Popular Political Podcasts Dataset

About the Dataset

Podcasts have exploded in popularity recently. Over the past decade, monthly podcast listeners have risen from just over 10% of the U.S. population to more than 40%—a reach on par with major social media platforms like Instagram.

Yet for all its reach and appeal, the medium has received little scholarly attention. Due in part the podcasting ecosystem’s decentralized structure and large file sizes, academics and analysts have carried out relatively little research on how podcasting shapes political conversations and policy debates.

In an effort to facilitate better real-time analysis of this influential medium, we have developed and released the Popular Political Podcast Dataset, which directs you to more than 40,000 episodes from 79 of the most prominent political podcast series. The dataset updates three times a week with new episode information.

The Popular Political Podcast Dataset is part of (and a bibliography for) research conducted at the Brookings Institution and elsewhere, including:

• ''Policy recommendations for addressing content moderation in podcasts'', by Valerie Wirtschafter and Chris Meserole, April 18, 2022
• ''Popular podcasters spread Russian disinformation about Ukraine biolabs'', by Jessica Brandt, Valerie Wirtschafter, and Adya Danaditya, March 23, 2022
• ''Prominent political podcasters played key role in spreading the 'Big Lie''', by Valerie Wirtschafter and Chris Meserole, January 4, 2022
• ''The challenge of detecting misinformation in podcasting'', by Valerie Wirtschafter, August 25, 2021

To use data from the dataset, please cite: ''The Popular Political Podcast Dataset developed by Valerie Wirtschafter and Chris Meserole at the Brookings Institution'' or link back to this database. If you do use data from the database in your research, please share it with us, so that we can feature it here. If there are additional features you would like us to incorporate, please let us know as well.

The information and links provided on this site are provided for non-commercial, academic research purposes only. Database links are provided to publicly available third-party feeds and sites and will direct you away from this site. We are not responsible for the contents, transmissions and/or downloads of any linked site or any link contained in a linked site, or any changes or updates to such sites. We are similarly not responsible for the privacy practices or the content of such third-party sites. We are providing these links to you only as a convenience, and the inclusion of any link does not imply our endorsement of, nor our ownership of rights to, the site. You are responsible for ensuring that your research use of any third-party site content does not infringe on any third-party rights, including without limitation, copyright.

Use of this site and any information provided herein are subject to the above conditions. If you visit, read, click on links from or otherwise use this site, you accept these conditions.

Inclusion Criteria

To be included as a series in this dataset, (1) the show must be a punditry/talk show-style series, (2) the description must include the term ''policy'', ''politic'', ''news'', or ''current events'' or discuss political topics in recent episodes, and (3) the series must satisfy one of the following criteria:

1. Featured in Apple's Top 100 List in November 2020
2. Featured in Apple's Top 100 List in April 2021
3. Recommended by Apple in the ''You Might Also Like'' section of a show identified through either (1) or (2)

This video provides a demo of the inclusion criteria process for two shows in Apple's Top 100 and a few additional recommended series:

Dataset Variables

The dataset includes the following variables:

• Episode ID: This variable provides a unique ID for each episode, ranging from 1 to the total number of episodes in the dataset at the time of the last update.


• Show Name: This variable provides the name of the show included in our dataset. There are 79 unique shows included in the dataset. It is possible to filter on this variable, using the “Show Name” filter.


• Full Date: The variable provides the date the episode was posted online. It is possible to filter on this variable, using the date range options in the “Date Range” filter.


• Title and Description: These variables provide the title and description for the series episode. It is possible to filter on these variables, using the “Character String Search” filter. To search for multiple terms, users can utilize the “OR” operator (|). For example, a search for COVID related episodes might take the form: “covid|pandemic|coronavirus.” The search will the scan through both the Title and Description columns and return only episodes that include one or more of these terms. All character string searches should use lowercase spelling.


• Partisan Leaning: This variable classifies podcasts hosts as “More Liberal”, “More Conservative” or “Moderate” based off their estimated political ideology. We calculate estimated political ideology using a method that relies on Twitter following decisions as a signal of partisan preference. More information on this method, which relies on well-established Bayesian ideal point estimation, can be found here. Where we are unable to discern Partisan Leaning through this method or other obvious means (e.g., Congressional party affiliation), we tag the series as ''Unknown.'' These classifications represent our best estimate of partisan leaning, based on publicly available information. It is possible to filter on this variable using the “Partisan Leaning” filter.


• Entered Dataset: This variable details how the series entered our dataset. A series could enter our dataset in three different ways: (1) it appeared in Apple’s Top 100 lists during November 2020, at the time of the presidential election; (2) it appeared in Apple’s Top 100 list April 2021, when we began exploring possibilities for this research; or (3) it was recommended by Apple for listeners who enjoyed one or more of show categorized as popular in either (1) or (2). The purpose of this inclusion criteria is to not only include shows that garnered a broad audience at the time, but also mimic one popular way that users might stumble across new shows – via app recommender systems. It is possible to filter on this variable using the “Entered Dataset” filter.


• Audio URL: This variable provides the URL to the raw audio file for each episode. These URLs are publicly available online and hosted alongside episode metadata by various podcast hosting providers. Using this link, users may directly download the audio file to their personal devices and then transcribe the file for more in-depth content analysis.

This feature allows you to explore how the political podcasting space has changed over time, using a variety of different filters.

Date Range:

Character String Search:

Figure Title: