transportation modal share and its low car ownership level. A typical suburban city
however generally displays low performance in these areas. In addition, their high
energy and water consumption contribute to the lowering of their EIs.
From a socioeconomic perspective, Montréal presents a performance that is far
below average in terms of the unemployment rate (SEI3), household expenditures
for housing (SEI5), household income (SEI6), the health status of the population
(SEI8), and crime rate (SEI9). These performances are nonetheless nuanced by better
scores in low ratio of income distribution (SEI7) and public expenditure for recreation and cultural activities, which are generally higher (SEI10). In typical suburban
cities, opposite trends can be observed. Higher performances are apparent in terms of
the unemployment rate, household expenditures for housing, household income,
crime rate, and health.
The above observations present interesting evidence of compensation between
USIs at three levels: at the level of indicators comprising environmental and socioeconomic indices, where an index can often reflect an over-performance in one or
two indicators only; between the EI and SEI, which occurs regardless of the size of
the city; and between suburban and central cities, where, practically, the indicators
are diametrically opposite. This demonstrates how cities in a metropolitan region are
complementary, and justifies the relevance of research pursued at regional level.
On another related issue, the aggregation of indicators is based on either compensatory or non-compensatory aggregation methods (OECD 2008). In the first case,
a very high score in a given indicator may compensate for a low score in other
indicators. Following this logic, a city could have, for example, a very high score in
waste management with a low score in another area and still obtain high environmental performance. The main criticism of compensatory aggregation methods is
that they allow compensation between the indicators that comprise the overall index
that is being used. However, the use of these methods has the advantage of
preserving the actual value of the scores obtained by the city for each indicator.
In the case of non-compensatory aggregation methods, the indicator values are
not taken into account. The indicators are instead processed according to an ordinal
scale, which eliminates the influence of extreme values on the aggregation toward
-3
-2
-1
0
1
2
3
ISE1
ISE2
ISE3
ISE4
ISE5
ISE6
ISE7
ISE8
ISE9
ISE10
Montréal
Average City
-4
-2
0
2
4
IE1
IE2
IE3
IE4
IE5
IE6
IE7
IE8
IE9
IE10
IE11
Montréal
Average City
Fig. 10.1 Radar diagrams of EI and SEI for Montreal and an Average City in Québec
196
G. A. Tanguay and J. Rajaonson
however generally displays low performance in these areas. In addition, their high
energy and water consumption contribute to the lowering of their EIs.
From a socioeconomic perspective, Montréal presents a performance that is far
below average in terms of the unemployment rate (SEI3), household expenditures
for housing (SEI5), household income (SEI6), the health status of the population
(SEI8), and crime rate (SEI9). These performances are nonetheless nuanced by better
scores in low ratio of income distribution (SEI7) and public expenditure for recreation and cultural activities, which are generally higher (SEI10). In typical suburban
cities, opposite trends can be observed. Higher performances are apparent in terms of
the unemployment rate, household expenditures for housing, household income,
crime rate, and health.
The above observations present interesting evidence of compensation between
USIs at three levels: at the level of indicators comprising environmental and socioeconomic indices, where an index can often reflect an over-performance in one or
two indicators only; between the EI and SEI, which occurs regardless of the size of
the city; and between suburban and central cities, where, practically, the indicators
are diametrically opposite. This demonstrates how cities in a metropolitan region are
complementary, and justifies the relevance of research pursued at regional level.
On another related issue, the aggregation of indicators is based on either compensatory or non-compensatory aggregation methods (OECD 2008). In the first case,
a very high score in a given indicator may compensate for a low score in other
indicators. Following this logic, a city could have, for example, a very high score in
waste management with a low score in another area and still obtain high environmental performance. The main criticism of compensatory aggregation methods is
that they allow compensation between the indicators that comprise the overall index
that is being used. However, the use of these methods has the advantage of
preserving the actual value of the scores obtained by the city for each indicator.
In the case of non-compensatory aggregation methods, the indicator values are
not taken into account. The indicators are instead processed according to an ordinal
scale, which eliminates the influence of extreme values on the aggregation toward
-3
-2
-1
0
1
2
3
ISE1
ISE2
ISE3
ISE4
ISE5
ISE6
ISE7
ISE8
ISE9
ISE10
Montréal
Average City
-4
-2
0
2
4
IE1
IE2
IE3
IE4
IE5
IE6
IE7
IE8
IE9
IE10
IE11
Montréal
Average City
Fig. 10.1 Radar diagrams of EI and SEI for Montreal and an Average City in Québec
196
G. A. Tanguay and J. Rajaonson