I have built a contingency table from a column in a dataframe:
> dat <- data.frame(x = c(1, 1, 3))
> table(dat)
x
1 3
2 1
I want to add a "column" to the table for the missing factor levels (in this case "2") with a count of 0. The result should look like this:
> table(dat)
x
1 2 3
2 0 1
I have searched the site and found many similar questions, but while they use the term "table" in the question title and body, they all actually ask about dataframes.
You need to explicitly specify the levels, i.e.
table(factor(dat$x, levels = 1:3))
# 1 2 3
# 2 0 1
Since you seem to look for how the integer bins 1:max are filled, you could use tabulate instead.
tabulate(dat$x)
# [1] 2 0 1
When you create a factor variable, you can explicit all levels even the ones that do not appear in the data:
dat <- data.frame(x = c(1, 1, 3))
dat$x <- factor(dat$x, levels = 1:3)
table(dat)
Result:
x
1 2 3
2 0 1
For completeness, using data.table one could do it like this. I included an example on numeric values as well as factors.
dat <- data.frame(
x = c(1, 1, 3),
y = factor(c(1, 1, 3), levels = 1:3),
z = factor(c(1, 1, 3), levels = 1:3, labels = c("cat", "bird", "dog"))
)
# example with numeric values in column x
setDT(dat)[.(min(x):max(x)), .N, .EACHI, on = "x"]
# x N
# 1: 1 2
# 2: 2 0
# 3: 3 1
# example with factor column y
setDT(dat)[.(levels(y)), .N, .EACHI, on = "y"]
# y N
# 1: 1 2
# 2: 2 0
# 3: 3 1
# same as above, a factor with labels
setDT(dat)[.(levels(z)), .N, .EACHI, on = "z"]
# z N
# 1: cat 2
# 2: bird 0
# 3: dog 1
