# R code for chapter 36 Graphical summaries of data: Aggregation
# Book details:
# http://www.amazon.co.uk/Health-Science-Statistics-using-Commander/dp/190790431X
# 

load(url("http://www.robin-beaumont.co.uk/virtualclassroom/book2data/healthwagedata.rda"))
# or
# http://www.robin-beaumont.co.uk/virtualclassroom/book2data/healthwagedata.sav
ls() # shows the rda file has a mydataframe dataframe
summary(mydataframe)
# now for summary for each level of eployment time:
# R commander generated:
library(abind, pos=14)
library(e1071, pos=15)
numSummary(mydataframe[,"hourwage"], groups=mydataframe$yrsscale, 
  statistics=c("mean", "sd", "IQR", "quantiles"), quantiles=c(0,.25,.5,.75,1))
# end r commander generated

# boxplots
# show outlier case numbers
Boxplot(hourwage~yrsscale, data=mydataframe, id.method="y")
# do NOT show outlier case numbers
Boxplot(hourwage~yrsscale, data=mydataframe, id.method="none")
# plot of means, and standard errors:
with(mydataframe, plotMeans(hourwage, yrsscale, error.bars="se"))
############
#  if we have used txt input would needed to have re-odered factor levels
############
# instead of above strategy can aggregate data# first need to remove any NA'sav
## if you use the R commander menu option:
## Data->Active data set->Remove cases with missing data 
test1 <- na.omit(cleandataframe)
cleandataframe <- na.omit(mydataframe)
cleandataframe <- na.omit(mydataframe[,c("hourwage","yrsscale")])
## if you remove the 'include all variables' tick box you only get the selected columns:
cleandataframe <- na.omit(mydataframe[,c("hourwage","yrsscale")])
########### now aggregate
meanhourly_time <- aggregate(cleandataframe[,c("hourwage"), drop=FALSE], 
  by=list(yrsscale=cleandataframe$yrsscale), FUN=mean)
####
# now interactively change values for the categories from factor level labels to numbers
## not shown
#  now produce scatterplot
scatterplot(yrsscale~hourwage, reg.line=FALSE, smooth=FALSE, spread=FALSE, 
  id.method='mahal', id.n = 2, boxplots=FALSE, span=0.5, data=AggregatedData)


#### end