TY - JOUR
T1 - Creating neighborhood typologies of GIS-based data in the absence of neighborhood-based sampling
T2 - A factor and cluster analytic strategy
AU - Gershoff, Elizabeth T.
AU - Pedersen, Sara
AU - Lawrence Aber, J.
N1 - Funding Information:
Funding for this project was provided through grants from the Centers for Disease Control and Prevention (5R49/CCR218598) and the National Institute of Mental Health (5R01MH063685) awarded to the first and third authors.
PY - 2009/1
Y1 - 2009/1
N2 - This article describes an innovative means of identifying a neighborhood typology that can be used for analyses of individual-level data that were not obtained through neighborhood-based sampling. A two-step approach was employed. First, exploratory factor analysis was used to reduce the number of neighborhood indicators to five clear factors of neighborhood characteristics. Second, a cluster analytic procedure was used to identify neighborhood types based on the five factors. These analyses resulted in a parsimonious solution of five distinct neighborhood clusters, or types, that constituted a manageable number of categories that could be used for future analyses of individuals grouped within neighborhood types. This method is a promising way to conduct neighborhood impact analyses that maximize the ability of researchers to characterize neighborhoods accurately (without sampling at the neighborhood level) while retaining the ability to conduct analyses of participants grouped within types of neighborhoods.
AB - This article describes an innovative means of identifying a neighborhood typology that can be used for analyses of individual-level data that were not obtained through neighborhood-based sampling. A two-step approach was employed. First, exploratory factor analysis was used to reduce the number of neighborhood indicators to five clear factors of neighborhood characteristics. Second, a cluster analytic procedure was used to identify neighborhood types based on the five factors. These analyses resulted in a parsimonious solution of five distinct neighborhood clusters, or types, that constituted a manageable number of categories that could be used for future analyses of individuals grouped within neighborhood types. This method is a promising way to conduct neighborhood impact analyses that maximize the ability of researchers to characterize neighborhoods accurately (without sampling at the neighborhood level) while retaining the ability to conduct analyses of participants grouped within types of neighborhoods.
KW - Cluster analysis
KW - Neighborhood
KW - New York City
UR - http://www.scopus.com/inward/record.url?scp=61649121711&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=61649121711&partnerID=8YFLogxK
U2 - 10.1080/10852350802498458
DO - 10.1080/10852350802498458
M3 - Article
C2 - 19197673
AN - SCOPUS:61649121711
VL - 37
SP - 35
EP - 47
JO - Community Mental Health Review
JF - Community Mental Health Review
SN - 0270-3114
IS - 1
ER -