Subset tibble of four specific rows

Question

I have a tibble, named Alabama, with 31 rows and I want a new tibble made of only rows 1, 11, 21, 31. These rows have are in a column named Year and have the values 1970, 1980, 1990, 2000

I have tried

al_decades <- filter(Alabama, Year==1970 && Year==1980 && Year==1990 && Year==2000)

and

al_decades <- subset(Alabama, Year==1970 && Year==1980 && Year==1990 && Year==2000)

but neither has worked.

A single year can't be both equal to 1970 and 1980 at the same time. — MrFlick, Dec 03 '19 at 06:32

akrun · Accepted Answer · 2019-12-03T06:46:03.597

We need %in% instead of ==

library(dplyr)
new <- filter(Alabama, Year %in% c(1970, 1980, 1990, 2000))

Or instead of & (not &&), use | to check for multiple elements as 'Year' wouldn't have different years in the same row

new <- filter(Alabama, (Year == 1970)|(Year == 1980)|(Year == 1990)|(Year == 2000))

Or if we are only interested in the decades, then another option is

new <- filter(Alabama, Year %%10 == 0)

If we know the position, another option is slice

new <- Alabama %>%
          slice(c(1, 11, 21, 31))

Or use filter with row_number()

new <- Alabama %>%
           filter(row_number() %in% c(1, 11, 21, 31))

Or using data.table

library(data.table)
setDT(Alabama, key = 'Year')[J(c(1970, 1980, 1990, 2000))]

In base R, we can use subset

subset(Alabama, Year %in% c(1970, 1980, 1990, 2000))

this was such a thorough and well stated answer, thank you very much! — Jay Wehrman, Dec 03 '19 at 06:42

Subset tibble of four specific rows

1 Answers1