README.md

In CasAndreu/ccceR: Tools for analysis of social media data

ccceR

This is an R package that allow researchers of the Center for Communication and Civic Engagement (CCCE) at the University of Washington to analyze and study social media data.

Installing the Package

To install this ccceR package use the install_github() function of the devtools package.

install.packages("devtools")
library(devtools)
install_github("CasAndreu/ccceR")
library(ccceR)

Reading xls files from Crimson Hegaxon

If you are using an Excel file downloaded from Crimson Hexagon, you need to do some simple pre-processing tasks in order to have a dataset you can apply the ccceR functions (a dataset with meaningful variable names).

To read the xls file into R, you can use the package rio. If you don't have the package, install it first. Then load it.

install.packages("rio")
library(rio)

Read and pre-process the data in R

file_path <- "path/to/the/xls/file"
raw_dataset <- import(file_path)
colnames(raw_dataset) <- raw_dataset[1, ] # The name of the variables are in the first row
dataset <- raw_dataset[2:nrow(raw_dataset), ] # Getting rid of the first row of the dataset

Functions:

add_rt_dummy()

This function adds a RT dummy variable into the dataset to indicate which messages are RTs. If the message is a retweet, RT == 1.

new_dataset <- add_rt_dummy(dataset, "Contents")

add_link_dummy()

This function adds a dummy variable to the dataset indicating which messages contain at least one URL. If the message contains URLs, link_dummy == 1.

new_dataset <- add_link_dummy(dataset, "Contents")

add_tweet_id()

This function adds a variable to the datset with the tweet id (tweet_id).

new_dataset <- add_tweet_id(dataset, "URL")

add_links_url()

This function adds 2 variables to the dataset with "shortened" (links_short_url) and "full" (links_full_url) URLs in the text messages.

new_dataset <- add_links_url(dataset, "Contents")

add_image_dummy()

This function adds a dummy variable to the dataset (image_dummy) indicating which messages contain links to images.

new_dataset <- add_image_dummy(dataset, "links_full_url")

download_images()

This function downloads images in tweets to a directory given by the user, and it also adds a variable to the dataset indicating the id of the tweet images.

new_dataset_with_image_id <- download_images(dataset, "links_full_url", "path/to/a/directory")

as_data_frame()

This funciton transforms all variables of a dataset into character variables so datasets with list variables can be written out inot a CSV file.

new_dataset <- as_data_frame(dataset)
write.csv(new_dataset, "new_dataset.csv", row.names = FALSE) 

CasAndreu/ccceR documentation built on May 6, 2019, 9:28 a.m.