This repository contains the source data for my blog post on popularity spikes following guests' appearances on The Joe Rogan Experience (JRE). I collected these data at 9:50am NZST on September 18, 2018 using the R scripts in code/.

code/episodes.R scrapes the JRE podcast directory for a list of episode numbers, dates and titles, and cleans these data by, e.g., filling in missing episode numbers and removing non-ASCII characters from episode titles.

code/guests.R generates a (manually adjusted) list of guests who appear on each JRE episode identified in data/episodes.csv.

code/popularity.R downloads Google Trends data (based on web search interest in the United States) for each unique value of guest_name in data/guests.csv.

I ran each script in a fresh instance of jre-guests.Rproj.

I used the httr and rvest packages to scrape episode metadata, gtrendsR to download Google Trends data, zoo to compute rolling means, and various packages from the tidyverse to read, manipulate and write data. These dependencies can be installed by running

install.packages(c("httr", "rvest", "gtrendsR", "zoo", "tidyverse"))

at the R console. All other commands should be available through the base R installation.

This repository contains data on episodes of The Joe Rogan Experience. These episodes are the intellectual property of Joe Rogan and Talking Monkey Productions, and are protected by copyright. I believe that my use of this intellectual property is covered by fair use. Such use is intended to be educational and my derived commentary would be severely impaired without the copyrighted material.

Any use of the data contained in this repository is at the user's own legal risk. I take no responsibility for the external use of these data, nor for any errors that they contain.

The contents of code/ are licensed under the MIT license.