rr-faq

Convert continuous numeric values to discrete categories defined by intervals


I have a data frame with a continuous numeric variable, age in months (age_mnths). I want to make a new discrete variable, with age categories based on age intervals.

# Some example data
rota2 <- data.frame(age_mnth = 1:170)

I've created ifelse based procedure (below), but I believe there is a possibility for more elegant solution.

rota2$age_gr<-ifelse(rota2$age_mnth < 6, rr2 <- "0-5 mnths",

   ifelse(rota2$age_mnth > 5 & rota2$age_mnth < 12, rr2 <- "6-11 mnths",

          ifelse(rota2$age_mnth > 11 & rota2$age_mnth < 24, rr2 <- "12-23 mnths",

                 ifelse(rota2$age_mnth > 23 & rota2$age_mnth < 60, rr2 <- "24-59 mnths",

                        ifelse(rota2$age_mnth > 59 & rota2$age_mnth < 167, rr2 <- "5-14 yrs",

                              rr2 <- "adult")))))

I know there is cut function but I couldn't deal with it for my purpose to discretize / categorize.


Solution

  • If there is a reason you don't want to use cut then I don't understand why. cut will work fine for what you want to do

    # Some example data
    rota2 <- data.frame(age_mnth = 1:170)
    # Your way of doing things to compare against
    rota2$age_gr<-ifelse(rota2$age_mnth<6,rr2<-"0-5 mnths",
                         ifelse(rota2$age_mnth>5&rota2$age_mnth<12,rr2<-"6-11 mnths",
                                ifelse(rota2$age_mnth>11&rota2$age_mnth<24,rr2<-"12-23 mnths",
                                       ifelse(rota2$age_mnth>23&rota2$age_mnth<60,rr2<-"24-59 mnths",
                                              ifelse(rota2$age_mnth>59&rota2$age_mnth<167,rr2<-"5-14 yrs",
                                                     rr2<-"adult")))))
    
    # Using cut
    rota2$age_grcut <- cut(rota2$age_mnth, 
                           breaks = c(-Inf, 6, 12, 24, 60, 167, Inf), 
                           labels = c("0-5 mnths", "6-11 mnths", "12-23 mnths", "24-59 mnths", "5-14 yrs", "adult"), 
                           right = FALSE)