Additional detail concerning the survey and variables analyzed here are presented in Huang et al. 2009 [34 (link)]. This study is based on a subset of data from the 2001 California Health Interview Survey (CHIS). This large (N = 55,428 households) random digit dial telephone survey in California is administered in seven languages (English, Spanish, Mandarin, Cantonese, Vietnamese, Korean and Khmer) and had a response rate, based on the American Association for Public Opinion Research equation RR4 [35 ], of 43.3% with a cooperation rate of 63.7% (weighted to account for the sample design) and 77.1% (unweighted).
We studied residents of San Diego and Los Angeles counties where over 70% of survey respondents supplied the name of the nearest intersection to their residence (In LA County, 8728/12196 = 71.5%, and in SD County 1952/2672 = 73%). These addresses were geocoded to represent the location of each respondent for purposes of this analysis. After exclusion of respondents with missing or invalid data, 8506 respondents from LA and 1883 respondents from SD were used in the analysis. These two counties were the only ones with nearest intersection data available in CHIS 2001.
The paper has two sections. In the first, we characterize street connectivity based on GIS-derived measures from buffers around the nearest intersections to respondents homes. In the second section we used a combination of CHIS variables, Census data, and the street connectivity data in a model-based analysis to explore the relative contributions of street connectivity and other variables to active transportation (AT).
We studied residents of San Diego and Los Angeles counties where over 70% of survey respondents supplied the name of the nearest intersection to their residence (In LA County, 8728/12196 = 71.5%, and in SD County 1952/2672 = 73%). These addresses were geocoded to represent the location of each respondent for purposes of this analysis. After exclusion of respondents with missing or invalid data, 8506 respondents from LA and 1883 respondents from SD were used in the analysis. These two counties were the only ones with nearest intersection data available in CHIS 2001.
The paper has two sections. In the first, we characterize street connectivity based on GIS-derived measures from buffers around the nearest intersections to respondents homes. In the second section we used a combination of CHIS variables, Census data, and the street connectivity data in a model-based analysis to explore the relative contributions of street connectivity and other variables to active transportation (AT).
Free full text: Click here
