gifford
1997; Evans and Lepore 1992; McCarthy and Saegert
Other people in the interaction. It matters whether
other people in the interaction are family, friends,
neighbors, strangers, similar, or perceived as not dan-
gerous (Gifford 1997; Bonnes et al. 1991; Jain 1987).
Abilityto achieve different kinds of privacy. Selectiveso-
cial interaction, solitude, and anonymity are examples
ofdifferent kindsofprivacy. Sometimesaperson needs
solitude, but at other times a person may wish for vari-
ous kinds and degrees of social interaction (Altman
Primary or secondary environments. A primary envi-
ronment is an environment, such as home or work, in
which a person spends a great deal of time, relates to
other people in the environment on a personal level,
and engages in important personal activities. Primary
environmentsthat havehighdensities maycausea per-
son to feel a greater “psychological security” threat
than anywhere else. Thus, the feeling of crowding in a
high-density apartment is hypothesized to be greater
than that experienced in a high-density neighborhood
(Loo and Ong 1984; Stokols 1976).
Source of interference. If a social interference or threat
is perceived as deliberate and personal (originating
from a particular person), crowding is experienced
more intensely than when the interference is perceived
as impersonal. Similarly, a breach of interpersonal dis-
tancenormsmaycausea moreintense feelingof crowd-
ing when the breach seems intentional and not the
result of physical conditions beyond a person’s control
(Russell and Snodgrass 1987; Stokols 1976).
and sometimes found, that culture plays a moderating
role by defining coping mechanisms, acceptable dis-
tancesfor personal space, and appropriate densities for
particularsettings (Bechtel 1997; Jain 1987; Merry 1987;
Rapoport1977;Gillisetal.1986). However,itisnotclear
what the nature of that moderating role is and whether
it stands on its own or is linked with other variables
(Loo and Ong 1984). It is interesting to note in this re-
gard that there are no words in the Dutch language for
crowding and privacy (Stringer and Kremer 1987).
Cultural heterogeneity or homogeneity. There are cul-
tural and personal mechanisms for coping with high-
density situations. In situations in which there is much
heterogeneity, it is possible that these coping mecha-
nisms may fail because of differences in values and the
absence of shared norms of behavior (Merry 1987).
Perceived safety. In some situations, the presence of
many others may be perceived as a threat. In other
situations, it may be seen as affording safety (Bernard
1992; Jacobs 1961). This depends on the other variables
presented here.
Latent and manifest meanings or functions. Environ-
mental elements may have latent meanings that are
moreculturally determined than their manifest mean-
ings. Latent meanings relate to values, images, and
symbolic landscapes that may be positively or nega-
tively evaluated, regardless of density (Rapoport
Perceived density. The assumption is that a high per-
ceived density is more likely to be related to crowding
thana low perceiveddensity.Nevertheless, it isalso as-
sumed that there is no simple and direct relationship
betweenperceiveddensityandcrowding.Itis theinter-
action between perceived density and other variables
that may result in a negative evaluation (Gifford 1997;
Rapoport 1975).
Social or spatial density. Whether a specific high den-
sity is a function of large numbers of people or a func-
tionof alackof spacemay make adifference in people’s
reaction to that density (Baum and Paulus 1987).
Competitionover resources. The perceptionofwhether
there are sufficient resources (e.g., services, transporta-
tion, and open space) for all may affect feelings of
crowding (Jain 1987; Wohlwill 1985).
Duration and frequency. Exposure to high density can
be tolerated for a short time (e.g., an elevator ride), but
extended exposure increases the probability of experi-
encing a feeling of crowding (Loo and Ong 1984). On
the other hand, extended periods of high density may
be tolerable if a person knows how long exposure will
last (Roberts 1978; Sundstrom 1978).
Predictability.Adverse conditionswillgenerate more
stress when they are unpredictable (Sundstrom 1978).
Choice. Choice involves whether a situation can be
changed or whether a person is free to leave a situation
(Bonnes and Secchiaroli 1995; Loo and Ong 1984).
Behavioralfreedoms orconstraints. Towhatextentdoes
the presence of other people limit a person’s behavior
rather than offer more opportunities for social interac-
tion (Evans and Lepore 1992; Roberts 1978; Saegert
The Concept of Density 405
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
Goalachievement. Towhatextent doesthe presence of
other people inhibit or facilitate the achievement of a
person’sgoal(s)? Forexample,a highdensityat apoliti-
cal demonstration may not be evaluated as crowding
because the presence of many people who share a per-
son’s political goals contributes to the achievement of
those goals (Russell and Snodgrass 1987).
Adaptation. People who grow up in a high-density
building may adjust more easily to social stimuli and
mayprefer tolivein arelativelyhigh-density neighbor-
hood.Ahistory ofintense orfrequent socialstimulation
may result in greater tolerance for high-density condi-
tions (Sundstrom 1978) and an improved ability to cope
with high-density conditions. Adaptation may be ex-
pressed, for example, as satisfaction with a high-density
neighborhoodand its accessible physicalservicesorasa
wish to continue living in present conditions (Jain
Personal experience. There is some evidence from re-
lated research that people who experience a particular
situation (e.g., high-rise living) are more favorable to it
(Churchman and Ginsberg 1984a).
Other intervening variables that may affect the feel-
ing of crowding are personal characteristics such as
gender(Bechtel1997),age (Merry1987; Wohlwill1987),
socioeconomicstatus(Bonnes andSecchiaroli1995;Jain
1987), lifestyle (Merry 1987), stage in the life cycle
(Churchman and Ginsberg 1984b; Michelson 1977),
personalpreferences,expectations, attitudes, personal-
ity, and coping capabilities (Gifford 1997; Baum and
Paulus 1987; Altman 1975).
Attitudes Toward Density
To round out this discussion on the subjective as-
pectsof density,it isimportant tolookat thequestion of
whether people (excluding planning professionals or
decisionmakers) are willing to accept higher densities.
There is not much direct research on this question, but
there are some indications as to what the different atti-
tudes might be.
In many countries—particularly the United States
and Australia—the often unexamined assumption is
that most people (or even everyone) would like a
single-familyhomeandthatthisinherently implieslow
densities (Gordon and Richardson 1997; Mullins 1995;
Scully 1994; Audirac and Shermyen 1994). However,
Table2shows thatsingle-familyhomescanandalready
do exist at various levels of density. Accepting the as-
sumption that most people want a single-family home,
some have attempted to intensify this building type to
relatively higher densities. For example, using smaller
lots and varying the shape and proportions of lots can
double density from the typical seven to ten single-
family homes per net hectare to seventeen to nineteen
homes per hectare (Preiser 1992; Wentling 1991). En-
couraging the development of attached townhouses
per hectare (Alterman and Churchman 1998). Con-
structing passive solar houses at forty dwelling units
per hectare, instead of the typical twenty-five dwelling
unitsperhectare, hasbeenusedtoincreasenetdensities
in Great Britain (Owens 1992). Audirac and Smith
(1992) found that there are people who are willing to
tradelarge lot sizesformore access to parksandrecrea-
tionorfor amorecentralanddenserlocation.However,
the increases in density that result from these ap-
proachesarerelatively small. They cannot achieve the
majorgoals and presumed advantages ofhighdensity
that have been discussed in previous sections of this
Some suggest that low-density suburbs should be
changed, not totally rejected. Van der Ryn and Calt-
horpe (1986a) suggest that there is a need to intensify
the culture of suburbs rather than dismiss suburbs be-
causethey ostensibly lackculture. Troy (1996)similarly
argues for the improvement of suburbs for teenagers
rather than encouraging them to move to the cities.
However, these suggestions ignore the question of
whether any changes can be significant enough to
achieve the goals of intensifying culture or improving
the situation for teenagers. The list of the advantages
of high density suggests that they cannot be. Further-
more, coming to some sort of conclusion about the
meritsofrelatively low-density housing, even when it
hasbeen moderately intensified, requires revisiting the
literature that identifies the problems of low-density
suburbs for many groups in the population (Church-
man 1993).
The question then becomes whether people can be
erence for single-family homes and low density. Shlay
(1985) argues that expressed preferences for single-
family home ownership may actually be shorthand for
otherunrevealed preferences such as a preference for a
middle-classstatus, a family-centered lifestyle, or a ho-
mogeneous residential suburb. Doyle (nd) argues that
itmay bepossibleto separatethedesirable components
of housing bundles (or particular characteristics that
are often found together) and use those components to
create new housing types based on new bundles. If, for
example, many people associate higher densities with
fewer residential amenities (e.g., ownership, open
space, and parking), planning that offers these ameni-
ties at higher densities may be able to attract people to
higher density areas.
Woodhull (1992) contends that automobile traffic as-
sociatedwithhigh densityis dislikedmost bypeople. He
argues that, at present, much is done to make high-
406 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
density living unbearable by, for example, locating
high-density developments near freeways or not pro-
viding adequate services or green space. Converting a
significant portion of the spatial resources consumed
by the car to other land uses may increase the positive
effects of high density and reduce the negative effects.
A number of studies have found that density is not
necessarily the key issue for residents of multifamily
housing. A report prepared by Lehman and Associates
(1995) notes that the most important multifamily hous-
ing issues raised in public focus groups were building
design, neighborhood integration, landscaping, visual
appearance, and scale in terms of height, not density
perse. Planners inVancouverfound that thenumberof
unitsinamultifamilyprojectwas more important than
density to user satisfaction (Vancouver City Planning
Department 1978).
There are many people who choose to live in multi-
familyhigh-density housingfor a varietyof reasons,in-
cludinglocation, lifecyclestage, lifestyle(Mullins1995;
Michelson 1977), relative cost (Mullins 1995; Audirac
and Shermyen 1994), and the social benefits of neigh-
boring and of having a variety of friends for children
(Churchman and Ginsberg 1984a). Furthermore, there
are many people who do not have the luxury of choice,
eitherbecauseofeconomiccircumstances(Self 1997) or
because of social or legal restrictions on where they are
allowed to live.
Thereare some indications that people are willing to
accept intensification and higher levels of density un-
dercertainconditions. For example, a study conducted
in Vancouver found that residents in single-family ar-
eas were more willing to accept new infill high-density
housingifitwas clearly family housing, if it was an im-
provement over previous land uses, if it was accompa-
niedbyacommunityfacilitysuch asapark,andifitwas
locatedinareas that were already heterogeneous (Van-
Aviv, Israel, found that residents were more open to
gradual increases in scale—that is, increasing density
by adding two additional stories to existing two- or
three-story multifamily buildings—than to the con-
struction of new high-rise buildings of sixteen or more
stories (Churchman 1998). In Paramount, California, a
planning proposal for a high-density residential devel-
opment policy of 148 multifamily dwelling units per
acre was rejected by voters, but a proposal for a maxi-
mum of 54 dwelling units per acre was approved (Di-
Mento et al. 1997).
A number of researchers advocate low-rise, high-
density residential development as a good middle
ground that offers social, economic, and ecological ad-
vantages (Alterman and Churchman 1998; Goodchild
1994; Cooper Marcus and Sarkissian 1986). Low-rise,
high-density developments may be able to achieve
many of the advantages of high densities without the
disadvantages, depending on intervening contextual
variables and specific site, urban design, and land use
planning. Breheny (1992a) cautions that these types of
developments may require reducing open-space stan-
dards and the subsequent loss of green space. This is
one of the disadvantages of low-rise, high-density de-
velopment that needs to be prevented, if possible.
One way to attempt to ameliorate some of the nega-
tive aspects of high density may be to attempt to de-
crease perceived density. Any design or planning
action that lowers the level of sensory stimulation that
represents human activity or the potential for such ac-
worthexploring,butthatreceivesvirtuallyno attention
and Hunter (1994), is the use of underground space for
morethan just underground parking, infrastructure, or
son to hypothesize that many land uses, particularly
uses that are affected by windows and sunlight (e.g.,
cinemas, theaters, and conference centers), would be
acceptable candidates for underground development.
Increasingthe use of underground space wouldreduce
density and conserve land.
It should be noted that there may be situations in
whichan increaseinperceived densityisuseful. For ex-
ample, an increase in perceived density could be used
to enhance the perceived vitality and urban quality of
an environment. An increase in perceived density
should occur if actions opposite to those described
above (with the exception of violating privacy) are
Whatlessons can be learned given the complexity of
the meaning and use of the term density and the addi-
tion of the subjective terms perceived density and
crowding? At the most basic level, density measures
mustbe clearlyand explicitlydefinedsothatcommuni-
cationcan take placeand so thatwecan learn fromeach
the interrelationships between variables and factors
mustbeaddressedin researchon densityasitisin prac-
tice. Real-world complexity includes a subjective ele-
ment that is always present in people’s behaviors,
expectations, and attitudes (including those of decision-
makers,planningprofessionals, and researchers); thus,
itmust be taken into account. It is easier for planners to
affect density and perceived density than to affect the
subjective experience of crowding. However, planners
have no choice but to try to address the implications of
the intervening factors that are relevant to crowding.
Another lesson is that no one solution will meet the
needs of every situation, context, person, or group.
The Concept of Density 407
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
Therefore, a variety of solutions (different types of set-
tlements,neighborhoods,housing,and transportation)
are essential to meet the needs between and within
countries, regions, and towns. Solutions should be
based on an understanding of the differences in needs
andexpectationsofrelevant groups so that they can of-
fer choices that can meet these needs and expectations.
A final lesson is that much more research is needed on
thevariousaspectsandramificationsof different kinds
andlevels of density. This isparticularlytruefor the re-
lationship between objective density, perceived den-
sity, and positive or negative subjective evaluations.
Planners will continue to use the term density be-
cause it is too good to resist. This article synthesizes re-
searchand practice literature inanattempt to provide a
better understanding of the various ramifications of
density so that its use is more considered. The hope is
that a better understanding of density will result in
more effective density-related planning.
I wish to express my heartfelt thanks to the three anonymous re-
viewers who read the first version of the article with such care and
providedme withextremely helpfuland challengingcomments. The
initial basis for this literature review was undertaken as part of a
planning policy project commissioned by the Israel Ministry of the
Interior (Alterman and Churchman 1998). Michal Mitrany served
as research assistant for that literature review.
1.Many ofthe articlesreviewed here are literature reviews. Thus,
thisliteraturereview is even moreextensivethan the listofreferences
2.Another important kind of density,living density, relates to the
number of persons per room in a dwelling. In this case, density-
relatedproblemsinvolvequestionsabouthowahousehold is defined
andwhich rooms areincluded in thedensity calculation. Somecalcu-
lations include the kitchen and bathroom(s), others do not. This arti-
cle focuses solely on residential density and does not deal with the
questions of living density because living density is too important to
be included here in a minimal way. Gaps in present knowledge on
densityinclude anunderstanding of how living density and residen-
tialdensity are related and whether a positiveexperience at one den-
sitylevel can compensate for anegative experience at the other level.
It is also unclear which level of density is more important.
4.No indicationis givenas towhat heightis consideredproblem-
aticinthissense.Presumably,thecutoff point would vary and becon-
text related. Alterman and Churchman (1998) suggest a cutoff point
of greater than twelve stories for Israel.
Aiello, John, Donna Thompson, and Andrew Baum. 1985. Children,
crowding,andcontrol: Effects ofenvironmental stress onsocial be-
havior. In Habitats for children: The impacts of density.See Wohlwill
and van Vliet 1985.
Alexander, Ernest. 1993. Density measures: A review and analysis.
Journal of Architectural and Planning Research 10, 3: 181-202.
Alterman, Rachelle. 1997. The challenge of farmland preservation:
Lessonsfrom a six-nation comparison. Journal of theAmerican Plan-
ning Association 63, 2: 220-43.
Alterman, Rachelle, and Arza Churchman. 1998. Housing density: A
guide to increasing the efficiency of urban land use (in Hebrew). Haifa,
Israel: Technion, Center for Urban and Regional Studies.
Altman, Irwin. 1975. The environment and social behavior. Monterey,
CA: Brooks/Cole.
Trait, interactional, organismic and transactional perspectives. In
Audirac, Ivonne, and Anne Shermyen. 1994. Postmodern placebo or
remedy for suburban malaise. Journal of Planning Education and Re-
search 13, 3: 161-73.
Audirac, Ivonne, and Marc Smith. 1992. Urban form and residential
choice: Preference for urban density in Florida. Journal of Architec-
tural and Planning Research 9, 1: 19-32.
Bannister, David. 1992. Energy use, transport and settlement pat-
Barton, H. 1992. City transport: Strategies for sustainability. In Sus-
tainable development and urban form.See Breheny 1992b.
Baum, Andrew, and Paul Paulus. 1987. Crowding. In Handbook of en-
vironmental psychology.See Stokols and Altman 1987.
sand Oaks, CA: Sage.
Bernard, Yvonne. 1992. North American and European research on
BerridgeLewinberg Greenberg, Ltd. 1991a.Guidelines for thereurbani-
sation of Metropolitan Toronto. Toronto, Canada: Municipality of
Metropolitan Toronto Corporate Printing Services.
. 1991b. Study of the reurbanisation of Metropolitan Toronto. To-
ronto, Canada: Municipality of Metropolitan Toronto Corporate
Printing Services.
Bonnes, Mirilia, Marino Bonaiuto, and Anna Paola Ercolani. 1991.
Crowding and residential satisfaction in the urban environment.
Environment and Behavior 23, 5: 531-52.
Bonnes, Mirilia, and Gianfranco Secchiaroli. 1995. Environmental psy-
chology: A psycho-social introduction. London: Sage.
Breheny, Michael. 1996. Centrists, decentrists and compromisers:
Viewson the future of urban form. InThe compact city: A sustainable
urban form? See Jenks et al. 1996a.
. 1992a. The contradictions of compact city: A review. In Sus-
tainable development and urban form.See Breheny 1992b.
, ed. 1992b. Sustainable development and urban form. London:
Burton, Tony, and Lilli Matson. 1996. Urban footprints: Making best
useofurbanland and resources: Aruralperspective.In The compact
city: A sustainable urban form? See Jenks et al. 1996a.
Buttler,H.J. 1981. Equilibriumofa residential city, attributesof hous-
ing and land-use zoning. Urban Studies 18, 1: 23-40.
Calthorpe, Peter. 1993. The next American metropolis. New York:
Princeton Architectural Press.
. 1992. The pedestrian pocket: New strategies for suburban
growth. In Sustainable cities: Concepts and strategies for eco-city devel-
opment.See Walter et al. 1992.
Carmody, John, and Raymond Sterling. 1993. Underground space de-
sign. New York: Van Nostrand Reinhold.
Chaskin, Robert. 1998. Neighborhood as a unit of planning and ac-
Christoforidis, Alexander. 1993. New alternatives to the suburb:
Neo-traditional developments. Journal of Planning Literature 8, 4:
Churchman, Arza. Forthcoming. Women and the environment:
Questioned and unquestioned assumptions. In Theoretical perspec-
tives in environment-behavior research: Underlying assumptions, re-
search problems and methodologies, Seymour Wapner, John Demick,
408 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
.1998. A study of attitudes towards housing issues inthe Cityof Tel
Aviv(inHebrew).Haifa, Israel: Technion, CenterforUrbanandRe-
gional Studies.
. 1993. A differentiated perspective on urban quality of life:
Women, children and the elderly. In Perception and evaluation of en-
vironmental quality, Mirilia Bonnes, ed. Rome: UNESCO Pro-
gramme on Man and Biosphere.
Churchman, Arza, Rachelle Alterman, Yael Azmon, Ronit
Jerusalem: Israel Women’s Network.
Churchman,Arza,andOded Ginosar. Forthcoming. Atheoreticalba-
sis for the post-occupancy evaluation of neighborhoods. Journal of
Environmental Psychology.
Churchman,Arza, and Yona Ginsberg. 1984a. The image and experi-
enceof high-rise housing in Israel. Journal of Environmental Psychol-
ogy 4, 1: 27-41.
. 1984b. The use of behavioral science research in physical
planning: Some inherent limitations. Journal of Architectural and
Planning Research 1, 1: 57-66.
Cityof Newcastle uponTyne. 1993. Newcastleupon Tyneunitary devel-
opment plan. City of Newcastle upon Tyne.
CooperMarcus, Clare, and Wendy Sarkissian. 1986. Housing as ifpeo-
ple mattered. Berkeley: University of California Press.
DiMento, Joseph, Sherry Ryan, and Drusilla van Hengel. 1997. Local
government land use policy responses to the Century Free-
way/Transitway. Journal of Planning Education and Research 17, 2:
Doyle, D. Gregg. nd. Giving the people what they want: Residential
preference and increasing density. Unpublished paper.
Evans, Gary, and Sheldon Cohen. 1987. Environmental stress. In
Evans, Gary, and Stephen Lepore. 1992. Conceptual and analytic is-
suesincrowding research. JournalofEnvironmental Psychology 12,1:
American Planning Association 63, 1: 107-26.
Faludi,Andreas,and Arnold vanderValk. 1994. Ruleand order: Dutch
planning doctrine in the 20th century. Dordrecht, the Netherlands:
Kluwer Academic.
Fischer, Claude. 1976. The urban experience. New York: Harcourt,
Brace, Jovanovich.
Fleming,India, Andrew Baum, and LindaWeiss. 1987.Social density
andperceived control as mediators of crowding stressin highden-
sity residential neighborhoods. Journal of Personality and Social Psy-
chology 52: 899-906.
Forsyth, Ann. 1997. Five images of a suburb: Perspectives on a new
urbandevelopment. Journal of the American Planning Association 63,
1: 45-60.
Frost,Lionel,and Tony Dingle.1995.Sustaining suburbia: Anhistori-
cal perspective on Australia’s urban growth. In Australian cities: Is-
sues, strategies and policies for urban Australia in the 1990’s.See Troy
Gifford, Robert. 1997. Environmental psychology: Principles and prac-
tices. Boston: Allyn and Bacon.
Gillis,A. R., Madeline Richard, andJohn Hagan. 1986. Ethnic suscep-
tibilityto crowding: An empirical analysis.Environment and Behav-
ior 18, 6: 683-706.
Ginsberg, Yona, and Arza Churchman. 1984. Housing satisfaction
and intention to move: Their explanatory variables. Socio Economic
Planning Sciences 18, 6: 425-31.
Goodchild,Barry. 1994.Housing design,urban form and sustainable
development. Town Planning Review 65, 2: 143-58.
Gordon, Peter, and Harry Richardson. 1997. Are compact cities a de-
sirable planning goal? Journal of the American Planning Association
63, 1: 95-106.
Gowling, Derek, and Leith Penny. 1988. Urbanisation, planning and
administration in the London region: Processes of a metropolitan
culture. In Four metropolises in Western Europe, Hans van der Cam-
men, ed. Assen/Maastricht, the Netherlands: Van Gorcum.
Handy,Susan.1996.Understandingthe link between urbanformand
nonworktravel behavior. Journal of PlanningEducation and Research
15, 3: 183-98.
Haughton, Graham, and Colin Hunter. 1994. Sustainable cities. Lon-
don: Jessica Kingsley.
Hillman,Mayer.1996. In favor ofthecompact city. In Thecompactcity:
A sustainable urban form? See Jenks et al. 1996a.
Hitchcock,John.1994.A primer ontheuseof density in landuseplanning.
Papers on Planning and Design no. 41. Toronto, Canada: Program
in Planning, University of Toronto.
psychologyperspective in environmentalpsychology.In Handbook
of environmental psychology.See Stokols and Altman 1987.
Israel Ministry of the Interior. 1992. National master plan integrating
construction for development and immigrant absorption. Tel Aviv: Is-
rael Ministry of the Interior.
Jacobs,Allan, andDonald Appleyard. 1987. Toward an urban design
manifest. Journal of the American Planning Association 53, 1: 112-20.
Jacobs, Jane. 1961. The death and life of great American cities. New York:
Random House.
Jain, Uday. 1987. The psychological consequences of crowding. New
Delhi, India: Sage.
Jenks, Mike, Elizabeth Burton, and Katie Williams, eds. 1996a. The
compact city: A sustainable urban form? London: E & FN Spon.
. 1996b. Compact city and sustainability: An introduction. In
The compact city: A sustainable urban form? See Jenks et al. 1996a.
Jenks, Mike, Katie Williams, and Elizabeth Burton. 1996. A sustain-
able future through the compact city? Urban intensification in the
United Kingdom. Environments By Design 1, 1: 5-20.
Katz, Peter, ed. 1994. The new urbanism: Toward an architecture of com-
munity. New York: McGraw-Hill.
Lehman and Associates. 1995. Urban density study: General report. To-
ronto, Canada: Office for the Greater Toronto Area.
Loo, Chalsa, and Paul Ong. 1984. Crowding perceptions, attitudes,
andconsequencesamongtheChinese.Environment and Behavior 16,
1: 55-87.
Manshaden, Walter, and Marc de Schmidt. 1992. The northern wing
oftheRandstad: Suburbia versusthe city intheNetherlands. Nether-
lands Journal of Housing and the Built Environment 7, 2: 179-92.
MartinCounty.1994.MartinCounty comprehensive growth management
plan. Martin County, FL.
McCarthy, Dennis, and Susan Saegert. 1978. Residential density, so-
cial overload, and social withdrawal. Human Ecology 6, 3: 253-72.
Merry, Sally. 1987. Crowding, conflict and neighborhood regulation.
In Neighborhood and community environments, Irwin Altman and
Abraham Wandersman, eds. New York: Plenum.
Michelson, William. 1977. Environmental choice, human behavior and
residential satisfaction. New York: Oxford University Press.
Ministry of Housing, Spatial Planning and the Environment. 1996.
National report of the Netherlands on habitat. The Hague, the Nether-
Montgomery,Roger.1998.Is there stilllifeinThe death and life?Journal
of the American Planning Association 64, 3: 269-74.
Mullins, Patrick. 1995. Household, consumerism and metropolitan
development.In Australian cities: Issues, strategiesand policies for ur-
ban Australia in the 1990’s.See Troy 1995a.
Nasar, Jack. 1997. Neo-traditional development, auto-dependency
and sense of community. In Space design and management for place
making: EDRA 28, Maurice Amiel and Jacqueline Vischer, eds. Ed-
mond, OK: EDRA.
The Concept of Density 409
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
New York City Planning Commission. 1993. Shaping the city’s future:
New York City planning and zoning reports. New York: New York
City Planning Commission.
Newman, Peter, and Jeffrey Kenworthy. 1989. Cities and automobile de-
Orchard, Lionel. 1995. National urban policy in the 1990’s. In Austra-
liancities:Issues,strategiesand policies for urban Australiainthe1990’s.
See Troy 1995a.
Ottensmann, J. R. 1977. Urban sprawl, land values and the density of
development. Land Economics 53, 4: 389-400.
Owens, Susan. 1992. Energy, environmental sustainability and land
useplanning.InSustainable development andurbanform.SeeBreheny
Peel,Mark. 1995. Theurban debate: From“Los Angeles” tothe urban
village.InAustraliancities:Issues, strategies and policies forurbanAus-
tralia in the 1990’s.See Troy 1995a.
Pickup,Laurie.1984.Women’sgender-roleanditsinfluenceon travel
behavior. Built Environment 10, 1: 61-68.
DC: Dearborn Financial Publishing and Urban Land Institute.
. 1989. Density and urban sprawl. Land Economics 65, 3: 193-
Radberg,Johan. 1998. Ebenezer Howard’s dream: 100 yearsafter. Pa-
per presented at the 44th International Federation of Housing and
Planning World Congress, September 13-17, Lisbon, Portugal.
.1996. Towards a theory of good urban form. Paper presented
at the 14th conference of the International Association for People-
Environment Studies, July 30-August 3, Stockholm, Sweden.
Rapoport, Amos. 1997. The nature and role of neighborhoods. Urban
Design Studies 3: 93-118.
. 1977. Human aspects of urban form. Oxford: Pergamon.
. 1975. Toward a redefinition of density. Environment and Be-
havior 7, 2: 7-32.
Regional Municipality of York. 1994. York region official plan. York,
UK: Regional Municipality of York.
Reid, Fred. 1986. Real possibilities in the transit myths. In Sustainable
communities: A new design synthesis for cities, suburbs and towns, Sim
Van der Ryn and Peter Calthorpe, eds. San Francisco: Sierra Club
Roberts, Caralee. 1978. Stressful experiences in urban places: Some
implications for design. In Priorities for environmental design and re-
search:Part I, Sue Weidemann and James Anderson,eds. Washing-
ton, DC: Environment and Design Research Association.
Ruback, R. Barry, and Janak Pandey. 1992. Very hot and really
crowded: Quasi experimental investigations of Indian “tempos”.
Environment and Behavior 24, 4: 527-54.
Russell, James, and Jacalyn Snodgrass. 1987. Emotion and the envi-
ronment. In Handbook of environmental psychology.See Stokols and
Altman 1987.
opment and the implications for local planning practice. Journal of
Environmental Planning and Management 35, 1: 43-61.
Saegert,Susan. 1978. High density environments: Their personaland
socialconsequences. InHuman response to crowding, Andrew Baum
and Yakov Epstein, eds. Hillsdale, NJ: Lawrence Erlbaum.
Schmidt, Donald, R. Goldman, and Nickolaus Feimer. 1979. Percep-
tions of crowding: Predicting at the residence, neighborhood and
city levels. Environment and Behavior 11, 1: 105-30.
Scully, Vincent. 1994. The architecture of community. In The new ur-
banism: Toward an architecture of community, Peter Katz, ed. New
York: McGraw-Hill.
Self, Peter. 1997. Environmentalism and cities. Newsletter Urban Re-
search Program no. 32, Canberra: Research School of Social Sci-
ences, Australian National University.
Shannon,Gary, and EllenCromley. 1985. Settlementand density pat-
terns:Towardthe21st century. InHabitatsfor children: The impactsof
density.See Wohlwill and van Vliet 1985.
Shefer, Daniel, Shaul Amir, Amnon Frenkel, and Hubert Law-Yone.
1997.Master plan anddevelopment plan forthe Northern Region.Haifa,
Israel: Technion, Center for Urban and Regional Studies.
Shireman, William. 1992. How to use the market to reduce sprawl,
congestionand waste in our cities.In Sustainablecities: Concepts and
strategies for eco-city development.See Walter et al. 1992.
Shlay, Ann. 1985. Castles in the sky: Measuring housing and neigh-
borhood ideology. Environment and Behavior 17, 5: 593-626.
Simon, Joan, and Gerda Wekerle. 1987. Planning with scarce re-
sources:Theminiaturizationofan urban neighborhood. In Housing
andneighborhoods:Theoreticaland empirical contributions, Willemvan
Vliet, Harvey Choldin, William Michelson, and David Popenoe,
eds. Westport, CT: Greenwood.
Smyth, Joseph. 1992. The economic power of sustainable develop-
ment: Building the new American dream. In Sustainable cities: Con-
cepts and strategies for eco-city development.See Walter et al. 1992.
Stenhouse, Douglas. 1992. Energy conservation benefits of high-
densitymixed-use landdevelopment. In Sustainable cities: Concepts
and strategies for eco-city development.See Walter et al. 1992.
Stokols, Daniel. 1987. Conceptual strategies of environmental psy-
chology. In Handbook of environmental psychology.See Stokols and
Altman 1987.
. 1976. The experience of crowding in primary and secondary
environments. Environment and Behavior 8, 1: 49-86.
implications for future research. Psychological Review 79, 3: 275-78.
Stokols, Daniel, and Irwin Altman, eds. 1987. Handbook of environ-
mental psychology. Vol. 1. New York: John Wiley.
Stringer, Peter, and Andre Kremer. 1987. Environmental psychology
in the Netherlands. In Handbook of environmental psychology. Vol. 2,
Daniel Stokols and Irwin Altman, eds. New York: John Wiley.
Sundstrom, Eric. 1978. Crowding as a sequential process: Review of
researchon the effects of population density onhumans. In Human
response to crowding, Andrew Baum and Yakov Epstein, eds. Hills-
dale, NJ: Lawrence Erlbaum.
Thomas,Louise,and Will Cousins. 1996.Thecompact city: A success-
ful,desirable and achievable urbanform? In Thecompact city: Asus-
tainable urban form? See Jenks et al. 1996b.
Troy, Patrick, ed. 1996. The perils of urban consolidation. Sydney, Aus-
tralia: Federation.
. 1995a. Australian cities: Issues, strategies and policies for urban
. 1995b. Introduction. In Australian cities: Issues, strategies and
policies for urban Australia in the 1990’s.See Troy 1995a.
. 1992. The new feudalism. Urban Futures Journal 2, 2: 36-44.
van Andel, Joost. 1998. Children’s use of neighborhood public space
in relation to density. Paper presented at the 24th Congress of the
InternationalAssociationof Applied Psychology,August9-14, San
vanderCammen, Hans, RobGroeneweg,and Guus vandeHoef. 1988.
RandstadHolland:Growthandplanningof a ring of cities.InFourme-
tropolises in Western Europe, Hans van der Cammen, ed. Assen/
Maastricht, the Netherlands: Van Gorcum.
Van der Ryn, Sim. 1986. The suburban context. In Sustainable commu-
nities:A new design synthesis for cities, suburbsand towns.SeeVan der
Ryn and Calthorpe 1986b.
VanderRyn,Sim,and Peter Calthorpe. 1986a. Thenewsuburbanfab-
ric. In Sustainable communities: A new design synthesis for cities, sub-
urbs and towns.See Van der Ryn and Calthorpe 1986b.
, eds. 1986b. Sustainable communities: A new design synthesis for
cities, suburbs and towns. San Francisco: Sierra Club Books.
410 Journal of Planning Literature
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
vander Valk, Arnold,and Andreas Faludi.1992. Growth regionsand
the future of Dutch planning doctrine. In Sustainable development
and urban form.See Breheny 1992b.
vanVliet, Willem.1985. Therole ofhousing type,household density,
and neighborhood density in peer interaction and social adjust-
ment.InHabitatsforchildren:Theimpactsofdensity.SeeWohlwill and
van Vliet 1985.
VancouverCityPlanning Department. 1986.Newneighbors: How Van-
couver’s single family residents feel about higher density housing. Van-
couver, Canada: Vancouver City Planning Department.
. 1978. Housing families at high densities. Vancouver, Canada:
Vancouver City Planning Department.
Walter,Bob,LoisArkin,andRichardCrenshaw,eds. 1992. Sustainable
cities: Concepts and strategies for eco-city development. Los Angeles:
Eco-Home Media.
Wentling, J. W. 1991. Technics: Small lot housing typology. Progres-
sive Architecture 72, 6: 45-49.
Wilson, Georjeanna, and Mark Baldassare. 1996. Overall “sense of
community” in a suburban region: The effects of localism, privacy
and urbanization. Environment and Behavior 28, 1: 27-43.
Wohlwill,Joachim. 1987. Thephysical environment andthe develop-
mentof the child. In Handbook of environmental psychology.See Stok-
ols and Altman 1987.
. 1985. Residential density as a variable in child-development
research.In Habitatsfor children:The impactsof density.SeeWohlwill
and van Vliet 1985.
Wohlwill, Joachim, and Willem van Vliet, eds. 1985. Habitats for chil-
dren: The impacts of density. Hillsdale, NJ: Lawrence Erlbaum.
Wong,Aline, and StephenYeh. 1985. Housinga nation: 25Years of pub-
lic housing in Singapore. Singapore: Housing and Development
Woodhull, Joel. 1992. How alternative forms of development can re-
duce traffic congestion. In Sustainable cities: Concepts and strategies
for eco-city development.See Walter et al. 1992.
The Concept of Density 411
at ENGINEERING LIBRARY on March 24, 2009 http://jpl.sagepub.comDownloaded from
... It is well known that there is variety and complexity in the definition and meaning of density. Density is used as a metric by decision-makers from many different disciplines and professions, such as anthropology, architecture, ecology, economics, environmentbehaviour studies, planning, psychology, sociology, transportation and urban design (Churchman, 1999). Not surprisingly, there is no one accepted measure that is employed by everyone (Churchman, 1999;Forsyth, 2003). ...
... Density is used as a metric by decision-makers from many different disciplines and professions, such as anthropology, architecture, ecology, economics, environmentbehaviour studies, planning, psychology, sociology, transportation and urban design (Churchman, 1999). Not surprisingly, there is no one accepted measure that is employed by everyone (Churchman, 1999;Forsyth, 2003). In the built environment, 'density' mostly means the ratio of population and/or of built space to a given area of land. ...
... In calculations of density, for example, the numerator (the number of units) and/or denominator (usually the base land area) may differ; for example, the number of people per hectare vs. the number of dwellings in km2. Also, what is included and excluded in the calculation of some measures of density may vary; for example, the net density in one local authority may include a measure of pavements whereas another local authority may exclude it from their calculations (Churchman, 1999;DETR, 1998). ...
Over the last few decades, the emergence of various social problems within the urban neighbourhoods of cities has called for further research to consider the role of urban social sustainability. For example, the decline of face-to-face social interaction and social trust among residents, increased noise, limited mobility, and social conflicts of the housing crisis. Social life in Iraq has been changed due to transformations in both political and economic milieus, and the introduction of technologies to people's lifestyles. These have affected social values and, in turn, contributed to significant changes in the social environment, leading to a continuous reduction in social interaction. Yet, social considerations at different levels are still neglected in Iraq in urban developments. Improving social sustainability requires comprehensive analysis to identify the factors that affect social interaction among residents. Using multiple case studies, this research investigates the influence of factors relevant to social sustainability indicators (SSI), physical characteristics of the built environment (PCBE), and demographic factors (DF) on social interaction. This includes social indices, including neighbouring, social networks, and social relationships among residents in communal spaces within single-family houses neighbourhoods (SFHNs). Additionally, this research identifies the communal spaces used for regular and formal social gatherings in SFHNs in Basra, Iraq. To achieve this, primary data have been collected from three single-family housing neighbourhoods in Basra. A range of different qualitative and quantitative techniques is applied systematically. These include semi-structured interviews with experts, to determine the influential factors from a professional perspective and a residents' survey, involving users' daily life activities in communal spaces to identify the influential factors according to users. Also, socio-spatial practices, involving observation and behavioural mapping are used to understand users’ behavioural patterns and to identify the most commonly used communal spaces, and a fieldwork site survey is applied to explain the current situation concerning communal spaces. The findings demonstrate a number of factors, mostly concerning SSI (the sense of community, privacy, safety and security); PCBE (the provision and location of public utilities, open green spaces, communal spaces that are climate responsive designed, accessibility, maintenance), and DF (gender, education level, employment status and the presence of relatives living within the neighbourhood), have been found to affect social interaction and social indices within the selected case studies. The findings also demonstrate that unintentional communal spaces, such as the space in front of the main entrance of houses, accommodate most of the regular social interactions between residents, while worship facilities, such as mosques and hussainya, offer formal scheduled gatherings in the neighbourhoods. The design implications of these findings call for full consideration of these factors in the design of future sustainable housing neighbourhoods in Basra, with attention given to the design of unintentional communal spaces as actual places of contact among neighbours. This research contributes to international literature and knowledge and offers much-needed empirical evidence to inform the design of future sustainable SFHNs in Iraq. This is realised through the development of design recommendations based on empirical evidence, noting modifications to existing assumptions about the influential factors on social interaction among residents, and identifying the role of communal spaces in facilitating these interactions. It also contributes to future empirical research on social sustainability and social interaction about the effectiveness of a mixed-methods approach and the refinement of existing indicators and measures.
... Proponents of the compact city concept promote high-density (e.g., economic density, morphological density) and mixed-use developments (e.g., co-location of residential, commercial and retail uses) as the critical solutions to countervail the negative externalities of urban sprawl and to improve human development. Relevant empirical evidence indicates that compact urban form has a positive effect on a variety of outcomes, including improved accessibility, improved economic outcomes (e.g., productivity, innovation), reduced energy consumption, increased efficiency of public service delivery, and improved health, safety, subjective well-being, and social equity [9][10][11][12][13][14][15][16]. According to this evidence, we can infer that compact cities may promote human development. ...
... However, not everyone appreciates the compact city model. Opponents pointed out the negative aspects of compact cities, such as shortages in urban green spaces, failures in providing affordable housing, and overcrowding in the urban residential area [9,11,12,15,[17][18][19]. These negative aspects, in turn, suggest that compact cities may have a negative impact on human development. ...
... This could be a result of the mismatch in educational resource supply and demand generated by compact development. Scholars generally agree that families are closer to facilities in a compact city [11,12], but it is worth noting that compact city shape is associated with faster population growth [27]. The population is disproportionately concentrated in specific areas, but educational opportunities are often limited, making it difficult for people to access educational resources in compact cities. ...
Full-text available
The Human Development Index does not follow a normal distribution. For skewed distributions, finite mixture models can provide better estimates than fixed-effects models. In this paper, the relationship between compact cities and human development is investigated by employing a finite mixture model using panel data of Chinese prefecture-level cities. In contrast to the majority of the literature, which focuses exclusively on economic density, this article examines the impact of economic and morphological density on the level of human development. The results show that the compact development model has a negative impact on the level of human development and that the intensity of the impact varies for cities with different characteristics.
... The numerator (unit area) in this definition can be people, buildings, dwellings, activities, and the denominator (quantity) can be different units of land measurement such as acre, hectare, square mile, and square kilometer. As seen, density does not have a single and precise definition and measurement accepted (Churchman, 1999;Dovey & Pafka, 2014). Different assumptions in density measurements also make it impossible to draw a clear outline of what the boundaries of high and low density are. ...
... While net density, which includes only residential areas/private lands, is used as a control tool in planning, the experience of urban space is mostly associated with gross density including public spaces (Pafka, 2020). On the other hand, using the 'average density' value in density measurements may lead to ignoring the variations within the area (Churchman, 1999). The "measured density" using certain quantities may differ from the "perceived density" that results from experiencing urban space. ...
... In the era of modern urbanism, different levels of urban densities have been related to urban, environmental and economic issues including vitality, health, safety, creativity and sustainability (Dovey & Pafka, 2014;Turok & McGranahan, 2013). Stating the contextdependent feature of density, Churchman (1999) explained the potential advantages and disadvantages of high densities in terms of transportation system, physical infrastructure and urban form, environmental, economic, personal, physiological and social aspects. The role of density to create a healthy environment is ambiguous. ...
Full-text available
The present study aims to examine the relationship between urban vitality, healthy environment and density through the city of Istanbul, which is going through the Covid-19 outbreak. In this context, an online survey was conducted to measure the assessments of the residents living in districts with different density categories regarding the neighborhoods and the city they live in. The evaluations made by the citizens in the dimensions of vitality, mobility, safety, healthiness, cleanliness, orderliness were reduced to two main factors as “urban vitality” and “healthy environment” using Principal Components Analysis. Then, the evaluations regarding these six variables and two factors were subjected to cross-inquiries with the personal, residential and district characteristics. Urban residents were also asked to evaluate the city life before and after the Covid-19 outbreak. The main findings of the study reveal that there is a statistically significant difference between the density levels of the districts in terms of the perception of urban vitality and some sub-variables of healthy environment. Also, there is an observed change in the thoughts about urban life in Istanbul due to the outbreak.
... The compact city has been proven to play a role in promoting sustainable development, such as improved productivity [7], reduced pollution [8][9][10][11][12][13] and for having smaller ecological footprints [14] and better city health [15,16]. However, compact cities have also had some negative effects, such as a lack of urban green space, overcrowding, and social withdrawal [9,11,12,17,18]. ...
... The compact city has been proven to play a role in promoting sustainable development, such as improved productivity [7], reduced pollution [8][9][10][11][12][13] and for having smaller ecological footprints [14] and better city health [15,16]. However, compact cities have also had some negative effects, such as a lack of urban green space, overcrowding, and social withdrawal [9,11,12,17,18]. We discovered that these studies focus mainly on the impact of compactness on the quality of life of the group as a whole, ignoring the impact on specific groups, particularly the urban poor. ...
Full-text available
City shape is an essential reflection of spatial structure, but it has largely been ignored in urban form research. This study employs night-time satellite imagery to depict the scope of urban economic activity to investigate its impact on urban poverty. It is the first study to provide a comprehensive assessment of the mechanisms of city shape on urban poverty by using the fixed-effect estimate methodology for panel data of 285 Chinese cities from 2000 to 2018. The results showed that city compactness has an inverted U-shaped relationship with poverty incidence, which was verified by several robustness tests. Compactness can significantly attract more population into the city, and space costs and commuting costs are important influence channels. Furthermore, there exists heterogeneous nexus between city shape and urban poverty. Compactness has more significant poverty reduction effects in low-attractive cities with low productivity, low wages, and high illiteracy rates.
... Building height data is important for understanding population distribution (Balakrishnan, 2020;Xie, 2006) and energy use within a city (Resch et al., 2016). Information about the third dimension can also help formulate effective urban planning regulations since a major concern of most urban planning regulations is managing the intensity of the land use (Alexander et al., 1988;Churchman, 1999). Understanding this land use intensity at the scale of cities requires detailed building height data, for example, to quantify the total residential built-up space available per hectare or sq. ...
This paper describes an open-source method for generating megacity-scale building height maps without proprietary software or Differential Global Navigation Satellite System surveyed ground control points (GCPs). We use the open-source Satellite Stereo Pipeline (S2P) software along with four scenes of 2.5m resolution Cartosat-1 data for Bengaluru to demonstrate this. Digital Surface Models (DSMs) of 5 m resolution are generated using S2P, and terrain removal is achieved using 30m SRTM data resampled to 5m. The resulting normalized DSM is calibrated and validated using 1270 GCPs. These were generated by counting the number of occupiable floors of buildings and marking zero elevation ground points. The final building height map of Bengaluru covers a total area of about 1420 km ² , and across the four scenes, has RMSE values ranging from 2.8m to 3.9m—an error of approximately one floor. Furthermore, we implemented this workflow using stereo imagery for Mumbai, and the RMSE values obtained were comparable to those for Bengaluru. Hence the method we describe is a very cost-effective way of generating megacity-scale building height maps. The height maps generated using this method can be used to better understand numerous urban characteristics including land use intensity and population distribution and can play a crucial role in urban planning and policy making.
... It is a key concept in urban planning and building design for managing sustainability in the long term because it helps to describe, predict, and control land uses, urban morphology, and soil consumption, the nonrenewable resource par excellence. Density proves to be a promising concept for these types of analyses because it is objective, quantifiable, and neutral; neutral because it does not involve an a priori judgment, and as a ratio, it does not entail a positive or negative meaning [61]. Nevertheless, density represents a complex concept [62] since it is context-related and defined and applied differently for different sectors, disciplines, and countries [60,63]. ...
Full-text available
In the last decades, a tendency towards urban tissue densification has been observed to counteract the urban sprawl. Densification may be achieved through more compact built areas, preferring the vertical to the horizontal development of buildings but avoiding bulky high-rise building blocks. This strategy significantly affects several aspects of the microclimate and produces direct and indirect effects on human health and well-being. In this regard, air pollution and heat stress constitute two increasing threats to human health and well-being that need to be faced immediately. The involved phenomena are various, intertwined, and may lead to conflicting results. Hence, regenerating existing, well-structured, and stratified urban areas by densification is not an easy challenge. Urban ventilation may favor the mitigation of detrimental effects of air pollution and heat stress on human life. Therefore, a multidisciplinary methodology is presented for embedding urban ventilation performance evaluation into urban management and planning processes. The scope is to propose a framework for urban renewal plans that is citizens-centered and aims at improving their health and well-being in existing urban areas. The methodology builds upon the performance-based approach and is supported by the conceptual framework and the literature reviews provided through the paper.
Full-text available
Research shows that stress, a common problem in densely populated cities, can be relieved by exposure to the natural environment. As great significance has been attached to the relationship between the urban environment and public health, this paper aims to study the relationship and interaction between the perceived sensory dimensions of urban park green space, attention restoration , and state empathy. Therefore, we conducted an on-site questionnaire (PSD Scale) survey in four typical parks in Chengdu and recorded age, sex, daily stress, frequency of visits to parks, and other basic information from the respondents. In the survey and visit, we found that the group structure in the recreation area comprises chiefly of a few transient unfamiliar travelers and most long-haul neighborhood sightseers. Among long-haul vacationers, the greater part of them are moderately aged and older individuals in the encompassing local locations, whose lives are, for the most part, quick and proficient. Hence, to mirror the populace attributes of high-thickness metropolitan parks and to feature the agent bunches that have lived in the parks from here onward, indefinitely quite a while, the chosen bunches are somewhere in the range of 35 and 65 years old (half male and half female), so make sure there are no ailments, no drinking, and no late evenings in the earlier days, so a specific actual fundamental belief is kept up with. The main part of the exercise focused on the perceived dimension, state empathy, and attention restoration. The software SPSS24.0 was applied to the test of the validity and reliability of the perceived sensory dimension (PSD) Scale, and then the important correlation between the perceived sensory dimensions in the parks and vis-itors' attention restoration was analyzed through multiple linear regression. Finally, the moderating effect of state empathy was tested by PROCESS. The findings show that (1) only seven dimensions in the PSD Scale are effective; (2) Serene and Refuge in the perceived sensory dimensions have a significant effect on the restorative components of attention. (3) Except for the dimensions of Rich in Species and Refuge, empathy enhanced the moderation effect in the interaction between the other five dimensions of the Perceived Restorative Scale (PRS), especially in the interaction between the Social and PRS dimensions. However, this topic needs to be further explored to provide a scientific basis and design strategy for research on the restoration potential of urban park green space in high-density urban areas.
This chapter outlines the concept of compact city in a specific local context, which is the research viewpoint and conceptual framework of the whole book. This chapter begins with the discussion of sustainable urban form, and further draws forth the concept of compact city. Based on a summary of previous definitions of the compact city, in combination with the consideration of special urban conditions in the rapid-growing inland cities, the main principles of compact city in this book are defined. Moreover, policy measurements and state strategies that are the potential tools to achieve the compact city are reviewed.
Full-text available
A post-occupancy evaluation of a new urban neighbourhood developed by the city of Toronto. Living spaces and open spaces were miniaturized to meet goals of fitting into existing streetscapes and respond to austerity.
Full-text available
This book is about an art in which the Netherlands excels: strategic planning. Foreign observers will need little convincing of the merits of Dutch planning. They will want to know whether routine explanations (small country, industrious, disciplined people hardened by the perennial fight against the sea) hold any water, and they will want to know where to look for the bag of tricks of Dutch planners. Dutch readers need to be convinced first that planning in the Netherlands is indeed effective before contemplating how this has come about. Our message for both is that, to the extent that Dutch planners do live in what others are inclined to see as a planners' paradise, it is a paradise carefully constructed and maintained by the planners themselves. This smacks of Bernard Shaw describing a profession as a conspiracy against laity. However, all knowledge and all technologies are 'socially constructed', meaning that they are the products of people or groups pursuing often conflicting aims and coming to arrangements about what is to pass as 'true' and 'good'. So this takes away the odium of Dutch planners having their own agenda. Positioning ourselves We are in the business of interpreting Dutch planning, and at the same time committed to improving it. This makes us part of the situation which we describe. This situation is characterized by the existence of two divergent traditions, urban design and the social-science discipline called 'planologie'.
The preference for low density development patterns and single family housing has been well documented in the US. Currently urban policy assumes however that such patterns have negative fiscal, environmental, and quality of life effects. As an alternative, it has been proposed that development should occur more compactly and at higher densities with public park space being made available to substitute for large lot size. Using a survey of Florida residents, this paper first finds that they prefer low density housing. The paper then explores the willingness by consumers to trade-off lot size for more access to park space. A small clientele does exist willing to make such a trade-off. -Authors