Audience Capture: Rethinking Influencer-Audience Relationships within the Alternative Influence Network

Team Members

Daniel Jurg

Martin Trans

Marc Tuters

Giulia Costanzo

Janaynne Carvalho Do Amaral

Jenny Wielga

Mengdi Zhu

Marius Liedtke

Daniel Lundgaard

Xiao Shuwen

Zhi Lin

Links

• Poster 1

• Poster 2

Contents

Summary of Key Findings

This project aims to understand the reciprocal relationship between audiences and content creators in the development of political identity on YouTube. Challenging the traditional view that influencers primarily shape their audiences, this study focuses on how audiences may influence the political identity formation of content creators. In the context of alternative influencers on YouTube, this phenomenon is often referred to as ‘audience capture.’ To investigate this, our project examines the case of American political commentator and YouTube personality Dave Rubin, host of "The Rubin Report." We compare content production (including transcripts and video titles) with audience engagement (such as comments). The following observations summarize the key findings from our exploratory week during the DMI Summer School 2024.

• Understanding the process of audience capture requires considering the political economy and the financing of political media content on YouTube by various media corporations. While much research on YouTube focuses on influencer culture and the DIY creator economy, Rubin’s ideological transformation appears to correlate with changes in financial backing from various corporations and donation platforms. Rubin began at the progressive media network The Young Turks (TYT) but left after political disagreements in March 2015. He subsequently joined Verizon’s RYOT Network and Larry King’s Ora.TV, both of which do not display overt political leanings. In June 2016, Rubin transitioned to a crowd-funded financing model via Patreon and secured a sponsorship partnership with Learn Liberty, an economically libertarian and socially conservative initiative.

• Rubin presents his channel as a platform that does not take a specific political stance but allows different perspectives to come together, with the comment sections reflecting this search for political identity. While many discussions aim for non-partisanship, there is a clear ideological shift from “against Trump” to a Trump supporter, corresponding with the ideological leanings of the media companies financially supporting Rubin. We find that the 2016 election and Trump play a significant role in Rubin’s ideological transformation.

• In addition to discussions about Trump, gay marriage emerges as another key issue where Rubin’s political leanings increasingly conflict with those of his audience. Our analysis shows a shift in content from discussing the role of government in gay marriage towards questions around identity, with religious belief and age playing important roles.

• When testing for the statistical relationship between audience engagement and content production on The Rubin Report, we find that comments about gay marriage lead to a statistically significant increase in videos where gay marriage is mentioned, either in transcripts or video titles. This pattern also applies to mentions of Sam Harris, who Rubin himself has cited as an influential figure in his ideological transformation. Our analysis reveals that comments mentioning Sam Harris Granger-caused such mentions in video titles and transcripts on both a daily and weekly basis.

• Analyzing the composition of participatory audiences over the years, we observe that the same individuals do not consistently comment throughout the years; there is significant turnover within Rubin’s participatory community. A co-commenting analysis shows that Rubin initially attracted an audience that primarily engaged with the progressive TYT channel. However, by 2016, there was an influx of more conservatively engaged audiences.

• Finally, examining Rubin's own engagement in the comment section, we find that he replied most frequently during his time with TYT. Over time, the type of engagement diversified, but direct replies diminished. That being said, audience engagement with his comments has increased over time.

1. Introduction

Within the context of what has commonly been referred to as "YouTube radicalization," this project aims to find empirical evidence supporting the process described by Lewis (2018), where alternative influencers are “radicalized by their own audience’s engagement” (p. 6). Alternative influencers on platforms like YouTube have self-described this phenomenon as audience capture—defined as “a self-reinforcing feedback loop that involves telling one's audience what they want to hear and getting rewarded for it” (The Portal Wiki). This project is part of a broader effort to contextualize and conceptualize the process of audience capture within media and communication studies. It seeks to understand the relationship between audiences and content creators in the development of political identity on YouTube, with the goal of finding empirical evidence for this process.

The idea that influencers shape political opinion is not new (Katz & Lazarsfeld 1950). What is relatively new, however, is the diminishing power of the mass media to set the broader framework for opinion leaders. If mid-century political influencers developed their hegemonic liberal “vital” centrism from the news (Schlesinger 2009), these new influencers draw on alternative sources to deliver to their audiences “what the news won't show” (Lewis 2020). It has been frequently remarked that these newly empowered opinion leaders have been instrumental in expanding the spectrum of acceptable ideologies and policies (Lewis 2020. Finlayson 2021). If we use a simple depth heuristic to think of radical political discussions online (de Zeeuw and Tuters 2020), then we can think of these influencers as bringing to the surface ideological positions whose substance has already been extensively worked out in subcultural spaces, often far from the view of most users of social media. While we still tend to conceptualize political communications on models inherited from the broadcast era, there are decades of work in the field of reception studies that emphasized the active role of audiences in negotiating collective political projects from below. In many cases, alternative influencers are less the source of the radical political ideas underpinning culture war and partisan realignment than they are their amplifiers. Indeed, the very term 'influence' seems to imply a top-down and one-to-many orientation that fails to capture much of what takes place in some of the most dynamic political discussion spaces online.

Lewis (2018) claims that “The AIN (Alternative Influence Network) as a whole facilitates radicalization through social networking practices: (1) Audiences are able to easily move from mainstream to extreme content through guest appearances and other links, and (2) political influencers themselves often shift to more radical positions following interactions with other influencers or their own audiences” (p. 1). While Lewis provides extensive evidence for the former practice, she doesn’t provide evidence for the latter. This project explores ways to support the claim that alternative influencers can be captured by their audience.

2. Initial Data Sets

To study audience capture, this project focuses on the case study of Dave Rubin, an American political commentator, YouTube personality, and host of "The Rubin Report." Rubin began his career in comedy before transitioning to political commentary, initially aligning with progressive ideals during his time with "The Young Turks." Over time, his political views shifted rightward, and he now identifies as a libertarian-conservative. Rubin is known for his critiques of progressivism and the Democratic Party, and he has authored two books.

Based on popular press articles and YouTube videos, Rubin presents an excellent case study for the phenomenon of audience capture. In the article “Intellectual Dark Web's Descent into Paranoia and Trumpism,” Fisher discusses the fragmentation of the Intellectual Dark Web (IDW), a group initially united against ‘new-found’ progressive ideas. Fisher (2023) highlights how some members, like Sam Harris, distanced themselves from the group due to the increasing embrace of conspiracy theories and Trumpism of group members, specifically targeting figures like Dave Rubin and Bret Weinstein. The article offers initial insights into how Rubin might have been influenced to move further to the right, both by the guests he invited on his show and by the audiences he attracted. Additionally, we drew on a video essay by José titled “Audiences and Echo Chambers: A Look at The Rubin Report” (2018). In this video, José provides a detailed history of The Rubin Report and its evolving relationship with its audience, which proved particularly helpful in understanding the dynamics at play.

Using Dave Rubin as a case study, we constructed two datasets from a larger historical dataset created by Dutch journalists to study YouTube radicalization (Tokmetzis, 2019). The advantage of using this historical dataset over newly captured data is that it includes content that may have been removed since YouTube enhanced its content moderation, providing a less curated version of the platform's data. We queried the historical dataset for all videos and comments made on The Rubin Report channel, resulting in an initial dataset of 826,674 comments. During the analysis, we discovered that a small portion of the data (N = 64) contained concatenation errors that we were unable to correct within the project’s timeframe. Consequently, we removed these rows, leaving us with a final dataset of 826,610 comments. Additionally, we retrieved 1,862 videos posted by The Rubin Report from the historical dataset. For these videos, we used speech-to-text software to generate transcripts. Both datasets were then uploaded to the 4CAT: Capture & Analysis Toolkit (Peeters & Hagen, 2022) for further analysis. See Table 1 for an overview of the datasets.

Table 1: Overview of The Rubin Report Datasets

3. Research Questions

The following key research question guided the project:

1. How does the ideological content of alternative influencers evolve in response to audience engagement and feedback on YouTube?

Taking Dave Rubin’s The Rubin Report channel as a case-study, we explored the following sub questions:

1. Can we identify specific patterns or triggers that lead Dave Rubin to adopt more radical ideological political positions?

2. How can we quantify the influence of audience feedback on the radicalization process? Can we find a statistical significance in the impact of audience feedback on the content production?

3. How does the ideological composition of Rubin’s audience change over time?

4. How does Rubin engage with his audience in the comment section?

4. Methodology

To address our research questions, we distinguish between two primary approaches: (1) the interactional approach and (2) the community-composition approach. This section outlines our methods for each research question, organized into four parts: (1) Triggers and Patterns in Ideological Transformation, (2) Quantifying Audience Feedback, (3) Community Composition, and (4) Audience Engagement by the Content Creator.

Triggers and Patterns in Ideological Transformation

To address our first research question—identifying specific patterns or triggers that lead Dave Rubin to adopt more radical ideological positions—we employed an interactional approach by comparing content production (transcripts and video titles) with audience comments. Our approach combined both inductive and deductive methods. Inductively, we used Natural Language Processing (NLP) techniques within 4CAT to analyze the top 10 bigrams per year in both transcripts and comments, allowing us to compare shifts in discussions over time.

We first tokenized the transcripts per year, with twitter as the tokenizer type, grouping per item, filter for copy-language and stopwords iso-en, and rejecting the following additional terms: people, yeah, guy, time, lot, stuff, sort, talk, day, week, uh, talking, start, pretty, gonna, bit, video, person, kid. We then extracted the top 10 co-words with a window size of 2, filtering out additional terms: rubin, report, reuben, couple, official, trr, weeks, hey, guys, quick, hey, minute, outfitter, code, blah.

Deductively, we reviewed popular press coverage on Rubin to identify pre-established themes in his ideological development. A key resource in this analysis was José's video essay titled "Audiences and Echo Chambers: A Look at The Rubin Report," which provides a detailed history of The Rubin Report and its evolving relationship with its audience (José, 2018).

Following our initial large-scale quantitative interactional analysis to identify specific patterns or triggers, we developed a more focused approach for close-reading and statistical analysis. The bi-gram analysis revealed an intriguing conflict around the topic of 'gay marriage,' particularly given Rubin’s identity as a gay man and the increasingly conservative stance of his audience, which appears opposed to gay rights.

Using 4CAT, we filtered the comments dataset to isolate all comments about gay marriage from 2016 (748 items) and 2014 (105 items). To gain a clearer understanding of the discourse surrounding gay marriage, we selected the 30 most liked comments from each sample for in-depth analysis.

For the close-reading analysis, we organized the comments by categorizing them with different colored labels based on the themes they addressed. Comments that mention Dave Rubin, gay marriage, and related topics were highlighted in blue; comments discussing biological arguments were highlighted in green; religious references were highlighted in pink; comments expressing emotions were highlighted in red; and comments containing political or ideological beliefs were highlighted in olive green.

Quantifying Audience Feedback

To address our second research question, we began by presenting an overview of Rubin’s content engagement metrics to gauge audience feedback. We then conducted a series of time series analyses, including the Granger causality test and Impulse Response Function (IRF), to examine the bidirectional influence between comments, video titles, and video transcripts over time, focusing on specific persons and topics.

For the time series analysis, we concentrated on subsets of data related to either specific guests, such as Sam Harris, or particular topics, such as gay marriage. To create these subsets, we filtered the data to include only comments and videos where the guest's name or topic of interest appeared in one of the following: (1) a comment on a video, (2) the video title (e.g., “Interviewing Sam Harris about …” or “Today we are going to talk about gay marriage”), or (3) the video transcripts, as a guest or topic might be discussed extensively within a video even if not mentioned in the title.

Community Composition

For our composition-based approach to address the third research question, we employed two methods. First, leveraging our access to a large-scale historical dataset that included 1,738 additional political channels on YouTube (2008-2018), we conducted a co-commenting analysis. This analysis assessed, on a yearly basis, which other channels were frequented by The Rubin Report audience, focusing on the top 5 channels each year.

Second, we examined the continuity of commenting activity among individual users to illustrate the influx and outflux of audiences. We identified commenters who had not made any comments in the previous year and then tracked how many years they continued to comment on the channel. The results of these analyses are presented in the plot within the findings section.

Engagement of Content Creator

Finally, to address our last research question, we analyzed Dave Rubin's use of the comment section. Using 4CAT, we filtered the comments dataset to isolate all comments made by Dave Rubin, resulting in a dataset of 1,916 comments. In the first step, we randomly selected 10% of these comments to develop an inductive category system. The categories were adjusted as needed throughout the coding process. Ultimately, the category system comprised 10 categories. An overview of these categories, along with brief explanations and anchor examples, is provided in Table 2.

Name of category	Explanation	Example
Reply to a comment	When he is answering to a comment	+Simon Sez No, I didn't make it up. Hamas had to give authorization for the rally as cited by many newspapers, in Arabic even: http://www.businessinsider.com/r-gaza-jihadist-supporters-rally-against-france-praise-islamic-state-2015-1
Thank you for a comment	When he is answering thank you to a comment	Thanks, Kal! --- dave
Thank you	When he said thank you for watching or sharing or just thank you	Thanks for watching!
Welcoming	When he is welcoming new persons	Welcome!
Preview	When he gives a preview of what will be happening on his channel soon	Many of you asking about Tommy Robinson interview. It has been taped and will be up next week...
Unspecific	Comments which seem very unspecific; it is not visible if this comment does belong to another one or to what this comment is referring	Hah!
Explanation	Sometimes he is explaining his self, why he said sthg	Oh man, I said JFK but that quote is FDR! Sorry slightly jet lagged when we did this!
YouTube	Problems with YouTube	YouTube has told us there is a bug on the channel but they have no idea when it will stop and won't tell us any more...
Directing to another video	Links to other videos	Watch the full interview with Kyle Kashuv here: https://www.youtube.com/watch?v=cNi089rmj-4
Announcement	When he announces something, e.g. he gives an update	Quick Update: The charges were just dropped against the guy in the video.

5. Findings

This section presents the findings of our research, structured according to the methodology: (1) Triggers and Patterns in Ideological Transformation, (2) Quantifying Audience Feedback, (3) Community Composition, and (4) Audience Engagement by the Content Creator. The figures presented in the report are best viewed on the poster.

Triggers and Patterns in Ideological Transformation

Dave Rubin’s ideological development is closely linked to the business partnerships and financing structures of The Rubin Report. Initially, Rubin was part of the progressive media network The Young Turks (TYT), but he left in March 2015 following political disagreements. Afterward, he associated with Verizon’s RYOT Network and Larry King’s Ora.TV, both of which do not display overt political leanings. In June 2016, Rubin transitioned to a crowd-funded model, asking viewers to support him with monthly donations via Patreon. Shortly after, he announced a sponsorship partnership with Learn Liberty, an economically libertarian and socially conservative initiative funded by the billionaire Koch family (Lewis, 2018).

Rubin's departure from TYT was catalyzed by a contentious viral moment in October 2014, when the neuroscientist and philosopher Sam Harris stated in an episode of Real Time with Bill Maher that "Islam is the mother lode of bad ideas.” While TYT critically engaged with Harris’s statement and even hosted Harris on their show to debate the host Cenk Uyger, Rubin sided with Harris and subsequently left the network. During his time at Ora.TV, Rubin published videos warning of a Donald Trump presidency. After partnering with Learn Liberty and partly becoming financially dependent on audience donations, Rubin made light of Trump's electoral victory and declared a shift to a “new political center”.

Against this context, we present Fig. 1 and Fig. 2 as an initial overview of the most common bigrams in the transcripts and comments.

Figure 1: Bigrams from video transcripts from 2013 till 2018

Figure 2: Bigrams from comments from 2013 till 2018

Comparing transcripts and comments reveals that other prominent figures in the alternative community, such as Sam Harris, Ben Shapiro, and Jordan Peterson, are frequently mentioned in both the transcripts and the comments. They seem to play a key role in the development of ideas for Rubin and his audience. However, a notable discrepancy emerges regarding the topic of gay marriage and gay rights. While these issues are significant points of discussion in the transcripts, they are not prominently featured in the comments section.

Engaging in a close-reading of discussions on gay marriage by The Rubin Report in 2014, we find Rubin supports gay marriage, arguing it promotes equality and strengthens families without negatively impacting heterosexual marriages. Rubin argues against governmental interference with marriage and leaves it to religious institutions. In the comments around gay marriage we find that in 2014 the discussions are mostly connected to topics such as abortion and weed legalization, focusing mainly on the legislative role that the government should play.

The discussion about politics and religion deepens with commenters asserting that Jesus would not have accepted gay marriage. Other commenters expanded the conversation by addressing religious freedom and the relationship between civil unions and religious ceremonies. The question of whether churches should marry gay couples sparked a division of opinions among the audience. Many commenters linked anti-gay rights sentiments and homophobia to religious beliefs, advocating for a separation of law from religion. There is also a level of hate speech directed at Christians in the discussion. Commenters also mentioned gay children, particularly in the context of churches attempting to "save" them. In these discussions, religious environments are portrayed as selectively applying their rules when it comes to "saving" gay people. The audience continues to associate anti-gay rights positions and homophobia with religion, reinforcing calls for the separation of law and religion. However, this discussion also includes instances of hate speech directed at Christian individuals.

Audiences appear to disagree with how Dave Rubin discusses gay marriage, expressing that the LGBT rights movement does not end with marriage equality. Some commenters suggested that Dave should broaden the discussion to include issues such as attacks on gay people, bullying, and suicide. Comments from 2014 primarily focused on the intersection of government, religion, and politics, with some questioning the political identity of American politician Hillary Clinton. She was accused of opportunism in her support of gay rights, rather than genuinely holding progressive views.

By 2016, the conversation shifted to focus on both Dave Rubin and Ben Shapiro, with commenters speculating that they were concealing their true feelings and thoughts about gay marriage due to their personal beliefs or religious backgrounds. This led to a divided response among some who identify as gay men, with some supporting and others opposing Ben Shapiro's stance. One commenter linked gay identity with a progressive identity, reflecting a broader discussion on political identities and sexual orientation.

While some comments in 2016 still addressed the government's role in legalizing gay marriage and the separation of religion from politics, there was a noticeable shift towards debating political identities. Individuals with different gender, sexual, and political identities attempted to define what it means to be a "true" liberal or conservative, and what values each political party should uphold. Observations of the comments suggest that it is challenging for individuals with differing gender, sexual, and political identities to discuss the topic of gay marriage without bias. As a result, the audience expressed distrust in Dave Rubin's and his guests' perspectives, perceiving that they might be hiding their biases.

Quantifying Audience Feedback

Fig. 3 presents an overview of the engagement metrics of the Rubin Report.

Figure 3: Overview of likes, dislikes, and comments on The Rubin Report from 2013 till 2018.

The analysis of the engagement metrics highlights that 2014 was a massive year for Dave Rubin in terms of views, but disproportionately low in terms of engagement. Interestingly, 2014 also appears to be an outlier with regards to the various forms of engagement, as videos in 2014 received almost the same amount of dislikes as likes - a rare occurrence when reviewing averages for a year, and not just individual videos. This trend is however quickly changed, and already in 2015 the proportion of likes is much higher and more closely resembles patterns seen across YouTube.

Excluding 2014, Dave Rubin experiences continuous growth in terms of views and engagement, kick started by a massive jump from 2015 to 2016 where avg. views more than doubled, and avg. engagement more than tripled across comments, likes and dislikes, which correlates with him going solo, focusing more on Patreon and association with Learn Liberty. The channel does however not reach the same peak in terms of average number of views from 2014, but engagement rates on videos, and the proportion of likes to dislikes is magnitudes higher.

It is also worth noting that we are seeing evidence of some degree of saturation in 2018, with lower average views, comments, likes, and in particular dislikes (compared to 2017), however, more analysis and review of more recent data is needed to make final conclusions about potential peaks in terms of viewership for Dave Rubin.

Based on our bigram analysis we also looked for a statistical relationship between comments and transcripts. In the case of Sam Harris, our analysis showed that the comment mentioning Sam Harris Granger-caused such mentioning in video titles and transcripts on both a daily and weekly basis. In the case of gay marriage, we found that the frequency of mentioning gay marriage led to an increase in the mentioning of gay marriage in video titles and transcripts (Fig. 4 and Fig. 5).

Figure 4 & 5: Shock and response of comments vis-a-vis gay video titles and transcripts

Community Composition

Moving towards the question about audience composition, Fig. 5 presents an overview of the continuity of commenting activities of Rubin’s audience.

Figure 6: Continuity of commenting activities of Rubin’s audience

In Fig. 6, each color represents a unique year. Each line connects two years, presenting the continuity of commenting activities from a starting year to an ending year. Lines that begin and end within the same year, forming circles, represent individuals who only commented during that year and did not continue commenting in the following year. The plot suggests that most individuals did not continue to make comments across multiple years. In other words, the commenters largely consisted of different groups of individuals each year.

In addition to the lack of continuity of participatory audience in Rubin’s community, we also observe a shift in audience composition based on co-commenting practices (Fig. 7).

Figure 7: Top 5 co-commented channels per year of Rubin’s audiences.

Examining the other channels that Rubin’s audience engaged with, we observe a shift from predominantly commenting on The Young Turks (TYT) to increasingly commenting on more right-wing channels, such as Stefan Molyneux, Sargon of Akkad, Rebel Media, and The Daily Wire. This shift in audience engagement mirrors Rubin's own ideological journey and the growing alignment of his content with more conservative viewpoints. As Rubin's content evolved, his audience gradually transitioned from progressive channels to right-wing platforms, indicating a broader realignment within his viewership.

While we cannot definitively determine the prior views of these commenters—some may have already been critical of TYT—it is evident that Rubin's participatory audience increasingly consumed news from more right-wing sources.

Engagement of Content Creator

The Beeswarm Plot covering the categorized comments created by Dave Rubin can be found below. Each circle represents a comment. The size of the circle depends on the number of likes the comment has received. The color indicates the category of the comment.

Figure 8: Beeswarm plot – Categorized comments of Dave Rubin (for legend see poster).

Looking at Fig. 8, it becomes apparent that, in the beginning, Dave Rubin frequently replied to comments, often simply expressing gratitude, such as saying "thank you," or supporting statements with responses like "great point." However, this interaction decreased significantly, particularly after he left TYT in 2015. Since then, Rubin has made fewer comments overall, and his use of the comment section has shifted more towards providing explanations, directing viewers to other videos, making previews, and requesting donations.

Comments categorized as “Explanation,” “Preview,” “Announcements,” “YouTube,” and “Donation Request” tend to receive the most likes. Overall, the number of likes on Rubin’s comments has increased over the years, particularly since 2016, when he partnered with the media company Patreon.

6. Discussion

In this exploratory project, we aimed to understand the relationship between audiences and alternative news content creators in the development of political identity on YouTube. Challenging the traditional view of influencers, this project specifically focused on the role audiences play in shaping the political identity of content creators, a phenomenon termed ‘audience capture.’ Before engaging with the broader implications, we have also considered how Dave Rubin's ideological transformation corresponds with shifts in financial support from corporations and donation platforms. While audience capture emphasizes responsiveness to audience feedback, it is crucial to contextualize this feedback within the wider political economy of content creation on YouTube.

Our findings suggest significant feedback mechanisms that indicate how audience feedback may influence Rubin's ideological transformation. We discuss these findings in relation to further research questions and potential avenues for investigation. First, we identified two key topics that reveal a disconnect between Rubin and his audience: Trump and gay marriage. To maintain his audience, Rubin appears to have adjusted his position and communication strategies on these topics. Further exploration of these themes could provide deeper insights into how Rubin reconciled these differences. This would require a more comprehensive and coordinated approach to filtering comments and transcripts for accurate comparison, as well as a closer reading of the content related to gay marriage and Trump, especially given that Trump was not extensively addressed in this project.

Second, we found that Granger causality offers a valuable method for exploring the reciprocal relationship between audience feedback and content creation in a statistically sound manner. Future research could apply this, or similar statistical techniques, to investigate the reciprocal relationship by focusing on themes identified through bigram analysis of comments and transcripts. Although our analysis revealed a statistically significant impact of audience feedback on the frequency of certain concepts being mentioned, we were unable to assess how these concepts were discussed. It would be helpful to further expand on this analysis.

Finally, our community-based approach highlights the usefulness of co-commenting analysis and commenting continuity as indicators for assessing audience capture. We observed that a new group of individuals provides feedback almost every year, indicating a dynamic and evolving participatory community. Moreover, the ability to observe on which other videos and channels audiences have been commenting also provides a unique perspective on the audience of content creators. The concepts of community composition and continuity could be further developed to track audience capture more effectively.

7. Conclusion

Overall, while there is still much work to be done in refining our methods, our explorations during the DMI Summer School have uncovered important dynamics in the reciprocal relationship between audiences and content creators on YouTube. More importantly, we have tested methods that not only provide insights into audience capture but also suggest promising empirical avenues for studying this phenomenon further. As much of the attention remains focused on influencers and YouTube, we hope our work encourages more investigation into the role and impact of the YouTube audience.

8. References

Finlayson, Alan. 2021. “Neoliberalism, the Alt-Right and the Intellectual Dark Web.” Theory, Culture & Society, September, 026327642110367. https://doi.org/10.1177/02632764211036731.

Fisher, Anthony L. 2023. “The Intellectual Dark Web’s Descent Into Paranoia and Trumpism.” The Daily Beast, January 19, 2023, sec. politics. https://www.thedailybeast.com/intellectual-dark-webs-descent-into-paranoia-and-trumpism.

José, 2018. "Audiences and Echo Chambers, A Look at The Rubin Report". https://www.youtube.com/watch?v=IwyucS8DiiQ

Lewis, Becca. 2018. “Alternative Influence.” Data & Society Research Institute. https://datasociety.net/library/alternative-influence/.

Lewis, Becca. 2020. “‘This Is What the News Won’t Show You’: YouTube Creators and the Reactionary Politics of Micro-Celebrity.” Television & New Media 21 (2): 201–17. https://doi.org/10.1177/1527476419879919.

Peeters, Stijn, and Sal Hagen. 2022. “The 4CAT Capture and Analysis Toolkit: A Modular Tool for Transparent and Traceable Social Media Research.” Computational Communication Research 4 (2): 571–89. https://doi.org/10.5117/CCR2022.2.007.HAGE.

Schlesinger, Arthur M. 2009. The Vital Center: The Politics of Freedom. 6. printing. New Brunswick, NJ.: Transaction Publishers.

Tokmetzis, Dimitri. 2019. “How They Did It: Exposing Right-Wing Radicalization on YouTube.” Global Investigative Journalism Network. October 28, 2019. https://gijn.org/2019/10/28/how-they-did-it-exposing-right-wing-radicalization-on-youtube/.

Zeeuw, Daniël de, and Marc Tuters. 2020. “Teh Internet Is Serious Business.” Cultural Politics 16 (2): 214–32. https://doi.org/10.1215/17432197-8233406.

“Audience Capture.” n.d. The Portal Wiki. Accessed January 9, 2024. https://theportal.wiki/wiki/Audience_Capture.