Counting user types
Counts are the essential summary for categorical data. Since text is categorical, it's important to get comfortable computing counts. The twitter_data
is composed of complaints and non-complaints, as indicated by the complaint_label
column, and also includes a column indicating whether or not the user is verified (i.e., they have been confirmed by Twitter to be who they say they are) called usr_verified
. Note that column is of type <lgl>
, meaning logical. Do verified users complain more?
This exercise is part of the course
Introduction to Text Analysis in R
Exercise instructions
- Load the
tidyverse
package, which includesdplyr
andggplot2
. - Filter the data to only keep tweets that are complaints.
- Count the number of verified and non-verified users that have complained.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Load the tidyverse package
___(___)
twitter_data %>%
# Filter for just the complaints
___(___) %>%
# Count the number of verified and non-verified users
___(___)