03.03.2015 Views

2000115-Strengthening-Communities-with-Neighborhood-Data

2000115-Strengthening-Communities-with-Neighborhood-Data

2000115-Strengthening-Communities-with-Neighborhood-Data

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

376 <strong>Strengthening</strong> <strong>Communities</strong> <strong>with</strong> <strong>Neighborhood</strong> <strong>Data</strong><br />

how programs actually work matches the expectations of how and why<br />

they are supposed to work [popularized in the theory of change (Weiss<br />

1995), which itself is seen as emergent (Patton 2011)]. Here causal attribution<br />

of program impacts is replaced by “contribution analysis” (Mayne<br />

2001), in which the question is whether program investments are correlated<br />

<strong>with</strong> community change. 2 Recent research on neighborhood effects also<br />

addresses the problem of multiple connected causes through a theory<br />

of “contextual causality” (Sampson 2012). 3 This shift toward contribution<br />

analysis has implications for spatial data analysis because it implies<br />

a greater emphasis on, for instance, tracking indicators of contextual<br />

factors and identifying statistical correlations between program outputs<br />

and community outcomes rather than seeking to isolate causal connections<br />

between outputs and outcomes.<br />

As methods vary by data type, an overview of spatial data types is followed<br />

by a brief discussion of selected spatial methods and tools that are<br />

relevant to place-based evaluations.<br />

Spatial <strong>Data</strong><br />

Classic taxonomies of spatial data distinguish between area data, event<br />

points, spatially continuous data, and spatial interaction or flow data<br />

(Bailey and Gatrell 1995). 4 Area data consist of discrete areas <strong>with</strong> attribute<br />

variation between areas or discrete points (see the MAUP section<br />

above). A common way to define which areas are neighbors is a so-called<br />

spatial weights matrix (often identified as W in equations), which identifies<br />

neighboring areas for a given area based on connectivity criteria,<br />

such as queen border contiguity, nearest neighbors, or bands of network<br />

distance. Event points are points whose location is subject to uncertainty<br />

(such as crimes). Bandwidths around points are often used to define<br />

neighboring points. Spatially continuous or geo-statistical data describe<br />

phenomena that are continuously distributed over space and are measured<br />

at sample points (such as contaminated soil). Spatial interaction<br />

or flow data characterize movement between an origin and a destination,<br />

such as in social networks or walkable access to school. See Radil, Flint,<br />

and Tita (2010) for an application and LeSage and Pace (2010) for spatial<br />

econometrics methods for flow data.<br />

The focus in this essay is on area data, because much socioeconomic<br />

data are collected for fixed administrative units (such as census tracts).<br />

In this context, spatial econometric methods have been developed to

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!