Trjtdtk

I have a dataframe in R that looks like this

 ID APPROVAL_STEP APPROVAL_STATUS APPROVAL_DATE APPROVER
1234 STEP_A APPROVED 23-Jan-2019 John Smith
1234 STEP_B APPROVED 21-Jan-2019 Jane Doe

I need it to look like this

 ID STEP_A_STATUS STEP_A_APPROVAL_DATE STEP_A_APPROVER STEP_B_STATUS STEP_B_APPROVAL_DATE STEP_B_APPROVER
1234 APPROVED 23-Jan-2019 John Smith APPROVED 21-Jan-2019 Jane Doe

And of course, with the original dataframe, any of APPROVAL_STATUS, APPROVAL_DATE, or APPROVER can be NA.

What is the most elegant way to do this? I know how to do it by looping through the unique IDs, grabbing each row, creating new columns, etc.; but is there any way to do this in a more elegant way (e.g., using tidyverse)?

What you are asking is how to convert data from "long" form to "wide" form.

You can just use the reshape() function which is in the stats package.

#' I constructed your data frame here
x<-data.frame(ID=c(1234,1234),
 APPROVAL_STEP=c("STEP_A","STEP_B"),
 APPROVAL_STATUS=c("APPROVED","APPROVED"),
 APPROVAL_DATE=c("23-Jan-2019","21-Jan-2019"),
 APPROVER=c("John Smith","Jane Doe"))




#' Now to reshape the data

library(stats)
reshape(x, 
 timevar="APPROVAL_STEP",
 idvar="ID",
 sep="_", direction = "wide")

Your output should look like.

 ID APPROVAL_STATUS_STEP_A APPROVAL_DATE_STEP_A APPROVER_STEP_A APPROVAL_STATUS_STEP_B APPROVAL_DATE_STEP_B
1 1234 APPROVED 23-Jan-2019 John Smith APPROVED 21-Jan-2019
 APPROVER_STEP_B
1 Jane Doe

The column names is a cosmetic issue, but that's the gist of it.

Hope this is helpful!