Drop Residual Factors

Want to learn more? I recommend working through: R for Data Science, R Cookbook, and R Graphics Cookbook.

create the problem

# create a dataframe with a factor variable (state names)
df <- data.frame(states = state.name[1:5], score=seq(1:5))
# drop two of the observations
subdf <- subset(df, score <= 3)
# however, all the factors remain
[1] "Alabama"    "Alaska"     "Arizona"    "Arkansas"   "California"

the solution

# run factor() to remake the factors correctly
subdf$states <- factor(subdf$states)
# check to see it is correct
[1] "Alabama" "Alaska"  "Arizona"