Measuring The Sustainabili Ty of Cities An Analysis of The Use of Local Indicators
Measuring The Sustainabili Ty of Cities An Analysis of The Use of Local Indicators
Measuring The Sustainabili Ty of Cities An Analysis of The Use of Local Indicators
Ecological Indicators
journal homepage: www.elsevier.com/locate/ecolind
A R T I C L E I N F O A B S T R A C T
Article history: We analyze 17 studies of the use of urban sustainable development indicators (SDI) in developed
Received 12 February 2009 western countries. The analysis reveals a lack of consensus not only on the conceptual framework and
Received in revised form 10 July 2009 the approach favored, but also on the selection and optimal number of indicators. First, by performing
Accepted 18 July 2009
different classifications and categorizations of SDI we identify problems inherent in territorial practices
that use SDI. Second, we argue that the lack of consensus in several steps of the creation of SDI stems
Keywords: notably from the ambiguity in the definitions of sustainable development, objectives for the use of such
Cities
indicators, the selection method and the accessibility of qualitative and quantitative data. Third, based
Indicators
on the reviewed studies, we propose a selection strategy for SDI through which we demonstrate the need
Sustainable development
Environment to adopt a parsimonious list of SDI covering the sustainable development components and their
Local governance constituent categories as broadly as possible while minimizing the number of indicators retained.
ß 2009 Elsevier Ltd. All rights reserved.
1470-160X/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ecolind.2009.07.013
408 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
2.2.1. Aggregation
Aggregation can be spatial (e.g. progression from a regional
spatial scale to a provincial scale), temporal (e.g. from a monthly
interval to an annual interval) and thematic. The first two
aggregation modes apply to a given variable that is compared over
time or between territories with the same administrative status. In
contrast, thematic aggregation groups a set of indicators according
Fig. 1. Classic dimensions of sustainable development.
to different themes, and is used to produce indices. The aggregation
method of indicators selected determines the conceptual framework
used. This framework refers to a method or set of organization
In Section 2 we describe the indicators and indices of methods intended to design and structure a vision of sustainable
sustainable development, along with the related basic concepts. development (e.g. organization into three dimensions of sustainable
In Section 3, we present the 17 studies from which we extracted development: environment, economy and society).
the 188 SDI and interpret the classifications and categorizations of
these indicators. In the following section, we perform a more 2.2.2. Weighting
detailed empirical analysis of the SDI compiled and discuss the Weighting may entail attributing a greater value or contribu-
reasons for the lack of consensus we observe. In Section 5, we tion to one indicator or index than another. This approach has
develop a strategy for selecting SDI based on a review of previous drawn much criticism (Perret, 2002; Martinez-Alier et al., 1998)
studies, underlining the need to adopt a parsimonious list of SDI because it is an arbitrary process and no weighting structure can
with a broad coverage. The conclusion follows. rationally justify the attribution of a given weight (e.g greater
weight) to a given indicator. Nonetheless, Callon et al. (2001)
2. Indicators and indices: basic concepts suggests that any attempt to weight sustainable development
indicators take public participation in decision-making processes
This study investigates indicators of sustainable development into account.
in general rather than problems related to their aggregation,
weighting, etc. However, for simplification purposes, before 2.2.3. Threshold, critical value, target value, relative performance
beginning the analysis of the studies inventoried, we present The concept of threshold refers to a scientifically demonstrated
some basic concepts related to sustainable development measures reference value which causes the phenomenon described to
(e.g. indicators). First we differentiate indicators from indices. We change status and present notable discontinuities and structural
then briefly discuss important concepts that underlie the changes or irreversibilities. It applies mainly to environmental
construction of indicators and indices: aggregation, weighting, challenges (e.g. water pollution, soil contamination and air
thresholds and target values. pollution).
The critical value corresponds to a recognized, generally
2.1. Indicators vs indices arbitrary, reference value derived from standards. It takes into
account uncertainties and individual and/or group interests. For
2.1.1. Indicators example, the figure of 30% of income allotted to housing is often
First, it is important to clarify the nuance between indicators considered a critical value to determine whether a household falls
and data or variables observed. A datum or variable observed below the poverty level.
becomes an indicator only once its role in the evaluation of a The concept of target value allows objectives to be set while
phenomenon has been established. For example, the number of measuring efforts and associated costs (economic, social and
unemployed is a datum or key variable in economics. Once it is political).
determined that an increase in the number of unemployed The relative performance of a given territory refers to the
expresses negative economic performance for a given territory, reference values based on other territories of the same stature.
this number becomes an indicator. The datum or variable These benchmarks are consequently used to construct statistical
pertaining to the number of unemployed can then be standardized, distances relative to the best and the worst performers. Relative
transformed or formulated in any way (e.g. rate, level) that can performance is generally used in cases where indicators and
account for economic performance. indices have no scientifically established thresholds or consensual
critical values. Thresholds are then determined based on the
2.1.2. Indices experience of other territories.
An index (or composite indicator) is a synthesis of indicators. As mentioned above, in this study we concentrate on the
Formalization of an indicator that necessitates the aggregation of indicators themselves rather than on questions related to their
several data or variables results in an index. The use of indices in aggregation, weighting etc. The studies retained therefore serve to
the field of sustainable development facilitates the understanding demonstrate how indicators are chosen based on a broader
G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418 409
Table 1
Summary of the 17 studies.
1 Ambiente Italia Research 42 municipalities, EUR (13 with System of indicators based on a limited 10
Institute (2003) population over 350k; 18 with number of themes, resulting from a lengthy
population 100k to 350k; 11 with analysis on the European scale. Record for
population less than 100k). each indicator retained.
2 Corporate Knights (2007) 24 cities, CAN (population over 100k). Comparison of cities according to six 35
weighted categories of indicators.
3 Federation of Canadian 20 cities and regional municipalities, Comparison according to indicators sorted 72
Municipalities (2004) CAN (population over 100k). along 11 dimensions. Comparison over time.
4 Tomalty (2007) 27 municipalities in Ontario, CAN Comparison of municipalities according to 33
(nine with population over 400k; indicators that meet determined objectives,
nine with population 75k to 400k; associated with three main dimensions of
nine with population less than 75k). sustainable development.
5 SustainLane Report (2007) 50 cities, US (50 largest cities in US). Comparison of cities according to 15 15
weighted categories of indicators.
6 Jacksonville Community City of Jacksonville, US (Metropolitan Selection of indicators based on 15 criteria, 86
Council (2004) area population 1.3m) meets nine objectives.
7 Cowley et al. (2007) 20 cities, UK (population over 300k). Selection of indicators based on three objectives 13
that reflect the sustainability of a city.
8 Koller (2006) 31 cities, SWZ (population over 20k). Standardized indicators transformed into 30
deciles: cities are assigned a grade of between
1 and 10 by indicators. Results presented
according to a dashboard of cities.
9 Meier and Wachter (2005) 14 cities and 8 cantons, SW (cantons’ Use of five criteria in the choice of indicators. 35
population 500k to 1m.; cities’ Record for each indicator retained.
population 10k to 100k).
10 Fraser Basin Council (2000) Fraser Bassin, BC, CAN (region’s Choice of indicators based on 8 criteria 40
population of 2.77m). corresponding to four objectives.
11 Sustainable Calgary (2004) Calgary, Alberta, CAN (metropolitan Choice of indicators based on eight criteria. 36
area population, 1.6m). Discussion groups were formed to determine the
indicators that best correspond to their territory.
12 Boston Foundation and Boston, US (Metropolitan area Identification of several dimensions whose 28
Greater Boston’s Civil population, 4.4m) issues are part of a global vision of sustainable
Community (2007) development
13 Brazzini-Mourier (2006) Onex, SWZ (population 20k). Decision-making tool. Record for each 39
indicator retained.
14 Planque and Lazzeri (2006) 34 communes in Communauté du Summary of several experiences with 74
Pays d’Aix, FR (one with population territorial indicators, methodology and
of 135k; four with population over construction of a benchmark. Detailed record
10k; 29 with pop less than 10k) produced for each indicator.
15 Agence Régionale pour 67 cities, Midi-Pyrénées region, FR Record for each indicator retained. 27
l’Environnement (2001) (population over 5k).
16 ARCOLATINO.org (2004) Cities in Arco Latino9 Record for each indicator retained. 21
(population 200k to 400k).
17 Thomas (2003) n/a Compilation of indicators from OECD, Eurostat, 22
United Nations Commission for Sustainable
Development, European Structural Indicators.
Only 22 indicators out of several hundred are
common to these large entities. Analytical and
comparative study.
9
Arco Latino is an association of regional and departmental communities corresponding to islands, coastal and inland regions in the Western Mediterranean (Portugal,
Spain, France and Italy).
analysis, e.g. based on an index. In the following section we present have reflected a lack of consensus from the point of view of goals
the studies analyzed. and evaluation tools that justify the analysis of a potential
consensual grid of SDI. Finally, Campbell (1996) recommended
3. Description of surveyed studies working at the city level because, at this scale, it is easier to
determine the responsibilities of all parties involved, the actions
We initially compiled a sample of 23 studies that apply are easier to control, while parties’ contrasts are clearer or more
indicators related to sustainable development to one or more cities clear-cut.
or urban centers covering a broad array of western countries, Of the 23 initial studies, three were specific to the field of
provinces and states. We subsequently decided to limit our transport, whereas three others had a limited range of indicators
analysis to studies specifically covering urban indicators of intended strictly to measure quality of life.2 The final sample
sustainable development for four main reasons. therefore consisted of 17 studies, covering cities or municipalities
First, we believe that cities should be the object of closer of various sizes (mainly larger ones) in the US, Canada and Europe.
scrutiny since the majority of developed economies’ population The main characteristics of the studies are described in Table 1,
live in cities and that because of that they are exposed or are the which yields two main findings. First, the 17 studies use between
source of many SD related problems such as pollution, crime,
housing etc.. . . and this is particularly true in western countries 2
In fact, five studies addressed the concept of quality of life. However, only two of
(Thomas, 2002). Secondly, for some SDI considered, data was these studies used a sufficiently broad range of indicators to be considered as SDI.
available only for urban agglomerations of more than 5000 people. We therefore retained these two studies for this exercise, given our goal of
Third, mostly larger cities have adopted SD strategies or plans that comparing studies with similar objectives.
410 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
Table 2
Summary of frequency of use of indicators in 17 studies.
2005), the objectives may also represent a break point in the use of development. Even if such practices contribute to evaluating
SDI by playing a determining role in the approach retained. If most strengths, weaknesses and forecasts as part of an implementation
studies aim to integrate social, economic and environmental strategy of sustainable development, they are admittedly not
dimensions in the form of SDI in a general framework, others immune to political excesses of municipal administrations that
attempt to clarify one or more particular aspects by setting specific define ‘‘their’’ objectives which ‘‘their’’ SDI must meet.
objectives or priorities (Tomalty, 2007; Jacksonville Community Therefore, we retain three important conditions that allow a
Council, 2004). In some cases, the objectives may even emerge less superficial approach to the design of SDI. First, sustainable
from the identification of a particular challenge as a priority in a development should not be systematically redefined when
sustainable development perspective. For example, the SMART adapted to a particular territorial context. Moreover, the inter-
Transportation Ranking Report (Nemetz, 2007) strictly concerns pretations of SD should be consistent with the Brundtland report
transport and considers transport challenges to underpin problems by reflecting the economic, social and environmental components
related to sustainable development in large Canadian cities. as well as the overlaps between them. Second, the objectives
Another example is the application of indicators in the field of regarding the origin and formulation of an SDI chart should reflect
Sustainable Housing (Winston and Eastaway, 2008), which con- the existing tradeoff between consideration of specific factors for
siders the challenges related to housing as a priority in sustainable each city or municipality and the need to homogenize the
indicators such that they allow a fairer comparison between
municipal jurisdictions of the same stature. Concomitantly, these
objectives must take into account the target audience and the
ultimate use of the indicators as tools that support municipal
sustainable development strategies.
We therefore believe that an effective approach would consist
in identifying the integrated dimensions of sustainable develop-
ment as broadly as possible, while ensuring that possible over-
lapping between these dimensions are clearly specified. In
addition, the statement of objectives that SDI must meet should
entail a social construction process that can be conditioned by
social, historical, economic or local factors (Duchêne et al., 2002)
without impeding the homogenization of measures that would
allow comparison of territories of the same stature. This aspect
deserves serious consideration, mainly to deter different territorial
jurisdictions from creating ‘‘their own’’ indicators to meet
objectives that they define. In the end, we emphasize compar-
ability vs specific objectives for three main reasons. First,
sustainable development policies and strategies are often estab-
lished at the regional level (e.g. province). Therefore common
indicators facilitate the monitoring of realized progress for the
Fig. 3. Classification of 188 indicators counted. individual cities and/or for the regions themselves. In the end, the
412 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
Table 3
Indicators used four times or more in the 17 studies.
7 2. Users of mass transit (MT) Percentage of users of MT vs other means Sustainable Transport
of transport
3. Density of urban population Density of population per square kilometre Sustainable Demographics
in urban areas
4. Quantity of waste Quantity of household waste sent to landfills Livable Waste
(tonnage per year)
5. Crime rate Crime rate per 100,000 inhabitants Equitable Security
6. Mean or median household income per year Mean/median household income ($ per year) Economic Income and
expenses
7. Job creation for all sectors combined Number of new jobs per year for all sectors Economic Employment
6 8. CItizen participation in public affairs Rate of participation in public hearings Social Governance
9. Low income households Households with income below $10,000 Equitable Income and
per year expenses
10. Concentration of PM10 particles Average annual concentration in designated Livable Air
residential zones
11. Businesses with environmental certification Percentage of businesses with environmental Viable Businesses
certification
12. Quantity of waste recycled Quantity of waste recycled in tonnes/ Viable Waste
inhabitant/year
13. Daily water consumption per person Average daily water consumption per person Sustainable Water
per year
4 24. Average distance traveled per capita for all Average distance in km/per capita/yr traveled Sustainable Transport
means of transport combined for all means of transport combined
25. Victims of traffic accidents Rate of death and injuries caused by traffic Livable Transport
accidents per 1000 inhabitants
26. Green space per 1000 inhabitants Percentage of urban space dedicated to green Livable Ecosystem,
space per 1000 inhabitants heritage,
green space
27. Playgrounds (parks) per 1000 inhabitants Percentage of urban space dedicated to Livable Social and
playgrounds per 1000 inhabitants community services
28. Cultural events Annual number of cultural events Social Social and
community services
29. Average capacity of primary and secondary Average number of students per class at Equitable Education
school classes primary and secondary schools
30. Diversity of new housing built Proportion of housing starts for each Equitable Housing conditions
type of dwelling
31. Premature mortality rate Ratio of deaths among people under Sustainable Demographics
age 75 per 100,000 inhabitants
32. Ecological footprint Ecological footprint of municipality Sustainable Other
use of common SDI should lead to a better coordination of actions comparability and the adoption of common SDI is necessary to
for cities within a given region (Thomas, 2002). Second, compar- avoid that SDI become marketing tools for cities tempted by
able SDI are important because they allow cities of the same size to choosing indicators that make them look good only. We will take
have a common grid to share and apply successful tools and these facts into account to develop our approach to the creation of
measures (Ambiente Italia Research Institute, 2003). Third, SDI in the next section of this article.
G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418 413
Fig. 5. The relationship between data condensation and audience (Shields et al.,
2002).
4.2. Variation in the choice and number of indicators 5. SDI selection strategy
Even when a similar approach is applied, we have observed that In this section we establish a strategy for choosing a list of
two studies or more may use indicators that vary considerably in pertinent indicators. This strategy is based on: (1) the definition of
number and attributes. As mentioned above, in the 17 studies basic selection criteria derived from the tradeoff between parsimony
examined, the number of SDI varies between 10 and 86, and the and coverage of pertinent categories, and (2) the classifications and
number of indicators related to the environmental component sorts that we previously analyzed in Section 1. Further, we consider
exceeds that of the other components, thus implying a lesser the problems identified in Sections 3 and 4 when justifying our
consensus. approach and validating our recommendations.
We generally attributed these variations to the classification
process that is the way the indicators are structured or organized in 5.1. Selection criteria
a given system, as well as to access constraints. Accordingly, we
differentiate divergence between practices generally suggested by Determining the optimal choice and number of indicators
scientists and those endorsed by municipal administrations. If the inevitably requires that selection criteria be defined. We evaluated
latter tend to favor a less conceptual structure comprising fewer the number of selection criteria used in the 17 studies. In total,
indicators, intended to achieve simple and quantifiable objectives, more than half of the studies use fewer than three criteria, whereas
scientists prefer a minimum of aggregation and, if possible, one study identifies up to 14 selection criteria. Of a total of 68
simplification, in order to be faithful to the concepts. This contrast criteria noted, only six are frequently used. The criteria are found
is illustrated in Fig. 5. under the following headings: ‘‘credible,’’ ‘‘universality,’’ ‘‘data
414 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
requirements and availability,’’ ‘‘comprehensible,’’ ‘‘links with actions, changes in behavior, etc.) to the problems encountered. In
management’’ and ‘‘spatial and temporal scales of applicability.’’4 the DSR model, anthropic pressure on the environment is simply
In addition, a growing number of authors recognize the replaced by a driving force to better integrate social, economic and
importance of defining transparent and rigorous selection criteria institutional indicators. In the DPSIR model, driving forces behind
to increase the value and scientific credibility of indicators selected the use of transport or industry, for example, exert pressure on the
while ensuring that they can inform decision-making processes environment such as emission of pollutants, which in turn degrade
(Belnap, 1998; Slocombe, 1998; Dale and Beyler, 2001; Niemeijer the state of the environment. This has an impact on human health
and De Groot, 2008). A particularly widespread approach in social and the ecosystem, which triggers a reaction by society in the form
sciences, perhaps owing to the simplicity of its acronym, is worth of political measures that can apply to each component of the
mentioning. An indicator must be SMART, namely: model.
However, for such models to be effective, the number of
‘‘Specific – be clear and concise and avoid vague terms; indicators must be multiplied by three, four or even five. Each
Measurable – quantifiable indicators to measure progress; category of indicator (transport, energy, education, health, air, etc.)
Achievable (Assignable) – someone must be able to complete must be broken down into indicators of pressure, state and
the objective; Relevant (Realistic) – able to be interpreted; response in the case of a PSR model, for example. It is partly for this
within budget and time frame; Time-related – completed by a reason that the United Nations, which pioneered this type of
certain date; change measured by a certain date.’’ (United approach in 1990, abandoned it in 2006. In addition, it does not
Nations Statistical Institute for Asia and Pacific, 2007). allow attainment of the objectives for which the indicators were
Even if the application of such an approach remains subjective, created (Bell and Morse, 2008). Lastly, there is no indication that
as Dale and Beyler (2001) note, it can be used to validate the choice increasing the number of indicators allows one to better capture
and number of indicators. However, it should be considered as a the relative sustainability for one or more territories. On the
complement to the adoption of clear objectives or the development contrary, as we mentioned in Section 3 of this article, the optimal
of general principles. For example, the Renewed EU Sustainable number of indicators should consider the tradeoff between broad
Development Strategy rests on seven objectives and illustrates the coverage of the sustainable development components and the goal
holistic nature of the concept of sustainable development of parsimony of indicators.
perfectly. Note that it applies selection criteria to supplement
the objectives (on this subject, see Jones and Patterson, 2007). 5.3. Survey-based selection strategy for SDI (SuBSelec)
Similarly, the International Institute for Sustainable Development
developed ten principles based on Agenda 21, and specified a priori In this section, we propose an optimal selection strategy of a
selection criteria (see Hardi and Zdan, 1997). subset of indicators from among the 188 indicators compiled. We
then determine a general criterion of optimization along with the
5.2. Conceptual framework and classification method of indicators pertinent constraints. To do so, it is important to set clear
fundamental guidelines for compiling the list of indicators.
Whereas a selection criteria chart can be used to produce a Evidently, the SDI retained should collectively cover the four
refined list of indicators, it is not sufficient to produce a structure or components and 20 categories of sustainable development
classification model justified by previously established objectives presented in Table 2. Whereas we could conceivably select all
and by the processes and states to be measured. 188 indicators, we consider this a simplistic and inefficient
The use of a conceptual framework is thus a fundamental step strategy given redundancy factors and the loss of efficiency in
that can ensure that the indicators selected indeed cover the the public dissemination of information and results.5 A more
phenomena to be measured (Niemeijer and De Groot, 2008). There targeted strategy would consist in retaining one indicator per each
are no specific models that allow one to simplify or predict of the 20 categories defined. We could thus cover all components of
interactions that govern sustainable development (Abolina and sustainable development and all subcategories. Nonetheless,
Zilans, 2002). Further, there are many possible approaches to although this approach would reduce the number of indicators,
designing and organizing indicators, each of which has advantages it would ignore asymmetry in the importance and complexity of
and disadvantages (Hart, 1998). these components and categories. For example, Table 2 reveals that
The main conceptual frameworks are well documented by the income and expenses category tends to be particularly well
McLaren (1996). They include frameworks centered on objectives, represented in the studies. This is not surprising, and seems
challenges, sectors, components of sustainable development and desirable given that it is useful to consider not only average income
cause and effect relationships; the latter are also known as but also variance in the distribution of income (e.g. via percentages
‘‘pressure-status-response’’ models (PSR). McLaren (1996) con- of low income). Thus it is worth representing all components and
tends that although these approaches can meet the objectives for categories, although more weight should be given to certain items.
which they were designed, only the combination of two or more of Once this initial analysis is accepted, three central elements
these approaches can capture the complexity of sustainable emerge in the establishment of an optimal selection criterion for
development. SDI. First, we are seeking a parsimonious list of SDI that is easily
The PSR models and their derivatives (DPSIR: driving force- understandable and usable. This would allow us to simplify and
pressure-state-impact-reaction; DSR: driving force-state-reaction) synthesize the information to facilitate the dissemination of
are particularly useful. Such models combine different properties analyses. In addition, we aim to cover all the components and
attributed to indicators. Niemeijer and De Groot (2008) and categories of sustainable development established in Table 2.
Barcelo (1999) propose definitions of these properties. Accord- Lastly, we endeavor to retain SDI on which a consensus exists in the
ingly, the PSR model serves to group a series of indicators to literature. This point is crucial to establish solid conceptual bases in
represent anthropic pressure on the environment, the resulting the development of indicators and indices. Ultimately, we are
state of the environment and the reactions or responses (political, attempting to achieve a consensus to facilitate comparison
4 5
To standardize the selection criteria used in the 17 studies, we adopted the The latter element is important according to several lists of selection criteria,
classification developed by Niemeijer and De Groot (2008), using the categories the best known of which is that of the 10 Bellagio principles. See Bell and Morse
whose definition corresponds to those found in the studies we analyzed. (2008).
G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418 415
Table 4
The 29 indicators retained.
between studies and cities over time6. Given these objectives, we from Table 3: 32 indicators are used four times or more, 23 are used
have adopted the following general selection criterion: five times or more and 13 appear six times or more. Based on our
Survey-based selection strategy for SDI (SuBSelec): a selection of other selection constraints, it is clear that the 13 SDI used six or
SDI based on a review of studies intended to minimize the number more times do not sufficiently cover the components and
of indicators selected from among those compiled, subject to the categories presented in Table 2. For instance, these SDI do not
following constraints: (1) choose the most cited indicators; (2) pertain to essential categories such as education and housing. By
cover the components of sustainable development and the contrast, the 32 SDI used four times or more cover all of the
pertinent predetermined categories and (3) choose the simplest components and categories. Thus, given the SuBSelec criteria
SDI to facilitate data collection, understanding and dissemination. defined above, to produce the list, indicators will be chosen from
among those cited four times rather than those cited five times or
5.4. Application of the strategy more. In this case we will therefore choose between two typical
situations found in studies using SDI. The first is the situation
Given the classifications and sorts performed above, the where a ‘‘high’’ number of SDI are used, to allow greater coverage of
starting point of the selection process is to determine the most components and categories of sustainable development. This
commonly used indicators. Three ‘‘characteristics’’ thus emerge situation corresponds in this case to the 32 indicators cited four
times. In the second situation, greater parsimony is favored at the
expense of reduced coverage of components and subcategories of
6
We are aware that the search for a consensus may create certain gaps and errors sustainable development. In this paper we refer to cases where 23
in the selection of indicators. We are not attempting to criticize or develop indicators are cited five or more times.7 We thus aim to determine
indicators, but rather to select a list that would be parsimonious and representative
of knowledge at a given stage. However, we believe that the selection of indicators
7
should be relatively consistent to allow a comparison over time. We will revisit This type of situation is summarized well in the introduction of the work by Bell
temporal and spatial importance below. and Morse (2008).
416 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
Livable 20 5.8 5
8 Equitable 27.7 8 7
Aside from the fact that some find attempts to measure sustainable
Viable 11.2 3.2 4
development illusory, the inherent unavoidable subjectivity is one of the main
Sustainable 21.3 6.2 7
criticisms of the choice and use of SDI.
G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418 417
similar coverage to that obtained with the 188 compiled from the Arcolatino, 2004. 21 indicateurs pour un développement durable de l’Arc Latin.
Communication présentée au module de Développement Sustainable de l’École
17 studies examined. Doctorale Biologie des Systèmes Intégrés, Agriculture, Environnement. Mon-
tpellier, France: Agropolis International Agropolis. Novembre.
6. Conclusion Barcelo, M., 1999. Les indicateurs d’étalement urbain et de développement durable
en milieu métropolitain. Cahier 99-06 Montréal: Observatoire métropolitain de
la région de Montréal.
From concept to local practice, ongoing debates increasingly Basiago, A.D., 1999. Economic social and environmental sustainability in devel-
underline the need to measure sustainable development SD based opment theory and urban planning practice. The Environmentalist 19,
on indicators or evaluation criteria. One of the most common 145–161.
Bell, S., Morse, S., 2008. Sustainability Indicators: Measuring the Immeasurable?
applications consists in comparing municipalities, notably to second ed. Earthscan, London.
support local decision-making processes. Belnap, J., 1998. Environmental auditing: choosing indicators of natural resource
Problems related to the use and application of SDI primarily condition: a case study in Arches National Park, Utah. Environmental Manage-
ment 22, 635–642.
stem from (1) an overly broad definition of sustainable develop- Boston Foundation and Greater Boston’s Civil Community, 2007. A Summary of the
ment that gives rise to multiple interpretations, (2) the absence of Boston Indicators Report: 2004–2006. Boston: The Boston Foundation.
standard and universal classification methods or approaches to Brazzini-Mourier, S., 2006. Indicateurs communaux intégrés. Rapport de synthèse.
Onex, Genève: Ville d’Onex.
designing SDI, particularly at the municipal level, and (3)
Callon, M., Lascoumes, P., Barthe, Y., 2001. Agir dans un monde incertain. Essai sur la
constraints caused by the accessibility of data that preclude their démocratie technique. Éditions Seuil, Paris.
quantification and the specific qualification of indicators. These Camagni, R., 2002. On the concept of territorial competitiveness: sound or
misleading? Urban Studies 39 (13), 2395–2411.
problems engender considerable diversity in the conceptual
Campbell, S., 1996. Green cities, growing cities, just cities? Urban planning and the
frameworks, choices and optimal number of SDI across the 17 contradictions of sustainable development. Journal of the American Planning
studies examined. Association 62.
In this article, we address these problems by applying a method Corporate Knights, 2007. Canada’s Most Sustainable Cities. Special Report of March
6, 2007. Toronto: Corporate Knights.
based strictly on the empirical data found in the 17 studies that Cowley, C., Madden, P., Quinn, T., Winter, T., 2007. The sustainable cities index:
examined SDI. Accordingly, the main rationale of our approach is to ranking the largest 20 British cities. Forum for the Future Organisation, London,
find a method of selecting indicators that would allow the broadest UK.
Dale, V.H., Beyeler, S.C., 2001. Challenges in the development and use of ecological
possible coverage of the integrated components of sustainable indicators. Ecological Indicators 1, 3–10.
development and the categories that comprise them, while Duchêne, F., Journel, C.M., Coanus, T., 2002. The forgotten construction process of
minimizing the number of indicators retained. indicator sets: the social dimension of air pollution measurement in the Greater
Lyon Area (France). Ecological Indicators 2, 187–196.
This method, which we call SuBSelec, is a strategy that subjects Federation of Canadian Municipalities, 2004. Quality of Life in Canadian Commu-
the 188 indicators extracted from 17 studies to three parameters to nities. Ottawa: FCM.
reduce the number of indicators to an optimal level. This optimal Fraser Basin Council, 2000. Sustainability Indicators for the Fraser Basin. Fraser
Basin. BC: The Fraser Basin Council.
level thus results from the union between the most frequently used Hardi, P., Zdan, T., 1997. Assessing Sustainable Development: Principles in Practice.
indicators and those that maximally include the integrated The International Institute for Sustainable Development, Winnipeg.
components of sustainable development and their constituent Hart, M., 1998. Sustainable Community Indicators: Trainer’s Workshop. HART
Environmental, West Hartford, US.
categories. In total, 29 SDI were retained in our study, and clearly
Jacksonville Community Council, 2004. Quality of Life. Progress Report. Jacksonville,
reflect the initial distribution of the 188 SDI through the integrated US: http://www.jcci.org/.
components of sustainable development. Jones, P., Patterson, J., 2007. The development of a practical evaluation tool for urban
Therefore, the SuBSelec method brings new perspectives to the sustainability. Indoor and Built Environment 16 (3), 255–272.
Kahn, M.E., 2006. Green Cities: Urban Growth and the Environment. Brookings
debates related to the selection of urban SDI. We see it as a Institution Press, Washington, D.C..
preliminary step in a selection process which aims to propose to Koller, C., 2006. Le palmarès des villes romandes et le besoin de renforcer la
planners and decision makers a scientifically based and opera- statistique urbaine sur le plan suisse (méthodologie, sources, et résultats).
Revue économique et sociale 1, 101–116.
tional SDI grid. Legrand, N., Planche, S., Rabia, F., 2007. Intégration d’indicateurs de développement
Although we recognize the subjective nature of our approach, durable dans un outil d’aide à la décision. École des ingénieurs de la ville de
we believe that the classification and categorization exercises can Paris, Paris.
Levett, R., 1998. Sustainability indicators: integrating quality of life and environ-
allow the selection of recognized and complementary indicators mental protection. Journal of the Royal Statistical Society A161, 291–302.
while covering the various aspects of sustainable development as Nemetz, P. (chair), 2007. SMART Transportation Ranking Report. Vancouver: Apple-
broadly as possible. Moreover, we reach the same conclusion as ton Charitable Foundation.
Newman, P., Kenworthy, J., 1988. The transport energy trade-off: fuel efficient
Niemeijer and De Groot (2008) in that selection of indicators is
traffic versus fuel efficient cities. Transportation Research 22A (3), 163–174.
invariably subject to arbitrary decisions at one stage of the process Niemeijer, D., De Groot, R.S., 2008. A conceptual framework for selecting environ-
or another. mental indicators sets. Ecological Indicators 8, 14–25.
Martinez-Alier, J., Munda, G., O’Neill, J., 1998. Weak comparability of values as a
In addition, our analysis demonstrates that current practices
foundation for ecological economics. Ecological Economics 26, 277–286.
related to SDI cannot meet standard objectives. Thus, considering Meier, A., Wachter, D., 2005. Indicateurs centraux pour le développement durable
the contradiction between the need to obtain indicators that allow des villes et cantons. Rapport du Cercle Indicateurs. Berne, Switzerland: Office
comparison between jurisdictions and the desire to reflect local fédéral du développement territorial.
McLaren, V.W., 1996. Urban Sustainability Reporting. Journal of the American
concerns, it is probable that consensus on certain SDI is a Planning Association 62 (2), 184–203.
prerequisite to these objectives being met. Nonetheless, it should Perret, B., 2002. Indicateurs sociaux. Etat des lieux et perspectives. Les Papiers du
be acknowledged that this is a relatively new field that will surely CERC 2002-01, Conseil de l’emploi, des revenus et de la cohésion sociale,
Paris.
benefit from ongoing and future local initiatives. Planque B., Lazzeri, Y., 2006. Programme de recherche politiques Elaboration
d’indicateurs pour un système de suivi-évaluation du développement dur-
References able: tome 1: Principes et méthodologie de construction du référentiel
territoriales et développement durable. Paris: Ministère de l’Equipement,
Abolina, K., Zilans, A., 2002. Evaluation of Urban Sustainability in Specific Sectors in des Transports, du Logement, du Tourisme et de la Mer Direction Générale de
Latvia. Environment, Development and Sustainability 6, 299–314. l’Urbanisme, de l’Habitat et de la Construction, Secrétariat permanent du Plan
Agence Régionale pour l’Environnement, 2001. Diagnostic, développement durable Urbanisme Construction Architecture, Réponse APR programme D2RT.
urbain: tome 2: les indicateurs. Toulouse: Agence Régionale Pour l’Environne- Shields, D.J., Šolar, S.V., Martin, W.E., 2002. The role of values and objectives in
ment. communicating indicators of sustainability. Ecological Indicators 2 (2002),
Ambiente Italia Research Institute, 2003. European Common Indicators (ECI): 149–160.
Towards a Local Sustainability Profile. Final Project Report. Luxembourg: Office Slocombe, D.S., 1998. Forum: defining goals and criteria for ecosystem-based
for Official Publications of the European Communities. management. Environmental Management 483–493.
418 G.A. Tanguay et al. / Ecological Indicators 10 (2010) 407–418
SustainLane Report, 2007. The SustainLane 2006: US City Rankings. Document Needs of Countries. Inception/Regional Workshop on RETA6356: Improving
consulted on December 15, 2007 from: http://www.sustainlane.com/us-city- Administrative Data Sources for the Monitoring of the MDG Indicators. Bang-
rankings/. kok, Thailand.
Sustainable Calgary, 2004. State of Our City Report 2004. Calgary: City of Calgary. United Nations Commission on Sustainable Development, 2002. Report of the
Thomas, P-Y. (dir.), 2002. Indicateurs pour un développement durable urbain. World Summit on Sustainable Development. United Nations publication:
Bruxelles: Institut Bruxellois de gestion de l’environnement – IBGE. Dossier Johannesburg, South Africa, 26/08-4/09.
documentaire no 1. Octobre. Verry, D., Nicolas, J.-P., 2005. Indicateurs de mobilité durable: de l’état de l’art à la
Thomas, P-Y. (dir), 2003. Mesurer le développement durable en Belgique: Quel rôle définition des indicateurs dans le projet SIMBAD. Rapport du Laboratoire
pour les processus participatifs? Research paper for the Conseil fédéral du d’Économie des Transports pour le compte de la DRAST (Ministère de l’Équipe-
développement durable. Belgium: Institut pour le développement Sustainable ment) et de l’ADEME, Paris.
(IDD) and the Centrum voor duurzame ontwikkeling (CDO). Winston, N., Eastaway, M.P., 2008. Sustainable housing in the urban context:
Tomalty, R. (dir), 2007. The Ontario Urban Sustainability Report, 2007. Ottawa: The international sustainable development indicator sets and housing. Social Indi-
Pembina Institute. cators Research 87, 211–221.
United Nations Statistical Institute for Asia and Pacific, 2007. Building Adminis- World Commission on Environment and Development, 1987. From One Earth to
trative Data Systems for Statistical Purposes-Addressing Training Issues and One World: an Overview. Oxford: Oxford University Press.