Quantifying Urban Diversity: Multiple Spatial Measures of Physical, Social, and Economic Characteristics - Computational Approaches for Urban Environments - page 156

that is higher than if all commercial points (including those outside of the block
group) were used. While a more inclusive computation would be more robust, it has
been sacrificed with the intention of reducing the computational burden and with
the understanding that the final calculation is derived from the mean which aids in
mitigating the impact of these types of occurrences.
The final part of the methodology consists of a composite index, the Urban
Livability (UL) Index, which combines all four parameters into a single number. For
this composite index, the final score for each of the four parameters is combined.
In the case of the block length and mixed-use parameters, the negative of the values
is taken, since for these particular indices, lower scores are considered better than
higher scores. The final index is as follows:

UL i D d i C . a i / C b i C . m i / (7.2)

where UL i equals the final composite index for the selected block group, d i equals
the rescaled value of the selected block group for the density parameter, b i equals
the rescaled value for the selected block group for the block length parameter, a i
equals the rescaled value of the selected block group for the building age difference
parameter, and m i equals the rescaled value for the selected block group of the
mixed-use parameter. While this unweighted additive approach to the composite
index may appear simplistic, it corresponds most closely to the work of Jacobs, who
did not view one of the parameters to be any more important than the other three. For
the final calculation, these composite scores are again rescaled from 0 to 1 utilizing
thesameformatfromEq. 7.1 , resulting values closer to 0 having a less “livable”
environment across the four parameters than values closer to 1.
For this research, all of the parameters have been examined at the census block
group level of aggregation. Jacobs carefully notes that her four generators of
diversity operate on a fine-grained level (Jacobs 1992 , pp. 150-151). While it would
be ideal to examine each of these parameters on a block-by-block basis, there is
difficulty in doing so. The most challenging of the parameters to examine at such a
small geographic level is that of mixed use. Simple observation in most cities will
show us that having a wide variety of uses on a single block is highly uncommon.
Figure 7.2 shows an example from Northwest D.C. where uses are clearly delineated
by street block. Commercial uses are clearly aligned along U Street, while T Street
along with 15th and 16th Streets are all residential in nature.
It is far more common for uses to be mixed by adjacent blocks, such as a
residential block adjacent to an office block and a retail block. Thus, it becomes
necessary to have a level of aggregation that captures this mixture of uses. Census
block groups provide a convenient, yet sufficiently fine-grained level of aggregation
for this study, as shown by the number, 433, needed to cover the entire city.
The most apparent downside to utilizing block groups is their irregular size.
This creates a challenging situation for comparing certain parameters across block
groups - again, mixed uses present a challenge. For areas with particularly high
population counts, block groups are smaller. For especially dense populations, the
block group may only contain a single residential complex. This creates problems