OK. Then using aggregate():
data$yes <- ifelse(data$response=="yes", 1, 0) data$no <- ifelse(data$response=="no", 1, 0) dataresp <- aggregate(cbind(no, yes)~region+district, data,
sum)
dataresp[,3:4] <- dataresp[,3:4]/rowSums(dataresp[,3:4]) # or dataresp[,3:4] <- prop.table(as.matrix(dataresp[,3:4]),
1)
dataresp
region district no yes 1 A d 0.5 0.5 2 A e 0.0 1.0 3 B f 0.5 0.5 4 B g 0.5 0.5 5 C h 0.5 0.5 6 C i 0.0 1.0 7 C j 1.0 0.0 David From: Peter Maclean [mailto:pmaclean2011 at yahoo.com] Sent: Sunday, November 10, 2013 12:52 PM To: dcarlson at tamu.edu Subject: Re: [R] Cross Tabulation Thanks. But I am creating?lots of tables and I need Regions and Districts to appear so as to avoid to much editing. ? Peter Maclean Department of Economics UDSM On Sunday, November 10, 2013 12:32 PM, David Carlson
<dcarlson at tamu.edu> wrote:
The simplest would be to create a variable combining region and district:
data$region_district <- with(data, paste(region, district)) prop.table(xtabs(~region_district+response, data), 1)
? ? ? ? ? ? ? response
region_district? no yes
? ? ? ? ? ? A d 0.5 0.5
? ? ? ? ? ? A e 0.0 1.0
? ? ? ? ? ? B f 0.5 0.5
? ? ? ? ? ? B g 0.5 0.5
? ? ? ? ? ? C h 0.5 0.5
? ? ? ? ? ? C i 0.0 1.0
? ? ? ? ? ? C j 1.0 0.0
-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352
-----Original Message-----
From: r-help-bounces at r-project.org
[mailto:r-help-bounces at r-project.org] On Behalf Of Peter Maclean
Sent: Sunday, November 10, 2013 12:06 AM
To: r-help at r-project.org
Subject: Re: [R] Cross Tabulation
#Would like to create a cross-table (Region, district, response)
and
#(Region, district, cost. The flat table function does not look
so good
region? <- c("A","A","A","A","B","B", "B", "B", "C","C", "C",
"C")
district <- c("d","d","e","e","f","f", "g", "g", "h","h", "i",
"j")
response <- c("yes", "no", "yes", "yes", "no", "yes", "yes",
"no", "yes", "no", "yes","no")
cost? <-? runif(12, 5.0, 9)
var <- c("region", "response", "district")
data <- data.frame(region, district, response, cost)
var1 <- c("region", "district", "response")
var2 <- c("region", "district", "cost")
data1 <- data[var1]
#This look okay
with(data, aggregate(x=cost, by=list(region, district),
FUN="mean"))
#This does not look good
#How do i remove the NaN or create a better one
prop.table(ftable(data1, exclude = c(NA, NaN)), 1)
prop.table(ftable(xtabs(~region + district+ response,
data=data)),1)
Peter Maclean
Department of Economics
UDSM
??? [[alternative HTML version deleted]]
______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible
code.