New York leukemia data taken from the data sets supporting Waller and Gotway 2004 (the data should be loaded by running example(NY_data) to demonstrate spatial data import techniques)



A data frame with 281 observations on the following 12 variables, and the binary coded spatial weights used in the source.

  • AREANAME name of census tract

  • AREAKEY unique FIPS code for each tract

  • X x-coordinate of tract centroid (in km)

  • Y y-coordinate of tract centroid (in km)

  • POP8 population size (1980 U.S. Census)

  • TRACTCAS number of cases 1978-1982

  • PROPCAS proportion of cases per tract

  • PCTOWNHOME percentage of people in each tract owning their own home

  • PCTAGE65P percentage of people in each tract aged 65 or more

  • Z ransformed propoprtions

  • AVGIDIST average distance between centroid and TCE sites

  • PEXPOSURE "exposure potential": inverse distance between each census tract centroid and the nearest TCE site, IDIST, transformed via log(100*IDIST)



The examples section shows how the DBF files from the book website for Chapter 9 were converted into the nydata data frame and the listw_NY spatial weights list. The shapes directory includes the original version of the UTM18 census tract boundaries imported from BNA format ( before the OGR/GDAL BNA driver was available. The NY8_utm18 shapefile was constructed using a bna2mif converter and converted to shapefile format after adding data using writeOGR. The new file NY8_bna_utm18.gpkg has been constructed from the original BNA file, but read using the OGR BNA driver with GEOS support. The NY8 shapefile includes invalid polygons, but because the OGR BNA driver may have GEOS support (used here), the tract polygon objects are valid.


Waller, L. and C. Gotway (2004) Applied Spatial Statistics for Public Health Data. New York: John Wiley and Sons.


## NY leukemia
if (FALSE) {
nydata <- read.dbf(system.file("misc/nydata.dbf", package="spData")[1])
coordinates(nydata) <- c("X", "Y")

nyadjmat <- as.matrix(read.dbf(system.file("misc/nyadjwts.dbf",
ID <- as.character(names(read.dbf(system.file("misc/nyadjwts.dbf",
identical(substring(ID, 2, 10), substring(as.character(nydata$AREAKEY), 2, 10))

nyadjlw <- mat2listw(nyadjmat, as.character(nydata$AREAKEY))
listw_NY <- nb2listw(nyadjlw$neighbours, style="B")