polarHistogram(df, family = NULL, columnNames = NULL, binSize = 1, spaceItem = 0.2, spaceFamily = 1.2, innerRadius = 0.3, outerRadius = 1, guides = c(10, 20, 40, 80), alphaStart = -0.3, circleProportion = 0.8, direction = "inwards", familyLabels = FALSE)
The data frame is expected to have at least four columns: family, item, score and value. The three first columns are categorical variables, the fourth column contains non-negative values. See the vignette for an example. Each bar represents the proportion of scores for one item. Items can be grouped by families. The resulting graph can be busy and might be better off saved as a pdf, with ggsave("myGraph.pdf").
a ggplot object
library(reshape2) set.seed(42) nFamily<-20 nItemPerFamily<-sample(1:6,nFamily,replace=TRUE) nValues<-3 randomWord<-function(n,nLetters=5) replicate(n,paste(sample(letters,nLetters,replace=TRUE),sep='',collapse='')) df<-data.frame( family=rep(randomWord(nFamily),nItemPerFamily), item=randomWord(sum(nItemPerFamily),3)) df<-cbind(df,as.data.frame(matrix(runif(nrow(df)*nValues),nrow=nrow(df),ncol=nValues))) df<-melt(df,c("family","item"),variable.name="score") # from wide to long print(head(df))family item score value 1 xdzyc jko V1 0.9799076 2 xdzyc psk V1 0.6438641 3 xdzyc xzg V1 0.5825784 4 xdzyc sxp V1 0.6158710 5 xdzyc qyw V1 0.9251403 6 xdzyc pvc V1 0.3900229