How to create a column with a quartile rank?

Question

I have a table called tableOne in R like this:

idNum        binaryVariable        salePrice 2               1                    55.56 4               0                    88.33 15              0                     4.45 87              1                    35.77 ...            ...                    ... 

I'd like to take the values produced from: summary(tableOne$salePrice) to create four quartiles by salePrice. I'd then like to create a column tableOne$quartile with which quartile each rows salePrice is in. It would look like:

idNum        binaryVariable            salePrice      quartile     2               1                    55.56            3     4               0                    88.33            4     15              0                     4.45            1     87              1                    35.77            2      ...            ...                    ...            ...   

Any suggestions?

Answer 1

This should do it:

tableOne <- within(tableOne, quartile <- as.integer(cut(salesPrice, quantile(salesPrice, probs=0:4/4), include.lowest=TRUE))) 

...Some details:

The within function is great for calculating new columns. You don't have to refer to columns as tableOne$salesPrice etc.

tableOne <- within(tableOne, quartile <- <<<some expression>>>) 

The quantile function calculates the quantiles (or in your case, quartiles). 0:4/4 evaluates to c(0, 0.25, 0.50, 0.75, 1).

Finally the cut function splits your data into those quartiles. But you get a factor with weird names, so as.integer turns it into groups 1,2,3,4.

Try ?within etc to learn more about the functions mentioned here...

Answer 2

A data.table approach

    library(data.table)     tableOne <- setDT(tableOne)[, quartile := cut(salesPrice, quantile(salesPrice, probs=0:4/4), include.lowest=TRUE, labels=FALSE)]