How Black Twitter influenced Black electoral opinion during the 2020 presidential election

As of April 2021, roughly 23 percent of all adults and 29 percent of Black adults in the U.S. reported using Twitter. While a low number compared to the 69 percent of Americans and 74 percent of Black Americans who reported using Facebook, Twitter influences everything from pop culture to politics, and has been pivotal in fostering communities of like-minded individuals. Black Twitter is one of the most notable online sub-communities and has a particular knack for creating viral tweets and hashtags that raise awareness about social and cultural issues.

New research by co-author Nia Atkins examines Black Twitter’s influence on several fronts. One is its ability to set the news agenda for mainstream media outlets—such as CNN, the New York Times, and USA Today—on issues affecting Black communities. Another is Black Twitter’s influence over Black voters within the realm of electoral politics, specifically during elections. While her research has concluded that Black Twitter, in the aggregate, lacks agenda-setting power over the mainstream media, it also suggests that Black Twitter may impact the discussions and electoral opinions of Black voters. Thus, Black Twitter could have a particular socio-political role—one which deserves additional attention from media outlets and political candidates interested in reaching Black voters.

The making of Black Twitter

In the modern pre-Twitter world, getting involved politically often consisted of in-person interaction—through methods such as door-to-door canvassing, petitioning, and passing out fliers. Other pre-social media organizing practices took place over the phone. One example of early phone-based organizing is the phone tree—a popular pre-email method in which, for example, five people would each call five people, each of whom would call an additional five people (and so on). While these face-to-face and phone-based methods are still in use today, social media platforms have allowed people historically excluded from mainstream political organizations—such as Black people and other marginalized groups—to engage in large-scale political discourse and advocate for what they believe in.

Since its inception, Twitter has provided a decentralized mechanism through which people can immediately and continuously share their points of view both within specific communities and with wider society. In fact, as the Federal Trade Commission argues in an ongoing legal case against Facebook, Twitter’s design fosters topic-based interaction with strangers, while conversely, platforms such as Instagram and Facebook facilitate connections with friends and family.

While Twitter’s role as a forum for the discussion of news, pop culture, and politics is not new, the use of Twitter for such a purpose by Black people has gained attention and interest from observers. It has been framed by its collective compilation of Black American voices within Twitter, commonly known for its appeal to socio-political commentary and critique on a variety of issues, and as felt by Black people. Though other identity-based communities exist on Twitter, Black Twitter remains the most well-known of such communities due to its ongoing success in grabbing the attention of outsiders. The ubiquitous expression, “Black Lives Matter,” for example, is a slogan and hashtag that was popularized by Black Americans on Twitter through this specific online community in the aftermath of the 2014 police shooting of Michael Brown, a Black teenager. But can such online, collective action, like the one leading to the hashtag “Black Lives Matter” more generally influence Black electoral opinion and the mainstream news agenda? Recent research on the Black Twitter community may provide some preliminary evidence regarding the answers to such questions.

The impact of Black Twitter

As part of her research, Atkins designed a research methodology that incorporated presidential candidate polling numbers among Black voters and Black Twitter sentiment within the context of the 2020 Democratic primary election. Sentiments were calculated using a lexicon-based sentiment analysis package called “sentimentr” developed by data scientist Tyler Rinker. This electoral contest was chosen as an appropriate case study due to the overwhelming majority of Black voters who supported the Democratic party over the Republican party, the relevance of social media and Twitter in particular to recent presidential elections, and the growing pertinence of Black Twitter to the cultural and socio-political zeitgeist. Sentiment analysis on a sample of tweets during this period was conducted to gauge Black Twitter’s feelings about the field of candidates. More specifically, the average sentiment score—a number falling between -1 and 1—was calculated for all tweets per month per candidate. Finally, the relationship between monthly candidate sentiment scores and monthly candidate polling numbers was assessed through linear regression.

Regarding the latter question of the previous section, the influence of Black Twitter discussions on the mainstream media agenda was also evaluated based on each platform’s discussion of two policy issue areas—criminal justice reform and reparations for American slavery—during the 2020 Democratic primary presidential election. These two policy areas were selected on account of their topical relevance to the general political conversation at the time, following renewed discussions of Black Lives Matter and anti-racism, as well as their specific salience to Black communities. The frequencies of a set of criminal justice reform and reparations-oriented keywords and phrases like “reparations,” “mass incarceration,” and “prison reform” were compared between a sample of Black Twitter tweets and a sample of articles from three mainstream newspapers. Finally, the relationship between the two variables was evaluated using a linear regression model.

Black Twitter is a politically noteworthy force

The results of the first regression model that measured Black Twitter sentiment and polling numbers indicate that a one-point increase in Black Twitter sentiment in one month resulted in a 12.1 percentage point increase in a candidate’s polling numbers among Black voters the following month. Thus, this model provides some evidence of a relationship between Black Twitter sentiment and Black voter support. It suggests that Black Twitter discourse around political candidates could impact broader Black electoral opinion to a potentially election-altering scale.

The second regression model, which focused on the policy issues, found a negligible relationship between Black Twitter issue-based conversation and mainstream media issue-based reporting. The results indicate that mainstream news organizations may not find policy discussions on Black Twitter newsworthy or may simply fail to monitor such conversations at all. Whatever the mechanism, there wasn’t much social media impact on policy discussions relevant to Black Americans.

Both of these data results assert preliminary evidence regarding the influence of Black Twitter when it comes to Black voting behavior and mainstream news agendas. Despite its reputation as a tool for raising socio-political and cultural awareness, Black Twitter appears to be generally ineffective at impacting the reporting of mainstream news outlets on such socio-political issues. This inefficacy was revealed specifically during a presidential primary election, a period in which policy issue-based dialogues were particularly relevant. Nevertheless, Black Twitter’s apparent relationship to the opinions of Black voters should encourage conventional news outlets to—at minimum—observe Black Twitter, particularly during presidential elections, for some insight into both the candidates that Black voters may support and the policies they may discuss. Ideally, however, these conclusions will also motivate the media to go even further by incorporating policy-oriented topics of Black Twitter conversation into their news coverage.

While Black Twitter’s relationship to Black voting patterns is a much less popular topic of discussion, this research sheds light on the correlative nature of such a relationship. If Black Twitter sentiment can impact polling numbers among Black voters by as much as twelve percentage points, the medium should be considered a relevant actor within the realm of electoral politics. Whether the shift in polling numbers is caused by Black Twitter itself or not, the conclusions suggest that Black Twitter may predict this shift.

These research findings demonstrate how one Twitter community of marginalized people can play a role in shaping the political landscape. With these conclusions in mind, journalists to voters to policymakers should take notice of Black Twitter users as a politically noteworthy force during election times. What the study’s findings indicate is their potential as voters or a populace of concerned citizens to predict shifts in Black voters’ support of a certain politician. Such findings demonstrate the influence of online platforms and should also encourage future candidates to better assess social media discourse, especially among Black online users, over the course of their campaigns as a supplementary gauge of wider demographic support.

Facebook is a general, unrestricted donor to the Brookings Institution. The findings, interpretations, and conclusions posted in this piece are solely those of the authors and are not influenced by any donation.