Lessons From the YouTube and Politics Workshop

kevin munger
Apr 18, 2019 · 6 min read
Vloggin

The cultural import of YouTube is difficult to overstate. The site has long boasted that they deliver over 1 billion hours of video content every day, and reporting on the “creators” who make a living producing videos and interacting with fans continues to shock and amuse.

But the political relevance of YouTube is still largely unknown. As a political scientist who studies social media, I really wanted to learn more about this. Thanks to support from the Princeton Center for the Study of Democratic Politics, I arranged (with Andy Guess) a workshop on the topic.

After a full day of methodological presentations and panel discussions, I came away with few answers but a host of questions, questions that I think scholars of political communication and media would like to see answered. I know that *I’d* like to read this research!

(This is not a summary of the workshop; any errors or bad ideas that follow are mine alone.)


What, exactly, is YouTube?

The most striking finding is how little we all knew. To date, social media adoption in the US has tended to be fastest among the tech-savvy — young adults, sure, but also scholars of online communication.

The uses of social media have now become more heterogeneous and rapidly changing. The reason is simple. A far broader swath of the US population uses social media regularly, and some of the most avid users come from populations that are unlikely to conduct academic research: teenagers, and people explicitly opposed to mainstream institutions of knowledge production.

As a result, more descriptive research is desperately needed. We need in-depth, qualitative description to surface trends and model their dynamics. And we need to comprehensive quantitative description to compare the importance of different uses of YouTube, and to track those changes over time.

How do we analyze video?

The technological challenges of turning videos into data that can be analyzed statistically are being solved — we saw a demo of Chris Lucas and Dean Knox’s feelR package, and it seems awesome. But video is distinct from other media formats in another way that poses challenges for media effects scholars: it takes much longer to deliver the same amount of content.

Delivering media stimuli online presents issues of audience attention. Furthermore, many online samples do not contain many of the very young or the very old, the people who might respond most strongly to novel political stimuli. We heard from Eunji Kim, who suggests addressing this problem by taking the media stimuli directly to research subjects, traveling around with iPads to ensure sufficient variation in age, rural identity and partisanship.

Getting data from YouTube itself turns out to be quite easy. A new Python API, constructed by Megan Brown and Leon Yin at the NYU SMaPP lab, is accessible and delivers only the data that social scientists might use.

More abstractly, the production function of video media is distinct from other media. Previously, high fixed costs for equipment had restricted the amount of video media, but near-ubiquitous laptop and smartphone cameras allow anyone to create videos they can distribute on YouTube. YouTube has a huge advantage in the video hosting space, with direct access to Google’s software (ad tech, recommendation systems) and hardware (distributed cloud storage, proprietary fiber optic cables) that makes competition from other video hosting sites implausible at scale.

This means that a single person can easily create multiple hours of video content every day. Even the most prolific writer cannot come close to that degree of efficiency.

Every consumer has some constraint on the amount of media they can consume per day. This massive increase in production efficiency changes the business model for media “firms”: if they can attract a small but devoted following, they can earn a profit. On the consumer side, even avid media consumers can fill their days with content from an incredibly small number of individuals. Scholars of the internet have long predicted that this media democratization is coming, but, as Matthew Hindman explains in The Internet Trap, the economics of the internet have instead produced centralization. The economics of YouTube itself, within the platform, may invert that equation.

Is YouTube a Social Network?

Scholars have developed a variety of tools for analyzing social networks, but it is unclear if YouTube will prove amenable to this style of analysis. It lacks the explicit network ties of other platforms, and many people use it as a straightforward way to consume video media.

The connections between YouTube videos often come from other sources. In analysis we saw from Cody Buntain, YouTube was the most shared domain on Twitter among accounts identified as being associated with the Russian IRA, and it is the second most shared domain on Reddit. This emphasizes the necessity of cross-platform research to tackle these problems.

But media reports certainly indicate that YouTube is a social hub for at least some groups of people. Consider the recent competition between Swedish comedian PewDiePie and Indian record company T-Series to become the most-subscribed channel on YouTube. Each side rallied huge networks of supporters and encouraged their fans to promote their channels. These are not passive consumers of comedy or music videos, but rather members of a community engaging in mediated social interaction.

Is YouTube used for Politics?

For traditional politics, the answer seems to be no — most politicians and mainstream political media use YouTube simply to host videos that were created for some other purpose, or recordings of speeches or rallies. People certainly watch these videos, but not in large enough numbers (or in novel enough ways) to merit in-depth analysis.

There are, however, a prominent network of far-right influencers who use the platform for political and cultural analysis. Primarily motivated by an opposition to the media, feminism and “social justice,” the popularity of these figures is comparable to that of the mainstream media. In some topic areas, they dominate the discussion, appearing in a high percentage of search results.

A larger conversation has to do with the nature of “politics” on YouTube. In Jaime Settle’s Frenemies, she theorizes that scrolling the Facebook NewsFeed tends to increase users’ conception of the political, as they get more information about the relationship between sociocultural signifiers and partisan identity.

If YouTube-as-social-network is primarily consumed by young people (who are the least interested in politics and have the weakest partisan identification), it is implausible that they are receiving explicitly partisan or electoral information. However, the social component — particularly the competition between communities who support rival YouTubers — can intersect with contemporary political issues.

PewDiePie, for example, has generated a number of controversies and conflicts with the mainstream media about his use of racist slurs and casual attitude towards sharing content created by far-right YouTubers. As ethnographer Crystal Abidin describes, however, many of PewDiePie’s fans interpreted the controversy as “a struggle between Influencers and legacy media more generally.” The cleavage of the struggle places this wildly popular figure on the same side as the far right in a fight against what they see as censorship. Content creators like PewDiePie can be wildly popular and earn huge sums of money, but they are still vulnerable to the whims of non-transparent, privately-held platforms that operate like private governments.

These frames — the connection between “internet humor” and far-right radicalization; the debate over censorship and free speech; the nature of precarious online work — are all political, and may become more salient to electoral politics.


YouTube and Politics

As the old academic saying goes…more research is needed on this topic! From descriptive data to economic models, from novel theories to application of previous analytical frameworks, there’s a ton of interesting work to be done.

kevin munger

Written by

Professor at Penn State. Internet and Politics. www.kevinmunger.com

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade