Geoscience Reference
In-Depth Information
13.1.1
The State of Social/Spatial Modeling
Everything happens somewhere: examining social life as extricated from the
influence of the built environment results in an unrealistic view. Yet the methods
available for understanding the clustering and dispersion of a set of individual social
networks in geography are limited, as social network and urban spatial models have
matured in separate domains, and are analyzed in separate spheres, through social
network analysis and geographic information systems (GIS), respectively (Andris
2011
). Network methods are also rarely used by those who study city form (Sevtsuk
and Mekonnen
2012
). Social networks represent influences and social capital as
graph configurations of nodes (agents) and links (e.g., edges) between nodes where
primary metrics are connectivity and embeddedness; alternatively, spatial (e.g., GIS)
models are represented in a contiguous topological plane, where adjacency and
proximity are primary metrics (Andris
2011
).
As a result, social/spatial phenomena are often explained separately by those
inclined toward computational sociology or geography, respectively. One example
is the study of obesity, where social networks (Christakis and Fowler
2007
) and city
form (Papas et al.
2007
) are examined as causal factors, but not in the same study. To
obtain a clearer picture of the mechanisms surrounding obesity, one should consider
social ties and the built environment as coincidental factors - as these influences
can compound. Similarly, research showing how students use a college campus
in space and time via WiFi usage describes the flexibility of meeting places due
to mobile computing (Sevtsuk et al.
2009
), but could be extended to assess social
gatherings in time and place, as do Eagle et al. (
2009
) on the same college campus,
during a similar time period. Eagle et al. (
2009
) show the temporal social patterns
of dyadic (pairwise) relationships in terms of calls, SMS messages and colocation,
and alludes to the role of the campus in providing the backdrop for social groups
and pairs. When combined with Sevtsuk et al. (
2009
), this study could provide the
social ties within a spatial setting to uncover where friends meet, where they travel
on the campus and how these factors can be leveraged to create a better campus
environment.
This is not to say that datasets on interpersonal communication and movement
have not been embedded into geography; analyses of interplace networks of social
flows such as commodities, telecommunications, migration, and commuting are
common in computational urban research (examples abound). Yet, these represent
place-to-place aggregate flows instead of person-to-person flows and thus do not
directly express the decisions of individuals. Small-scale examples of spatially
embedded social networks describe gang membership (Radil et al.
2010
; Papachris-
tos et al.
2013
), transportation (Frei and Axhausen
2011
; Arentze et al.
2012
),
and epidemiology (Emch et al.
2012
). We take these initiatives a step further by
creating a general method that can respond to patterns of human socialization in a
built environment. These studies can elucidate where and when (different types of)
relationships form and could be used to advise architects, urban, and transportation
planners in creating places that support and create social connectivity.
Search WWH ::
Custom Search