← SASnRSubset variables

Introduction

When a dataset has many columns and we only need a few, we select just those variables — the column-wise counterpart of row filtering.
SAS offers keep to pick variables and drop to remove them. Both support name lists and colon-prefix patterns like h: for “every variable starting with h”.
In R, tidyverse uses select() with a rich mini-language: bare names, negative names, starts_with(), ends_with() and friends. Base R uses [, cols] subscripting with names or name patterns.
This lesson shows four common variable-subset tasks on the CLASS dataset: keep specific columns, drop specific columns, drop by prefix pattern, drop by suffix pattern.
After running, we have multiple subset datasets with only the chosen columns.

Subset variables

SAS (Base SAS)+

R (tidyverse)+

library(tidyverse)

class<-tribble(
  ~Name,~Sex,~Age,~Height,~Weight,
  "Alfred","M",14,69,112.5,
  "Alice","F",13,56.5,84,
  "Barbara","F",13,65.3,98,
  "Carol","F",14,62.8,102.5,
  "Henry","M",14,63.5,102.5,
  "James","M",12,57.3,83,
)

subset <- select(class, Name, Sex, Age)


subset <- select(class, -Height, -Weight)

subset<-select(class,-starts_with(c("H","W")))

subset<-select(class,-ends_with("t"))

To keep specific columns in tidyverse, we use select() with the column names: select(class, Name, Sex, Age).
To drop columns instead, we prefix the names with a minus sign: select(class, -Height, -Weight). Reads as “everything from class except Height and Weight”.
For pattern-based selection, select() accepts helper functions: starts_with("H"), ends_with("t"), contains("_"), matches(regex).
We combine helpers with minus for negative patterns: select(class, -starts_with(c("H", "W"))) drops every variable starting with H or W.
This mini-language makes column selection very expressive without the clutter of name lists.

Happy learning!

Search lessons

Introduction

Subset variables