Question

I have a factor variable represented by the histogram bins with values: ‘660-664’ , … , ‘740-744′ , 745-749’ ..

How can I map the factor variable to its mean value, e.g mapping ‘660-664’ to 662?

Basically, what I’m looking for is the inverse of the “cut” function.

Answer 1

You can make use of the plot = FALSE argument from hist to extract the breaks, then use that to get your midpoints:

set.seed(1)
x <- sample(300, 30)
x
#  [1]  80 112 171 270  60 266 278 194 184  18 296  52 198 111 221 142
# [17] 204 281 108 219 262 290 182  35  74 107   4 105 237  93
temp <- hist(x, plot = FALSE)$breaks
temp
# [1]   0  50 100 150 200 250 300
rowMeans(cbind(head(temp, -1),
               tail(temp, -1)))
# [1]  25  75 125 175 225 275

---

Update: Calculating the mean from a character string of ranges

Judging by your comments, you might be looking for something more like this:

myVec <- c("735-739", "715-719", "690-694", "695-699", "695-699",
           "670-674", "720-724", "705-709", "685-689")
myVec
# [1] "735-739" "715-719" "690-694" "695-699" "695-699" "670-674"
# [7] "720-724" "705-709" "685-689"
sapply(strsplit(myVec, "-"), function(x) mean(as.numeric(x)))
# [1] 737 717 692 697 697 672 722 707 687