An overview of the steps that we took may be viewed here.

Exportation of data from INFO-MAP and formatting
The initial step to performing the data analysis is to download the data set from the INFO-MAP software package and format such that it may be read by your statistical software. Instructions on how to do this may be found here.

Reading the data into statistical software package
Two data files are to be used in the analysis of this data set (Bailey and Gatrell, 1995):

1. Larnyxy.txt - Larynx cancer data
2. LungCancer.txt - Lung cancer data

Reading in the data from a text file(remember that both files are need to be used):
cases _ spoints (scan ("cases.xy"))

Plotting a dot plot:
pointmap (cases)

Drawing and adding a polygon:
cases.poly _ getpoly()
polymap (cases.poly, add=T)

We need to define a distance scale at which to evaluate the K-functions. As a rule, it is useful only to evaluate the K-function up to about one-third of the linear extent of the polygon. At distances larger than this the K-function estimate becomes inefficient (we (and the book) suggest the range from 0 to 2000). The S function seq is used to generate a vector of distances for the K-function.
s _ seq (from = 0, to = 2000, len = 20)
Kcase _ khat (cases, cases.poly, s)  (the same for the "control" data set)
SEcc _ secal (cases, cont, cases.poly, s)

Now the difference between the K-functions can be plotted on a graph together with an approximate 95% confidence envelope calculated as +/-2SE:
matplot (s, 2*cbind (SEcc, - SEcc), type = "l", lty = 1, col = 1)
lines (s, Kcase - Kcont, lwd = 2)
title ("K-function difference and +/- 2SE envelope")

The code that we used:

1. Load the library of splancs:

> library(splancs)

2. Read in the point datasets:

Larynx.pts<-spoints(scan("/home/ssa8/Larnyxy.txt"))
LungCancer.pts<-spoints(scan("/home/ssa8/LungCancer.txt"))

3. Draw the point map of the datasets (two ways- separate maps
for each data set and one with both data sets):

a) 2 maps:
 
par (mfrow = c (1,2))
par (pty = "s")
pointmap (LungCancer.pts)
title(main="Lung cancer cases", xlab="Easting", ylab="Northing")
pointmap (Larynx.pts)
title(main="Larynx cancer cases", xlab="Easting", ylab="Northing")

b) 1 map, 2 data sets:

par (mfrow = c (1,1))
par (pty = "s")
pointmap(LungCancer.pts,pch="|", col = 1)
pointmap(Larynx.pts, add = T, pch = "*", col = 2)
title(main="Lung and larynx cancer cases", xlab="Easting", ylab="Northing")
legend(x=346000, y=413000, c("Lung cancer", "Larynx cancer"), lty = 1, col = 1:2)

4. Draw a polygon on the point map, and create a polygon map:

> Larynx.poly<-getpoly()     #only once

OPTIONAL: To add the polygon map to the point plot:

> polymap(Larynx.poly, add = T)

5. Get the K-hat:

s_seq(from=0, to =2000, len=20)
Larynx.khat_khat(Larynx.pts,Larynx.poly,s)
LungCancer.khat_khat(LungCancer.pts,Larynx.poly,s)

6. Get Difference and plot:

SEcc_secal(Larynx.pts,LungCancer.pts,Larynx.poly,s)
matplot(s,2*cbind(SEcc,-SEcc),type='l',lty=1,col=1)
lines(s,Larynx.khat-LungCancer.khat,lwd=2)
title("Difference between K functions for larynx and lung cancers",
  xlab="distance", ylab="Difference in K functions and limits")

1. T. C. Bailey, A. C. Gatrell (1995): Interactive Spatial Data Analysis, Longman/Wiley, New York.
2. B.S.Rowlingson, P.J.Diggle (1993): SPLANCS: SPATIAL POINT PATTERN ANALYSIS CODE IN S-PLUS, Computers and Geosciences, 19, 5, pp. 627- 655.