Data frames represent a table of data. Recall our cats data frame:
cats <- data.frame(coat = c("calico", "black", "tabby"),
weight = c(2.1, 5.0, 3.2),
likes_string = c(1, 0, 1))
cats coat weight likes_string
1 calico 2.1 1
2 black 5.0 0
3 tabby 3.2 1Columns of a data frame are vectors of different types, each of the same length, that are organized by belonging to the same table.
In our cats example, we have a character, a numeric, and a logical column/variable. As we have seen already, each column of data.frame is a vector.
$There are several ways to extract an individual column in a data frame, including using the $ notation for extracting columns by name:
cats$coat[1] "calico" "black" "tabby" This approach returns the column as a vector.
[,]A column can also be accessed using the square bracket notation with two dimensions:
cats[, 1][1] "calico" "black" "tabby" The syntax df[i, j] to extract the entry in the ith row and the jth column from the data frame called df. For example, df[3, 1] will extract the single entry in the third row and first column.
A blank i or j tells R to extract all of the rows or columns, respectively. That is, df[, 1] will extract all rows for the 1st column, and df[2, ] will extract the second row across all columns.
df[, j] will extract the jth column from the data frame called df as a vector.
df[i, ] will extract the ith row from the data frame called df as a data frame.
For example, the following code extracts the data from the second column of cats as a vector
cats[, 2][1] 2.1 5.0 3.2and the following code extracts the second row of cats as a data frame:
cats[2, ] coat weight likes_string
2 black 5 0[]Note, to extract the \(j\)th column, you can alternatively use the single-dimension square bracket syntax: df[j].
cats[2] weight
1 2.1
2 5.0
3 3.2This syntax also works with named indexing.
cats["weight"] weight
1 2.1
2 5.0
3 3.2This, however, returns a one-dimensional data frame, rather than a vector.
[[]]Finally, another way to extract a column from a data frame is using the double bracket syntax [[]]:
cats[[2]][1] 2.1 5.0 3.2This also works with named indexing:
cats[["weight"]][1] 2.1 5.0 3.2Can you see how the output of indexing with this [[]] syntax differs from the single bracket syntax []?
The double bracket syntax [[]] returns the column as a vector, whereas the single bracket syntax [] returns the column as a single-column data frame.
Like vectors, data frames have column names, which can be accessed with the names() function.
names(cats)[1] "coat" "weight" "likes_string"If you want to rename the second column of cats, you can assign a new name to the second element of names(cats).
names(cats)[2] <- "weight_kg"
cats coat weight_kg likes_string
1 calico 2.1 1
2 black 5.0 0
3 tabby 3.2 1In the dplyr lesson, you will learn another method of extracting columns of a data frame and renaming them using the select() and rename() functions from the dplyr package in the tidyverse suite.
The square bracket syntaxes (but not the $ syntax) options allow us to select multiple columns at once:
Can you guess what the following code will return?
cats[1:2] coat weight_kg
1 calico 2.1
2 black 5.0
3 tabby 3.2cats[c(3, 1), 3][1] 1 1Like for vectors, we can also subset using logical operations:
The following code will select only the rows where the cat’s weight is at least 3
cats$weight > 3[1] FALSE TRUE TRUEcats[cats$weight > 3, ] coat weight_kg likes_string
2 black 5.0 0
3 tabby 3.2 1The following code will select only the rows where the cat’s coat is tabby or calico and returns just the values in the “weight” column:
cats[cats$coat %in% c("tabby", "calico"), "weight"]NULLLet’s move away from cats to a more interesting data example: gapminder
gapminder <- read.csv("data/gapminder_data.csv")The gapminder dataset contains information on the population, life expectancy, GDP per capita for a range of country-year combinations:
head(gapminder) country year pop continent lifeExp gdpPercap
1 Afghanistan 1952 8425333 Asia 28.801 779.4453
2 Afghanistan 1957 9240934 Asia 30.332 820.8530
3 Afghanistan 1962 10267083 Asia 31.997 853.1007
4 Afghanistan 1967 11537966 Asia 34.020 836.1971
5 Afghanistan 1972 13079460 Asia 36.088 739.9811
6 Afghanistan 1977 14880372 Asia 38.438 786.1134dim(gapminder)[1] 1704 6