Question

I am working on a tutorial r question set where I need to create a "new" variable that characterizes "old" in the following manner. Define a tract to have a low rate, less than the median for the data set, medium rate between medium and 75 percentile, and high if the rate is higher than 75 percentile.

What would the r code look like? Not sure how to pull "old" variable out of the data set and define a high, medium and low tract.

Answer 1

Suppose D is the dataset.

The code is as follows:

x <- D$old ### where old is the "old" variable you have referred to.

(Now, assuming that the median is the median of the "old" variable, we can proceed as below)

m <- median(x)

p75 <- quantile(x,0.75)

n <- length(x)

new <- rep(0,length=n)

for(i in 1:n)

{

if(x[i] <= m) new[i] <- "low"

if(x[i]>m & x[i] <= p75) <- "medium"

if(x[i] > p75) <- "high"

}

Note: The resulting variable "new" will be of the type "character", which is a categorical type of variable. To convert it into numeric form, the following code can be used:

new1 <- as.factor(new)

Hope this helps!