Breaking the algorithmic stranglehold and how journalism might lead the way to political change
As R. McChesney wrote in The Problem of the Media, the media is “at the center of struggles for power and control in any society, and they are arguably even more vital players in democratic nations.”
The media landscape is in a state of constant evolution, bringing about unexpected developments with profound implications for democratic governance and politics. The advent of new media has fundamentally transformed the functioning of government institutions, the communication strategies of political leaders, the conduct of elections, and the nature of citizen engagement.
In the ethos of Civic Science’s mission to bridge divides, and inspire positive societal change through interdisciplinary efforts, I am launching a brief series on this Substack to examine the intersection of new media, data science, psychology, and political science and what that portends for our democracy. This is the third post in this series.
In a previous Substack piece, I examined how new media algorithms essentially rig the game for voters by trapping them in echo chambers that reinforce their beliefs while sidelining rigorous fact-checking and diverse perspectives. But what can break this algorithmic stranglehold, and how might journalism pave the way for change?
A study led by Talia Stroud from The University of Texas at Austin and Joshua Tucker from New York University, demonstrated the significant sway algorithms hold over what populates our feeds. The study found that tweaking essential elements of social media algorithms shifted the content users encountered, but notably didn't budge users’ political stance. “We now know just how influential the algorithm is in shaping people’s on-platform experiences, but we also know that changing the algorithm for even a few months isn’t likely to change people’s political attitudes,” Stroud and Tucker said in a joint statement. Their findings aren’t surprising. Research repeatedly demonstrates that voting is driven by emotion. It's not that facts are irrelevant; it's that they don’t shift beliefs.
One organization, People’s Action, is exploring a novel approach to shifting political attitudes through what they call “deep canvassing.” Its strategy? Engage in patient, nonjudgmental conversations about hot-button issues like race, immigration, health care, and the economy. The goal is to prompt a reconsideration of beliefs.
The early results are striking, finding that more than 3 percent of those they’ve spoken with have switched their support. Even more impressive, 8.5 percent of independent women and 4.9 percent of all women have signaled a shift towards more progressive policies. This method is astonishingly effective—102 times more so than conventional campaign tactics. And these impactful conversations took, on average, just 15 minutes.
The organization attributes its success to its so-called "deep canvassing" — canvassing that emphasizes genuine and meaningful conversations. The canvassing begins by asking simple, human questions: How are you doing? How are you holding up in the current economic landscape? These aren't just polite inquiries; canvassers genuinely listen. This approach gradually fosters a true dialogue, with canvassers and participants sharing personal stories and discussing what’s at stake for themselves and their communities in the upcoming election. By the end of these exchanges, a deeper internal conflict often emerges—why the person feels frustrated, uncertain about their vote, or skeptical of certain policies. At this point, the canvasser helps place these concerns into a broader context. They don’t directly try to persuade anyone to change their mind about a candidate or issue. Instead, they’ve discovered that when people begin to recognize the dissonance between their beliefs and what they truly desire, their views tend to shift—many end up embracing a more progressive perspective.
The contrasts between “deep canvassing” and typical political coverage are strikingly obvious. Typical political coverage is often sound-bite heavy and horse-race driven. The dominance of the sound bite is often blamed for the dismal state of America's political discourse. Former governor Michael Dukakis, reflecting on the 1988 presidential campaign, lamented, "If you couldn't say it in less than 10 seconds, it wasn't heard because it wasn't aired."
In today's media environment, the narrative is increasingly sculpted by a new breed of commentators who deconstruct and remix stories through their analysis. As Daniel C. Hallin wrote in Sound Bite Democracy, “now such words, rather than simply being reproduced and transmitted to the audience, are treated as raw material to be taken apart, combined with other sounds and images, and woven into a new narrative. Greater use is made of outside material, such as "expert" opinion intended to put the candidates' statements and actions into perspective, and "visuals," including both film and graphics. Unlike their predecessors, today's TV journalists generally impose on all of these elements the unity of a clear story line.” What this means is that the stories we spin about politics drift further from their original truths. We end up in a world where facts are so reconfigured that their origins become almost unrecognizable. Although Hallin's observations are from 1992, they remain exceptionally relevant today, especially when viewed in the context of new media.
As I’ve discussed before, about 32% of Americans aged 18 to 29 now turn to TikTok for their news, and that number is steadily rising. Much of TikTok’s news content is reflective, whether it’s explainer videos from mainstream news outlets that attempt to give viewers more context about the news of the day, or independent pundits who purport to counter those outlets’ “biased” or “elitist” worldviews. The app, locked in a constant battle for the viewer’s attention, allows a maximum post length of 3 minutes – severely restricting attempts to elaborate on views and put them in sufficient context. What results is an alienation of facts. Take Kamala Harris’s coconut meme for example, does anyone know where it came from? When was it said? And in what context? My guess is rarely.
In the current media landscape, the genuine dialogue and people-first approach that People’s Action cites as key to its success is notably missing. While the absence is obvious, the path to a solution is anything but straightforward.
Consider this: In 1992 the Charlotte Observer revamped its election coverage to focus on what voters truly cared about. It introduced the “citizens agenda,” a straightforward concept: ask voters what issues they want candidates to address. By prioritizing this feedback over typical campaign drama, the Observer aimed to shape coverage around genuine voter concerns rather than the usual noise of controversies and race updates.
The “citizens agenda” approach has been adopted by various newsrooms across the U.S., with reports showing it shifts the focus from game-centric coverage to more transparent, relevant journalism tailored to specific audiences. The approach, as one report noted, teaches journalists “how to build trust, how to move from problem-focused to solutions-oriented stories, how to be transparent, how to truly listen to communities in new ways, how to nurture different audiences, and how to encourage “good conflict” instead of polarizing products.” While little data exists on the impact the approach has on voters, the parallels between the “citizens agenda” approach and People’s Action’s approach are apparent – both center community members, listen to their issues, and focus on policy rather than politics.
In today’s world, where conspiracy theories and polarization seem to dominate the political landscape, it’s worth pondering how things might change if media platforms—especially the new ones—adopted a more citizen-centered journalistic approach. Could this shift help us escape the grip of algorithmic division?
 | A guest post by
|