Muted Orange

storyteller series (1).png

Process & Practice Reflections

Narrative Strategy for Racial Equity

Melinda Weekes-Laidlow

Guest Author

Muted Salmon

Virtual reality is in the eye of the beholder. Filmmaker and entrepreneur Chris Milk optimistically calls VR an "empathy machine," while the Wall Street Journal warns that VR's "hype is about to come crashing down." Across this spectrum of opinion, the only thing everyone can agree on is that it's a newly accessible medium offering novel experiences for audiences.

Do these experiences lead to empathy? If you define empathy as, "knowing and feeling what another person knows and feels," of course some VR stories do. So do some books, radio programs, and movies - each in their own way. Saying VR is an empathy machine is like saying a paintbrush is an art machine. There are myriad ways to generate aspects of empathy with media, just like there are infinite ways to create forms of art with tools.

>Saying VR is an empathy machine is like saying a paintbrush is an art machine.

## What's Special about VR?
The term "[affordance](https://medium.com/r/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FAffordance)" refers to how an object or environment offers the opportunity for us to perform an action. A door handle affords pulling, a knob affords being twisted, and a button affords being pushed.

Every medium has its affordances. Books use the unit of words, which are symbolic and lacking in sensory information. Words afford a writer the ability to express abstract ideas, describe the nebulous internal thoughts of a character, and leverage the subjective imagination of a reader. Radio is pure sound, so it fosters intimacy through sensorial minimalism, and affords multi-tasking in today's busy world. Video is audiovisual rectangular content, which has afforded directors like Stanley Kubrick control of our attention within a scene without each viewer looking wherever they want. Another acclaimed director, Alfred Hitchcock, often referred to his style of video as "pure cinema" - using camera movement, editing, music and sound to tell stories that would be impossible in any other medium.

![BLOG-VR-Img1](https://harmony-strapi-storage.s3.amazonaws.com/BLOG_VR_Img1_f6bc499d76.png)

How would the infamous shower scene from Psycho be directed in VR? What is "pure VR"? What are VR stories that would be impossible in any other medium? The unique storytelling units of VR are space and 1st person perspective.

![VR-Blog-Img2](https://harmony-strapi-storage.s3.amazonaws.com/VR_Blog_Img2_6c0994aaea.png)

Two of the most discussed experiences in VR use these spatial and 1st person affordances to pave new avenues for empathy. The Syrian refugee crisis has been described in articles, radio shows, TV news broadcasts, films, and more, but in Clouds Over Sidra you are transported in VR to experience the space - from the huge scale of the camp to the tiny room of Sidra and her family. You look down at Sidra's young brother crawling around the floor, and you look over at her family huddled over dinner in the corner of the room. You learn and feel what Sidra knows and feels by empathizing through space and perspective.

In 6x9: An Immersive Experience of Solitary Confinement, we experience something similarly novel. Words, photos, and videos about solitary confinement afford avenues of empathy, but 6x9 virtually places you in a narrow cell, looking around for things to do, and witnessing hallucinations in 1st person. You learn how isolated and desperate inmates are in such a limited space, and you feel their claustrophobic mental strife by seeing through their delirious perspective.

>You learn and feel what Sidra knows and feels by empathizing through space and perspective.

![VR-Blog-Img3](https://harmony-strapi-storage.s3.amazonaws.com/VR_Blog_Img3_0c7705c96b.jpeg)

## Emotion is the Key Ingredient
Whether text, audio, video, or VR, there will be little empathy for characters without an emotional rise in audiences. In a media landscape of breathtaking lines of poetry, unforgettable chorus lines we sing along to, and Pixar movies that leave us sobbing uncontrollably over a talking fish's challenges, how is VR uniquely pulling on our heartstrings?

One simple and obvious answer is that VR is new, and new experiences can be emotionally exciting. There is a dismissive framing of [novelty as a gimmick](https://medium.com/r/?url=http%3A%2F%2Fwww.pcgamesn.com%2Fhow-why-and-when-vr-will-fail), but with infinite potential stories to be told, there are infinite ways we can newly experience scenarios through the affordances of space and 1st person perspective. These novel experiences can add to our emotional understanding of situations and empathy for people involved.

Another channel of emotional arousal specific to VR is its hold on our attention. The more of our sensory bandwidth a medium uses, the more it will consume our attention automatically. VR surrounds us in audiovisual information, which results in a faster ramp up to immersion in an experience that is harder to be distracted from. And until VR starts integrating pop-up advertising & multi-tasking (which is coming fast!), audiences are focusing on a single-task for several minutes. In today's attention economy, this is extremely rare and valuable.

We're frequently inundated by the cacophony of social media, notifications, and a "[fear of missing out](https://medium.com/r/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FFear_of_missing_out)." Along comes a medium where we are immediately in another world, stuck in there, and focused on one thing. Critics say VR is isolating, but the rapid, persistent immersion in a singular experience is cathartic and emotionally powerful for a society battling constant distraction.

## Goals of Empathy
Empathic stories across all mediums are important and powerful. They help us learn and feel for other people, but the impact may sometimes be contrary to our long-term societal goals.

Many stories revolving around social issues guide our empathy to a specific destination. The story directs us to empathize with a different gender, race, nationality, or some "other" group of people. There are good people in pain and bad people causing the pain. The good people are the empathic destination, the bad people are inhuman.

But some stories makes us more empathic people in general, growing our capacity to empathize overall. They contain emotionally complex characters with ambiguous morals, training readers to imagine and infer what a character thinks and feels. These stories exercise readers' empathy like a muscle, building empathic capacity for anyone in any situation. This critical difference between empathic destination and empathic capacity was demonstrated in a [study by Emanuele Castano](https://medium.com/r/?url=http%3A%2F%2Fwell.blogs.nytimes.com%2F2013%2F10%2F03%2Fi-know-how-youre-feeling-i-read-chekhov%2F) showing how reading literary fiction builds more empathy than reading popular fiction. Empathic destination is direct and specific to a particular situation, while empathic capacity is broadly applicable and cultivates a more unified and empathic society.

Empathic capacity makes us more empathic to those involved in the next crisis, to the annoying colleague, to the homeless people we pass daily on the street, and most importantly to the people most different from ourselves.

>Empathic destination is direct and specific to a particular situation, while empathic capacity is broadly applicable and cultivates a more unified and empathic society.

Isn't that the ultimate goal of these empathic stories, VR and otherwise? We want to bring people together, to decrease the rife polarization in our society, and to change hearts and minds to change the world. Stories direct our perspective through abstract words, minimal sounds, and virtual space to remind us that we can think, feel, and act beyond ourselves for everyone's benefit.

Google_Daydream_1042.jpg

VR: Going Beyond the Empathy Machine

Harmony Labs

wired.png

Does more engagement mean your message is resonating with more people? For politicians campaigning on social media, focusing too much on shares and likes might be a mistake.

Viral Political Ads May Not Be As Persuasive As You Think

Bright Blue

We recently wrapped a project with advocates, activists, and organizers in the Upper Midwest to map the current landscape of civic information consumption. The fragmentation of media systems has left many wondering: how do people in the U.S. get information in what many perceive to be local media deserts? Using the [Narrative Observatory](https://narrativeobservatory.org/), we explored how audiences engage and consume this info in the Upper Midwest — and where untapped engagement opportunities may exist to reach them. We are excited to showcase the findings in this new [public report](https://docsend.com/view/r39933mm4bjjungu).

The biggest takeaways?

-   There are two types of people who consume media online — people who **read** and people who **watch**.
-   Media consumption is not segmented by topics or social issues, but instead by people who consume news vs. people who **don’t** consume news.
-   Coverage of local issues comes from a **national POV**. When local content emerges, it still comes from major news sites.
-   The op-ed of the 21st century is in a diverse set of places. To reach people with civic info, look not to news, but to **non-news ecosystems**.
-   People search for civic information relevant to their **day-to-day lives** (emissions testing, post office, school closures) — not  for broad, macro-level civic topics.

If you are interested in watching a live presentation of these findings, or in contributing to our ongoing efforts to explore the data around civic information, please [get in touch](https://www.harmonylabs.org/contact).

civic-info-blog.png

Exploring civic information ecosystems online to identify "the new op-ed"

Media Consumption in the American Midwest

Deep Salmon

Over the last few years, we have learned that health is a deeply felt, universal experience intersected by a number of systems — political, racial, economic. From food to insurance to fitness to vaccinations, we all have different understandings of what “health” is, seen through the values we hold and the specific stories we consume. Given these differences, how might we move towards a future where people in the U.S. believe that healthcare systems are not only accessible and available to all, but also recognize and rectify past and current injustices?

As the nation’s largest philanthropy dedicated solely to health,  [The Robert Wood Johnson Foundation](https://www.rwjf.org/)  (RWJF) has been working to help achieve health equity and expand opportunities to pursue the best health possible. In 2021, RWJF enlisted our support to help with their narrative efforts around this goal. Using the  [Narrative Observatory @Harmony Labs](https://narrativeobservatory.org/), we’re building an audience-narrative architecture that allows RWJF and its partners to develop content that will change the narrative on health equity for key audiences.

We are excited to share this  [public report](https://docsend.com/view/r2xpjfisb5usnsmb)  about our initial research findings on audiences and the health equity narrative landscapes they inhabit as a result of their media consumption choices. These findings are already helping us surface the most promising story opportunities for testing and, ultimately, the kinds of content RWJF and its partners might make to guide audiences toward a more equitable health future.

Our next steps, working alongside  [Story Strategy Group](https://storystrategy.co/)  and a diverse group of storytellers, will be to create content we can test to accomplish this goal. Follow this space for results of this testing in the upcoming months, as well as more about how we approached this research and the insights gained over the course of this project.

in-search-of-healthy-narratives.png

Creating an audience-narrative architecture for health equity

In Search of Healthy Narratives

Deep Blue

> “Bias in research has real-world pains that disproportionately affect people that look like me. Dark-skinned people. Women… Having the equity lens to know that people respond to different research methods differently — it’s important. It’s expensive. It’s time-consuming. But it’s necessary… to ensure that we receive accurate results, that we treat all groups the same, we don’t lump folks together, and that we’re better able to interpret the data.” -Sian Lewis

At the center of Harmony Labs is research. Our research utilizes the [Narrative Observatory](https://narrativeobservatory.org/), data infrastructure about U.S. media consumption, to provide strategists and changemakers with insights necessary to create effective, constructive stories. Since late 2021, we’ve been hard at work building an [audience-narrative architecture](https://medium.com/harmony-labs/in-search-of-healthy-narratives-889aa256d94c) for the Robert Wood Johnson Foundation’s (RWJF) efforts to advance health equity in the United States.

This work requires an unflinching commitment to _data equity_ in research design and modeling, and to implement standards of practice that clearly and thoroughly engage with what data equity _should_ be. But barriers like a lack of incentive, a lack of resources, and a lack of concrete steps have led to data equity being historically omitted from most research best practices, and rectifying harmful narratives is often rooted in rectifying harmful research design. [Janay Cody, PhD](https://www.harmonylabs.org/janay-c-cody) and [Sian Lewis, MS](https://www.harmonylabs.org/sian-lewis), both data scientists and equity consultants as well as members of the Harmony Labs Advisory Board, are currently working as equity advisors for our project with RWJF. In a recent conversation, Science Director [Riki Conrey, PhD](https://www.harmonylabs.org/riki-conrey) spoke with both Dr. Cody and Lewis to discuss:

What should  data equity actually look like? How can we, and other researchers, enact practices that hold us accountable?

We hope the insights from their conversation can act as a resource and conversation starter for all researchers, changemakers, advocates, and storytellers.

> “Data equity sits right in the middle … of strategy and science.” -Janay Cody, PhD

## Data equity is:

- Being conscious of outcomes — research and the tools created from it must, by design, afford outcomes that uplift and empower communities, rather than disparage or drive existing, harmful stereotypes.
- Instituted throughout the lifecycle of any data product — from actual practitioners doing the work, to collection, to interpretation, to analysis.
- Seeking “sameness” in measurement — which may not always mean sameness in method to ensure that concepts are captured equitably across different groups or cultural contexts.
- Interrogating assumptions in real-world settings — interrogating whether models are appropriate, or if existing parametric assumptions are actually relevant/useable.
- Informing research design — having the right kinds of conversations across an array of perspectives to ensure the right questions are being asked of the data.
- Building the work into a project’s budget — respecting time, effort, and resources.
- Practical — real choices have to be made in designing the work, and trade-offs will necessarily occur.
- Participatory action research — partnering with the communities that are actually going to be affected by the data.

> “I actually [talk] to the folks who are affected by the models that I’m building many times during the process. It is a part of my work. I literally build it into all my plans. It’s a part of the budget. It’s an essential part of it. I don’t think that folks have that lens at all. And I wish that they did. I think that a lot of the equity issues we went into would be alleviated if people took the time and spent the resources to do this.” -Sian Lewis

## Data equity is not:

- Diversity. Diversity is representation, but equity is a **measurable** improvement in outcomes.
- Exclusively linked to the end product and how it affects people — data equity is instituted from the **ground up**.
- An afterthought or “filter” to lay across data measurement.

> “Data equity is not diversity. Diversity is making sure there is representation of people from different backgrounds. That’s great. That is not equity. Equity is ensuring a measurable improvement in outcome, such that typical characteristics that we care about, like race and gender and age, don’t statistically determine particular outcomes.” -Janay Cody, PhD

## Modeling

Algorithms can be biased by design. They are simplifications of the world and use historical data to predict the future — essentially, mathematical stereotypes. Delivering models with an equity lens is hard, but possible, and often rooted in **clarity** and **documentation**.

- Be clear about the purpose of the model and what it should be specifically designed to do.
- Trade-offs will be necessary — improving the accuracy of those who typically are **unseen** means a decline in the accuracy of individuals who typically are seen the most.
- Documentation from start to finish — don’t gatekeep your models or silo your expertise, make the instructions clear and easily accessible.
- Be iterative — models should be updated and checked regularly over time.

> “[Data equity] is … empowering because it finally gives a voice to people. I like to describe myself as the wind. You don’t see me, you just feel my effects… My goal is to give the evidence to the people who need it to actually say the thing that they need to say, make the claim, change the conversation… My goal is not to be seen. It is to be felt through the insights that I provide that are useful for people who need to accomplish a task. That’s very different than the way we think about research and academia and the way we were trained.” -Janay Cody, PhD

## Making Design Decisions

Listen to Dr. Cody walk through a recent example in which the structure of an experiment was questioned and restructured after applying an equity lens to the research design decisions. The transcript is available below.

![Frame 1244 Large.jpeg](https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/1364878768&color=%23ff5500&auto_play=false&hide_related=false&show_comments=true&show_user=true&show_reposts=false&show_teaser=true&visual=true)

_“Oftentimes groups have to balance wanting to demonstrate their impact with actually wanting to make one. In digital campaigns, one of the ways in which they target is through zip codes, and for Black folks, and I’m thinking of Black folks in Pennsylvania, 3% of the zip codes encompassed the majority of that population. When it comes to designing a test, an experiment, to be able to demonstrate your impact in an election or at the end of a campaign, that’s very difficult to do with your zip code, the possible universe of zip codes being [shrunk] so much. So a lot of the things that I’ve had to do is talk about how we’re going to design an experiment. You can create your target universe based upon having a Black population, for example, that is, [one] that makes up at least 30% of the total population of the zip code, where there is a minimum of at least 5,000 respondents, and then you would find treatment and control that way so that you are not taking out most of your sample for an experiment. … There’s this other possibility of making the decision to not to do an RCT and instead choosing an alternative method, like a quasi experiment where you construct your treatment group strategically instead of randomly, and then you construct a comparison group through either propensity score matching, regression discontinuity… if time is on your side, you do … different estimations. But it’s the very real design question of, ‘Do I do an experiment at all, as I’m trying to balance: I want to make an impact in a Black community, but I also need to be able to demonstrate it to my funders and to people who want to support the scaling up of this particular initiative.’”_

Though far from comprehensive, these foundational principles of data equity across research design and modeling, as well as an understanding of what data equity is _not_, have informed and are informing our current and ongoing work with RWJF. Follow this space for more updates concerning both this project and others.

&nbsp;

----------

&nbsp;

**_Dr. Janay Cody_** _is a behavioral data scientist and data equity advisor. Dr. Cody empowers results-driven leaders with equitable data policies, inclusive research, and measurement strategies to manage their impact._

_Dr. Cody is highly regarded for the analytic solutions she created for philanthropies including_ [_the Kettering Foundation_](https://www.kettering.org/wp-content/uploads/9_CONNECTIONS2007_DobleCody.pdf)_; government research teams including the_ [_Administrative Office of the Courts on behalf of the Judicial Council of California_](https://www.courts.ca.gov/documents/PTC_phase_II_web.pdf) _and_ [_the National Assessment of Educational Progress_](https://www.nationsreportcard.gov/reading_math_2015/#?grade=4)_; and data and tech companies including_ [_Catalist_](https://catalist.us/who-we-are/)_, Movement Labs,_ [_Harmony Labs_](https://www.harmonylabs.org/people)_, and Analyst Institute._

_Dr. Cody received her PhD in political science from the University of Notre Dame, du lac._

&nbsp;

----------

&nbsp;

**_Sian Lewis, MS_** _is a data scientist and researcher who serves as the Head of Research at A — B (_[_www.abpartners.co_](http://www.abpartners.co/)_). Also, she serves as a political data scientist by employing predictive analytics on Democratic political campaigns. Her research focuses on narratives, inoculation, and persuasion. She formerly served as the data science and analytics team manager for the Department of Defense Healthcare Management System Modernization project (DHMSM) at Booz Allen Hamilton. Her recent accolades include the 2021 Society of Women Engineers Spark award, the 2021 Black Engineer of the Year Modern Day Technology Leader Award, the 2020 Women of Color in STEM award, and the 2019 DC FemTech award._

data-equity.jpg

Professional data equity experts on how they practice

Equity in Research… For Equity in Health

Harmony Grey

The work of advocacy is commonly conceived as an effort to influence or win people over with a different point of view —often our own point of view—to change how people conceptualize themselves, others, or the world we share, in support of some important goal. The pioneers and perfecters of advocacy as mass influence were the missionary church, military intelligence, and consumer capitalism. These institutions still linger in the language of advocacy. We speak of conversions, engagement, strategy, tactics, targets, impact, our base and opposition, of “[winning hearts and minds](https://en.wikipedia.org/wiki/Winning_hearts_and_minds).” Implicit in this language is a view of human beings as instruments of larger powers and purposes, tools that must be moved, tuned, or activated by forces external to their will. And this implicit view is often at odds with the explicit goals of advocacy, especially when those goals are deep, complex, and long term, like changing how we relate to planet and place, for instance, in response to the climate crisis. Instrumentalize people often enough, and their capacity for independent discernment or autonomous moral action will certainly diminish. Is there another way?

We think so.

Some work we recently completed with support from the [Robert Wood Johnson Foundation](https://www.rwjf.org/) in collaboration with [Story Strategy Group](https://www.storystrategy.co/) points to a different way to think about advocacy. The work was about how people in the U.S. encounter health and equity in the media; the health and health equity narratives that arise from audience encounters with media; and how media content might help shape and shift those narratives. There were a couple big surprises.

But before we dig in, we need first to ask what audiences are, and what role they play in advocacy work. Google tells us that audiences are “the assembled spectators or listeners at a public event.” Once upon a time, audiences formed when people assembled in person, in the same time and place. Nowadays, we tend to assemble in, with, and through media, virtually and asynchronously, often and everywhere. This can make defining what is coherent and distinctive about an audience challenging. In our work, we’ve found that values are excellent predictors for how people assemble into media audiences. This makes intuitive sense. Values are the things we ascribe particular importance to. They’re like goals or happy endings. So, of course, we seek out and engage with media that bring to life our happy endings, that reflect and reify what we value.

For this reason, in our work, we use values (specifically clusters of values derived from [Shalom Schwartz's theory of basic human values](https://scholarworks.gvsu.edu/cgi/viewcontent.cgi?article=1116&context=orpc)) to segment people into audiences, and we use audiences to explore and describe how people engage with media. At the highest level, our segmentation comprises four audiences, placed on the x-y axes to the right. Audiences in the top two quadrants hold values related to exploring and creating, while audiences in the bottom two quadrants value protecting and preserving what they have. Audiences to the left tend to orient more toward the collective, whereas audiences to the right skew more individualistic. Like all segmentations, our audience segmentation gives advocates a way to reason about a reality where every human being is unique, and to strategize about how to speak with broad groups of people who share values, attitudes, or behaviors. Unlike segmentations based on demography or political support, however, this audience segmentation has the added benefit of corresponding to how people actually exist inside of today’s media culture. So, when the time comes to make a campaign or to buy media, it can tell us about the content, voices, and platforms that are already resonating with audiences.

![image (20).png](https://harmony-strapi-storage.s3.amazonaws.com/image_20_e7634ff9a7.png)

Now that you have the basics of how we think about audiences, let’s return to the two big audience-related surprises in our recent health equity project. The first surprise was that the social media content we created actually transported all U.S. audiences to endorse a narrative on health equity that:

1. Acknowledged racial inequities in our current healthcare system;
2. Demanded we address those inequities; and
3. Imagined a future where patients of all races felt respected and had access to the care they needed and wanted.

The particular surprise here was that our content transported even the audience we call Don’t Tread on Me. This is the audience whose values center authority, and which tends, on average, not to acknowledge that white people are treated better than people of other races in healthcare settings. So it caught our attention when Don’t Tread on Me not only acknowledged the harms of medical racism, but also endorsed change, even despite perceived personal trade-offs. Stories featuring choice and control served as narrative entry points for Don’t Tread on Me. Each of our other three audiences had different entry points, which you can learn more about in our [full report](/project/health-equity) on this work, which includes the stories and social media content we used for testing.

But it wasn’t only the entry points to the narrative that differed by audience. Different audiences also gravitated more strongly toward different parts of the narrative. We could see this, because for testing we broke the narrative down into the three parts described above. These three parts roughly correspond to the beginning, middle, and ending of the positive narrative on health equity this project seeks to grow. Breaking the narrative down in this way allows us to move beyond testing for simple message agreement, to start observing more about how people respond to stories in their day-to-day media interactions on different platforms and devices. This differentiation in the way audiences gravitated toward narrative beginnings, middles, and endings was the other big surprise from the project. And this is the finding we think suggests a different way to think about mass advocacy.

So what exactly did we find? When we averaged the effects observed in audiences from treating them with social media content, we found these effects generally corresponded to the values and health-related narratives that define each audience. Which is to say, for example, the Don’t Tread on Me audience gravitated most strongly to the second part of the narrative, the part that demands change. This accords with Don’t Tread on Me’s value on authority and strong leadership, their orientation toward action, especially in the face of adversity and against perceived opposition or enemies. The audience we call Tough Cookies, by contrast, which values social order, safety and security, was most attuned to the threat of inequity in healthcare, and so gravitated most strongly to the first, problem-focused part of the narrative. Whereas the imaginative, autonomy-focused audience If You Say So gravitated most strongly to the third part of the narrative, which imagined an equitable future.

![image (20).png](https://harmony-strapi-storage.s3.amazonaws.com/image_20_dba532bd7f.png)

Perceiving a problem, demanding change, and imagining pathways to effect change are essential components in any narrative that supports a new future. And inside of this narrative structure, there is a place and potential contribution for each of our four audiences, one which does not require a change in their basic values, but rather brings those values into the service of a new future as features, not bugs, as strengths to power the different parts of the process of change. This is good news, because every society is bound to contain a diversity of values. Not everyone will live all the time in the values space of Community, Universalism, and Benevolence, which define the audience we call People Power. And advertising, at least at the scale of most advocacy work, is probably not powerful enough to occasion durable changes in what people value.

When commercial game designers design a game, they often think about and test how best to integrate the behaviors of all potential players, even the lurkers and loners who shun goal-oriented play. This is not only more inclusive, it’s also good business, insofar as it creates a game that accommodates a broader base of players. In the same way, we imagine a model for mass advocacy that looks less like advertising and more like organizing. Less like dividing the world into us versus them and engineering the “us” to consent to a limited set of viewpoints, and more like designing narrative spaces that reflect how people with different experiences and values collectively see themselves, others, and the world we share. In this way, more people will feel more at home in the narratives that are likely to afford a new future, and also have a role to play in propagating those narratives.

What would this look like in practice? This we need to continue to explore and discover, together with more partners working not just on equity in healthcare, an issue most people in the U.S. can agree requires attention. But we already know some of the questions we need to ask to evolve our practice, to move us away from the notion that there is always some percentage of the population we need to instrumentalize or “fix,” in order to realize some brave new world. For instance:

- Can we get curious about people who hold values that diverge from our own? And how do we structure research for advocacy that supports and sustains this curiosity, even in the face of challenging findings?
- Can we imagine and hold a vision for the future that does not require a fundamental shift in or erasure of the values of any audience? How can we express this vision in narratives that are both authentic to what we value and likely to resonate with other audiences?
- Can we create content that is an invitation for more than one audience to participate in the future we envision, and that succeeds in organizing them to support and propagate the narratives required to realize it? Can we do this without generating or perpetuating harm, and without working against our larger, long-term goals?

Asking these questions and evolving our practice may just be the difference between actually making progress on the issues we care about in a way that brings everyone along, or continuing as a people to fragment and polarize as our planet burns. If you’d like to join us in our exploration of what an audience-centered model for mass advocacy could look like, and how it could start to enact the positive, pluralistic, inclusive society much of the work we support strives to bring into being, [please be in touch](/contact).

Twitter Cover V5.png

Audience-Centered Advocacy : Toward a New Model?

Brian has a record of establishing innovative organizations at the intersection of tech and social impact. Before Harmony Labs, he served as managing director at [Institute of Play](http://www.instituteofplay.org/). He also served on the core founding teams for [Quest to Learn](http://www.q2l.org/), a New York City public school that leverages game design to make school engaging and culturally relevant for young people; and [GlassLab Games](https://www.glasslabgames.org/), a Silicon Valley development shop transforming digital games, like [SimCity](http://www.simcity.com/), into classroom learning environments with real-time formative assessments built in. Brian was trained as a poet. He has worked as a speechwriter, a management consultant, a professor, cook, and gardener, and built by hand a timber-frame cabin in the woods, where he spent five years reading, writing, and meditating.

Brian Waniewski

Executive Director

The Harmony Labs Director of Science, Riki comes to the team with an extensive background in data science for social good. Most known for her work on narrative research projects like [Story at Scale](https://www.storyatscale.org/) and the [Peoria Project](https://thepeoriaproject.org/), Riki has joined Harmony Labs to help researchers connect artists to the insights they need about their audiences. Riki holds a PhD from Northwestern University in Social Psychology but only because data science wasn’t invented yet when she was being educated.



Riki Conrey

 Science Director

Harmony Orange

2022 was a period of significant growth at Harmony Labs. We continued to evolve our research approach, which produced valuable new audience-based insights, narrative and network analyses, and empirical validation of cultural strategy and content
for dozens of partners, both old and new.

At the center of this work was the Narrative Observatory: data infrastructure purpose-built to help artists, activists, and advocates tell better stories through and with media. In the coming weeks, we look forward to officially launching an [all-new version of this tool](https://narrativeobservatory.org), culminating 4 years of research and prototyping with support from the Bill & Melinda Gates Foundation and the Robert Wood Johnson Foundation. 

Building from these tools and in close collaboration with our community, we are proud of what we accomplished together and have highlighted some standout insights from our research below.  

## The Narrative Observatory
In the last year alone, the Narrative Observatory has been used for innovative new research on guns, immigration, denuclearization, health equity, news consumption, and more. We’ve selected some standout insights below: 


### It is possible to tell a story that engages and moves all four of our values-based audiences by giving people a role in a possible future.
[Robert Wood Johnson Foundation](https://www.rwjf.org/) enlisted our support to build an audience-narrative architecture that could help change the narrative on health equity. In partnership with Story Strategy Group, we built and tested hundreds of new storytelling approaches, some of which even transported our hard to move Don’t Tread on Me audience towards a target narrative of health equity. Explore the complete findings [here](https://narrativeobservatory.org/strategy-validation). 

### Immigration-related content online can be divided into two major categories: stories about immigration and stories about immigrants.
This finding was one of many uncovered during our partnership with [United We Dream](https://unitedwedreamaction.org/about/) to study how Latinx people encounter stories about immigration in U.S. media, and our [full research](https://harmonylabs.org/project/immigration-stories) is already helping inform communication strategies for pro-immigration advocacy. 

### Audiences already know a world that includes nuclear weapons is risky, thanks to video games and entertainment TV—but they don’t, however, believe a nuclear apocalypse is avoidable.
We completed [new research](https://harmonylabs.org/project/rewriting-nukes) with our legacy partner [Nuclear Threat Initiative](https://www.nti.org/) (NTI) on denuclearization in the U.S. and created audience-specific cultural briefs for a group of artists and media makers to retell the nuclear story.

### For Latino audiences who consume YouTube, the highest reaching content cluster—reaching 18% of this audience—features “hot takes” style news analysis, punditry, or “rants” with edgy, assertive, and overtly masculine qualities. 
We partnered with [EquisLabs](https://www.weareequis.us/en-US/labs) and [Cultural Engagement Lab](https://www.engagementlab.org/) to research the polarizing content targeted at Latinos and identify channels with possible in-roads (or obstacles) for narrative intervention. The major finding was [four distinct content ecosystems](https://harmonylabs.org/project/info-ecosystems), each with unique engagement opportunities and challenges. 

### Coverage of local information and issues comes mostly from a national perspective. Local content does emerge, but mostly from within a major national lens.
We completed [new research](https://harmonylabs.org/project/civic-info) to better understand the current landscape of civic information consumption for those who aren’t already consuming news, and where untapped engagement opportunities may exist to reach them. 

### 3.5% of all adults in the U.S. consume gun-related content on YouTube daily. 
In partnership with the [Brown-Lifespan Center for Digital Health](https://digitalhealth.med.brown.edu/), we generated a [groundbreaking report](https://harmonylabs.org/project/gun-media) on U.S. gun culture online, also published in the [_Preventive Medicine_](https://www.sciencedirect.com/science/article/pii/S0091743522003073) journal. This research offered a first look at the narrative landscape around guns and firearms in America, with the goal of laying the groundwork for future intervention and sustainable firearm safety.

## Communities & Tools for Impact

- We continued to make our one-of-a-kind media data available to academics via our Research Network and a safe, secure analysis platform, adding new researchers from [George Washington University](https://www.gwu.edu/), [Stanford University](https://www.stanford.edu/), [University of Utah](https://www.utah.edu/), [Hertie School](https://www.hertie-school.org/en/), and [Rethink Media](https://rethinkmedia.org/).

- We collaborated with a number of different artist communities through our work with [The Butterfly Lab for Immigrant Narrative Strategy](https://www.raceforward.org/practice/arts-narrative-culture/butterfly-lab-immigrant-narrative-strategy) and [Narrative Initiative](https://narrativeinitiative.org/). Through these partnerships, we helped cohorts of artists, educators, and advocates create new storytelling strategies to counter problematic cultural narratives. 

- We grew our [Advisory Board](https://harmonylabs.org/about#our-team), including engaging them to help strengthen our [equitable and participatory research practices](https://harmonylabs.org/news/equity-in-research-for-equity-in-health). We also expanded our Board of Directors with two new members, Maryl Widdows and Omar Wasow.

- Finally, a key part of our mission remains sharing our work widely and publicly, and in the last year we expanded our efforts to increase the visibility of our work through our blog, [webinars](https://us02web.zoom.us/webinar/register/WN_U9s73r0tRS-Mk5AeNMERgA), [roundtables](https://innox.berkeley.edu/event/innovation-x-roundtable-algorithms-and-ai-for-media-in-the-21st-century/), [keynotes](https://pages.awscloud.com/NAMER-event-OE-imagine-nonprofit-unlock-ai-ml-2022-reg.html?Languages=German), and new websites. Helping make this possible were creative collaborations with [Base Design](https://www.basedesign.com/) and [Pedro Sanches](https://wip.llc/).

&nbsp;

The work above was only made possible with generous support from our funders, our partners, and everyone who we were lucky enough to collaborate with over the last year. 

We’re always looking for ways to expand our work with new partners, and we invite you to reach out if you have an idea we might pursue together to further our mission to use science, data, and creativity to research and reshape our relationship with media.



Harmon.jpg

Researching & reshaping our relationship to media

A New Year for Harmony Labs

Muted Green

We recently hosted a deep-dive into the 4-values based audience at the heart of the [Narrative Observatory @ Harmony Labs](https://narrativeobservatory.org/). Below is an abbreviated recording meant to serve as a foundational resource on these audiences and how they can be useful for narrative and culture change work.

![Work-Harmony-Labs.png](https://player.vimeo.com/video/734517489?h=e8dc26fcbf "Watch now: Science Director Riki Conrey presents Harmony Labs’ audience architecture.")

In this recording, our Harmony Labs Science Director [Riki Conrey](https://harmonylabs.org/riki-conrey) walks through:

-   Our values-based audience architecture
-   How our audience segmentation incorporates the intersectionality of demography, values, and media consumption
-   A sample of the primary media consumed by each of the four audiences
-   Opportunities that can exist through content creation or storytelling to engage audiences at the edges of your base

You can also find the full recording of our live webinar from July 20, 2022 [here](https://vimeo.com/731841630/5e7285d773), featuring our three panelists: [Janay Cody, PhD](https://www.linkedin.com/in/janaycody) (resident IF YOU SAY SO); [Michael Frias](https://www.linkedin.com/in/michael-frias-7a14247b) (resident DON’T TREAD ON ME); and [Gretchen Barton](https://www.linkedin.com/in/gretchen-barton-3690231) (resident TOUGH COOKIE). Please comment here or reach out to hello@harmonylabs.org with any remaining questions you might have.

&nbsp;

----------

&nbsp;

Coming soon is more new work covering issues like guns, civic information, health equity, reparations and anti-racism, Asian-American identity, and more. Follow us on [LinkedIn](https://www.linkedin.com/company/harmonylabs), [Twitter](https://twitter.com/harmonylabs), or [Medium](https://medium.com/harmony-labs) for the latest.

We are looking for new, innovative partners to join the Narrative Observatory. Have an interesting project related to audience, narrative, or story? Looking to supercharge your storytelling efforts for culture change? [Let’s talk](https://harmonylabs.org/contact).

Maps-Slide-Updated.001.jpeg

A deep-dive into our values-based audiences for narrative and cultural change

ICYMI: Audiences 101

Light Purple

Harmony Labs is on a mission to create a world where media systems support democratic culture and healthy, happy people. Over the last decade media systems have evolved to be outrage machines, reproducing and reifying division and difference, in ways that are starting to pose a threat to the conduct of civic life. These civic divides now extend to every  issue of import — immigration, racial justice, vaccines, public education, and more — as they play out in the media. Given the magnitude and complexity of these challenges, accomplishing our mission entails more work than any one institution can undertake alone — our work is collaborative by necessity and by design.

That’s why we are incredibly grateful to welcome a new  [Advisory Board](https://harmonylabs.org/people/#advisors)  to Harmony Labs. This group of 30+ talented professionals brings a diverse set of skills and experience to our organization, representing philanthropy, academia, advocacy, media, tech, and more. They’re already super-charging our work.

Most urgently, this new Advisory Board will help us shape the future of the  [Narrative Observatory](http://narrativeobservatory.org/). These advisors will also serve as ambassadors of our work, helping garner new connections and opportunities, as we look to grow our community and deepen our impact.

We’d like to extend our deepest appreciation to all of our new advisors. They are an indispensable source of input, feedback, and inspiration. As we strive to reform and transform media systems, we look forward to each of them challenging and enriching our work in essential ways. You can learn more about our work and our new Advisors  [here](https://harmonylabs.org/people/#advisors).

HLabsAdvisoryBoard.jpg

Introducing Our New Advisory Board

Tech-Policy-Press.png

Platform Recommendations to Empower Influencers Driving Civic Engagement

huffpost.png

The Neuroscience of Decision-Making in 2016

webinar test.png

Join us on 2/28/23 @ 1 pm ET for a deep dive into recent research focused on improving healthcare and public health.

Health Equity Narratives Webinar

Light Pink

art.jpg

Storytelling strategist Liz Manne on content-forward narrative change for health equity

The Art of Creative for Narrative Strategy

Harmony Cream

_While this project and research report were both completed in early 2022, we believe the findings are valuable as evergreen insights to help those developing interventions to reduce gun harm and violence in the U.S._

Firearms exist as a cornerstone of American life. A February 2022 [report](https://pubmed.ncbi.nlm.nih.gov/34928699/) shows nearly 7.5 million Americans became new firearm owners during the 1st year of the COVID pandemic. Firearm injury rates are skyrocketing, and most Americans — gun owners or not — misidentify the most common reasons for gun injury and death.

The result? Americans are increasingly united in their desire to reduce gun injury and violence, but are deeply divided on how to do so. Reasons for this are vast, but likely symptomatic of the social issue of “gun culture” and its myriad narratives. Guns define heroes and villains, create and resolve conflict, and establish the stories we engage with every day. But from news to entertainment to casual hobbies, how Americans approach and perceive guns — and firearm injury and safety — differs.

To answer the question of “who sees guns in what ways,” [Harmony Labs](http://harmonylabs.org/) and the [Brown-Lifespan Center for Digital Health](https://digitalhealth.med.brown.edu/) partnered to generate a first look at the online narrative landscape that exists around guns and firearms in America — with the goal of uncovering Americans’ narratives around not just guns themselves, but also firearm injury and violence. By exploring the volume and categories through which Americans encounter stories online about guns and gun injury in their lives, we hope to offer a first step in creating a path forward that both acknowledges the existing “gun culture” while also improving safety.

_An important disclaimer: “FIREARM INJURY” is the problem we want to contribute to solving, but we consider any space in which stories about GUNS are told to be a space of rich, storytelling opportunity. Thus, we use the more general term (“GUNS”) here to talk about the cultural spaces in which firearm injury stories specifically could gain traction._

## Our approach
 
To understand Americans’ perspectives on guns, we framed this as a question of **narrative**. The objective was not to find out where the gun “problem” was, but to know where in the media Americans were encountering guns, as a first step to understanding how public narratives about guns are influencing peoples’ private lives.

Searching for where people consume gun content in this pilot project, we looked to a microcosm of all American culture: YouTube.

[70% of all American adults](https://www.pewresearch.org/fact-tank/2019/12/04/10-facts-about-americans-and-youtube/) see YouTube content every day .

[The Narrative Observatory @Harmony Labs](https://narrativeobservatory.org/) works with data donated by our commercial partners, audience data from research panels of individuals who have opted in to share their browsing behavior donated by Nielsen, and content data that records the actual stories delivered in the media donated by PeakMetrics. For this project, we used audience browsing behavior to identify which gun-related YouTube videos they watched and what gun-related searches they submitted to Google, Yahoo, and Bing. The research below concerns YouTube viewing and internet search behavior by American adults between January 2020 and September 2021. The results of our work were recently published as a [peer-reviewed paper](https://www.sciencedirect.com/science/article/abs/pii/S0091743522003073) in _Preventative Medicine_, and we are excited to share publicly this [new report](https://docsend.com/view/2wu96mp3muxm7rwd).


## “Gun Culture” is, to be frank, everywhere

The media Americans encounter serves as a backdrop for exposure to information about guns.

7% of all YouTube users or about 3.5% of all American adults consume gun-related content on YouTube daily.

We partitioned this YouTube watch behavior into 7 naturally occurring “gun ecosystems” that we identified by grouping similar watch patterns together. While specific to YouTube, the ecosystems embody a wide arc of culture that might reflect how Americans encounter guns across other media and social media platforms and cultural zones.

The ecosystems were: **Hunting & Fishing, Guns & Gear, Guns 4 Fun, News & Hot Takes, Music, Movies,** and **Games**, each reaching different numbers and types of Americans with gun-related videos.

&nbsp; 

>![1_ZlAtvN6Cj8NEAXt2Dgg_Gw.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_Zl_Atv_N6_Cj8_NEA_Xt2_Dgg_Gw_9d31471b8c.webp)
>![1_K0ZS_s1ZgbhV9Yt9HpAGsw.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_K0_ZS_s1_Zgbh_V9_Yt9_Hp_A_Gsw_6fb16b8c39.webp)

>![1_adjPIroOS9-PdxKywoTOwA.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_adj_P_Iro_OS_9_Pdx_Kywo_T_Ow_A_b1da628255.webp)
>![1_lElhmSfD1W5r2qgd378Slg.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_l_Elhm_Sf_D1_W5r2qgd378_Slg_d1ae8dc347.webp)

>![1_aQnJgCshN7Wy1cOVayV7aA.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_a_Qn_Jg_Csh_N7_Wy1c_O_Vay_V7a_A_f943a62042.webp)
>![1_Mlw0g8gUqH0xR1WJen5Zog.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_Mlw0g8g_Uq_H0x_R1_W_Jen5_Zog_7df33835d6.webp)

>![1_3yOcRvXwJFL9LSKikW3N8g.webp](https://harmony-strapi-storage.s3.amazonaws.com/1_3y_Oc_Rv_Xw_JFL_9_LS_Kik_W3_N8g_b8942bd3f1.webp)


## An American narrative of “random attacks”

We then looked at what types of searches YouTube viewers and other users in each ecosystem conducted. Across these ecosystems, we found that 95% of searches about guns and gun harms were about mass shooting events or police-involved shootings rather than other, more common kinds of harms including suicide, accidents, and violence in the home. This indicates that folks are fearful of the rare or newsworthy, rather than of the common and much more present risks of gun ownership.

For example, consuming content from the News & Hot Takes gun ecosystem, the ecosystem with the largest reach often characterized by outrage, arguments, and strong opinions, is associated with searching for active shooters and shootings. People who watched News & Hot Takes guns content on YouTube were more likely than people from other ecosystems to search on the same day for information about shootings and active shooters.

**What this tells us:** News & Hot Takes is a part of sustaining a pervasive narrative about firearm-related harms as random, unpreventable attacks by strangers — Americans are not exposed to the personal, localized burden of, or risks for, firearm injury. This narrative may serve to normalize firearm-related harms, decrease the perceived risk and impact of these harms, and make preventable firearm injury more likely.


## Where do we go from here?

As Americans experience firearms on YouTube, online searches and associated results, and then circle back again, they build narratives about “gun culture” along the way.

The biggest question we are left with:

**What if ending firearm injury in the United States requires a rewrite of how we talk about “firearm injury” itself?**

By beginning to understand how firearms shape the national consciousness, as well as the behavior of Americans in relation to firearms online, we can start to form a common set of norms that lay the groundwork for intervention and sustainable firearm safety.

This work is a first step in what we hope is evolving, ongoing research and analysis. If you are interested in collaborating on or contributing to our ongoing efforts to explore the data around firearm injury and safety, please [get in touch](https://harmonylabs.org/contact).

&nbsp;

----------

&nbsp;

_**About Brown-Lifespan Center for Digital Health**_

_The Brown-Lifespan Center for Digital Health (CDH) is a hub where creative minds across diverse disciplines and perspectives can create and implement impactful digital health. CDH is a model of how an academic medical center can influence health by uniting technology, science, entrepreneurs, healthcare, and health systems._

&nbsp;

_**About Harmony Labs**_

_Harmony Labs is a 501c(3) non-profit on a mission to create a world where media systems support democratic culture and healthy, happy people. We’ve been doing audience, narrative, and story analysis for more than a decade, helping storytellers channel the immense power of story to shape the future. With the Narrative Observatory project, for the first time ever, we’re harnessing powerful industry relationships and an academic research network to develop data infrastructure purpose-built to identify and track narratives and story opportunities, and to learn about audiences across platforms._

guns-cover.png

Researching Americans' exposure to firearms in media as a step toward firearm safety intervention

Guns: An All-American Story

nbc-news.png

Young Latino men may be 'easy targets' for disinformation on immigration, report says

Bright Green

With the [Narrative Observatory](https://harmonylabs.org/narrative-observatory)  project we’re harnessing powerful industry relationships and an academic research network to develop data infrastructure purpose-built to identify and track narratives and story opportunities, and to learn about audiences across platforms. We use the data that flow through culture — the TV, films, news, social media, and music — to tell us where our audiences are; what kinds of stories they consume when they’re not thinking about social issues; what narratives they do encounter about issues like racial justice, climate change, or gender equity; and when a specific story represents a narrative opportunity to reach and resonate with them.

Under the hood is some fancy data dancing and advanced software engineering that makes everything go. This includes proprietary and hard-to-get data sets; a data infrastructure that prioritizes privacy, security and flexibility; industry-grade processing power; and a growing library of analytic tools and predictive models. The system we’ve designed is intended to maximize the impact of our small (but mighty!) engineering team so that we can focus on our mission and continue to build and link our best-in-class data holdings.

## What data?
For the past year, Harmony Labs has been working with commercial partners on  [data sharing agreements](https://medium.com/harmony-labs/growing-our-data-growing-our-network-545f327f0187)  across a broad spectrum of media platforms. As a 501(c)3 organization, bound to work for the public good, we have adopted a set of principles and practices around our use of data to: ensure we only gather data that is useful to  [our mission](https://harmonylabs.org/mission); anonymize any data that contains personally identifiable information (PII) at ingest; maintain robust security with regard to laws and regulations, limited access, and encryption; actively work with our partners and in-house data science team to provide the highest possible standards for scientific integrity, clearly communicating methods, assumptions, and practices.

With these principles and practices in mind, we’ve amassed over 50 terabytes of media content to support our Narrative Observatory work, including content scraped from websites and APIs; TV, web, and mobile consumption data from opt-in panels; song lyrics; and closed captions for nearly all U.S. broadcast television going back to 2014. The integration of so much heterogenous data into a single accessible, reliable platform presents unique value and some thorny engineering challenges.

![HLabsCultChange2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Cult_Change2_3397b34caa.jpg)

## How do we make this data useful?
The first step when working with data at this scale is to determine how we’ll receive it from our partners. In some cases there are established export processes while others have never shared their data before and require a more collaborative process. Transfer mechanisms range from  [S3](https://aws.amazon.com/s3/)  uploads, direct  [PostgreSQL](https://www.postgresql.org/)  connections, streaming live updates from  [RabbitMQ](https://www.rabbitmq.com/), and  [SFTP](https://en.wikipedia.org/wiki/SSH_File_Transfer_Protocol)  transfers.

With the data in our hands, we can then get to work sifting through sample data and documentation. This is a joint effort between our engineering and data science teams. As is typical when working with proprietary data, these teams attempt to optimize the data storage architecture to fit their own use cases. For our partners that don’t have externally-facing APIs, it’s rare that many considerations have been made — and rightly so — to design the storage of their data so as to make it easily accessible to external partners.

The biggest challenge we face when integrating new data is how do we maximize its usefulness to data scientists and our partners while maintaining our data ethics principles and practices. We like to think that no one has quite the valuable network of partners that we do and we’re confident no one is as dedicated to conscientiously selecting these data, coordinating them in such a way that increases their collective value, and making them accessible to our internal teams and our research partners in a safe and secure manner.

## What do we do with the data?
There is a saying in the world of software engineering: “death by 1,000 integrations.” As a software architecture adds integrations with external dependencies or data sources, the process for implementing those integrations needs to be as streamlined and similar as possible to ensure correctness, robustness, and timely availability. Typical software teams have external integrations for features like user authentication, text messaging, or software monitoring. In most cases those are constrained and separate integrations that barely interact. Our software ecosystem is made up of the opposite use case — we generate very little data internally except as derivative analysis products.

Knowing that external integrations will always be our primary use case, we’ve tried to optimize workflows and technologies around the following categories: security, correctness, and shared infrastructure systems.

A maxim we follow strictly is the  [principle of least privilege](https://en.wikipedia.org/wiki/Principle_of_least_privilege). All of Harmony Labs’ employees, partners, and software are only given access to the specific set of resources and permissions necessary to complete their pre-specified goals. Even our applications and supporting software are only able to interact with specific services and data, preventing unintentional cross-contamination and spillage vectors. This commitment to security and access controls, along with annual independent security audits — the most recent of which was successfully completed in December 2020 — increases the confidence of our partners who graciously share data with us.

Data is useless if it’s not correct and, because we are making statements about media content and consumption, the correctness of our data is absolutely critical for our partners. But, we don’t stop at simply running our extract, transform, load ([ETL](https://en.wikipedia.org/wiki/Extract,_transform,_load)) processes and hoping for the best. We’re also building periodic audit processes to account for partner data that is new or may have changed, or issues that may have arisen during initial ETL.

Shared infrastructure systems, or the standardization of compute, storage, and analysis infrastructure, allow us to continue active development while maintaining existing applications. All our production software runs on  [Kubernetes](https://kubernetes.io/)  and utilizes managed services from  [AWS](https://aws.amazon.com/)  wherever possible. We limit the amount of software we’re maintaining to only our necessary applications or open source software to enable better abstractions or monitoring and maintenance, we plan and complete deliberate software and service upgrades regularly in order to prevent urgent interruptions to long-running projects, and we use application templates and a single, shared method for our  [CI/CD](https://en.wikipedia.org/wiki/CI/CD)  processes. All of these practices together ensure that data access and linking is more repeatable and predictable for both software engineers and the users of the data.

## What’s Next?
In the past year this data infrastructure has powered the  [initial prototype for a Narrative Observatory](https://medium.com/harmony-labs/what-the-narrative-observatory-should-be-3d8ac31fdbcc)  and enabled public facing data analysis to help  [organizations better understand their audiences](https://harmonylabs.org/media-mapping),  [help inform storytelling about poverty and economic mobility in the U.S](http://obiaudiences.org/), and  [identify story opportunities for advocates of healthy A](https://medium.com/harmony-labs/the-most-common-narrative-about-tech-is-not-what-you-think-robot-vacuums-and-wishes-granted-ec9d0d5e28d2)I. The system we’ve designed is also flexible enough to accommodate our growing  [research network](https://harmonylabs.org/research).

To get here, our engineering team has overcome significant challenges with source data; from recombining terabytes of snippets of text that are delivered to us by line and out of order, to multiple schema and format changes over a partner’s historical backlog, to a request from a partner to completely rewrite our ingest app to pull data from Elasticsearch instead of PostgreSQL!

We’re confident that the systems and processes we’ve designed will continue to meet new sets of engineering challenges while ensuring that our infrastructure is increasingly secure, efficient, and, most importantly, accessible to those most ready to use these tools to effect real culture change. In the months ahead we’re working to expand our corpus of data to include more media types and audience dimensions, while continuing to innovate technically. Follow this space for updates on our progress or  [get in touch](https://harmonylabs.org/contact)  if you have any feedback.

HLabsCultchange1.jpg

Data Infrastructure for Culture Change

At the heart of the  [Narrative Observatory @ Harmony Labs](http://narrativeobservatory.org/)  are people. ​​By centering people in our research we can understand to what extent they encounter an issue, narrative, or type of content, like disinformation, in their media and cultural intake. It can help us to better understand vulnerabilities, opportunities, and risks, and to form hypotheses for narrative intervention with our partners. So, by the time our partners are ready to produce content, we know whom we should be speaking to, how, and where.

To make sense of the hundreds of thousands of people who participate in our opt-in research panels, we use audience “segments.” Traditionally this type of segmentation has been used to understand people in terms of their demography: age, race, and gender. That approach has been a necessity for many decades, because most of the data we’ve had access to only included those types of markers.  But those markers don’t define true audiences: groups of people who share the same values and go to the same places to consume the same stories.

We’ve found that values are the best predictor of cultural affinity because values represent the personal goals, story arcs and endings that inspire us. These values form the foundation for the  [four key audiences](https://narrativeobservatory.org/audiences/)  we use in our research — based on the  [universal human values from Shalom Shwartz](https://scholarworks.gvsu.edu/cgi/viewcontent.cgi?article=1116&context=orpc)  — and have another feature that makes them invaluable as a tool for understanding people: they make neighborhoods.

![HLabsAudHeart2.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Aud_Heart2_d943cfaa88.png "The 4-values distinct audiences from the Narrative Observatory @ Harmony Labs that we use to understand how people encounter issues, narratives, or types of content.")

Not everyone is in just one of the four audience quadrants. For example, Riki Conrey, our Science Director, is  [People Power](https://narrativeobservatory.org/audiences/people-power/)  with a strong dash of  [Tough Cookies](https://narrativeobservatory.org/audiences/tough-cookies/). She can’t really watch anything less cozy than the  _Great British Bake Off,_  but she reads the Washington  _Post_  and the New York  _Times_. Her values are audience-bridging values: tradition, family, responsibility, and tolerance. Our Communications Director, Paul Johnson, is also squarely in the People Power quadrant but ends up consuming a lot of [If You Say So](https://narrativeobservatory.org/audiences/if-you-say-so/)  media — and is desperate to find common ground in these story worlds with his teenage son.

When the shared spaces, or the spaces _between_  the audiences mean something, segmentation becomes less a reductionist exercise in stereotyping than a vocabulary for describing the diversity of people’s story choices. And the audiences come to form a true map, with direction, volume, and intensity, of those experiences.

Here’s a map of who watches Fox News on YouTube. Spoiler alert: it’s mainly  [Don’t Tread on Me](https://narrativeobservatory.org/audiences/dont-tread-on-me/).

![HLabsAudHeart3.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Aud_Heart3_defbb162ba.png)

Compare this with CNN on YouTube. A large density of the audience is People Power, but it also reaches down into Tough Cookies and all the way over into If You Say So. There’s even a tiny dot of interest in the same audience that watches Fox News.

![HLabsAndHeart4.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_And_Heart4_8177e507b9.png)

The center of gravity of Fox News on YouTube is in just one audience: Don’t Tread on Me. CNN has broader appeal with a center of gravity firmly in People Power. ABC News, with its homier, more local feel is a bridging opportunity. It appeals to people on the border of community-focused People Power and security-focused Tough Cookies.

![HLabsAndHeart5.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_And_Heart5_e6b5fe5e4a.png)

While these types of general insights, i.e. “who is consuming news,” can be interesting directionally, they aren’t as interesting as something even more specific, like a social issue we might want to influence. By centering our audience research on the specific stories people are consuming, we can start to help answer questions like: how and how often does my issue show up for audiences?; how do audiences interact with issue-relevant content?; are there healthy narratives out there my work can help grow? It can also provide a framework for evaluating the impact of the content we create.

Take for example the issue of “guns.” There’s a large swathe of people from Don’t Tread on Me through If You Say So being reached with stories about guns, some even breaching the border of People Power. In the context of the “values” for each of the audiences referred to earlier, that is the hemisphere of the audience that focuses on the “me” over the “we.”

![HLabsAndHeart6.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_And_Heart6_78f779fd9e.png)

The brightest spot here is in Don’t Tread on Me where “guns” often represent security and defense. But the trailing audience that extends up through If You Say So is encountering guns through many types of stories, especially Esports where weaponry is a top topic for streaming gamers.

If guns and weaponry bridge Don’t Tread on Me and If You Say So, what bridges If You Say So with their other neighbor, People Power? Popular music, especially Hip Hop.

![HLabsAudHeart7.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Aud_Heart7_2b581aaa7d.png)

We’ve spent much of the last 18 months identifying story opportunities and threats like these for our partners who are working to shift the cultural narratives that condition public understanding and constrain institutional action on issues like poverty, racial justice, immigration, or climate. For all of them, putting people in the center of mapping exercises like this one has changed the game. Instead of just creating campaigns that try to meet the audience where they are, there’s been a shift towards also meeting the audience where  _we_  are — in the cultural spaces that we authentically share.

HLabsAudHeart1.png

Audience @ Heart

science.jpeg

How Americans encounter guns: Mixed methods content analysis of YouTube and internet search data

Deep Pink

Earlier this year, work kicked off with winners of Mozilla’s Creative Media Awards,  [exploring the intersection of art, advocacy, and data science](https://medium.com/harmony-labs/designing-for-and-measuring-the-impact-of-art-for-advocacy-98e44c410d8f).

In the past few months we’ve been working with this group of artists, technologists, and activists to better understand the impact they want their projects to have and the audiences they are trying to reach.

We also stopped by  [Mozfest](http://mozillafestival.org/), Mozilla’s annual gathering for advocates of a more humane digital world, for a discussion about healthy AI narratives.

## Memeing at Mozfest

In a session provocatively titled:  [_Stop Scaring People About AI_](https://www.youtube.com/watch?v=7FTKedxtKog), we facilitated a discussion on the most common narratives about Artificial Intelligence in popular culture. While it’s not so simple as creating “happier” messages about AI, we do think there are healthier alternatives to the narrative status quo in this space.

We had a lightning round to identify recognizable examples of  [the most common narratives in tech](https://medium.com/harmony-labs/the-most-common-narrative-about-tech-is-not-what-you-think-robot-vacuums-and-wishes-granted-ec9d0d5e28d2)  and participants shared their own experiences with healthier narratives in pop culture.  _Big Hero 6,_ with its robot’s moral code, and even  _Pinocchio,_  and its lessons in transparency, were surfaced as stories that show technology evolving when humans are given agency over its design.

We put this new understanding of a healthy AI narrative into practice by making some memes! Some of these memes were definitely of the “you had to be there” variety, but they all proved to be an effective way of applying what we’d learned: telling stories that promote agency and empower people have the best chances of reaching and resonating with a general public.

![HLabsAIAdvocacy2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_AI_Advocacy2_5a73053a01.jpg "One of our favorite memes from our Mozfest session. Drake gets healthy AI.")

## Audiences for AI Narratives
Before narratives can be shifted in culture (via memes or media), storytellers first need to understand their audience. We use the data that flow through culture — the TV, films, news, social media, and music — to tell us where audiences are; what kinds of stories they consume when they’re not thinking about social issues; what narratives they do encounter about issues like AI; and when a specific story represents a narrative opportunity to reach and resonate with them.

The 2021 CMA cohort is a group of Black artists examining AI’s relationship with racial justice. To help them understand where their audiences are in culture and what they care about when not thinking about AI, we’ve focused on 3 audience segments: Black Americans, Technologists, and Persuadables, each based on our previous research report,  [_Beyond Demography: Black Audiences Online_](https://docsend.com/view/xafiesp86f9ju4yk). We looked at what platforms audiences are most likely to spend time on and for how long; what sites they visit for news and entertainment; top YouTube channels, tags and influencers; and more cultural touchpoints.

![HLabsAIAdvocacy3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_AI_Advocacy3_baaa0dc797.jpg "Top Youtube influencers from January, 2021.")

The goal with this audience research is to inspire new ideas during the content creation process and provide insight into how content might be distributed. This initial discussion with grantees also spurred new research questions from the creators, which our data science team will use to generate insights that can be integrated back into their media making practice.

## What’s next: Impact Plans + Project Previews

Next up is individual coaching sessions with our Impact Producers,  [Jennifer MacArthur](https://www.linkedin.com/in/jennifermacarthur/)  and  [Brett Gaylor](https://www.brettgaylor.com/), to develop individualized impact plans for each project that zero in on audience, story opportunities, and distribution strategies. We’re also working with teams to develop testing plans for their projects, so they can validate their initial hypotheses around the impact they are trying to achieve and any creative refinements they might make pre-launch to maximize the efficacy of their original project visions.

In our next update we’ll do a deep dive into the impressive slate of projects themselves and start to explore how media testing can be a valuable tool, not just for measurement after the fact, but for improving media  _during_  the creative process.

HLabsAIAdvocacy1.jpg

AI, Art & Advocacy

Complex questions continue to challenge our understanding of media systems and their influence on politics, society, and culture. What’s media’s role in shaping the deep cultural narratives that condition public understanding? How do these narratives constrain public capacity to address critical social issues, like poverty, the climate crisis, or immigration reform? How are narratives formed and how can they be tracked computationally?

As we pursued answers to these questions over the last few years, it became increasingly clear that existing approaches and infrastructure were not meeting the needs of the research community. In 2019, we started the  [Harmony Labs Research Network](http://harmonylabs.org/research).

The core of this network is a community of research and practice and a unique data platform that includes reliable access to proprietary and hard-to-get data sets, state-of-the-art analytical capacities, a growing network of developer applications, data and subject matter expertise from our in-house data scientists, and industry-grade data infrastructure and technical support.

With recent expansions of our data offering to include comprehensive online news, TV, and Facebook ad datasets (and  [more](https://harmonylabs.org/research)), the number of researchers leveraging this corpus of data to investigate healthy media systems has grown significantly in 2020, too. Here’s what they’re up to.

## 14 new projects across academia, news, & non-profits

![Hlabs11:10:202.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs11_10_202_7215e6b406.jpg)

### Analyzing networks of audience overlap across news sources
_University of Pennsylvania_

-   Project Leads:  [Sandra Gonzalez-Bailon](https://www.asc.upenn.edu/node/648)  (Associate Professor of Communications) and  [Tian Yang](https://www.asc.upenn.edu/people/students/tian-yang)  (Doctoral Candidate)
-   Department: Annenberg School of Communications
-   Dataset: TV and Web panel data

### Project Ratio: Quantifying the production and consumption of news
_University of Pennsylvania_

-   Project Leads: Duncan Watts (Professor), Homa Hosseinmardi (Postdoctoral Research Fellow), David Rothschild (Economist)
-   Department: Annenberg School of Communications, Department of Computer and Information Science
-   Dataset: TV and Web panel data; TV closed captioning; Online News
-   Publication: [_Evaluating the fake news problem at the scale of the information ecosystem_](https://advances.sciencemag.org/content/6/14/eaay3539)  in Science Advances

### Assessing the effectiveness of FB ad enforcement policies in Myanmar
_Carter Center_

-   Project Leads:  [Michael Baldassaro](https://www.linkedin.com/in/michael-baldassaro-2b440757/)  (Senior Advisor) ,  [Michael Scholtens](https://www.linkedin.com/in/michael-scholtens-99476438/)  (Data Analyst) and  [Dan Richardson](https://www.linkedin.com/in/daniel-richardson-b90b76144/)  (Program Assistant)
-   Department: Digital threats to democracy
-   Dataset: FB political ad data
-   Publication:  [_It’s time for Congress to regulate political advertising on social media_](https://thehill.com/blogs/congress-blog/politics/519933-its-time-for-congress-to-regulate-political-advertising-on)

### Creating a Facebook political ad dashboard for accurate insight into digital ad spending
_Center for Responsive Politics_

-   Project Leads: Sarah Bryner (Director of Research and Strategy) and  [Jacob Hileman](https://www.linkedin.com/in/jacob-hileman-4a1058171/)  (Information Technology Director)
-   Department: Political Ads Research
-   Dataset: FB political ad data; using machine learning to add value to related content for use by network partners.

### Finding common ground in groups with ideological differences: using language as a signal
_University of Maryland_

-   Project Leads:  [Philip Resnik](http://users.umiacs.umd.edu/~resnik/)  (Professor)
-   Department: Department of Linguistics and the Institute for Advanced Computer Studies (UMIACS)
-   Dataset: Civic engagement data; Online news data

### Identifying unethical election advertising and monitoring a platform’s response.
_Global Witness_

-   Project Leads:  [Sam Leon](https://www.linkedin.com/in/sam-leon-4b072b68/?originalSubdomain=uk)
-   Department: Digital threats to democracy campaign
-   Dataset: FB political ad data

### Exploring advertiser transparency and microtargeting in online political ads
_University of Southern California_

-   Project Leads:  [Shomik Jain](https://www.linkedin.com/in/shomikjain/)  (Masters Student) &  [Abby Wood](https://gould.usc.edu/faculty/?id=71046)  (Associate Professor of Law, Political Science and Public Policy)
-   Department: Gould School of Law
-   Dataset: FB Political Ads
-   Publication:  [_Facebook Political Ads And Accountability: Outside Groups Are Most Negative, Especially When Microtargeting Or Hiding Donors_](https://arxiv.org/abs/2010.01730)

### Monitoring hateful and decisive discourse around elections
_Institute for Strategic Dialogue (ISD)_

-   Project Leads:  [Chloe Colliver](https://www.isdglobal.org/isd_team/chloe-colliver/)  (Head of Digital Policy and Strategy) and  [Cooper Gatewood](https://www.isdglobal.org/isd_team/cooper-gatewood/)  (Digital Research Manager)
-   Dataset: FB political ads

### Investigating the influence of ad delivery algorithms
_Northeastern University_

-   Project Leads:  [Alan Mislove](https://www.khoury.northeastern.edu/people/alan-mislove/) (Professor of Computer Science) and  [Piotr Sapieżyński](https://www.sapiezynski.com/)  (Associate Research Scientist)
-   Department: Khoury College of Computer Science
-   Dataset: FB political ad data

### Understanding the dynamics of information dissemination and consumption for COVID-19 and the 2020 election
_Northeastern University_

-   Project Leads:  [David Lazer](https://lazerlab.net/) (Professor), Damian Ruck (Postdoctoral Research Fellow), Briony Swire Thompson (Senior Research Scientist)
-   Department: Political Science, Khoury College of Computer Sciences, Network Science Institute
-   Dataset: Online news data

### Building a FB political ad dashboard with crowd sourced data
_New York University_

-   Project Leads:  [Jeremy Merrill](http://jeremybmerrill.com/) (freelance journalist, published at ProPublica, The Markup)
-   Dataset: FB political ad data
-   Publication:  [_“Trumpcare” Does Not Exist. Nevertheless Facebook and Google Cash In on Misleading Ads for “Garbage” Health Insurance._](https://www.propublica.org/article/trumpcare-does-not-exist-nevertheless-facebook-and-google-cash-in-on-misleading-ads-for-garbage-health-insurance)

### Investigating political advertising tactics of the major candidates and third-party groups
_Politico_

-   Project Leads:  [Andrew McGill](https://www.linkedin.com/in/andrewmcgill/)  (Editor of Interactive News),  [Alan James Vestal](https://www.linkedin.com/in/allan-james-vestal-95a24813/)  (Senior News Applications Developer),  [Mark Scott](https://www.linkedin.com/in/markscott82/)  (Chief Technology Correspondent)
-   Dataset: FB political ads

### Assessing the impact of local news deserts on consumption of political news
_Stanford University_

-   Project Leads:  [Greg Martin](https://www.google.com/search?q=greg+martin+stanford&oq=greg+&aqs=chrome.0.69i59j69i57j46l5j69i60.1115j0j4&sourceid=chrome&ie=UTF-8)  (Assistant Professor) and  [Shoshana Vasserman](https://shoshanavasserman.com/)  (Assistant Professor)
-   Department: Stanford Graduate School of Business
-   Dataset: Civic engagement data; Online news data

### A tool for customizable, daily updates on new political advertisements
_Who Targets Me_

-   Project Leads:  [Sam Jeffers](https://whotargets.me/en/about-who-targets-me/)  (Executive Director) and Kevin Carmody (Developer)
-   Publication: [“_Introducing Decoder”_](https://whotargets.me/en/introducing-decoder/)

## What’s Next
The work underway by these groups has the potential to unlock important insights about the mechanics of media influence and help inform the design of healthier media systems in the future. We’re looking forward to working with each of these groups to ensure their findings have the widest possible impact, ranging from traditional academic outputs to new formats like blogs, websites, and the press. Watch this space for updates.

In the coming months we’re also hard at work sourcing additional data sets that are valuable to these types of research efforts and additional research partners to leverage them in their work.  [Contact us](mailto:research@harmonylabs.org)  if that might be you.

HLabs11:10:201.jpg

Growing Our Data, Growing Our Network

A few months ago we started a news narratives project to examine how the media frames events of national interest in the U.S. Events like #MeToo allegations against prominent public figures, extreme weather occurrences, and mass shootings. We chose these types of events, in part, for practical reasons: since they tend to recur, they afford ample material for ongoing analysis. Mostly though we chose them, because they connect to some of the deepest political and cultural fault lines running through public life today — gender roles, climate change, personal freedoms versus group rights — and therefore might provide insight into the news media’s relationship to these fault lines. Does the media tend to act as neutral observer, perpetuator, instigator?

We started with mass shootings, a choice which recent events in California and Pittsburgh have, sadly, made poignant. For the purposes of this analysis, we defined a mass shooting as an event in the U.S. where at least ten people were killed in a single incident with guns. We looked at news content from the last two years, from the top two hundred most trafficked online news sites, according to Internet Explorer usage data. And we chose four events for analysis: the school shooting in Santa Fe, Texas in May 2018; the Parkland, Florida school shooting in February 2018; the shooting at a concert in Las Vegas in October 2017; and the nightclub shooting in Orlando, Florida in June 2016. Four events is a small sample, so please take our observations as a starting point for broader, deeper research. For us, they represent something to test against how the news media treats future mass shootings. Given the increasing prevalence of mass shootings generally, and the lack of politically viable mitigating measures, additional analysis opportunities seem like a woeful inevitability.

First, a few words about our method of analysis, then on to the findings, organized in five headlines.

## Method
For each of these events, on the day they occurred and for eight days afterward, we subjected news article headlines and associated article body text to two rounds of clustering.  The first clustering determined which articles were about the shooting event, and the second round surfaced clusters of articles with similar narrative content within each day’s coverage of the shooting. We came up with forty-one discrete narratives, which we used to hand-label each secondary cluster. We then characterized and grouped those forty-one narratives into seven macro-narratives, which serve as broad topic categories. In doing this, we discovered that some narratives had distinct manifestations as pro-gun, anti-gun or neutral. So we also coded by hand each article according to its general stance on guns.

It should be noted that the second round of machine clustering produced a lot of noise. On average, around 40% of the total articles produced on any day were effectively unclustered. This sounds like a high number. Anecdotally, based on a spot surveying on unclustered headlines, we believe that these unclustered articles could, largely, be clustered within the forty-one narratives that surfaced, and we are working on technical improvements for our clustering. As is, this method became much less noisy when we reduced the number of media sources from 200 to 50. In the future, for this kind of analysis, in addition to making technical improvements, we may adopt a more limited source list, given that the top 50 sources account for roughly 67% of total news consumption, according to our data, with the next 150 sources adding only a marginal increase of 17%.

### 1. There are consistent narrative patterns in the development and decay of mass shooting events in the news.
For the first day or two, the most prominent media macro-narrative is an elaboration of what happened, which we call “Just the Facts.” This is overtaken in prominence on day two and beyond by macro-narratives that focus either on “The Shooter” — motives, early warning signs, and other biographical excavations — or on “The Community.” This suggests a transition to an examination of why a shooting may have happened. In the last two shootings, the community macro-narrative outweighs the shooter-focused macro-narrative, which may be a result of greater sensitivity on the part of the media to the role shooter-focused narratives play in inspiring copycats, an effect which has been documented as “media contagion” in numerous studies (like  [this one](https://www.tandfonline.com/doi/abs/10.1080/01924036.2018.1437458?needAccess=true&journalCode=rcac20)  published by Jason R. Silva and Joel A. Capellan in early 2018) and publicized in awareness raising efforts like  [No Notoriety](https://nonotoriety.com/)  and  [Do Not Name Them](http://www.dontnamethem.org/). Alongside these three macro-narratives, there is a stream of four other macro-narratives, comparable in prominence to one another, having to do with: a demonization of “the other,” whether the perpetrator or the political opposition (“Tribalism”); reflections on how the event is playing out in the media (“Media Reality”); proposals for political and technical solutions (“Solutions and Science”); and observations on how shootings are expressions of larger U.S. cultural dynamics (“Culture Caused”). All media activity tends to have decayed to nothing about 10 days after the shooting. Beneath the time series below, we present these macro narratives and the forty-one narratives they comprise in order of prevalence, expressed as a percentage of total new stories across all four media events. And we include a sample headline and publisher from the Santa Fe shooting for each narrative, in order to illustrate the kinds of stories each narrative describes.

![HLabs11:20:182.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_182_790d0538ac.jpg)

![](https://miro.medium.com/max/1400/1*QsMw6kCE1YEXMoxq9Amg8w.png)![HLabs11:20:183.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_183_9a310a795a.jpg)

![](https://miro.medium.com/max/1400/1*Lp2_o8Cp_gcVXGh4-fNvJA.png)![HLabs11:20:184.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_184_dca504060d.jpg)

![](https://miro.medium.com/max/1400/1*4ungwR3j4tGO16-bX2NDtA.png)![HLabs11:20:185.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_185_48b1e0ae6d.jpg)

### 2. The Parkland shooting was different.
After three days of fairly typical narrative activity, on day 4, there is an upswing in the following three macro-narratives: tribalism, meta-media reflections, and solutions. This upswing continues through Day 7. A full week after the shooting, we see a total volume of news stories comparable to the first day of coverage, a significant deviation from the standard narrative decay pattern. Why? Just about every media outlet has weighed in with speculation here. Many of them have called out how the Parkland student community quickly coalesced behind a call for action, and how that call was just as quickly channeled into a nationally organized effort, headlined by a handful of strong student voices. Certainly, the upswing in the tribalism macro-narrative seems to have drawn fuel from the crisis actor conspiracy theories those strong student voices elicited. And reflections on this media phenomenon, in turn, accounted for a portion of the increase in the meta-media macro-narrative, along with the now commonplace, ceremonial unpacking of President Trump’s tweets. The student calls for concrete, specific legislative action were prominent within the solutions narrative upswing, but there were also calls from pro-gun activists, for instance, to arm teachers or distribute bullet-proof backpacks and other military-grade gear. This is another way in which Parkland differed from previous shootings. We saw pro-gun stories evoking solutions narratives, typically the narrative terrain of neutral or anti-gun stories. And this kind of activity persisted in the Sante Fe shooting that followed Parkland, which may suggest a strategic innovation for pro-gun message makers. One more note here: in the Santa Fe shooting that followed Parkland, a new narrative cluster arose, containing references to and commentary from Parkland and the Parkland students.

![](https://miro.medium.com/max/1400/1*rBalhRI67v0GGfPVEsmZwA.png)![HLabs11:20:186.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_186_369f5475ee.jpg)

### 3. There are many more neutral and anti-gun stories generally, but pro-gun stories appear more steadily, not only around the time of mass shootings.
Neutral and anti-gun stories occur over quite a short time period following each mass shooting, with the exception of Parkland. The role of mass shootings in provoking public interaction with the topic of reforming our relationship to guns is clear. Gun-related stories that do occur between shooting events tend to be pro-gun, focused on positive expressions of gun culture and other matters relating to second amendment rights. By way of illustration, we looked at gun-related stories in the period between January 2017 and the Las Vegas shooting in October and found that six out of the top ten gun-related article publishers in this time tended to publish mostly pro-gun stories or to have the promotion of guns as their express mission, namely:  [rightedition.com](http://rightedition.com/),  [thetruthaboutguns.com](http://thetruthaboutguns.com/),  [foxnews.com](http://foxnews.com/),  [ammoland.com](http://ammoland.com/),  [breitbart.com](http://breitbart.com/), and  [guns.com](http://guns.com/).

![Hlabs11:20:187.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs11_20_187_7e601d052f.jpg)

### 4. There is consistency across events in the way anti-gun and pro-gun stories deploy framing narratives.
Anti-gun stories tend to invoke solutions or community macro-narratives, whereas pro-gun stories tend to invoke tribalism and culture macro-narratives. This will come as no surprise to anyone familiar with current messaging tendencies on the right and left in the U.S.

![](https://miro.medium.com/max/1400/1*bvHyjs1b6ZGaxNDSJ9hEvw.png)![Hlabs11:20:188.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs11_20_188_de4ee7923a.jpg)

### 5. There is consistency across events in the story stances media outlets tend to publish, and in the framing narratives they favor.
In addition, mass-shooting related stories published by pro-gun leaning outlets often enjoy a less prominent placement than analogous stories in anti-gun publications. Here is one typical publishing pattern in pro-gun outlets, for instance: front-page stories reporting on the basics of the event on the day of its occurrence and a day or two thereafter, followed by less prominent, more scattered coverage on subsequent days, consisting mostly in critiques of anti-gun activity. By way of illustration, below are homepages from breitbart.com, drudgereport.com, foxnews.com, and cnn.com, on days one through five after the Las Vegas shooting. Surveying these pages illustrates just how different a picture of the world the same body of occurrence can be used to create via narrative framing.  _(editor’s note: we have updated the language in this section and the charts below for clarity based on initial feedback from readers.)_

![](https://miro.medium.com/max/1400/1*cJ5PGg4b08FYMdMD-ajk5A.png)![HLabs11:20:189.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_189_78eceb7959.jpg)

![](https://miro.medium.com/max/1400/1*oez0-ayqGM82CFx6_4nlyw.png)![HLabs11:20:1810.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_1810_cc7a3f7891.jpg)

![](https://miro.medium.com/max/1400/1*jaWQfeL01QpukuwI9pWkwA.png)![HLabs11:20:1811.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs11_20_1811_3fe2342812.jpg)

It is remarkable just how quickly and consistently different events get poured into a few simple framing narratives. Obviously, there are many reasons for this stark reduction of complexity, from the speed of today’s media environment, to the structure of the newsroom. But given how much of our daily lives occur in media, and how few counterbalancing cultural forces exist to the kind of communication media elicits, this complexity reduction may have outsized consequences. The  [recent work of Amanda Ripley](https://thewholestory.solutionsjournalism.org/complicating-the-narratives-b91ea06ddf63)  and the  [Solutions Journalism Network](https://www.solutionsjournalism.org/)  suggests that it may be, at least in part, to blame for the fierce intractability of so many of the partisan conflicts which clog public discourse and predispose us not to listen to one another.

What if news media instead opened a space for deep inquiry into the events of the day, where complex questions could be asked, such as: what is it about the structure of our communities that make possible, again and again, mass murder? Or what is it about how we are with one another as individuals, or within institutions, that makes it such a challenge for us to agree on — or even agree to test out — solutions to such an obviously egregious ill? Wrestling with and arriving at answers to questions like these would be hard to sum up in a headline, more process than product. Then again, it may provide a space for collective curiosity to gain ground and flourish.

This project was undertaken, in part, to test a new news content dataset we’re developing at Harmony Labs, known as  [Project Ratio](https://harmonylabs.org/ratio).

HarmonyLabs11:20:181.jpg

The Mass Shooting

Anatomy of a News Media Event

Bright Yellow

The social problems artists and activists often seek to address are complex and resist change. If we are interested in optimizing creative work for social change effects, Harmony Labs believes an opportunity exists to integrate data-driven insights around audience, narratives, and story opportunities. Innovative approaches to measuring — and increasing — impact will further strengthen media making practices.

Over the last four years we’ve had the opportunity to explore this intersection of art, advocacy, and data science in support of the  [Mozilla Foundation](https://foundation.mozilla.org/en/)’s  [Creative Media Awards program](https://foundation.mozilla.org/en/what-we-fund/awards/), where we’ve been lucky enough to share some of our  [research](https://docsend.com/view/zdn5h5r8z3cez8ud)  and impact design practices with an array of gifted media makers. Last week, we were particularly excited to kick off work with the 2021 cohort, which includes eight new projects  [examining AI’s relationship with racial justice](https://foundation.mozilla.org/en/blog/announcing-8-projects-examining-ais-relationship-with-racial-justice/). From Mozilla’s blog post announcing this year’s awardees:

> Today, the AI systems in our everyday lives can perpetuate and amplify biases that have long existed offline: Recommendation algorithms promote racist messages. Facial recognition systems misidentify Black faces. And voice assistants like Alexa and Siri struggle to understand Black voices. As the AI in consumer technology grows more sophisticated and prevalent, problems like these will grow even more complex.

This talented group of artists, technologists, and activists are developing a range of projects to tackle these issues, from apps, to short films, to new datasets.

Harmony’s role, led by Impact Producers  [Jennifer MacArthur](https://www.linkedin.com/in/jennifermacarthur/)  and  [Brett Gaylor](https://www.brettgaylor.com/), is working directly with awardees to provide a landscape analysis of key audiences for their work and current narrative and story opportunities for AI. We’ll be hosting a series of workshops with subject matter experts to help inform the creative process, an impact design curriculum to maximize the efficacy of their work, and media effects testing to help awardees fine-tune the final products.

In our time working with Mozilla on this program, we’ve been fortunate to support dozens of brilliant artists and projects, and we always end up learning as much from them as they do from us. Given the caliber of this year’s cohort and the critical nature of the issues they will be exploring, this year is shaping up to be no exception. We will provide progress updates as we go and, most importantly, let you know when these projects launch later this year.

HLabsArtAdvocacy.jpg

Designing for and measuring the impact of art for advocacy

The Association for Psychological Science (APS) held its 27th annual convention in New York, May 21–24. Hundreds of presentations offered the latest insights moving psychological science forward. APS, and conferences like it, are a great way to connect researchers across different points of view within psychology.  **But the vision for connection across traditional boundaries took on a larger scope than ever.**

### A call to action:
[Roberta Golinkoff and Kathryn Hirsh-Pasek](https://www.youtube.com/watch?v=8SXO3SNjKcI), two of this year’s James McKeen Cattell Fellow Awardees, put it most clearly when they implored the audience to take their research “to the streets!” Appeals to connect to non-academic audiences, and concrete avenues for doing so, kept popping up throughout the long weekend.

![Work-Harmony-Labs.png](https://www.youtube.com/embed/8SXO3SNjKcI "Roberta Golinkoff and Kathryn Hirsh-Pasek at APS15")

### Reflecting ‘the real world’: Scientists, topics, and methods.
Why should residents of the ivory tower be concerned with what happens on the streets, much less fight to bring their hard-won results in front of untrained audiences? Drs. Golinkoff and Hirsh-Pasek gave several classic reasons: To put the knowledge gleaned from the lab to work solving problems and improving people’s lives; to make the outputs of science visible in the public; because work is often funded by public sources, and the public should benefit from their investment when it’s possible to do so directly.

### Plus, it’s good science:
In his invited talk, Dr. Robert Sellers centered his talk on  [why psychological science should (and must!) care about racial and ethnic diversity](http://www.psychologicalscience.org/index.php/convention/why-diversity-should-matter-to-psychological-science.html). The populations that typically compose psychology studies — those Western, Educated, Industrialized, Rich, and Democratic (WEIRD) student samples — do not represent the world. The scientific method is based on being objective, but acknowledges unique perspectives. Science benefits from a diversity of perspectives earned through formal training, but also from the life experience of working scientists.

> The populations that typically compose psychology studies — those Western, Educated, Industrialized, Rich, and Democratic (WEIRD) student samples — do not represent the world.

In addition to scientists that reflect the population, other presenters addressed an interest research designs that reflect the real world (have higher ecological validity), especially for social interactions. In different talks and symposia,  [Ayelet Fishbach](http://faculty.chicagobooth.edu/ayelet.fishbach/research/),  [Dave Amodio](http://amodiolab.org/people), and  [Valerie Purdie-Vaughns](http://www.columbia.edu/cu/psychology/vpvaughns/people.html)  all discussed how social context has been neglected in social psychology research because of the overwhelmingly complexity of variables, data, and time constraints to hand-code all the content. They also addressed how this is changing: there’s more data about real world social interactions, better tools to capture it, and methods to help researchers handle big datasets and phenomena of interest.

> We have more data than ever before about real world social interactions, better tools to capture it, and stronger methods to help researchers handle big datasets and phenomena of interes

### How can scientists be empowered to put those ideas like those shared into practice?
Thankfully methods and opportunities to operationalize lofty goals emerged throughout in this year’s APS programming to provide guidance.

### Develop tools:
Researchers in  [Elizabeth Page-Gould’](http://www.sprgtoronto.org/faculty/)s lab offered one example in studies of media.  [Amanda Sharples](http://www.sprgtoronto.org/students/name/amanda-sharples/)  studies the effect of media exposure on attitudes for outgroup members, and reflected on the challenges of relying on participants’ self-reported measures. Lab member  [Sabrina Thai](http://www.sprgtoronto.org/students/name/sabrina-thai/) is developing an open source app to help improve real-time collection of data — a sophisticated, convenient way to make experience diaries more accurate.

![HarmonyLabs08:10:151.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Harmony_Labs08_10_151_c9e01cb74e.jpg)

### Expand their methods repertoire (and use new tools):
In symposia like “Diversifying methodological choices in the field of psychological science” and “Computational methods for linguistic insights into personality and behavior,” researchers advocated for using both inductive and deductive reasoning to process big data sets, natural language processing to extract meaning from text, and Bayesian (rather than frequentist) techniques to make causal inferences. Free tools like JASP, outfitted with a familiar, user-friendly interface and easy-to-interpret output, help adoption.

> Interest in and acceptance of academic-corporate collaborations has risen in recent years.

### Seek out and connect with non-student populations:
The ability to capture and model social interactions using social media data is a new possibility that allows researchers to overcome some of the barriers of studying complex social interactions due to time and observability constraints.In discussing research into on 100MM+ gamers interactions and conversations, [Jeremy “Lyte” Lin of Riot Games](http://leagueoflegends.wikia.com/wiki/Jeffrey_%27Lyte%27_Lin)  announced his company’s openness to proposals from academics. Indeed, there is increasing interest in and acceptance of academic-corporate collaborations.

> HI offers a space for cross-pollination.

### What’s the role of HI Labs?
With our interdisciplinary staff and commitment to ‘real-world’ application [Harmony Institute](http://harmony-institute.org/work/case-studies/)  is in a unique position to respond to the challenges issued at APS. In addition to serving as a potential site of academic research and collaboration, HI offers an exploratory space where experts from fields as different as neuroscience and documentary filmmaking can connect, learn from one another, and and better understand the role of media in the human experience.

HarmonyLabs08:10:151.jpg

Closing the loop between lab and field: Reflections on APS 2015

Light Blue

Last year we started down a new path to define what a future of Beneficial Media might look like, a future where media systems were actually good for people and good for society. To start, we convened hundreds of leaders from across media in a conversation not about what was wrong, but about what could be right. We set out to identify what values would define that future, what might be needed, what might be useful. As so much has shifted over the last six months so have our plans. The coalition we had hoped to launch to further define and advocate for Beneficial Media has, for the moment, been replaced by this livestream series to continue that conversation and work. But it is also a response to the moment and an expression of the need to engage in the question of what media will be like and should be like in a post-COVID world.

[](https://www.crowdcast.io/e/beneficial-media-future-052720)

![HLabsLivestream1.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Livestream1_abbbd1b93e.jpg)

With over a hundred people attending, chatting, asking questions, it was every bit as vibrant as we hoped. Our guests  [Talia Stroud](https://twitter.com/taliastroud?lang=en)  from the University of Texas,  [Aimee Woodall](https://twitter.com/aimeewoodall)  from The Black Sheep Agency, and  [Justin Purnell](https://twitter.com/jpurnell)  from NBCUniversal led us on a conversation that took us through our current search for content that Justin described as a “flight to quality” to a deeper conversation about what changes we’re experiencing might be durable. Talia described the potential of “attending to compassion and empathy” that had been the norm in the early weeks of this pandemic but that could be an experience we design for in the future. She also identified a deficit in local information systems that speaks to a need for the revival of local information — a topic we will return to in a future livestream discussion.

Questions from the community took us in the direction of new business models, new incentives for quality, and the need for media system to help honor and guide us to expertise, especially in times of crisis.

We hope you’ll watch the recording  [here](https://www.crowdcast.io/e/beneficial-media-future-052720) and tune in for our next  [session](https://www.crowdcast.io/e/misinformation-06242020)  on 6/24 where we will be talking mis/dis-information with  [Claire Wardle](https://twitter.com/cward1e)  ([First Draft](https://firstdraftnews.org/)),  [Ashley Bryant](https://twitter.com/_alynnbryant)  ([A/B Partners](https://www.abpartners.co/)), and  [Lisa Kaplan](https://twitter.com/lisackaplan)  ([Alethea Group](https://www.aletheagroup.com/)).

[](https://www.crowdcast.io/e/misinformation-06242020/register)

![HLabslivestream2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labslivestream2_0b0d1ca6c0.jpg)

Register  [here](https://www.crowdcast.io/e/misinformation-06242020)  for our next Beneficial Media Futures livestream on 6/24.

These are conversations that we will continue to host and if you have suggestions for possible topics, drop them in the comments below. We hope these discussions can be the beginning of a broader movement toward Beneficial Media.

Much more to come. Be well.

HLabsLivestream1.jpg

From a “Flight to Quality” to “Attending to Compassion and Empathy”: a Recap of Harmony Labs’ 5/27 Livestream Discussion

Nature-comms.jpg

Audience preferences are predicted by temporal reliability of neural processing

Muted Blue

by  [Equis Research](https://www.equisresearch.us/),  [Harmony Labs](https://harmonylabs.org/), and  [PredictWise](http://predictwise.com/)

&nbsp;

Digital media and cultural content have exploded over the last decade leaving advocates and organizers guessing about who is where. Creating coherent communication strategies to engage any specific audience in the U.S. is challenging by itself. But developing smart, focused strategies for distinct subpopulations, like the Latinx population, requires a nuanced understanding. In this critical time, we need to know more about where Latinx audiences spend their time consuming news and information both online and offline so we can create a more sophisticated set of approaches to reach and engage them. That is why  [Equis Research](https://www.equisresearch.us/)  partnered with  [Harmony Labs](https://harmonylabs.org/)  and  [PredictWise](http://predictwise.com/)  to better understand the media consumption habits of America’s fastest growing part of the population. The result is a comprehensive view of Latinx media consumption that we are excited to make publicly available in this  [new report](https://static1.squarespace.com/static/58595e5af5e23119f38fa37b/t/5e851485c9453c34f3784f6b/1585779845615/Harmony+Equis+v7+-+Final.pdf).

[](https://static1.squarespace.com/static/58595e5af5e23119f38fa37b/t/5e851485c9453c34f3784f6b/1585779845615/Harmony+Equis+v7+-+Final.pdf)
![HLabsLatinx1.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Latinx1_562c6188cc.jpg)

Why? While television is a critical medium to continue reaching certain subsegments of the Latinx population, we know that a growing share of the community engages with news and entertainment content online. According to The Pew Research Center, on a typical weekday, three-quarters of U.S. Latinos get their news from internet sources,  [nearly equal to the share who do so from television](https://www.pewresearch.org/fact-tank/2018/01/11/among-u-s-latinos-the-internet-now-rivals-television-as-a-source-for-news/). And among the younger Latinx community, there are a growing number of cord-cutters with a preference for getting information online. As increasing amounts of investment go toward digital media, knowing where and how Latinx people are online presents many opportunities to reach Latinx voters where they are.

## Key Takeaways
![HLabsLatinx2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Latinx2_c7346408fe.jpg)

Full report available  [here](https://static1.squarespace.com/static/58595e5af5e23119f38fa37b/t/5e851485c9453c34f3784f6b/1585779845615/Harmony+Equis+v7+-+Final.pdf), but key takeaways include:

### The internet is not just for young people
This research looked at users’ desktop behavior and, overall, Latinx audiences are active on social media. But, they differ from the national population in the platforms where we can reach them with ads. YouTube and Google are more important digital ad platforms than Facebook. YouTube is especially important. Latinx audiences spend twice as much time there than do Non-Latinx adults.

Importantly, however, there are three very different types of Latinx internet users. Younger people spend a lot of time making and consuming video, especially e-sports. Slightly older users, especially women, connect with others through sites like Facebook. Busy families use the web to bank, shop, work, and learn.

Three distinct Latinx web audiences as defined by minutes spent on their top four websites.

![HLabsLatinx3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Latinx3_5b903e66f7.jpg)

### On television, to reach Latinx audiences, reach beyond the news.
News makes up a small proportion of TV consumption for everyone. For Latinx audiences, this proportion is even smaller (~11%) as they turn to TV for entertainment vs. information.

![HLabsLatinx4.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Latinx4_4c6a351ff1.jpg)

News consumption as a % of total consumption

### When they are watching TV news, they often tune into Spanish-language programming.
Even among Latinx adults who prefer entertainment content, talk shows like Despierta America and mainstream media like CNN Newsroom have substantial reach.

### For news online, it’s all about the mainstream media.
Online and on TV, favorite English-speaking news sources include mainstream news from outlets like Yahoo, CNN, ABC, and MSNBC.

### Latinx TV audiences vary with respect to language preference and focus on news.
There are four very different TV audiences within the Latinx community. Older Spanish-speaking Latinx adults watch news, telenovelas, and talk shows. Spanish-speaking families watch some news and a lot of kids programming. Older English-speaking Latinx adults watch a lot of news, but the largest and youngest audience watches English- language drama, film, and sitcoms, not news.

![HLabsLatinx5.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Latinx5_b30be9a121.jpg)

Four distinct Latinx TV audiences as defined by minutes spent watching their top ten programs.

## Putting This into Practice

At Equis our work is rooted in the needs of state and national partners, with whom we work to ensure our research and learnings can improve the efficacy of engagement tactics to reach and engage Latinx voters and increase civic participation. Here is how we imagine the insights from this report might be applied to real world engagement and targeting efforts:

### 1. Build a digital advertising strategy with YouTube, Facebook, and Google Search Ads as cornerstones.

Latinx audiences use more social media overall than non-Latinx audiences. For all three distinct audience segments, the top sites by far were Google, YouTube, then Facebook. Since all three platforms permit digital advertising, they should form the core of any Latinx digital advertising strategy. But the three platforms serve different audiences and require different approaches to targeting.

**Center YouTube.**  Latinx adults overall spend about twice as much time on YouTube as on Facebook. Younger Latinx adults (30% of adults under 30 are high YouTube users) in particular tend to spend a huge amount of time on the platform. Although YouTube doesn’t offer precise targeting based on third-party individual-level audience data, it is a cornerstone of the Latinx online engagement given the propensity of Latinx people who engage with the platform regularly.

**Find precision on Facebook**. Facebook still dominates online time for one segment of Latinx — adults who use the internet primarily to connect with other people. Those folks will, presumably, be even more engaged with social platforms during the COVID crisis, and Facebook allows us to use lists of individuals for super-precise targeting.

**Get reach with Google Ads.** 41% of Latinx adults don’t spend a lot of time on YouTube or Facebook. They are using the internet to do their banking, work from home, get support for their families, and learn. To reach this group, we can use Google Ads, especially in Spanish, to reach them during their search experience.

### 2. Explore a cultural strategy for young Latinx voters that centers e-sports and e-gaming.
Segmenting by consumption behavior rather than demographics revealed new places of opportunity. Twitch, Roblox (live streaming platforms for gamers) and e-sports platforms are extremely popular with the same segment of the young Latinx audience that spends a huge amount of time on YouTube. Activating young voters in this cycle means not just mobilizing them through GOTV but  _persuading_  them that their civic engagement is important. Reaching this young audience where they are, especially in the online gaming and e-sports world — may be a rich medium for that critical culture changing work.

### 3. Build a TV and press strategy with a focus on entertainment
Most of the TV content Latinx audiences consume is entertainment. 41% of Latinx adults are most effectively reached during English-language programming like drama, sitcoms, and film rather than news. Even the preferred online and TV news sources for Latinx audiences tend to be lighter in tone and more centrist than what non-Latinx audiences are consuming. Latinx audiences spend proportionally less time than non-Latinx audiences watching news from Fox News, for instance, and more on Univision’s Despierta America and local news from ABC, NBC, and CBS affiliates. Online, they are more likely to be found on Yahoo, CNN, and MSNBC than the Washington Post. Latinx audiences expect and prefer their news to contain lifestyle content as well as newsy information. As we are shaping our content creation strategies, we can use the tones of these preferred channels to make content engaging, entertaining, and informative.

## Conclusion
Our hope is that these insights, and the others contained in the  [full report](https://static1.squarespace.com/static/58595e5af5e23119f38fa37b/t/5e851485c9453c34f3784f6b/1585779845615/Harmony+Equis+v7+-+Final.pdf), can provide a helpful map of the Latinx audience from which many others can draw their own impactful routes to engagement and activation. In short, we hope it brings a little more clarity and guidance to possible approaches in what are certainly unclear times.

If you are interested in learning more about this report or have ideas for leveraging the findings in your work, we encourage you to  [get in touch](https://harmonylabs.org/contact)  and start a conversation.

&nbsp;

---

### Methodology
Survey-based approaches to collecting media consumption information are inadequate. Self-reported survey data about personal media consumption habits continue to yield unreliable estimates on media consumption patterns, [as demonstrated by recent research spearheaded by PredictWise](http://researchdmr.com/BehavioralSurveyData). Instead, in this study we relied on novel behavioral data: two opt-in consumer panels that passively track real-time desktop browsing and TV watching patterns of thousands of Latinx adults without the bias of traditional self-reported surveys. We explored television and online consumption habits including time spent on social media on Facebook, Twitter, Reddit, and YouTube. We dug into consumption by program/station, website, social media platform, and, where possible, we broke down Latinx consumption by age, gender, education, and region/state.

Panelists on both panels volunteer for data collection just like survey panelists do, and the data we analyzed were anonymized. The 2019 television-viewing panel includes monthly individual-level consumption data of 90,000 Americans in approximately 50,000 households. We analyzed data only for adults, of whom about 11,500 self-report Latinx ethnicity. To create the television consumption data, panelists’ televisions track the name of the program and station being watched on a minute-by-minute basis, for both live and digitally recorded, or “TIVoed,” national television (i.e., excluding local programming).

The 2019 internet data include monthly web use data (desktop only) of about 65,000 adult users nationwide, about 8,000 of whom self-identify as Latinx. Panelists’ web browsers track, on a second-by-second basis, the URL being visited; the duration and metadata of each pageview is recorded. In multi-person households, panelists manually record who is browsing.

&nbsp; 

---

### About Equis Research
[Equis Research](https://www.equisresearch.us/)  was started with the vision of creating a better understanding of Latinx communities, investing in innovative approaches to engage them (specifically using digital, data and tech tools).

### About Harmony Labs
[Harmony Labs](https://harmonylabs.org/)  builds communities and tools to reform and transform media systems. Our mission is to create a world where media systems support healthy, democratic culture and healthy, happy people. We’re organizing partners and publics to identify and gain consensus on the mechanics of media influence, and the values and design principles that underlie beneficial media. And, we’re building solid, sustainable interventions and innovations that start moving us in the direction of media that can serve the public good.

### About PredictWise
[PredictWise](http://predictwise.com/)  provides audience solutions for our digital first information environment. Our audiences are generated from massive survey and mobile device data collected over years, and provide individual-level targets that are completely tailored to any campaign. PredictWise ID resolution technology allows our clients to reach top targets instantaneously, no matter where they are, guaranteeing more yield on all relevant metrics. Our core ethos is to help our clients serve the right message to the right individual.

equis-cover.png

Mapping media consumption across TV, the internet, and the news

What Media Matters for Latinx Americans?

Over the last few months, we’ve been working with the  [Nuclear Threat Initiative](https://www.nti.org/)  (NTI) to identify narratives that support a nuclear weapons-free future. We teamed up with  [Kirk Cheyfitz](https://politicalnarrative.us/)  and Worthy Strategy Group who did deep qualitative research with audiences and storytellers to find out just how the stories NTI wants to tell can work.

For an in-depth look at this project and the story opportunities that emerged from it, there will be a community briefing on September 13th. You can register  [here](https://us06web.zoom.us/webinar/register/WN_4IXou_MvSpaJsru7cOrnHw).

In advance of that bigger conversation, we wanted to share one important finding from this project that can be used in narrative work on any issue.

## Put a Future in It, Friend
This abbreviates to PFIF, which is somewhat less compelling than KISS: Keep it Simple Stupid, but it was the single most important insight gained in the quantitative testing for this project. Across the dozens of concepts tested,  audiences only moved on denuclearization if the story concept included a vision for a positive future.

Here is just one example from this testing: the first version moved the audience in the desired direction, the second did not.

![HLabssellend1.jpeg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labssellend1_1e7b268825.jpeg)

This concept, with its positive future, moved the target audience.

![Hlabssellend2.jpeg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabssellend2_fa17f49294.jpeg)

This concept, without a positive future, did not move the target audience.

The same story told two ways works when we close with the opportunity for a better future. The same was true for another concept, this one created by a professional with real ad experience. The concept here is beautiful and compelling, but it didn’t move the audience.

![HLabssellend3.png](https://harmony-strapi-storage.s3.amazonaws.com/H_Labssellend3_101f02be11.png)

But, when we turned that question mark about “2030?” into a certainty that we have the opportunity for a plentiful future, the audience moved.

![Hlabssellend4.jpeg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabssellend4_5bf777e65d.jpeg)

When the tone of the creative was general frustration — a pretty common approach in advocacy media — the audience stayed put.

![hlabssellsend5.png](https://harmony-strapi-storage.s3.amazonaws.com/hlabssellsend5_e763e29aaa.png)

But,  when the PFIF insight was applied, and there was an emphasis on learning from the past, the audience moved.

![Hlabssellend6.jpeg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabssellend6_2749d4d69e.jpeg)

## What does it mean for a story to “work?”
Not every creative concept moved the audience, but the most successful ones consistently depicted the future.  Part of the reason was the definition established ahead of time of what “works.” Often, advocates use immediate action as a metric for success: clickthroughs, emoji reactions, or donations.

Scaring people might be effective in getting them to act right now, but in narrative change, persistent, long-term commitment to a healthier future is typically the goal. So the measurement for what “worked” was based on answers to  three key questions,  asked via survey after encountering these stories:

-   “**I can imagine a future** **where every country eventually eliminates all nuclear weapons.**” [Strongly Agree — Strongly Disagree]
-   “**I want to live in a world where** **no country, including the United States, has any nuclear weapons.**” [Strongly Agree — Strongly Disagree]
-   “**I personally can do something** **that gets us closer to a world without nuclear weapons.**” [Strongly Agree — Strongly Disagree]

We combined these answers into an “index” of endorsement for the overall narrative and scores on that index were compared for people who saw the story concepts to scores to people who saw neutral control media, that is media not about denuclearization. Looking at how the neutral control sample answered these three questions helps explain why stories about nukes need a future so badly.

Desire for a nuke-free future is already high. While conventional advertisers may need to convince consumers they want a product (like tooth whitening gel), nuclear disarmament advocates have already made that “sale”: almost 80% of the control sample in this study agreed that they want no one (including the US) to have nukes.  But just half of the survey respondents in the control sample reported being able to _imagine_ a world without nuclear weapons, and only about a third believed they could be a part of the change.

“Narrative change” might have become a buzzword—and its precise meaning is still being discussed—but everyone agrees that a narrative is made of individual stories and that a story has a beginning, a middle, and an end. Narrative change communications can use some of the same tools as advertising (like the testing we used here), but they should keep in mind they aren’t selling the problem—climate change, poverty, nuclear weapons, or unregulated artificial intelligence.  They need to be selling an ending audiences want to “buy into.” Naming the problem has its place but selling the ending—the future we can achieve together—is where the change is likely to come.

cherry-blossom.jpg

Dispatches from Data Science: Selling the Ending

Light Yellow

During our first livestream session on Beneficial Media futures post-pandemic misinformation and disinformation were key topics that we chose to anchor our 6/24 panel (you can watch a replay  [here](https://www.crowdcast.io/e/misinformation-06242020)). Misinformation and disinformation undermine our ability to make sense of the world around us and, in a crisis, challenge our ability to build consensus to respond effectively. What are the roots of these very different issues in media, and what can we do about them?

[](https://www.crowdcast.io/e/misinformation-06242020)

![HLabs2livestream1.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs2livestream1_8670f83c14.jpg)

With a vibrant audience contributing to the conversation from the get-go, our guests Ashley Bryant from  [A/B Partners](https://www.abpartners.co/), Lisa Kaplan from  [Alethea Media](https://www.aletheagroup.com/), and Claire Wardle from  [First Draft](https://firstdraftnews.org/)  led us on a discussion that started with Claire explaining the essential distinctions between misinformation and disinformation and Ashley bringing us some history on leveraging media to build power in disempowered communities. We explored why these issues felt so much more impactful now and who should be responsible for solving them, specifically in the context of the TikTok spoof of President Trump’s campaign rally and the burgeoning Facebook and Instagram advertising boycott. Lisa invited us to think about where responsible interface design might overlap with regulation and commercial pressure. In the end, we found some optimistic ground about what could be possible in a Beneficial Media future — even in a moment where these challenges felt existential.

Next Wednesday, August 12th at 2 pm ET, we’ll be co-hosting our next conversation with  [A/B Partners](https://www.abpartners.co/)  to release some new research on media consumption in Black communities we created together. We’ll use that report as a jumping off point for a deeper conversation about race and racism in media and how Beneficial Media can and must be anti-racist. You can  [**register here**](https://www.crowdcast.io/e/antiracism-in-media-08122020/register)  to attend. These are conversations that we will continue to host and if you have suggestions for possible topics, drop them in the comments below. We hope these discussions can be the beginning of a broader movement toward Beneficial Media.

More to come. Stay safe and healthy.

HLabs2livestream1.jpg

Misinformation, Disinformation, and the Role of the 4th Estate: a Recap of Harmony Labs’ 6/24 Livestream Discussion

This weekend’s [PyGotham](https://pygotham.org/2015/talks/schedule/) conference renewed an appreciation for all things Python in NYC. Python is a programming language known for being concise and easy for humans to read. [As the R-using sole holdout at Harmony Institute, I’m still getting used to Python’s conventions and approaches to data gathering, analysis, and presentation. PyGotham was a chance for me to see what others are doing, and get excited about new possibilities.]

The range of topics, industries, and goals represented by different users at PyGotham was impressive. From detecting [sarcasm in speech](https://pygotham.org/2015/talks/152/using-python-for-sarcasm/) to developing a database of [NYC boiler information](https://pygotham.org/2015/talks/143/collecting-and-organizing/), presenters discussed how they’re using the seemingly infinite flexibility Python to attack their problems. The audience also had a diverse range of expertise required, from the newer users wrapping heads around constructing a web app (Thanks, [Kat Chuang](https://pygotham.org/2015/talks/219/one-two-buckle-your-shoe/)) or a data pipeline (Thanks, [Joe Cabrera](https://pygotham.org/2015/talks/202/building-a-data-processin/)), to more experts debating the relative merits of running applications on cloud container services (Thanks, [Jeff Uthaichai and Chris Becker](https://pygotham.org/2015/talks/163/docker-containers-in-the/)) and scaling automated recommendations in real time (Thanks, [Brian Muller](https://pygotham.org/2015/talks/116/untangling-twisted-how-we/)).

A constant theme throughout the weekend was the uphill battle of data organizing. **Data is everywhere, but generating real insight from any dataset demands a lot of specific effort**. At PyGotham, much discussion swirled around working with city and federal governments’ woefully inaccessible data and out of touch systems. Several presenters reflected on the chaos of receiving different, idiosyncratic file formats; approaches to extracting data from PDFs came up repeatedly. While it’s not unique to Python users, the Python packages and solutions available for scraping, organizing, and using data are sophisticated. Finding ways around these stumbling blocks was also cause for celebration — presenters spoke about the package BeautifulSoup with reverence, and applause broke out for [Steven Slotterback](https://pygotham.org/2015/talks/143/collecting-and-organizing/)’s workaround for grabbing NYC boiler data without having to fill out a captcha every time.

&nbsp;

---

&nbsp;

_To learn more about Python and the specific topics discussed,you can find recorded conference talks at [http://www.pyvideo.org/.](http://www.pyvideo.org/)_
_Special thanks to the Women in Machine Learning and Data Science group, who extended the invitation!_

HarmonyLabs08:20:151.jpg

Researcher Corinne Brenner was invited to attend PyGotham 2015, here are some of her reflections.

PyGotham 2015

science.png

Evaluating the fake news problem at the scale of the information ecosystem

People make and reflect culture by telling stories. Patterns exist in the stories people tell, even when those stories pertain to different social issues or topics. The patterns that emerge from the shows we stream, the music we listen to, the games we play, and the news we share, we call narratives. Think of narratives as the averages of many, many stories, the things that hold true among them. Stories are like tiles in a narrative’s mosaic, so to speak, the stars that make up a narrative constellation.

Many organizations and media makers are working hard to shift the deep cultural narratives that shape public understanding and inspire action on all kinds of social issues like poverty, immigration, and climate. Often this work lacks a shared understanding of what narrative is, how to identify and characterize narrative reliably, and how narrative connects to people with the power to make change. We’re hopeful that this is a place Harmony Labs can help. It’s the motivation behind our  [Narrative Observatory](https://harmonylabs.org/narrative-observatory).

We’ll be digging into the technical specifics of the Narrative Observatory in upcoming blog posts. For now, here are a few thoughts in video format on narratives and narrative change, starting with some definitions:

![Work-Harmony-Labs.png](https://player.vimeo.com/video/448129295?h=e4dbdb33f7)

Building on the definitions in  _What’s a Narrative,_  we can start to reliably identify, track, and measure narratives across media types over long time scales using a combination of computational text analysis and human reading and annotation. At least that’s the work we’ve been knee deep in for the last six months. Narratives can also be connected to the people or audiences who consume, create, and amplify them in culture. For part 2 in our video series, we take a closer look at identifying narratives through the lens of climate skepticism:

![Work-Harmony-Labs.png](https://player.vimeo.com/video/448131403?h=19db0609db)

Once narratives and audiences have been identified or mapped, then we can start working to understand how to produce change. Using a narrative landscape as a guide can yield new opportunities to reach and engage your audience. In the final video of this series, we look at the issue of Artificial Intelligence and examine how various creative interventions influenced public perception:

![Work-Harmony-Labs.png](https://player.vimeo.com/video/448140993?h=04de7b6304)

## Building a Narrative Observatory
We’re excited about the potential that this basic approach holds for foundations, advocacy organizations, and anyone interested in understanding culture to make better media. And we are already developing a new data platform to identify, measure, and track narratives related to social issues, over long time scales and across multiple cultural domains; learn about the people or audiences who participate in these narratives; and provide narrative change makers a framework for content effects testing.

We call it the Narrative Observatory. You can learn more about this project on our  [website](https://harmonylabs.org/narrative-observatory), explore one of our first outputs  [here](https://obiaudiences.org/), or follow along here, on this blog, as we document our journey, and explore together how to solve some of the technical, definitional, and practical challenges that bedevil narrative change research and measurement.

HLabsNarrative1.jpg

What’s a Narrative?

At Harmony Labs we’ve been thinking a lot about technology. While some tech innovations have clearly changed the world for the worse, they’ve also produced 90% effective  [COVID vaccines in less than a year](https://www.marketplace.org/shows/marketplace-tech/covid-19-technology-moderna-pfizer-messenger-rna-coronavirus/). There’s no going back. We need tech, but we need it to work for us, not against us. And we need the stories we tell about tech to center us — the humans — so that people don’t feel powerless to advocate for “healthy” tech, now and in the future.

The Mozilla Foundation foregrounds this kind of human agency in their call for  [trustworthy AI](https://foundation.mozilla.org/en/internet-health/trustworthy-artificial-intelligence/):

> We need to move towards a world of AI that is helpful — rather than harmful — to human beings. For us, this means two things: human agency is a core part of how AI is built and integrated and corporate accountability is real and enforced.

Some AI-focused media narratives foreground human agency as well, like the Augmented Intelligence narrative we identified in  [this recent analysis](https://docsend.com/view/zdn5h5r8z3cez8ud). Augmented Intelligence presents technology as augmenting our intelligence, but capable of being unfair and downright dangerous. We, humans are responsible for designing and regulating it, so that it helps us and doesn’t hurt us. This analysis looked at survey and online news data, but we wanted to know how often people encounter this narrative in other forms of media.

![Work-Harmony-Labs.png](https://www.youtube.com/embed/fn3KWM1kuAw "Boston Dynamics robot dancing to  Do You Love Me?")

So, we looked at TV, starting with more than 80,000 adults living in more than 39,000 households in the US who participate in a TV consumption panel. We connected all the shows they watched on TV in November 2020 with the closed captioning transcripts of those TV shows and the ads that accompanied them. (Drawing those connections is a technical monster. If you want to know more about the data science, please don’t hesitate to  [reach out](https://harmonylabs.org/contact).)

What we found were lots of ads for robot vacuums. Oh, the robot vacuums. These little friends make magical maps of our houses, clean while we’re out, talk to us on our phones, and — most importantly — never get tangled up in their cords or discommode us in any other way.

![HlabsRumba2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs_Rumba2_8b1107e5f2.jpg "Advertisement for the Roomba**®** Robot Vacuum from iRobot")

These robot vacuum ads convey a narrative we call Wishes Granted. In this narrative, technology is a tool made by humans. But instead of having risks as well as benefits, it operates like a magic wand, solving our problems quickly and without worry. This narrative shows up not just in robot vacuum and other smart home tech ads, but also in ads for coffee makers, laundry detergent, and even menstrual pads.

Surprisingly, **at least 67% of the content audiences consumed about technology — artificial intelligence, machine learning, and algorithms included — was in Wishes Granted-type ads.**  That’s 5 times as much exposure to Wishes Granted as to any other tech-related narrative!

![HLabsRumba3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Rumba3_8933370ef5.jpg)
At any given time, audiences are consuming many different narratives, and storytellers will be most successful if they understand that narrative landscape as a whole. A big motivation for us in building the  [Narrative Observatory](https://harmonylabs.org/narrative-observatory)  has been to create a way to help our partners understand this narrative landscape and spot the story opportunities and threats.

The sheer dominance of the Wishes Granted narrative has the potential to create resistance to the more nuanced Augmented Intelligence narrative, by constantly reinforcing the consequence-free magic of technology. But it is also doing some of our narrative work for us. It’s inspiring audiences with the tremendous promise of technology and artificial intelligence to make our lives better. So all we have to do is to move audiences a bit, to understand that tech comes with risk, and that we humans have the power to control as well as to invent.

&nbsp;

---

&nbsp;

_If you are a researcher who would like to access the data used in this analysis, and other datasets too, please consider becoming part of our_ [_research network_](https://harmonylabs.org/research)_. We’re always happy to geek out on methods, so please_ [_reach out_](https://harmonylabs.org/contact) _to us to learn more about our models and how we do what we do._

HLabsRumba1.jpg

The Most Common Narrative About Tech Is Not What You Think: Robot Vacuums and Wishes Granted

Understanding people’s media habits gives us a deeper view into who they are, the values they hold, and how to best engage them. So, whether you like K-Pop or UFC (or maybe what’s in your  [fridge](https://www.nytimes.com/interactive/2020/10/27/upshot/biden-trump-poll-quiz.html)?) can actually say a lot about how you see the world and the issues you care about.

For many professional communicators, audience segmentation is pretty standard stuff. Traditionally, demographics have served as a proxy for audience, because data about age, race, and gender are relatively easy to get. But, strategy for culture change requires understanding not just audiences’ demographics, but their cultures: the TV, news, games, music, food, film, and fashion people regularly engage with.

As digital media and cultural content have exploded in volume and variety, it’s become increasingly difficult to know where and how to reliably reach and engage audiences, and the traditional approaches just don’t tell us enough.

## Don’t start with the story you want to tell, start with stories your audience is already consuming
The people we want to reach don’t spend their time thinking about us and our issue. So what stories do they live inside? What voices do they value? How can we speak to them in ways that resonate? Using behavioral cultural consumption data — actual minute-by-minute records of what people watch and click on — allows us to know exactly the platforms, genres, and content specific audiences are actually engaging with, and to characterize engagement behaviors and content with respect to narrative and other important story dimensions.

In practice, our approach to audience analysis seeks to answer these core questions for storytellers:

-   **Who are you talking to?** The different audiences which differ with respect to core values and also cultural consumption and the narratives they hold about core social issues, like climate, immigration, or education.
-   **Where are they in media and culture?**  The genres, platforms, media formats, and voices that engage each audience, so that storytellers can find and reach them where they are.
-   **What do they care about?**  Their core values or personal goals — e.g. help others, stay safe, get rich, have fun — and how those values show up in different cultural milieus.

Recently we launched a  [new website](http://obiaudiences.org/)  that takes this approach and aims to help organizations and media makers change the stories we tell ourselves about poverty and economic mobility in the U.S.

[](https://obiaudiences.org/)

![HLabsMusic2.gif](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Music2_0ac88ea175.gif "Audience profiles at  narrativeobservatory.org")

This project is a deep-dive into 4 distinct audiences in the U.S. through the lens of the media they are encountering. For those interested in the data science behind this work, here’s how we did it:

-   We started by reanalyzing some existing  [audiences created by GOOD](https://onedrive.live.com/?authkey=%21ACSPUeSyVAEdByE&cid=624A7F93617FBDB7&id=624A7F93617FBDB7%21147&parId=624A7F93617FBDB7%21129&o=OneUp). This gave us baseline survey questions and working audiences.
-   Next, using demography and geography, we fuzzy-matched the working audiences with behavioral data capturing online and television habits of 50,000+ Americans and read media for each audience related to poverty, COVID, and BLM. To validate what we found, we duplicated this process on a second media consumption data set.
-   These readings informed additional survey questions, connected to core values and attitudes for race, gender, place, and class. By combining values — It’s more important to me to be helpful than successful — with poverty beliefs — Poor people manage their money well — we were striving to capture in our audiences both poverty perspectives and information to predict how each audience is likely to respond to COVID, BLM, or whatever other crises 2020 brings.
-   Then we conducted a voter-file matched survey with ~2,900 respondents, and analyzed survey results to update our audiences and generate an 8-question audience classifier.
-   Finally, we re-read the behavioral data regarding online and television consumption to enrich each audience profile and associate them with poverty narratives in news and Twitter. So all the sample Pinterest pins, streaming videos, tweets, news articles, and other media are actually derived from data.

To see what kind of stories and media this approach surfaces, step inside a “[day in the media life](https://obiaudiences.org/feed/)” of one of these four audiences and explore the real-life media artifacts collected from 50,000+ Americans between January and October 2020.

## So, is your audience more Tiger King or Hamilton?
Starting the storytelling process by getting to know your audience — particularly those audiences who don’t currently occupy a lot of the narrative space on an issue (because they’re busy with the things they care about) — gives us hints about the channels, influencers, and formats for stories that can move people. Using core values as the frame for understanding those audiences gives us a lot of hints about how to tell stories that engage and persuade those audiences.

![HLabsMusic3.gif](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Music3_fe3a942b24.gif)

In short, understanding your audience through their affinity for  _Tiger King_  or  _Hamilton_, or through the lens of countless other cultural touchstones, will make for better stories.

![HLabsMusic4.gif](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Music4_224529164c.gif)

If you want to find out what audience you are, we even created a  [nifty quiz](https://obiaudiences.org/?quiz)  to find out.

&nbsp;

---

&nbsp; 

_If you are a researcher who would like to access the data used in this analysis, and other datasets too, please consider becoming part of our_ [_research network_](https://harmonylabs.org/research)_. We’re always happy to geek out on methods, so please_ [_reach out_](https://harmonylabs.org/contact) _to us to learn more about our models and how we do what we do. Special thanks to_ [_Pedro Sanches_](https://www.psanches.com/) _for the wonderful design work on this project._

posty.png

What do Hamilton, Post Malone, Dolly Parton and Axl Rose have in common?


Measuring the impact of games is a complex challenge. We’ve  [considered why games may be effective from a psychological perspective](http://harmony-institute.org/therippleeffect/2012/09/05/social-impact-games-new-challenges-for-change/)  and spoken to designers about the  [delicate balance between accuracy and fun](http://harmony-institute.org/therippleeffect/2012/12/20/anti-speaker-recap-history-and-videogames/). Still, fundamental questions remain. What types of people are playing games, and to what effect?

Games are a huge component of the entertainment marketplace —  [Industry analysts project](http://www.reuters.com/article/2012/06/01/us-videogameshow-e3-show-factbox-idUSBRE8501IN20120601)  that gaming revenue, including sales from mobile devices would reach $78.5 billion in 2012. However, detailed social research on gamers is still an emerging field. In order to better understand some of the issues at play, we spoke with  [Cornelia Brunner](http://cct.edc.org/people/brunner-cornelia)  of the Center for Children and Technology.

Brunner has worked in education for over 40 years, starting at Sesame Street and moving on to experimental and progressive classrooms that incorporate new media and technology. Pivoting off our work in game evaluation, she explained her ongoing research on games in educational settings. We asked Brunner, what would an inclusive, meaningful game look like?


![HarmonyLabs07:21:151.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Harmony_Labs07_21_151_e6a88a7bda.jpg)

One of Brunner’s insights, culled from ongoing psychological research with students, is that different people conceive of technology in different ways. More specifically, she has observed that people access technologies through  [gendered perspectives](http://youtu.be/TmROmy5jT80). Brunner terms the perspectives “butch” and “femme” to emphasize the social construction of gender roles and avoid confusion with biological sex. Importantly, members of either sex can interact with technology from either perspective, in varying degrees.

Generalizing through extended content analysis, Brunner has formalized a set of dichotomies that can be applied to technology while noting that these categories are fluid and subject to change over time. Butch perspectives prefer centralized technology solutions that transcend limitations, while femme perspectives want diffuse artifacts that help share and communicate. For example, when asked to draw a picture of the World Wide Web, the butch tendency was to draw a representation of a browser window, while the representative femme illustration resembled a network map.

This framework is particularly useful for analyzing video games. Brunner identified two gendered dimensions that apply to many games: collision-arrangement and shooter-puzzle.

![HarmonyLabs07:21:152.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Harmony_Labs07_21_152_fb62f26070.jpg)

The red letters (F,B) refer to femme and butch perspectives.

An interesting consequence of video game development is that it has been easier to develop video games in the butch-dominant quadrant of collision and shooter. Tetris, perhaps the best pure example of a game based upon the principles of arrangement, stands as a historical outlier to this theme. Only relatively recently in the history of gaming have simulations been powerful enough to begin to realistically model complex social interactions as seen in The Sims (2000) or more recently in games like  [Prom Week](http://promweek.soe.ucsc.edu/?p=101)  (2012). Brunner also noted that mobile technology, such as smartphones and tablets incorporate more femme traits like expressive, exploratory, focused on communication than butch characteristics.

This understanding of gendered perspectives provides researchers a tool for designing and evaluating games that appeal across a broader range of perspectives.

This post was originally published on our blog,  [The Ripple Effect](http://harmony-institute.org/therippleeffect/).

_Image: Player interface from Prom Week | University of California, Santa Cruz_

controller.jpeg

Inclusive approaches to gaming: A conversation with Cornelia Brunner

An important part of our work prototyping the  [Narrative Observatory](https://harmonylabs.org/narrative-observatory)  is to understand potential users, and to derive from that understanding a product strategy and sustainability model. So for a couple months this fall, we conducted 32 interviews across 22 organizations, spanning the non-profit and media sectors. Many thanks to everyone who participated!

![HLabsNarrativeO1.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Narrative_O1_5e4efc35fe.jpg)

From these interviews, we created four user types — funder, media maker, in-house strategist, strategy consultant — each with associated jobs to be done, pains, and potential gains. Below are three headlines from our findings, followed by the product direction we think they suggest. You can also explore  [this more detailed synthesis of jobs, pains, and gains](https://www.dropbox.com/s/q99xqh80ziircaj/Narrative-Observatory-Jobs-Pains-Gains-201120.pdf?dl=0).

## When it comes to narrative work, there is a disparity between what people say would be valuable and what they report actually spending time on.
17% of potential gains pertain to better narrative research and using narratives to identify strategic story opportunities; and 9% of potential gains pertain to being able to attribute narratives to particular media sources. However, only 8% — 10% of reported jobs focus on narrative research, identification, and tracking, and 5% on media monitoring. Therefore, a Narrative Observatory focused primarily on describing narrative ebb and flow may not see broad or rapid adoption.

## Across all user types, a clear need emerged for cultural understanding of audiences, in order to develop content, target, buy media, and test effects.
43% of jobs, 22% of pains, and 26% of potential gains overall pertain to cultural audiences. Even people who considered narrative research and tracking too technically burdensome still spend time trying to understand audiences culturally, and to integrate this understanding into content development. Therefore, a Narrative Observatory with entry points focused on the cultural profiling of audiences may foster broad, rapid adoption.

## When practitioners do work with narrative, they’re mostly looking for immediately actionable story opportunities; funders have different needs.
About 18% of reported narrative-related jobs overall pertain to content planning and execution. However, measurement (19%), funding (18%), and coordination (12%) pains for narrative work abound. People feel pressure to take on the burdensome cost and technical complexity of acquiring and wielding data to show narrative impact. And we repeatedly hear how narrative work lacks definitional consistency, clear metrics, and shared monitoring and evaluation frameworks. Therefore, a Narrative Observatory outputting story-based opportunity and threat analysis will be most useful and readily adopted by practitioners; while funders may find utility in supporting a narrative identification, analysis, and tracking layer that can provide a shared basis for investment coordination, monitoring, and evaluation, and reduce perceived reporting burdens among grantees.

These findings suggest a Narrative Observatory with the following capabilities or layers:

-   **Audience identification and cultural profiling.**  Who should we be talking to? What do they care about? Where are they in media and culture?
-   **Narrative discovery, measurement, tracking.** What narratives are associated with our social issue and exposed to our audiences? What’s the prevalence of those narratives over time and across media types, platforms, sources, and other key dimensions? Is there a target narrative that can guide the stories we develop?
-   **Ongoing story threats and opportunities.** What stories engage and move audiences toward our target narrative now? What story threats need to be addressed? What hypotheses can we derive from observation or testing with regard to setting, character, tone, voice, world, or representations of past, present, and future?

And the following customers:

-   **Agencies**  with cultural strategists helping brands and non-profits engage social issues
-   **Media makers and brands**  developing and distributing impact-oriented content
-   **Advocacy non-profits** working on issues in terms of culture or narrative change
-   **Foundations**  with issue-specific narrative and media investments

In order to appeal to practitioners — the key to broad adoption — the Narrative Observatory probably needs to foreground the audience and story opportunity layers. For most customers, the narrative layer may offer an expedient way to refer to significant story patterns vis a vis audiences or content hypotheses, but the narrative layer will not integrate well into day-to-day work, nor drive interaction with the Narrative Observatory. The only users likely to focus on the narrative layer are funders, who will interact with it like a stock index, observing how the narrative “market” moves relative to their investments. For funders with narrative change portfolios, or issue-based funder coalitions, narrative definitions, target narrative identification, and narrative modeling and measurement may serve the additional purpose of helping to coordinate between organizations and among grantees.

> **“The biggest challenge is information overload and underload. We need the ability to zoom out and really look at the whole media landscape rather than just the part you’re most familiar with — that’s the piece that’s most difficult.”**
> 
> **– Eli P., Civic Signals**

This raises a conundrum. While the narrative layer is the most time and resource intensive to conceive, build and operate — and probably the most differentiating, since  narrative models must be custom-built for each issue  — it is the least in demand or valued by three out of four customer types.

We have received initial funding from the Bill & Melinda Gates Foundation to prototype a narrative layer for the issue of poverty and economic mobility in the U.S. This funding also aims to help chart a course toward market-based sustainability, as the Narrative Observatory expands to additional social issues, like climate, education reform, or immigration. But the upfront cost of developing narrative models appears out of balance with the market opportunity suggested by our research. So philanthropic investment for additional issue-specific narrative models remains a necessity, and may provide important network effects for driving Narrative Observatory adoption. Without this investment, audience and story-based feature development will take priority over narrative features, which may ultimately yield a weaker, less differentiated product.

> **“When you build a movement you are supposed to build on top of what’s already there. You don’t know what the long-term looks like. You’re just hitting current events and moving with the news, which is the speed of light. The long-form of how to track a movement over 10–15 years is something I see lacking in the narrative space today.”**
> 
> – Rohit K., SB Digital

This suggests a revenue model where foundations trigger the Narrative Observatory to expand into a new issue by funding the development of new narrative models, plus whatever support grantees need to make full use of related audience and story insights. General per issue insight on audience, narrative, and story live online for anyone to access. And tiered subscriptions are available for custom audience work and/or ongoing reporting on story threats and opportunities.

We developed  [this website](https://obiaudiences.org/)  to support the Bill & Melinda Gates Foundation grantees working on poverty and economic mobility narrative change, and also to test our hypotheses about cultural audience information being the most useful entry point to the Narrative Observatory. Please have a look, play around, and, if you like, participate in our user testing by completing  [this survey](https://harmonylabs.typeform.com/to/ERIbeXXw).

> **“We need more concrete evidence of the impact of our narrative change work. A tool like the Narrative Observatory could contribute to that. It would be a more concrete way to capture impact that is long-term and focused.”**
> 
> - Perspective Fund

The development of the Narrative Observatory is a community undertaking, meant to serve people working on culture change. So please let us know what you think, what we got right or wrong, using  [this survey](https://harmonylabs.typeform.com/to/u10najpW), so we can refine our direction and make something you actually need! You can also follow this blog or  [subscribe to our newsletter](https://harmonylabs.org/contact/#contact-subscribe)  to stay in the loop on all things Narrative Observatory.

&nbsp;

---

&nbsp;

_The full report on our user research findings is_ [_available here_](https://www.dropbox.com/s/kj7d6d6cmbyyixs/Harmony%20Labs_Narrative%20Observatory_User%20Research_Emerging%20Insights%20External_Fall%202020.pdf?dl=0)_. This report is based on research funded in part by the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Bill & Melinda Gates Foundation. Special thanks to Allie Mahler and Amanda Yogendran from_ [_Community By Design_](http://www.communityxdesign.com/) _for leading user research and business planning for the Narrative Observatory._

obi.jpg

What the Narrative Observatory Should Be

When we think about disinformation and the upcoming election, we should not think primarily about “fake news,” nor the hoaxes that routinely spring from the web’s dark loins. Rather, we should focus on the strategies intelligence agencies and autocrats have used for decades, if not centuries, to retain and project their power, to remove the conditions opposition needs to take root and flourish. Strategies like exacerbating existing social divisions by promoting particular types of media content within particular communities. Strategies into which our current media systems and personal digital dependencies have breathed new life. This is the message contained in the steady drip of disinformation stories the media has recently produced — like  [this one](https://www.usatoday.com/in-depth/news/investigations/2020/02/24/2020-election-disinformation-presidential-race/4725496002/)  from  _USA Today_  or  [this one](https://www.theatlantic.com/magazine/archive/2020/03/the-2020-disinformation-war/605530/)  from  _The Atlantic —_ schooled by organizations like  [FirstDraft](https://firstdraftnews.org/), which has been promoting a more inclusive, more accurate view of disinformation for years. This upgrade to mainstream understanding is good news.

[](https://firstdraftnews.org/latest/information-disorder-the-techniques-we-saw-in-2016-have-evolved/)

![HLabs03:16:201.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs03_16_201_e47f351f74.jpg "Mis/Dis-information definitions from First Draft’s _Essential Guide to Understanding Information Disorder_")

There has not yet been a corresponding upgrade in our notion of what to do about disinformation, however. The same articles that observe this new reality — of persistent, global-local disinformation networks, of post-truth, of polarized echo chambers, or whatever — tend to dissolve in nostalgic paeans to “trust and truth,” to bygone days when a few experts or organizations established the “objective facts” that bound together we, the people. Alternately, these articles profile an ever-changing set of socio-technical initiatives to stem the supply of disinformation, to remove, fact-check, label, or otherwise disadvantage content deemed ill-intentioned, questionable, or low-quality.  [Supply-side interventions](https://www.rand.org/research/projects/truth-decay/fighting-disinformation.html)  like these are bound to play a role in whatever coping mechanisms we, the people ultimately develop, especially given the responsibility that platforms and their opaque algorithms bear for our exposure to information generally. But both the supply-side response and “trust and truth” nostalgia seem born of the same ongoing infatuation with technology and technocrats that may have gotten us into this mess in the first place. So what else can be done?

There was a moment in 2017 when media literacy seemed to be an answer. At that time, Facebook, California lawmakers, the James L. Knight Foundation, the American Library Association, and others were announcing initiatives to educate us back to civic health. These are detailed in the Data & Society report, “[The Promises, Challenges, and Futures of Media Literacy](https://datasociety.net/pubs/oh/DataAndSociety_Media_Literacy_2018.pdf),” a clear-eyed contextualization of media literacy and a caution against viewing it as a panacea for disinformation. Report authors Monica Bulger and Patrick Davison observe that media literacy “suffers from all the issues plaguing education generally,” such as wide variations in goals and intended outcomes, politicization, outdated underlying mental models, and a lack of a national evidence base. Nevertheless, Bulger and Davison offer a few ideas to focus media literacy’s future, for instance, developing and testing curricula that go beyond individual critical thinking outcomes to seek changes in community norms and behaviors. Media literacy theorist, Renee Hobbs arrives at a similar recommendation in her 2010 white paper, “[Digital and Media Literacy: A Plan of Action](https://mediaeducationlab.com/sites/default/files/Hobbs%2520Digital%2520and%2520Media%2520Literacy%2520Plan%2520of%2520Action_0_0.pdf),” but goes a step further. Hobbs suggests we “engage the entertainment industry’s creative community in an entertainment-education initiative to raise visibility and create shared social norms” regarding pro-social media behaviors. These two ideas point the way to a possible response that draws on, but ultimately extends beyond media literacy’s traditional bounds. What if we enlisted the very media systems and content — Silicon Valley and Hollywood — that disinformation exploits to build the community norms required for resistance and resilience in a post-truth information environment?

The idea of using media systems and content to update or influence community norms at scale is not new. In fact, it has proven remarkably effective across many communities and national contexts over the last few decades. Think “[Sesame Street](https://www.theatlantic.com/education/archive/2015/06/sesame-street-preschool-education/396056/)” as a response to falling literacy rates among low-income urban youth in the 1960’s. Consider how the gay porn industry, under activist pressure in the 1990’s,  [made condom use the norm](https://books.google.com/books?id=MBlMDwAAQBAJ&pg=PA91&lpg=PA91)  in a community beset by HIV-AIDS. There’s the radio drama “Musekeweya,” which Rwandans have credited with helping  [heal ethnic divides](https://www.npr.org/2018/04/16/602872309/romeo-juliet-in-kigali-how-a-soap-opera-sought-to-change-behavior-in-rwanda)  after the 1994 genocide. South Africa’s “Soul City” television series fostered a significant  [shift in popular opposition to domestic violence](https://www.ncbi.nlm.nih.gov/pubmed/16006028). Viewers of MTV Africa’s television show,  _Shuga_, are twice as  [likely to get tested for HIV](http://www.mtvstayingalive.org/blog/2016/06/new-world-bank-study-reveals-that-mtv-shuga-viewers-are-twice-as-likely-to-get-tested-for-hiv/). And, more recently, Hulu’s “East Los High” led to  [measurable improvements in knowledge of reproductive health practices](https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2016.303072)  in its young LatinX audience.

What we haven’t seen are these types of content and norm-changing strategies employed at scale to address citizenship and political participation in a post-truth information environment.  People are starting to call for it though, as in  [this recent report](https://www.ned.org/wp-content/uploads/2020/02/Firming-Up-Democracys-Soft-Underbelly-Authoritarian-Influence-and-Media-Vulnerability-Lucas.pdf)  from the National Endowment for Democracy, which observes, “Norm building is a common feature of governmental and nongovernmental efforts to pursue the public good in democratic societies. . . . We need to apply such norm building to the information space.” We would like to add our voices to these calls. In fact, we’ve already started experimenting.

Recently, we produced a couple podcast pilots. Our means were minimal, and therefore the scale of our experiments was small. Our intention was to use these pilots to help people better imagine what might be possible, with more partners and resources. The first podcast pilot, “[Anomaly](https://soundcloud.com/cerealmade/anomaly-welcome-to-keensboro),” is a narrative fiction set in a near dystopian future. It follows 17-year old Kory Hernandez, as she travels from her family home in Queens to rural Illinois to wait out the wars and environmental hazards of East Coast USA.

The world and the narrative that unfolds serve as scaffolds for characters and situations designed to draw young audiences in, then model — through positive and negative social responses and outcomes — the attitudes and behaviors conducive to resilience in a disinformation-rich environment. The pilot also seeks to model — in the contrasts between key rural and urban characters — the plausibility of balancing small-town community-oriented mindsets  with more open, expansive definitions of what it means to be American. These strategies were developed by applying an audience-centered social change framework to common media literacy skills and other related habits of mind, in collaboration with researchers, producers, talent, and young people, in a process analogous to what was used for  [VR Action Lab](https://harmonylabs.org/vr-action-lab).

![Hlabs03:16:202.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs03_16_202_b5a263f414.jpg)

The second podcast pilot, “The Audience,” is a more traditional, journalistic look at what’s happening at the cutting edges of our media systems, and how these developments will affect mainstream culture. This podcast has a host, and each episode has a theme, starting with “[The Revolution Will Be Synthesized](https://soundcloud.com/harmonylabs/the-revolution-will-be-synthesized-1/s-P7yfxEaiQTx),” around which we arrange a provocative collage of interviews, found and archival audio, and listener interactions.

![Work-Harmony-Labs.png](https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/782565250&color=%23ff5500&auto_play=false&hide_related=false&show_comments=true&show_user=true&show_reposts=false&show_teaser=true&visual=true)

We’re aiming for an unusual audience mix here: older folks (age 55+) and young people (ages 12–18). Our assumption is that each of these groups generally has something important to offer the other in terms of building media systems resilience: on the one side, greater fluency with emerging technologies and, on the other, a larger store of life experience from which to interpret new developments. The podcast will be supported by a few different curricula, designed specifically for informal learning institutions, like public libraries and community centers, as well as middle schools and high schools. These curricula will touch lightly on traditional media literacies, like critical thinking. Predominantly, however, they will focus on media making as a context for developing an embodied awareness of media’s effects on oneself, others, and communities, and then using this awareness to generate and reinforce the norms and behaviors likely to grow civic resilience.

![Hlabs03:16:203.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs03_16_203_83e88f903b.jpg)

The kind of content we envision and have already started experimenting with goes beyond “awareness-raising” or “tackling tough issues,” which is how entertainment-education has typically been used in the United States. And it is distinct from worthy projects like integrating scientifically accurate information into existing storylines, or increasing visibility for under- or mis-represented groups. It draws on a tradition that flows through a range of disciplines and fields — like health communication, behavioral science, psychology — and creative people, like playwright  [Bertolt Brecht](https://en.wikipedia.org/wiki/Lehrst%C3%BCcke), who viewed theater — his particular medium — as a revolutionary venue for participation and discovery powerful enough to reconstitute a body politic. The power present day media systems put at our disposal would have been unimaginable to Brecht. If, as recent media stories suggest, we are already in another period of significant cultural reconstitution, then we had better start figuring out how we intend to use that power.

Here are some questions we’ll be asking, as we continue this work and move beyond small podcast pilots to a series on one of the streaming services perhaps? :)

-   How do we balance design for outcomes with traditional, highly collaborative creative processes in a way that yields awesome, super engaging content?
-   How do we integrate audience participation and continuous feedback into content development?
-   How do we reach a large enough audience to achieve changes in norms, attitudes, and behaviors at scale?
-   How can we study, at scale over long time periods, the effects of the content we produce, given the seemingly insurmountable challenge of attributing those effects to particular media interventions?
-   What business models, capital structures, and partnerships are needed to support, sustain, and grow this work?
-   Will you join us?

&nbsp;

---

### Asch Harwood
Asch is the founder of  [Red Hook Media Lab](https://www.redhookmedialab.com/)  and executive producer for “Anomaly.” He has spent the last decade working at the intersection of democracy, design, and technology at organizations like the Council on Foreign Relations and UNICEF.

### Brian Waniewski
Brian is the Executive Director of  [Harmony Labs](http://harmonylabs.org/)  and executive producer for “The Audience.” Before Harmony Labs, he served as managing director at Institute of Play. He also served on the core founding teams for  [Quest to Learn](http://www.q2l.org/), a New York City public school that leverages game design to make school engaging and culturally relevant for young people; and  [GlassLab Games](https://www.glasslabgames.org/), a Silicon Valley development shop transforming digital games, like  [SimCity](http://www.simcity.com/), into classroom learning environments with real-time formative assessments built in.

Hlabs03:16:203.jpg

What to Do About Disinformation — Content for Citizen Resilience!

Asch Harwood

Proimm1.jpg

Race Forward’s Butterfly Lab in conversation with Harmony Labs

Art for Pro-Immigration Narratives

As we near the end of this month, we’d like to share some of the exciting projects we’ve worked on recently — and some of our team’s own media interests, too.

If you’d like regular updates like these,  [subscribe](https://harmonylabs.us7.list-manage.com/subscribe?u=182a49cea013a769c494b1fef&id=c6203bbb0e)  to get them delivered directly to your inbox.

### Updates from the Lab

-   We recently published a new  [report](https://medium.com/harmony-labs/finding-the-immigrant-story-a1aab067c704)  with United We Dream Action that looked at Latinx encounters with immigration stories in the media, which was covered in both  [The New York Times](https://www.nytimes.com/2022/04/12/us/politics/latinos-voters-midterms.html)  and  [NBC News](https://www.nbcnews.com/news/latino/young-latino-men-may-easy-targets-disinformation-immigration-report-sa-rcna24288?cid=sm_npd_nn_tw_ma). This work builds off previous  [research](https://harmonylabs.medium.com/el-nuevo-noticiero-6495f349ee6d)  with Equis Labs and Cultural Engagement Lab.
-   Our Science Director Riki Conrey spoke with Jeff Chang, Senior Advisor at Race Forward, about how our research was used to create effective media interventions for hard-to-reach audiences. Listen to the audio recording or read the transcript  [here](https://medium.com/harmony-labs/art-for-pro-immigration-narratives-34707a20f931).
-   Another partner leveraging our research to inform a data-driven storytelling approach is the West Virginia Community Development Hub.  [Read more](https://medium.com/harmony-labs/when-data-met-art-redefining-narratives-of-poverty-through-research-driven-storytelling-23145a64cf82)  about how they used the  [Narrative Observatory](https://narrativeobservatory.org/)  to transform their outreach efforts.
-   Women’s Way, a gender equity organization, and Purpose, a social impact agency, recently used the Narrative Observatory to test public narratives about poverty and economic mobility.  [Read more](https://medium.com/harmony-labs/imagining-a-world-without-poverty-34dc786c11a3)  about their findings.

### What We're Into

-   When we’re not researching media, we’re avidly consuming it, notably the podcast Hidden Brain. The recent  [episode](https://open.spotify.com/episode/1LkSFkYMqN4Si9bvQJzoKF?context=spotify%3Ashow%3A20Gf4IAauFrfj7RBkjcWxh&dl_branch=1&si=MY7syMR8R0eFhifP3j3dzQ)  “How to Change the World” explores how data and psychology can be used to analyze the efficacy of violent versus nonviolent revolutions throughout history.
-   Believe it or not, some of us enjoy time away from screens. We love a good book, like the latest from ethics philosopher and information systems expert Shoshana Zuboff:  [The Age of Surveillance Capitalism](https://bookshop.org/books/the-age-of-surveillance-capitalism-the-fight-for-a-human-future-at-the-new-frontier-of-power/9781541758001), a revealing look at how corporations turn human experience into data, and the consequences of these structures.
-   Speaking of corporations using our personal data, a group of Artificial Intelligence researchers, some of them employees of Meta, recently used 7M+ public, geolocated social media photos to construct an “[underground fashion map](https://www.vice.com/en/article/bvn594/facebook-ai-researchers-built-a-fashion-map-with-your-social-media-photos).”
-   We’ve recently discovered Logic, a print magazine trying to disrupt the discourse around tech, via  [this fascinating conversation](https://logicmag.io/beacons/dis-info-studies-andre-brock-jr/)  with André Brock, Jr. and J. Khadijah Abdurahman that traces the history of disinformation on the Black internet.
-   Proving that the internet can still be a weird and wonderful place, we’ve been loving an  [Instagram account](https://www.instagram.com/depthsofwikipedia/?hl=en)  that highlights the strangest, most obscure, and most interesting pages from Wikipedia (which, of course, has its own  [Wikipedia page](https://en.wikipedia.org/wiki/Depths_of_Wikipedia)).

We are looking for new, innovative partners to join the  [Narrative Observatory](https://narrativeobservatory.org/). Have an interesting project related to audience, narrative, or story? Looking to supercharge your storytelling efforts for culture change?  [Let’s talk](https://harmonylabs.org/contact).

pexels-berk-ozdemir-3779822 (2).jpg

Harmony’s latest projects and success stories

Monthly Round Up: April 2022

We’ve been hearing from more and more people in our community about an urgent need to restore faith in democracy and the democratic process. So we’ve been thinking hard about what Harmony Labs, as a media research studio, can do to answer this need. Helping media systems better support democratic culture is part of our mission, after all, and we’ve been reflecting on the the relationship between media and democracy since long before the  [January 6th insurrection](https://medium.com/harmony-labs/new-story-homes-wanted-a-case-for-smaller-nobler-lies-6ca02f9f339).

So this month we started imagining a pipeline of insights and subscription-based reporting on how democracy shows up in the media for different audiences in the U.S., along with narrative opportunities to increase in the relevance of democracy as a topic in the media and pro-democracy narratives for key audiences. We share a few general, early insights here to gauge interest in this kind of pipeline and reporting. So if this is something you feel is needed and could use in your own work, please  [email us](mailto:hello@harmonylabs.org).

## Democracy in the news: all about elections

It turns out that democracy is only  _sometimes_ a very big part of our culture, as reflected in the media. About 15% of all people who read any news around election day 2020 read something about democracy — but the scope of democracy stories is limited to voting and elections.

This chart shows the reach of the word  _democracy_  in news read over the last two years as imputed from our opt-in media consumption research panels.Spikes appear around Election Day and the insurrection, and quickly fall away outside of those events.

&nbsp;

![democracymedia.png](https://harmony-strapi-storage.s3.amazonaws.com/democracymedia_86a17a2130.png)

&nbsp;

Certainly, voting is an important cornerstone of a healthy democracy, but the idea of democracy in America seems to be anchored around Election Day, and Election Day only. That means that every two years, we have to recreate, reconstruct, and rebuild pro-democracy narratives in America over and over again.

## Democratic culture in the U.S. is not monolithic

How do we knit together stories about democracy toward a narrative that isn’t dependent on elections or the calendar year? Narrative starts with audience, and audiences are defined by so much more than their political affiliations. We have to look at people’s larger cultural experiences and understand their values, starting with our  [four values-based audiences from the Narrative Observatory](https://narrativeobservatory.org/audiences/). People’s understanding of “democracy” is so much more than whether they vote left or right, but which democratic values resonate with the personal goals and values that shape their choices, as well as their media consumption.

These are all the news outlets that brought democracy content to our audiences in the last two years. If  _The Epoch Times_  and  _The New York Times_  are both bringing stories about a subject to audiences, it’s a good bet that the subject is the same, but the themes and values in those stories are not.

&nbsp;

![democracymedia2.png](https://harmony-strapi-storage.s3.amazonaws.com/democracymedia2_04c9f8e931.png)

## Narrative opportunities for democracy differ by audience

For each of our four audiences, we want to learn which “democracy” stories are resonating with them. Not only do people consume different news sources, but they follow and are activated by different “democracy” stories.

[Don't Tread on Me](https://narrativeobservatory.org/audiences/dont-tread-on-me/), for example, contains people that are typically considered the heart of the conservative political base. They are also an audience where a commitment to and concern for democracy is often more prevalent than in the progressive base. For them, democracy means  _not socialism_  and trusting authority, stories that they follow closely in the media.

&nbsp;

![democracymedia3.png](https://harmony-strapi-storage.s3.amazonaws.com/democracymedia3_70648fd71b.png)

&nbsp;

The audience we call  [If You Say So](https://narrativeobservatory.org/audiences/if-you-say-so/)  is often recognized as made up of the least regular voters, frequently distrustful of politicians and any institutional system. Though they express progressive political values, they tend to lag in consuming democracy content.

What the data tell us is that the key audiences that we hope to engage in democracy are not that interested in the currently available stories: stories about elections. Among their trusted sources, If You Say So rarely encounters the word “democracy,” so they may be the hardest to reach, least bought-in, and just generally so disengaged with news that it’s hard to imagine staging narrative interventions in conventional places.

They are, however, interested in  _access to the polls_, an interest that is in line with their overall pursuit of autonomy and independence. Once we know what democracy stories the harder-to-reach audiences are actually engaged with, we can more easily build interventions that speak to the issues they care most about and around topics that matter most to them.

&nbsp;

![democracymedia4.png](https://harmony-strapi-storage.s3.amazonaws.com/democracymedia4_f6e0ae0df9.png)

## The democracy narratives we need

Right now, we talk about democracy when it’s time to vote — but how can we tell a story of democracy that’s ubiquitous, with a much larger narrative about America, one that is talked about all the time? It’s clear from this quick look at democracy in the news that democracy isn’t a singular idea that audiences spend their time thinking about. What they do engage with — and what might be able to move them — are the fundamental components of democracy that reflect their core values.

In the coming months, using our Narrative Observatory, a  [one-of-a-kind data tool](https://narrativeobservatory.org/)  which connects the minute-by-minute media behavior of 300,000+ people to media content and survey data, we hope to unpack more about democracy in these audiences in the following research phases:

-   **Audience Reach**  (Q2, 2022 | Monthly): How much and where does each audience consume content that is actually about “democracy” as a topic?
-   **Topic Modeling & Network Graphing**  (Q3, 2022 | Quarterly): How much and where does each audience engage with the component parts of democracy?
-   **Content Scoring**  (Q4, 2022 | Quarterly) : Where on the spectrum of democratic to anti-democratic does specific content fall?
-   **Opportunities & Assessment**  (As Needed): What are the narrative opportunities and what specific content interventions were most (or least) effective and moving audiences?

To do this, we’re actively looking for partners and funders to help us shape our research agenda, produce a regular report, and attract subscribers to support and sustain this work. If you’re interested in partnering with us in this work, or would like to subscribe to this new report, please  [get in touch](https://www.harmonylabs.org/contact).

capitol.jpg

Using data to derive the pro-democracy narratives we need

Democracy in the Media

by  [EquisLabs](https://www.equislabs.us/),  [Harmony Labs](https://harmonylabs.org/), and  [Cultural Engagement Lab](https://www.engagementlab.org/)

&nbsp;

According to a recent  [Pew Research Report](https://www.pewresearch.org/internet/2021/04/07/social-media-use-in-2021/), YouTube is used by 81% of U.S. adults and 85% of adults who identify as Hispanic. To get a better understanding of the types of content Latino audiences were engaging with on the platform, EquisLabs, Harmony Labs, and the Cultural Engagement Lab set out to analyze over 250,000 days of YouTube consumption between January 1, 2020 and May 31, 2021.

This new research expands upon work from the 2020 partnership between Harmony Labs and Equis Research which provided a  [comprehensive view](https://docsend.com/view/x6986v8mdwdf7iuk)  of Latinx media consumption in America. This work provides advocates and organizers a deeper, more powerful framework upon which engagement and persuasion efforts can be better constructed on YouTube specifically. We are excited to share the results of our research publicly in this  [new report](https://static1.squarespace.com/static/5c993096f4a30b0001b73418/t/61e89eb9d4ab9913bc309804/1642634940820/elnuevonoticiero.pdf).

## YouTube Rules
Since 2020, EquisLabs has monitored YouTube channels that traffic in polarizing content targeted at Latinos in both English and Spanish. 47 YouTube channels were analyzed to provide a comprehensive characterization of the narrative landscape within which viewers receive mis- and disinformation online. Of these 47, the channel with the most reach was, not surprisingly, Fox News, consumed every day by 4 out of every 1,000 Hispanic audience members.

&nbsp;

![HLabsNuevo2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo2_fa2c40a714.jpg)

## Recurrent Themes
In analyzing these channels, consistent themes began to emerge (and recur). For instance, “big government” frequently appears as orchestrating the collusion of five antagonists intent on brainwashing, silencing, and dividing a presupposed “us” (meaning everyday people “like you and me.”)

&nbsp;

![HLabsNuevo3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo3_82ffaff10f.jpg)

&nbsp;

This theme of “antagonists” vs “us” plays out in ways that are distinctly Latin, for example: Cuba and Venezuela offer concerns around big government, socialism, the surveillance state, and the Black Lives Matter movement (portrayed as a paramilitary group). At home in the context of right-leaning political content, a significant number of these channels follow these recurring themes with a primary goal of invoking viewers’ common sense and online research skills to identify collusion and threats.

## 4 Unique Content Ecosystems
To further understand what other channels associated with these 47 channels of interest, we identified all the other YouTube channels audiences engaged with and located channels that tended to cluster together by audience interaction. What emerged were 4 unique “content ecosystems” we’ve characterized as Right Hive, News Plus, En Exclusiva, and Flex.

### Provocative Right-Leaning News & Hot Takes

&nbsp;

![HLabsNuevo4.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo4_0efc5df321.jpg)

&nbsp;

Right Hive is best characterized as provocative and highly-charged. About 4% of the Latinx YouTube audience was found to consume content from this ecosystem, with core channels being  [Fox News](https://www.youtube.com/user/FoxNewsChannel),  [Timcast](https://www.youtube.com/channel/UCLwNTXWEjVd2qIHLcXxQWxA), and the  [Next News Network](https://www.youtube.com/user/NextNewsNetwork)  — and 1.12% of our audience specifically engaged with one of the 47 channels of interest there. This ecosystem is not, however, exclusively focused on right-learning news content — we found that many channels featured content about animals, food, and religion.

### Informative Content Across The Political Spectrum

&nbsp;

![HLabsNuevo5.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo5_dab72676fd.jpg)

&nbsp;

News Plus is defined as encompassing informative, mainstream news content across the entire political spectrum, viewed by 13% of our Latinx audience with primary channels like  [Fleccas Talks](https://www.youtube.com/channel/UCIpwPuJsrboNnf200oV8cWQ),  [Rebel News](https://www.youtube.com/c/RebelMediaTV), and  [TheDC Shorts](https://www.youtube.com/user/DailyCallerVideo)  — and 0.08% of our audience engaged with one of the 47 channels of interest there. The most practical characterization of this ecosystem is a behavior of all news all the time, including “hot-take” content from both right- and left-leaning audiences.

### Highly-Charged, Latino-Targeted Spanish Language Content

&nbsp;

![HLabsNuevo6.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo6_3670a9ce46.jpg)

&nbsp;

En Exclusiva, reached of 2% of our Latinx audience — and 0.24% of our audience engaged with one of our 47 channels of interest there — is an ecosystem defined by highly-charged Spanish language news and entertainment, occupied by news channels like  [Noticias Telemundo](https://www.youtube.com/channel/UCRwA1NUcUnwsly35ikGhp0A)  as well as family drama content such as  [Caso Cerrado](https://www.youtube.com/watch?v=kIhccxDb3GY&ab_channel=CasoCerrado).

### “Hot Takes” In News Analysis And Overtly Masculine Punditry

&nbsp;

![HLabsNuevo7.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nuevo7_0fd681daa0.jpg)

&nbsp;

The Flex ecosystem is dedicated to news analysis, punditry, or “rant”-focused content that exhibits both humorous and overtly masculine qualities, most notably The Joe Rogan Experience ([PowerfulJRE](https://www.youtube.com/c/joerogan)). This ecosystem has highest reach of our Latinx audience at 18%, and 0.27% of our audience engaged with our 47 channels of interest there, and the users in this ecosystem have a characteristic of consuming specific and controversial media that has less obvious overlap with content in the other ecosystems.

## Putting This Work Into Practice
Summarized findings of opportunities for each ecosystem are as follows:

-   **RIGHT HIVE**  — Ultimately the hardest to penetrate and quite challenging to counter-narrate directly, this ecosystem is essentially an echo chamber and the most limited of the 4 in opportunities to stage content interventions.
-   **NEWS PLUS**  — This ecosystem is a critical space to stage content that crosses political divides — coupled with its sizable reach, it is the primary critical space where organizers and practitioners can place impactful content.
-   **EN EXCLUSIVA**  — The En Exclusiva ecosystem offers content similarities to the Right Hive ecosystem with the benefit of targeting Spanish-first audiences, making it a unique opportunity to reach Right Hive audiences via Spanish-language content.
-   **FLEX**  — The ecosystem with the highest reach, Flex invites content that is humor-focused or with a “hot-take” quality, elevating many worldviews of Right Hive to levels that enjoy better engagement and are generally more entertaining.

Application of these prescriptive research findings above can inform the development of effective, powerful content strategy — in other words, the findings of the research efforts are best put into practice when advocates and media makers utilize the work to test content hypotheses “in the wild.”

## Conclusion
Ultimately these insights are presented publicly with the hopes of providing actionable insights and resources for content creators, change-makers, and strategists en route to building impactful methods of intervention. If you are curious about these learnings or have your own ideas for leveraging these findings, we encourage you to read the full report  [here](https://static1.squarespace.com/static/5c993096f4a30b0001b73418/t/61e89eb9d4ab9913bc309804/1642634940820/elnuevonoticiero.pdf)  and  [get in touch](https://harmonylabs.org/contact)  to start a conversation.

## Methodology

Using minute-by-minute media behaviors of over 300,000 people in the U.S., identifying roughly 10,000 users who self-identified as Hispanic or regularly consumed Spanish-language YouTube content, we isolated a single Latinx YouTube audience to serve as the focus of our analysis.

When identifying content ecosystems, we chose our characterized groupings by examining the strongest clusters of audience interaction, quantifying the ecosystems that have the overall highest daily reach across YouTube as separate from ecosystems with the highest channel reach within our 47 identified channels of interest.

A key step in our process required us to exclude music channels from our analysis, as none of our 47 channels of interest focus on music. Despite music channels occupying the highest reach with Hispanic YouTube viewers, we chose to focus on channels that would best represent opportunities for narrative intervention, rather than prioritizing “reach” as a primary indicator of opportunity.

&nbsp;

---

### About EquisLabs
_[EquisLabs](https://www.equislabs.us/)  is working to create a better understanding of Latino communities, invest in innovative approaches to reach and engage them, and uplift the Latinx leaders of tomorrow who will implement and carry forward this important work. Equis is a leader in digital innovation, focused on message testing and experiments, anti-disinformation tracking, and digital capacity-building for grassroots groups. Equis is also committed to developing Latino leaders across the progressive space, serving as a convener of organizations and individuals leading Latinx-focused programs._

### About Cultural Engagement Lab
_[Cultural Engagement Lab](https://www.engagementlab.org/)  (CEL) ignites change by collaborating with artists and creatives to create compelling and inspiring stories that power advocacy, electoral, and civic engagement efforts, especially those that seek to engage a multi-racial audience. Led by a dynamic team of women of color artists and political strategists who understand the power of culture to shift narratives and create lasting change, CEL engages seasoned artists and creatives to lend their storytelling talents to meaningful engagement. CEL is the 501(c)(4) sister organization of The Center for Cultural Power._

### **About Harmony Labs**
_[Harmony Labs](http://harmonylabs.org/)  is a 501c(3) non-profit on a mission to create a world where media systems support democratic culture and healthy, happy people. We’ve been doing audience, narrative, and story analysis for more than a decade, helping storytellers channel the immense power of story to shape the future. With the  [Narrative Observatory](http://narrativeobservatory.org/) project, for the first time ever, we’re harnessing powerful industry relationships and an academic research network to develop data infrastructure purpose-built to identify and track narratives and story opportunities ,and to learn about audiences across platforms._

Equis-ecosystems.jpeg

New research mapping 4 information ecosystems for Latinos on YouTube

El Nuevo Noticiero

Harmony White

![kirk-blog1.jpg](https://harmony-strapi-storage.s3.amazonaws.com/kirk_blog1_231117a213.jpg)

_Over the last year, Harmony Labs has been developing an audience-narrative architecture to help the Robert Wood Johnson Foundation (RWJF) and its partners find “narrative strategies” and actual storytelling that would activate people in the U.S., especially healthcare professionals, to dismantle structural racism in America’s healthcare and public health systems. The foundation of this work was research on audiences and the health equity narrative landscapes they inhabit. This initial research helped surface the most promising opportunities for content testing and validation, from which our project partner [Story Strategy Group](https://www.storystrategy.co/) convened groups of professional storytellers, anti-racism experts, community leaders, and content strategists to develop new stories that might transport audiences toward a target narrative of health equity. This piece is one in a three-part series from some of the subject matter experts and storytellers who worked on this project. You can explore the full findings [here](https://harmonylabs.org/project/health-equity)._

_This blog was originally published on [Medium](https://medium.com/harmony-labs/healthcares-narrative-emergency-ad6bb4a4626b)_

&nbsp;

---

&nbsp;

Science has discovered that the human brain produces deep stories to make sense of the world and form memories — to explain how everything works and why. The most widely shared of these stories determine our culture and the structure of our societies. They are held in our unconscious, center emotions rather than information, and defy rational argument. (Check out “[The Science of Winning with Stories](https://politicalnarrative.us/narrative/2020/10/8/the-science-of-winning-with-stories-using-agency-urgency-and-community).”)

For better and worse, such stories create the limits of what we believe to be good, bad, possible, and impossible. Making a more just world requires understanding the deep stories most people hold and replacing many of them with better, truer stories.

Learning how to use storytelling as a tool for persuasion and change has been my work for a few decades — as a journalist, a writer, a marketer, and the founder of an international content advertising agency called Story Worldwide, which I sold in 2015. Since then I’ve been working exclusively with progressive causes and, occasionally, candidates to change our world for the better.

In 2021, the Robert Wood Johnson Foundation (RWJF) engaged [Story Strategy Group](https://www.storystrategy.co/) (SSG) and Harmony Labs to find “narrative strategies” and actual storytelling that would activate Americans, especially healthcare professionals, to dismantle structural racism in America’s healthcare and public health systems. That summer, I joined the SSG team on the Health Equity Narrative project; HEN, as we called it.

I’ve worked on many ambitious narrative change projects ([youth voting and civic participation](https://allianceforyouthorganizing.org/midwest-culture-lab/), [gender justice](https://www.storyatscale.org/), [nuclear disarmament](https://www.nti.org/nti-nuclear-narrative-and-audience-research/), [partnering organizers and artists to change the culture](https://www.culturesurge.com/learn-more/), among others). The HEN project, like several before it, at first seemed overwhelming — trying to advance racial justice in a massive healthcare system that touches and is touched by virtually every aspect of American society. It seemed like an assignment to change the whole world.

Complicating the task, my own deep story of America’s healthcare systems was at odds with the project’s mission: I believed the way to achieve racial equity in healthcare was to focus on economics, not race, by making healthcare free or, at the least, easily affordable for everyone. In our profit-centered system, health and well-being are retail commodities with a price tag way beyond the reach of many. In my deep story, most if not all racial disparity in health outcomes would disappear if we only had a national health system where high-quality care was available and affordable to everyone as a human right.

I now know that my own deep story of the U.S. healthcare system is widely shared by a lot of Americans, including many healthcare professionals. It felt completely right and true to me, but it wasn’t the whole story. Ultimately, I had to make the same sort of mental journey that is the HEN project’s goal for a majority of Americans. I had to embrace a narrative of how systemic racism in healthcare, in and of itself, shortens the lives and increases the suffering of people of color no matter what their economic status is. The true and powerful story we need, I realized on a personal as well as a societal level, is the one that narrates how we can and will eradicate racist systems in healthcare because systemic racism, all by itself, will keep killing people of color even if we manage to end all economic inequality everywhere and forever.

I organize my work with a flexible but rigorous process for finding and validating effective narrative strategies. It’s a multi-disciplinary process. Originally created by my content agency in 2005, it’s been tested and evolved since for dozens of clients — global corporate marketers and social causes and campaigns. It revolves around creating a “Story Platform,” my name for the core narrative threads that connect a culture-defining collection of deep stories. A properly constructed Story Platform has the potential to change culture by powering a new group of deep stories to alter or replace an existing collection. The “[Quick Guide to Narrative Strategy & the Story Platform Process](https://docs.google.com/document/d/1WH-x575aWW6UuQMPWQk_NklNN5qOxVrwRpxlDFwpdKU/edit)” can provide background.

The process is grounded in the classic practices of journalism and the social sciences. It centers detailed observation of people and culture. It begins by asking fundamental questions: “What exactly is going on? What systemic forces are involved. Who’s winning? Who’s losing? How is this everyday reality portrayed in the deep stories of the audiences we want ro reach?” Answering requires a wide search through histories, novels, movies, music, news stories, scholarly and scientific journals, opinion surveys, court decisions, and so on. The process also requires in-depth interviews with representative samples of the audiences we want to reach, as well as with people who know something important from first-hand experience.

Our early research showed that many Americans suspect, as I did, that the way to improve healthcare systems for people of color and everyone else is to change the economics. Surveys and in-depth interviews show Americans are frustrated by the high cost of medicine, resentful of long waits, big bills, and poor results, and fairly knowledgeable about the “social determinants” of health — socioeconomic status, education, housing and neighborhood, health insurance, environmental factors, employment, and more. All these factors are tied to money, of course, so all disproportionately disadvantage people of color in a racist society.

Along with accurately identifying the unaffordability of healthcare, [polling](https://www.icpsr.umich.edu/web/HMCA/studies/37633) says roughly 8 in 10 Americans agree with the inaccurate statement that “the biggest reason people in America become unhealthy” is “they make poor choices.” It struck me immediately that most Americans use this same story to explain poverty. Essentially, if superficially, majorities agree counterfactually that both lack of money and lack of health are largely “your own fault” in America.

This belief — “The problem is money, not race!” — turned up in the comments sections of major medical journals, where administrators and physicians wrote to question and deny the accuracy of studies and opinion pieces documenting racist structures in healthcare. Another sign of the healthcare industry’s refusal to think about racist systems in medicine is that the word “racism” wasn’t even a search term in PubMed, the government’s national medical library, until 2013.

A startling turning point for me came when I discovered that the denials and willful blindness inside the healthcare establishment persist in the face of convincing medical research over the last couple decades that proves skin color — not economics, not education, not any other socioeconomic determinant — is the only explanation for at least part of the large racial disparities in health outcomes in the U.S.

Experts recently summed this up in the “Annual Reviews of Public Health,” concluding dryly, “…[E]mpirical analyses have revealed the persistence of racial differences in health even after adjustment for socioeconomic status….” The article, “[Racism and Health: Evidence and Needed Research](https://www.annualreviews.org/doi/abs/10.1146/annurev-publhealth-040218-043750),” is one among many peer-reviewed studies documenting that when researchers compare people whose incomes and education, for example, are the same, Black people, Native people, and many other people of color still are found to receive worse care and have worse outcomes than white people.

To define and quantify “worse care” and “worse outcomes,” the science of medicine and public health focuses on life-long outcomes across population groups. The study of racist structures in healthcare systems relies on measurements of “mortality and morbidity,” the medical terms that include life and longevity; illness and its consequences; death and its causes; suffering and quality of life. These are not peripheral concerns; they are involved in the deepest joys and fears of human existence and the personal stories underneath the statistics are deeply moving.

Other team members provided data about the American people’s prevailing deep stories of healthcare. [Gretchen Barton](https://worthystrategygroup.com/) conducted revealing psychological interviews with a representative sample of Americans. [Melinda Weekes-Laidlow](https://melindaweekes.com/) assembled and facilitated group discussions, primarily with people of color working in healthcare and communications. (See Melinda’s story of her experiences on the HEN project.) I conducted in-depth interviews with RWJF staff and attended a three-day convening of RWJF healthcare grantees.

The revelation that came from these findings transformed my own deep story of America’s healthcare system. The medical research, together with the emotional stories of lives shortened and damaged by racist practices, all pointed directly to a specific core story that had to be told to begin dismantling structural racism in healthcare. I had long understood that unaffordable care is a terrible injustice, especially in the world’s richest country. Now, I could see that even in a far-off utopia where every American has the means, education, and access to obtain uniformly first-class healthcare, people of color would still live shorter lives with more preventable ill health and needless pain and suffering than white people. This massive injustice would continue, it was clear, for as long as systemic racist policies and practices are permitted to exist in our healthcare systems. This is the reality of healthcare in present-day America.

Having wrapped my head around this reality and having understood the deep stories about healthcare in the minds of most Americans, I could proceed to creating the Story Platform for a new health equity narrative.

I like to say a Story Platform is a bit of data-driven poetry composed to serve as the emotional heart of many different stories that, being connected at the core, will have a cumulative impact on changing the culture in a particular way. Research findings over the past few years have contributed to defining properties that all Story Platforms must have. For example, they must include a story of how the future will be better once the change being sought has been realized.

In this project in particular, methodical research and careful listening told us a majority of Americans hold deep stories in which racism is not a primary factor in health disparities. So, before we could persuade anyone to get involved in changing the system, we would have to persuade them, modifying and overcoming their own deep stories about bad personal choices and lack of money, that the system really is racist and racist policies and practices alone are doing lethal harm to many people of color.

My notional Story Platform for this project turned out to be:

We will imagine a future
and win a system that gives us each
an equal chance to live
the long, healthy lives we want
and guarantees none of us
ever gets sick or dies
because of our bank account …
or our race or ethnicity.

This brought us to the final step in the process, crafting and testing very compact stories based on everything we had learned. The crafting part is all about the art of storytelling. The testing part, on the other hand, is all science and the tests were overseen by Riki Conrey, director of science at Harmony Labs.

Harmony Labs divides people in the U.S. into four major audience segments. As an alternative to segmentation only by demographics (such as race, age, and gender) or political support for a particular party, Harmony Labs uses values-based segmentation that derives from Shalom Schwartz’s [theory of basic human values](https://scholarworks.gvsu.edu/orpc/vol2/iss1/11/). Each audience is identified with a distinct set of primary values (shown in the chart). These values intersect and correlate with all kinds of other features conventionally used to segment audiences like race, age, and gender. ([Explained further by Harmony](https://harmonylabs.org/article/audience-heart).)

![audiences.png](https://harmony-strapi-storage.s3.amazonaws.com/audiences_e125c582b1.png)

Altogether, SSG and Harmony created and tested 105 pieces of content in three rounds of testing. For the final round, SSG engaged five creative teams. Each of four was assigned to produce content for a single audience segment. My team alone was asked to produce content intended to cross borders and connect with all audiences. We went looking for the supposedly impossible-to-find American unicorn of common ground and, using the Story Platform above, we were able to do so.

“Team” may be grandiose: it’s me and [Ellen Jacob](https://ellenjacob.com/about), a veteran creative director in advertising, book and magazine publishing, an artist-photographer and an activist. (Ellen and I are married and work together often.)

In round three, a total of 28 pieces of content were produced as Facebook posts and tested for both persuasion and engagement. Testing showed that 11 of those 28 posts persuaded significant numbers of people that race was part of healthcare’s problem and that each of us has a role in changing that. Eight of those 11 persuasive posts, including the “top performer” across all audience segments, were produced by Ellen and me.

The most successful post (below) features a photo of two Native American women to illustrate a first-person story about a sister who is “hit hard” by diabetes. The post connects their personal struggle to the healthcare system’s structural racism, calling the system itself “untrustworthy.” The future is imagined as an opportunity to “win a healthcare system that earns everyone’s trust.” All in all, the story told in the post adheres to every point of the Story Platform. We collaborated on the post with the Native-woman-led narrative organization [IllumiNative](https://illuminative.org/).

Harmony’s science director noted that the post produced double-digit movement toward agreement that race systemically determines quality of care, yet the words “race” and “racism” never appear.

God knows, our entire society needs drastic change. In the area of healthcare, the need is extremely urgent because racist structures are literally killing people of color everyday, while a profit-making model does the same to poor people of all races and ethnicities. It is no small matter, then, to understand better how to use storytelling to reliably persuade Americans of this lethal reality and to empower and activate them to work for a better future for everyone.

![kirk-blog2.png](https://harmony-strapi-storage.s3.amazonaws.com/kirk_blog2_4abcaddf25.png "This is the Facebook post that performed best in moving all audiences to embrace our target narrative.
")



storyteller series2.png

How fixing my own flawed story about racism in healthcare systems helped find a pathway to a better, truer story for many.

Healthcare’s Narrative Emergency

Kirk Cheyfitz

![HLabsNarObsrv1.gif](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Nar_Obsrv1_0271b7e4c9.gif)

Media systems have become outrage machines. They sort people into partisan echo chambers. They reproduce and reify divisions and differences, in ways that threaten the conduct of civic life and make advocacy work — even around immediate, life-or-death issues, like pandemic response or climate — virtually impossible. Advocates, organizers, and storytellers need new approaches to reach the right people in a way that interrupts the escalation in othering, alienation, and violence, and makes it possible to envision and work together to realize a positive pluralistic future.

The new  [Narrative Observatory @Harmony Labs](http://narrativeobservatory.org/)  aims to inform these much needed new approaches, solving some of the technical, definitional, and practical challenges bedeviling narrative and cultural strategy. Made possible with generous initial support from the  [Bill & Melinda Gates Foundation](http://gatesfoundation.org/), this first-of-its-kind data tool is already being used by dozens of organizations to navigate today’s media minefield and reach the public on issues that matter.

At the heart of the Narrative Observatory are opt-in audience panel data, donated by companies like  [Comscore](https://www.comscore.com/)  and  [Nielsen](https://global.nielsen.com/global/en/), giving us a window into how 300,000+ people in the U.S. interact with media, across platforms and devices. Which means that, unlike existing media monitoring tools, we’re able to focus on analyzing the actual media actual people engage with rather than on total tweets or all articles or bot traffic. Using these audiences as a foundation, thanks to media content data donated by partners like  [TVEyes](https://tveyes.com/)  and  [PeakMetrics](https://www.peakmetrics.com/), we analyzed over 500,000 online news stories, 1,000,000 tweets, and 600,000 TV airings, between January 2020 and May 2021, to understand audiences relative to their place in culture and to identify, measure, and track narratives within audiences over long time scales. For the first iteration of the tool, we’ve focused on the narratives of poverty & economic mobility in the U.S.

## **A Values-based Approach**

The Narrative Observatory’s data infrastructure connects insights and analysis about  [values-based audiences](https://narrativeobservatory.org/audiences/)  to the story patterns or narratives that affect how audiences understand and engage with social issues and the future.

![HlabsNarObsrv2.gif](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs_Nar_Obsrv2_6021e3884a.gif)

We identified  [_4 main value-based audiences_](https://narrativeobservatory.org/audiences/)  that became the foundation for this work. Key audience-based findings include:

-   Because each audience has fundamentally different values, even if they share the same beliefs about the causes of poverty, they are likely to have different attitudes about it, and different perspectives on possible solutions.
-   Using the universal human values framework of  [Shalom Schwartz](https://en.wikipedia.org/wiki/Theory_of_Basic_Human_Values)  means there are intrinsic relationships between values that allow us to see “neighborhoods,” or opportunity zones to bridge differences — and hopefully over time — heal some of these real and perceived divides.

## **Narratives of Poverty in Media and Culture**

To our audience panel data, we join terabytes of content data: TV transcripts, online news articles, song lyrics, YouTube video transcripts, etc. This allows us to look across media types for patterns in the actual content people are choosing to consume, and identify and measure narratives that inform their understanding of issues

![HlabsNarObsrv3.gif](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs_Nar_Obsrv3_6bfc1cd9a9.gif)

We identified  [_8 dominant narratives for poverty and economic mobility_](https://narrativeobservatory.org/narratives/)  through this approach. Some other key narrative findings include:

-   Most of the media content people interact with daily does not concern poverty or economic mobility.
-   News is the place where people most often engage with the issue of poverty and economic mobility in media.
-   Most content about poverty and economic mobility doesn’t feature people as protagonists, and people experiencing poverty almost never feature. Ultimately, there is no “north star,” or healthy narrative that exists in the current narrative landscape.

## **The Stories People Watch and Share**

We subjected relevant media artifacts to an additional layer of qualitative analysis to surface important  [story opportunities, threats, and strategically important features](https://narrativeobservatory.org/feed/). Drilling down to the specific stories, we attach both the audiences who have interacted with them and any narratives the stories trigger.
![HlabsNarObsrv4.gif](https://harmony-strapi-storage.s3.amazonaws.com/Hlabs_Nar_Obsrv4_4c21a99b6b.gif)
![](https://miro.medium.com/max/1400/1*edcguiHY0aWcKoj2uBlFAw.gif)

These individual story examples can be leveraged as inspiration at the beginning of the creative process to better understand the cultural worlds our audiences already inhabit so that media can be more effectively designed to reach into these places. Key story findings include:

-   Different audiences go to the same places to consume the same stories. In some cases, these can be opportunities to bridge divides between people with different core values and in other cases, they can be seen as threats to audiences vulnerable to mis- or dis-information or other forms of persuasion.
-   By observing the content audiences are already choosing to engage with, media makers can build hypotheses about the kind of content that’s likely to resonate and start testing everything from world building, to character development, to story sequence, tone, and aesthetics, before full content production begins.

## **Built for the Public**

The Narrative Observatory is a first-of-its-kind public interest data resource for anyone using narrative or cultural strategy. It’s already providing storytellers, strategists, advocates, and funders deep insight into audiences as they relate to narratives and story threats and opportunities.

We invite you to explore the tool and share your feedback. We’re aiming for monthly updates, and we are in the process of expanding it to additional issues beyond poverty and economic mobility. Follow this space or  [get in touch](https://harmonylabs.org/contact)  for future updates.

NO Launch 2023 V4 (1).png

Exploring the terrain of deeply held narratives about how the world works, what matters, and what’s possible.

Narrative Observatory @ Harmony Labs

By  [United We Dream Action](https://unitedwedreamaction.org/about/)  and  [Harmony Labs](https://harmonylabs.org/)

&nbsp;

[United We Dream Action](https://unitedwedreamaction.org/about/)  (UWDA) is the largest immigrant youth-led network in the United States with a goal of creating a political and organizing home for Latinx people and immigrants broadly to enact the changes our communities need and deserve, while advancing a vision of racial and immigrant justice. Since its founding, UWDA has been dedicated to broadening the story of who immigrants are, looking to push the bounds of public perception about immigrants, placing immigrants front and center, and taking back the power to make change. UWDA sees the anti-immigrant movement as disinformation, one that creates false narratives about immigrants and spread lies to push racist and dehumanizing policies that separate families, deny people the full rights of citizenship, and funnel billions of dollars into harmful enforcement agencies to detain and deport our friends, families, and neighbors.

UWDA partnered with  [Harmony Labs](https://harmonylabs.org/)  to answer a key question: how are young Latinx people encountering stories about immigration in U.S. media? For this work, Harmony Labs analyzed the content that Latinx audiences are currently consuming, using opt-in internet and television panel data between January 1, 2021 and August 31, 2021, touching 300,000+ people in the U.S. across TV, online news, online search, and YouTube.

This work will inform communication strategies for practitioners working in pro-immigration advocacy. By identifying real media that audiences engage with and the real opportunities that exist within them, from placement to messaging to influencers, we can tell more effective stories in the places that matter. We are excited to make our findings available in this  [new public report](https://docsend.com/view/buhmqnkm7r5qb7qt).

## High-level profiles
Our analysis looked at where audiences are in culture (content types, influential voices, brands), how they engage with stories about immigrants and immigration, and what story opportunities and threats exist to effectively engage Latinx individuals around issues of immigration. UWDA identified young Latinx aged 18–35 as their point of focus, and to address this audience, we segmented the audience into three demographic groups:

1. Latinx men and women 36+
2. Latinx women 18–35
3. Latinx men 18–35

We looked at different platforms (TV, online news, online search, and YouTube) and were able to create unique pictures of where and how these three audience segments engage differently with media and common patterns across different platforms. In particular, YouTube, which is viewed by  [70% of Americans](https://www.pewresearch.org/fact-tank/2019/12/04/10-facts-about-americans-and-youtube/)  every day, provided an opportunity to cluster similar channels together, telling us more about both the content itself and the audiences with the highest reach inside of them.

From these higher-level portraits of people’s broad cultural interests, we were able to understand which audiences were encountering potentially polarizing and anti-immigration stories — and where.

## Latinx Men and Women 36+: Practical & Tuned-In
**Media Profile:** An age group that includes millennials, online behavior for the 36+ age bracket typically includes an emphasis on watching news, shopping deals, travel, and home repair.

**Immigration Content:** The audience most likely to encounter polarizing and anti-immigration stories are Latinx ages 36+, especially Latinx men aged 36+. They encounter these stories primarily through right-wing news, especially on TV, where they are seeing the highest number of immigration stories from sources like  _OAN News_ and  _Fox and Friends_, and on YouTube, where they are the largest occupants of the  **News** and  **“Docutainment”** clusters.

![findingimstory2.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory2_5a8db3ba39.png)

![findingimstory3.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory3_51d57181f8.png)

## Latinx Women 18–35: Caring & Community-driven
**Media Profile:**  Latinx women under 35 engage more with a wide variety of cultural channels. More than any other audience group, they use TV and film to interrogate the complexity of human relationships,

**Immigration Content:** Latinx women aged 18–35 are most likely to seek out stories about immigrants, not just immigration, in a way that closely aligns to United We Dream Action’s narrative goals. They are most likely to encounter these stories in their entertainment choices, particularly TV. They are also searching for information about immigration and the immigrant experience, frequently looking to learn about the health and well-being of immigrants, as well as the best law schools for immigration law.

![findingimstory4.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory4_4ec31a2b1d.png)

![findingimstory5.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory5_f110cb3321.png)

## Latinx Men 18–35: Entrepreneurial & Individualistic
**Media Profile:** Latinx men under 35 inhabit a very insular, virtual world, spending a huge proportion of their time online engaging with fantasy and gaming. When they engage with the broader world, it’s often through humor and goofiness. Philip DeFranco, Joe Rogan, and Mr. Beast are “political” voices with high reach in this group. Overall, this age group is interested in entrepreneurialism and a “hustle” mentality, especially related to issues that affect them personally. This interest was seen in two particular news stories heavily consumed by young men: GameStop and COVID-related school closures.

**Immigration Content:** Latinx men aged 18–35 men aren’t consuming much media about immigration or immigrants, and when they are, it’s incidental and rarely anti-immigration or anti-immigrant. When their major influencers, including Philip DeFranco and Joe Rogan, do speak about immigration, they usually lean towards pro-immigrant language. This content vacuum, however, puts them at risk of passively consuming anti-immigrant content as bystanders.

![findingimstory6.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory6_93002832f5.png)
![findingimstory7.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory7_92fccfe5dc.png)

## Immigration vs Immigrants

A key insight we found very early on in this research was identifying types of stories in the landscape, separated by (sometimes subtle, sometimes not-so-subtle) differences in story subject, language used, and “target” audience. Immigration-related content online can be divided into two major categories: stories about  **immigration**  and stories about  **immigrants**.

![findingimstory8.png](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory8_07b3df4af2.png)

**Immigration stories**  are about immigration systems, including patterns, rules, and enforcement.

**Immigrant stories**  are about individuals and their culture, heritage, and experiences of being from one country and living in another.

This difference is a primary feature of the media landscape for immigration, one that helps make explicit how different stories can exist from starkly different intentions, causing starkly different impacts. Latinx women under 35, for instance, are most likely to seek out stories about immigrants, not just immigration. These “human” forward stories focus on people and their experiences, as well as the overall well-being of the immigrant community. These stories are read by the base audience for pro-immigration content, signaling a  **type** of story that should be told if we are to redirect audiences toward a pro-immigrant narrative.

## Where do we go from here?

![findingimstory9.jpg](https://harmony-strapi-storage.s3.amazonaws.com/findingimstory9_8c11e7e4db.jpg)

To introduce young men to a cultural landscape with immigrant experience stories, communicators on this topic should look to expand the reach of pro-immigrant content that speaks to the experiences of immigrants and their lives. A key next step is to place this content in spaces currently devoid of this conversation where young Latinx men are currently consuming media, pulling the focus from news and policy, or from fun, virtual worlds, and moving it toward experiences and interpersonal stories of immigrants themselves.

The majority of United We Dream’s membership is young Latinx women, and the fact that this group is most engaged in stories about immigrants is a testament to their narrative building. It’s also clear that older Latinx audiences are consuming the most anti-immigrant content, and as such, UWDA plans on expanding efforts to reach these audiences through their  [Reclaim The Web program](https://theamericanonews.com/2020/10/30/immigrant-youth-disinformation/)  on Whatsapp.

One of the notable gaps in this work is the content vacuum that exists for young Latinx men. This vacuum of immigrant content along with their concerns about the economy, and penchant for individualism, make them easy targets for racialized disinformation about the role of immigrants in this country.

For UWDA, they see this research as showing how urgently needed their work is, as creators of culturally relevant content, to intervene and to fight back against racialized disinformation and the radicalization of young Latinx men that can occur in this content vacuum by bad actors. Given the susceptibility of this specific audience, UWDA plans to further authentically engage with this audience both politically and culturally, using this analysis to better engage a growing demographic. UWDA will be applying these learnings into content and anti-disinformation programming to better engage this growing audience, meeting this audience where they are, and experimenting with key influencers and content in the gaming community.

&nbsp;

---

### About United We Dream Action
[United We Dream Action](https://action.unitedwedream.org/)  (UWDA) is the largest immigrant youth-led community in the country. We create welcoming spaces for young people — regardless of immigration status — to support, engage, and empower them to make their voice heard and win! Whether organizing in the streets, building cutting edge technology systems, opening doors for LGBTQ immigrant youth, clearing pathways to education, stopping deportations or creating alliances across social movements, United We Dream puts undocumented immigrant youth in the driver’s seat to strategize, innovate and win.

### About Harmony Labs
[Harmony Labs](http://harmonylabs.org/)  is a 501c(3) non-profit on a mission to create a world where media systems support democratic culture and healthy, happy people. We’ve been doing audience, narrative, and story analysis for more than a decade, helping storytellers channel the immense power of story to shape the future. With the  [Narrative Observatory](http://narrativeobservatory.org/) project, for the first time ever, we’re harnessing powerful industry relationships and an academic research network to develop data infrastructure purpose-built to identify and track narratives and story opportunities ,and to learn about audiences across platforms.

FINAL_Harmonylabs_0316_TB (1).001.jpeg

Latinx encounters with immigration stories in the US media

Finding the Immigrant Story

Muted Pink

The intersection of art, advocacy, and data science creates an opportunity for the work of social change to be both creative and informed by research. Over the last four years,  [Harmony Labs](https://harmonylabs.org/)  has been privileged to work alongside the Mozilla Foundation’s  [Creative Media Award program](https://foundation.mozilla.org/en/what-we-fund/awards/)  awardees in building projects that live within this intersectional space.

For the  [2021 cohort](https://medium.com/harmony-labs/designing-for-and-measuring-the-impact-of-art-for-advocacy-98e44c410d8f), Mozilla provided funding for Black artists using art, tech, and media to spotlight how Artificial Intelligence (AI) can reinforce — or disrupt — systems of oppression. We worked directly with awardees to provide: research on key audiences for their work and current narrative and story opportunities for AI, an impact design curriculum to help inform their creative process and maximize the efficacy of their work, and media effects testing to help awardees fine-tune the final products.

## The 2021 Projects
Led by Impact Producers  [Jennifer MacArthur](https://www.linkedin.com/in/jennifermacarthur/)  and  [Brett Gaylor](https://www.brettgaylor.com/), the awardees participated in a series of webinars and individual coaching sessions to produce impact plans for their projects that honed in on target audiences, story opportunities, and distribution strategies. There were also opportunities for this talented group of artists, technologists, and activists to share, collaborate, and provide feedback on one another’s projects. The resulting work spans web, VR, film, and tech, and we are excited to share highlights from 8 of these projects here:

### Melalogic 
A [skin health resource](https://www.melalogic.com/) for Black people from trusted professionals who look like them. Users can participate in building the Black Skin Health AI Dataset, a public resource to help power future dermatology practices and pre-diagnoses.

### Binary Calculations Are Inadequate
From transmedia artist  [Stephanie Dinkins](https://twitter.com/StephDink), [this project](https://binarycalculationsareinadequate.org/) is an both an app and a workshop series that seeks to push data-driven algorithms that “increasingly control our daily lives” to become less reductive and more complex, driven by kindness and generosity.

&nbsp;

![HlabAIRJ2.png](https://harmony-strapi-storage.s3.amazonaws.com/Hlab_AIRJ_2_70585e05ca.png "A still from the website for  Binary Calculations Are Inadequate from Stephanie Dinkins.")

### Future Wake
An [interactive website and art project](https://www.futurewake.com/#/) about future fatal encounters with the police, creating stories generated from AI based on 20 years of historical data to bring more transparency about police-related fatalities and  [predictive policing](https://www.futurewake.com/#/data-and-methods?tab=1).

### Afro Algorithms
From creator [Anatola Araba](http://www.anatolaaraba.com/),  [_Afro Algorithms_](https://www.afroalgorithms.com/) is a 3D animated short film in the Afrofuturist genre that tells the story of “Aero, the first AI leader of Earth, as she discovers who she really is — beyond her programming.” Exploring topics of AI and bias, Aero realizes there are gaps in her worldview that include experiences of the historically marginalized and oppressed.

&nbsp;

![HlabAIRJ3.png](https://harmony-strapi-storage.s3.amazonaws.com/Hlab_AIRJ_3_0a8a275db1.png "A still from the website for the Afro Algorithms fillm from Anatola Araba.")

### Hope
Hope is a [collection of immersive journalism and Afrofuturism artwork](https://foundation.mozilla.org/en/blog/hope-documentary-traces-the-beginnings-of-new-black-led-uk-task-force-to-tackle-ai-bias-and-discrimination/) from artist and civic tech entrepreneur Tracey Bowen. Hope imagines two possible futures: one where bias and discrimination have been banished from technology, and one where it runs rampant — it presents the issue of AI bias and discrimination in a way that educates and raises awareness.

### Dark Matters
[Johann Diedrick](http://www.johanndiedrick.com/)  built an immersive, [interactive website](https://darkmatters.ml/) that utilizes 3D modeling, sound, and storytelling to spotlight the absence of Black speech in datasets that train voice assistants like Siri, Alexa, and Google Home. This absence creates new forms of racial exclusion and bias — and thus, Black speech is forced to code switch for “AI ears.”

&nbsp;

![HlabAIRJ4.png](https://harmony-strapi-storage.s3.amazonaws.com/Hlab_AIRJ_4_c14aff8915.png "A still from Johann Diedrick's interactive website Dark Matters.")

### Points of View
This [XR project](https://www.verse.com/video/62914-pov-tribeca-web-xr/) is described by its creator Alton Glass as a “hyper-digital sci-fi virtual reality series immersed in a near future Los Angeles where personal data is the new currency and weaponized A.I. Police drones enforce the law.” The result is an immersive experience that gives viewers a close look at the dangers of AI-biased predictive policing and surveillance.

&nbsp;

![HlabAIRJ5.png](https://harmony-strapi-storage.s3.amazonaws.com/Hlab_AIRJ_5_cf64cc36ad.png "A still from the “hyper-digital sci-fi virtual reality series” Points of View from Alton Glass.")

### Black Arts + Culture: Generative Traditions With AI & Design in Carnival
[This project](https://carnival-ai.com/), from [Vernelle Noel, Nettrice Gaskins, and Valencia James](https://carnival-ai.com/?page_id=37), uses the Trinidad Carnival as a foundation for exploration in AI, art, and design. By combining AI with digital heritage, art, and culture, the project aims to “celebrate the creativity, resiliency, and vibrancy of Black and Caribbean histories and cultures.”

## The Future of Data-Informed Art

Taking a data-driven approach to this type of work can be valuable not just in measuring impact after launch, but also for improving media during the creative development process. Where possible, we also worked with the artists to develop testing plans that can help validate their initial impact hypotheses and inform creative refinements before releasing their works to the public.

Over the course of the last four years supporting this work, it’s become clear that data can play an important role without limiting the creative process. In fact, the potential impact of ambitious programs and projects like these is only enhanced through a commitment to research, development, and effects testing throughout the process. Other key ingredients to high-impact programs and projects include:

-   **Find a strong theme.** For positive cohesion amongst artists and strategists, a core subject (in this case, AI’s relationship to racial justice) around which collaborators can unite helps enrich analysis and engagement.
-   **Incorporate audience & narrative strategy.** Making sure creative projects are received by the “right” audiences and tell the most effective stories is critical for impact and lasting change.
-   **Build work that supports a theory of change.** In other words, measurement of the effectiveness of this cohort is best achieved when viewed against the scaffolding of a larger, ambitious, change-focused mission.

The ultimate goal is for these pieces to provide the necessary scaffolding for creativity to flourish. We’ve been very fortunate to work with such a brilliant group of artists, advocates, media makers, and technologists and through their creative and singular approaches to tackling issues like racial justice, we always end up learning as much from they as they do from us. We want to extend a heartfelt thank-you to all of the creators, and to the Mozilla Foundation, for allowing Harmony Labs to join you on this important creative journey!

If you’re interested in this kind of work, we urge you to  [get in touch.](https://harmonylabs.org/contact)

HlabAIRJ1.png

How Harmony Labs Supported Black Artists Studying AI in Mozilla’s 2021 Creative Media Awards Program

AI Advocacy, Impact Design, & Racial Justice

“Welfare makes people lazy.”

Persistent narratives surrounding poverty and economic mobility are the result of story patterns that repeatedly reinforce those narratives. When audiences engage with stories in media, those stories can reinforce the themes present in other stories. Audiences repeatedly engaging with similar themes across different stories cause narratives to be born, and over time those narratives can reproduce, spread, and change. The Narrative Observatory @HarmonyLabs has identified eight distinct narratives about poverty and economic mobility. Story interventions are necessary to counteract the persistently problematic narratives and transform the public discussion about poverty so that a permanent solution can be achieved.

[Women’s Way](https://womensway.org/)  is a Philadelphia-based organization dedicated to the advancement of women, girls, and gender equity. In collaboration with  [Purpose](https://www.purpose.com/), a social impact agency, and  [Harmony Labs](https://harmonylabs.org/), a media research lab, Women’s Way was interested in a specific inquiry: how to motivate individuals to recognize gender-based economic inequalities and take action to overcome them. If stories can persuade people, they were interested in knowing which stories work and told by whom.

Starting with the public narratives about poverty from the Narrative Observatory @Harmony Labs, this project focused on two specific narratives to inform the conception and execution of the storytelling intervention testing:

-   [Not Like Us](https://narrativeobservatory.org/narratives/not-like-us/)  — wherein “poor people” become an  _other_  through innuendo, association, and coded rhetoric that conspires to cast people in poverty as bogeymen threatening society. This narrative generally comes across as matter-of-fact, often rooted in fear and guided by an ethos of perpetuating laziness.
-   [The Grind](https://narrativeobservatory.org/narratives/the-grind/)  — wherein people should do “whatever it takes” to transcend poverty, namely perseverance, hard work, and a “little luck.” This narrative generally assumes an aspirational and resolute tone.

The  [Bill & Melinda Gates Foundation](https://www.gatesfoundation.org/), whose funding paved the way for the creation of the Narrative Observatory and the analysis of narratives like those above, partnered with Harmony Labs and Purpose to support Women’s Way in staging a series of tests to answer the overarching hypothesis:  _Women telling real stories about their struggles to overcome financial hardships can influence people to believe in solutions for poverty._

## Setting Up The Test
To test this hypothesis, Women’s Way looked to understand how different stories about a woman with lived experience with economic insecurity might persuade audiences about ending poverty for good.

To do this, Harmony Labs and Purpose told the Women’s Way hypothesis through three lenses. The stories showed a woman, Barbie, talking about the path she took to get out of poverty:

**A.** Through Hard Work

**B.** Through Financial Literacy

**C.** Through Declaring Bankruptcy

&nbsp;

![Worldw:oPOV2.png](https://harmony-strapi-storage.s3.amazonaws.com/Worldw_o_POV_2_c63c27d590.png)

&nbsp;

**Harmony Lab then implemented** a randomized controlled trial (RCT) to test the efficacy of these stories. About 1600 online respondents viewed one of these three stories, the test, or a piece of unrelated media, the control. They then took a survey indicating their level of agreement with three statements:

_“Government has a responsibility to ensure that people can earn enough to get themselves out of poverty for good.”_

_“I can imagine how a world without poverty would work.”_

_“The best way to end poverty for good is to provide more support to those in need.”_

We then conducted data analysis to compare the survey responses of online respondents who viewed the stories about Barbie to the people who viewed the unrelated control, which represents an audience’s baseline position. Through this analysis, we can tell if a piece of media “moved” an audience from their baseline.

## Key Results

&nbsp;

![Worldw:oPOV3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Worldw_o_POV_3_a858bb5183.jpg)

&nbsp;

**Key Result #1: The content affected people[^first].** There was likely a positive impact  **overall** from the story strips. Across the three versions of the story and all survey questions, there was an increase in agreement[^second]. compared to the control.

&nbsp;

![Worldw:oPOV4.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Worldw_o_POV_4_152a18929e.jpg)

&nbsp;

**Key Result #2: Concrete interventions had a higher impact[^third].** Both the “Financial Literacy” and “Bankruptcy” stories had positive effects[^fourth]. The “Hard Work” story, which is most in line with the “Grind” narrative, had a vanishingly small impact.  _(Note the 2nd chart above — the “Hard Work” bar_ **_is_** _on the graph, but is so small it does not reach beyond the axis.)_ This leads us to believe that a story about “more of the same,” just working harder, will probably not change audiences’ views about systemic solutions to poverty.

&nbsp;

![Worldw:oPOV5.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Worldw_o_POV_5_65b3b2ff58.jpg)

&nbsp;

**Key Result #3: Female and male audiences were persuaded differently[^fifth].** The overall impact of these three stories is stronger for women, with the “Financial Literacy” story resulting in a statistically significant impact. Stories about women told by women appear to have a stronger impact on women, and they should be deployed with that differential impact in mind.

## What’s Next?
This test with Women’s Way, Purpose, and Harmony Labs is just one example of the ways to put the findings of the  [Narrative Observatory](https://narrativeobservatory.org/)  to work. In this instance, the three story options invited audiences to respond to common narratives (like “Not Like Us” or “The Grind”) that traditionally appear in public media — but by featuring an actual individual telling her lived experience, audiences were given an additional element to consider: person-to-person empathy and specific interventions beyond “hard work” that may help to eradicate poverty.

This suggests a tactic available to stage interventions in poverty narratives. Powerful stories can (and do) lead to concrete persuasion, and this insight offers possible routes for communications strategies for any organization with a mission of building economic mobility.

Although the work of narrative strategy happens at the level of audience and story, it is important to continually understand the entire narrative landscape to gauge how the stories we tell will land and affect audiences, and inform a data-driven approach to storytelling and content creation. Narratives in public media are ever-changing which means that this kind of work is never “finished.” If you are interested in staging a test like this one, or in the general work of audience and narrative intervention, please  [get in touch](https://harmonylabs.org/contact).

&nbsp;

[^first]: Gray border indicates a 90% confidence level
[^second]: 90% confidence level
[^third]: Gray borders indicate a 90% confidence level
[^fourth]: 90% confidence level
[^fifth]: Black border indicates a 95% confidence level. Gray border indicates a 90% confidence level. No borders indicate that statistical confidence is below the 90% level.

&nbsp;

---

&nbsp;

_This blog is based on research funded in part by the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Bill & Melinda Gates Foundation._

Worldw:oPOV1.png

Testing Storytelling Intervention Strategies Against Narratives of Economic Mobility

Imagining a World Without Poverty

Women's Way

womensway.org

Purpose

purpose.com

Is it possible to activate the U.S. public toward a nuclear-free future? Many experts and advocates believe that, currently, Americans are disengaged from stories about nuclear weapons, yet what we have found recently is that between 10%-30% of people who watch TV see something about nuclear weapons every day. The lack of engagement in nuclear disarmament despite a heavy volume of exposure to stories about nuclear weapons seems to suggest there is an element missing. Based on our analysis, these stories represent nuclear war as a dangerous, apocalyptic, and unavoidable outcome. What we need, then, is a narrative intervention to make denuclearization a realistic and believable future.

![Frame 1244 Large.jpeg](https://www.youtube.com/embed/R-feQRKCW5U)

The  [Nuclear Threat Initiative](https://www.nti.org/)  (NTI) partnered with Harmony Labs to find the existing narratives being told to U.S. audiences about nuclear weapons and identify opportunities to re-represent the nuclear story. In our  [previous work](https://medium.com/harmony-labs/dispatches-from-data-science-selling-the-ending-d9660f299fd3)  with NTI, we did qualitative and quantitative research to identify the best way to build narratives that support a nuclear weapons-free future. In this early work, we unearthed a major insight: “Put a future in it.”

&nbsp;

![rickmorty1.png](https://harmony-strapi-storage.s3.amazonaws.com/rickmorty1_210479f125.png)

&nbsp;

In other words, when communications practitioners are selling change of any kind, they should provide an ending that audiences want to “buy into”: a hopeful future we can achieve together. This became a recurring theme of our recent research with NTI where we discovered that the most persuadable audiences, defined below, are currently receiving a hopeless future, but that there are strong opportunities for narrative interventions. We’re excited to share the findings in this  [new public report](https://docsend.com/view/xeykmc3th97gfhxk).

## Who’s actually persuadable?

Harmony Labs tested user responses to media about nukes in randomized controlled trials (RCT) and defined the persuadable audience as the group most likely to change their opinion after receiving a narrative intervention designed to persuade them that a nuclear-free future is attainable. We then used demographic data to “expand” these findings so that we could see who in our opt-in media consumption panel data — covering 300,000 people in the United States — is likely to be similarly persuadable. We used the  [4 values-based audiences from our Narrative Observatory](https://narrativeobservatory.org/audiences/)  to better understand where audiences fall on the “persuadability” spectrum.

&nbsp;

![rickmorty2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/rickmorty2_8b762925d2.jpg)

![rickmorty3.png](https://harmony-strapi-storage.s3.amazonaws.com/rickmorty3_7a7cd8d61e.png)

This grid-mapping shows who NTI’s target audiences should be. The most persuadable audiences are  [People Power](https://narrativeobservatory.org/audiences/people-power/),  [If You Say So](https://narrativeobservatory.org/audiences/if-you-say-so/)  and  [Tough Cookies](https://narrativeobservatory.org/audiences/tough-cookies/).  [Don't Tread on Me](https://narrativeobservatory.org/audiences/dont-tread-on-me/)  are considered unpersuadable because they tend to disagree with pursuing a nuclear weapons-free future and narrative interventions tend to not make an impact. Alternatively, there is also an existing group of people who are already bought into a denuclearized future: People Power. These audiences are considered “unpersuadable” because they’re already on board with actively moving toward the target narrative. Therefore, we focus our persuasion efforts on If You Say So and Tough Cookies.

## It’s all fun and games till…

Persuadable audiences aren’t defined by how they engage with denuclearization because they don’t — yet. Instead, these audiences are defined by what they care about now, where they spend their time, and the kinds of stories that already resonate with them — letting us meet them where they’re already at in media.

YouTube is a key opportunity platform. About 1% of YouTube users are consuming nuke content every day, and gaming is the most popular genre for persuadables on YouTube — 3 out of every 1,000 adult YouTube users in the most persuadable group consume nuke-related gaming content every day. The content in this space pointed us toward a widespread, salient narrative about nukes that regularly reaches persuadables:

**Nuclear weapons are capable of world-ending destruction, unhinged individuals will cause our downfall — and there is no other future.**

This narrative is most often seen by If You Say So in gaming content like  _Minecraft_,  _Call of Duty_,  _Fallout_, even extending itself to entertainment TV featuring nukes. In cartoons targeted for adults, the cartoonish villainy of nuclear attack is reinforced in shows like  _South Park_  and  _Rick and Morty_, often combined with non-human or other-worldly encounters.

&nbsp;

![rickmorty4.png](https://harmony-strapi-storage.s3.amazonaws.com/rickmorty4_56d31c66dd.png)

![RMrevised.jpg](https://harmony-strapi-storage.s3.amazonaws.com/R_Mrevised_e156f4f6d2.jpg)

&nbsp;

Persuadable  Tough Cookies  represent additional cultural spaces in our Opportunity Zone. They spend more time watching TV than the other audiences: sitcoms like  _black-ish_,  _The Fresh Prince of Bel-Air_, and  _Martin_, shows that reflect the family focus of this audience. Religiosity is also a key opportunity for them, and faith-based content has significant reach here with sites like Bible Study Tool and Bible Gateway being amongst the most popular.

&nbsp;

![rickmorty6.jpg](https://harmony-strapi-storage.s3.amazonaws.com/rickmorty6_3b3fd2037c.jpg)

## Making the apocalypse avoidable

Persuadable audiences are already bought into the urgency and nature of the threat, but they aren’t convinced yet that denuclearization is possible. Interventions here need to tell a story of a more hopeful future: an avoidable apocalypse.

Moving forward, the next step for NTI and all advocates in the space is to make content that speaks to audiences within the “opportunity zone” and take the existing aware-but-cynical nuclear weapons narrative — currently present in gaming and entertainment — and add an achievable nuclear weapon-free ending.

In an upcoming blog, we will showcase our process of building cultural briefs for creators to help them understand the persuadable audiences’ cultural landscapes. These cultural briefs are intended to be used by artists and media makers to advise on form, style, and notable examples, to begin retelling the story in the right places. Follow this space for the next update, and if you are interested in research like this project, or in the general work of audience and narrative intervention, please  [get in touch](https://harmonylabs.org/contact).

&nbsp;

---

## Terminology & scope

For this report, Harmony Labs used opt-in internet and television panel data between January 1, 2020 and October 31, 2021, touching 300,000+ people in the U.S. and offering a minute-by-minute view into the content audiences care about, wherever they consume or create it.

Online news content was from 1/1/21 to 8/31/21. YouTube content was analyzed form 1/1/20 to 9/30/21. Online search content was from 1/1/20 to 10/31/21. TV was from 6/1/21 to 8/31/21. The earlier starts for online search and YouTube compared to online news accounts allowed for more nuclear content, and the different ending dates was due to different windows of complete data. The timeframe of TV data was relatively short because of the large amount of TV transcript data available for any given time period.

Nuclear content is deﬁned as any found media consumed that uses the keywords “nuke” and “nuclear.” We searched for these keywords on YouTube video titles, descriptions, and tags, online news article text, online search text queries, and TV transcripts. We excluded found media when the text surrounding the keyword indicated that the media was unrelated, such as when “nuclear” was followed by “family.”

Untitled.png

Rewriting the nuclear story for U.S. audiences

The Unhinged Apocalypse of Fallout & Rick and Morty

NYT.jpeg

How Young Latinos’ Online and TV Habits Offer Political Clues for 2022

Cultural and social transformation requires deep, interpersonal work often rooted in storytelling at the community level. The kinds of systemic changes that can transform a region or nation — such as policy changes — are influenced by narrative shift, by a change in our collective belief about what will work to resolve persistent challenges. Narrative shift starts with audiences and ends with powerful stories that can shape (and reshape) the future. In 2020,  [Bill & Melinda Gates Foundation](https://www.gatesfoundation.org/)  and other funders launched an ambitious program to provide 28 grants of $100,000 each in the Voices for Economic Opportunity Grand Challenge to a wide array of organizations that will work to correct mistaken assumptions and improve understanding of the barriers to economic mobility through the stories of those who experience poverty. Through this program, participants had access to data infrastructure provided by  [Harmony Labs](https://harmonylabs.org/), a media research lab, and practitioner strategy provided by  [Purpose](https://www.purpose.com/), a social movement incubator, to ultimately create and launch media and campaigns intended to support a framework of policy intervention.

Among this cohort of participants was the  [West Virginia Community Development Hub](https://wvhub.org/)  (The Hub), an organization with the mission to support West Virginians with the tools, training, and resources needed to create the kinds of high-impact solutions that lead to greater economic mobility for all. The leaders in this organization believe that economy building begins with communities, and their work is driven by the understanding that local residents are subject-matter experts on what is needed to unify communities and grow economies.

The Hub produced  [a series of short documentaries](https://wvhub.org/redefinedwv/)  showcasing community leaders who are making impactful change from the ground up in West Virginia using research from Harmony Labs. The Hub had two primary goals: (1) shifting the belief that West Virginia communities are at times hopeless and without leaders to the belief that anyone in West Virginia can lead positive change, and (2) helping the rest of America see West Virginians as they truly are — resilient, resourceful, and innovative. This work ultimately redefined stereotypical, long-held public assumptions about West Virginian culture by presenting transformative stories showcasing community leaders expressing their self-agency to make impactful change, even in the face of significant challenges, and the vibrant cultures that these leaders helped shape.

The Hub’s work shows how effective, researched storytelling can build a powerful foundation for narrative change.

&nbsp;

![DatametArt2.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art2_4ef95f9d6b.jpg "A still from the documentary short about The Vintage Theatre Company.")

## Rethink Your Base — When Data Meets Audience
Emma Pepper, Director of Strategic Network Communications at The Hub, began this project by addressing a core challenge: how to reflect The Hub’s audience back to themselves and to fellow West Virginians in a way that would motivate more people to both take initiative and scale up their efforts in facilitating increased economic mobility at the local level. To start, they engaged the audience work developed by Harmony Labs that was provided to participants in the Gates Foundation program, focusing attention around  [the four values-distinct audiences](https://obiaudiences.org/audiences/)  that have become a focal point of the  [Narrative Observatory](https://narrativeobservatory.org/). Emma and her team used this audience framework as the foundation to build the organization’s impact initiative. While The Hub previously defined their audience in terms of geography, by county or state-specific location, the values-based audience schema brought a new methodology and vocabulary that enabled the team to target their audience more accurately, more effectively precipitating a narrative change — or shift in core beliefs.

Emma and her team recognized that the audiences who were organically drawn to the organization’s work most closely exhibited shared values with Harmony Labs’  [“Tough Cookies](https://obiaudiences.org/audiences/tough-cookies/)” audience. The “Tough Cookies” audience characterization provided more than just basic demographic information like age, gender, race, political affiliation, etc. — the Harmony Labs’ framework for audience profiling allowed The Hub more insight into important voices (Dolly Parton, Reba McEntire, and Donald Trump), key brands that reflected a desire for the cozy pleasures of home (Betty Crocker, Lowe’s, and Purina), and also major news networks where the audiences consumed information (Fox News, CNN, and CBS News).

Emma and her team dug deeper into the Tough Cookies’ psyche through the news and TV they chose for themselves. A “stick-to-it” energy defines Tough Cookies, an audience that plays by the rules and believes in  [The Grind](https://narrativeobservatory.com/narratives/the-grind/)  narrative of economic opportunity: follow the rules, keep working, and eventually your ship might come in. This idea of commitment to carrying on in the face of adversity felt true to the team’s experience of their own audiences, and so did the strong focus on family, faith, and service. Emma and her team imagined how these insights from the data showed up in their own experiences with West Virginians and found an opportunity in something they heard over and over in the work: “I want to give back to make this a better place for future generations. I have hope that, working together, we can make a difference.”

While the Tough Cookies audience profile created a roadmap for The Hub team to understand their audience in a new light, the research shows national trends that are not strongly nuanced by locally-expressed attitudes toward poverty (and moving out of poverty). The region’s historic dependence on natural resource extraction as a means to prosperity creates unique, yet commonly shared, attitudes toward poverty for local residents.

The Harmony Labs research needed to be paired with “translators” who understood how Tough Cookies values were being expressed locally, so the team at The Hub engaged fellow staff members who provide front-line services to West Virginia residents. To explore explicit applications of language, they also conducted A/B testing through Facebook advertisements targeted to West Virginians to ensure that the Tough Cookies values were translated in a way that resonated with local audiences and felt authentic to their lived experience.

&nbsp;

![DatametArt3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art3_e77e1e5f73.jpg "A still from the documentary short about The Welch News in McDowell County, WV.")

## Creating a Setting
Beyond intellectually knowing what the audience was about, the cultural symbolism that The Hub found in the Harmony Labs data shaped the  _setting_  for their stories. Storytelling about Appalachia is deeply intertwined with imagery that came about as part of advocacy efforts for the War on Poverty championed by Lyndon B. Johnson in the 1960s. During this time, photographers were dispatched to Appalachian communities such as West Virginia and Kentucky to capture images of rural poverty. The resulting images were primarily black and white photographs of white residents wearing torn and dirty clothing and living in dilapidated housing. The formula for these images has been replicated so many times over the following decades that a simple Google image search of “Appalachian poverty” yields thousands of examples.

The Hub team knew they needed to show West Virginia as it actually is.

Seeing the look and feel of the entertainment TV shows that Tough Cookies choose for themselves offered a creative opportunity to flip the script. These shows often feature the outdoors, but portray the browns laced with gold. That’s not an accident — TV companies have done plenty of research on the look and feel that their audiences like — but Tough Cookies prefer a little sparkle in their entertainment and a hint of miracle in their stories.

&nbsp;

![DatametArt4.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art4_7a2ab8e354.jpg "TV shows consumed by the Tough Cookies audience.")

&nbsp;

In response, The Hub filmed their localized community documentaries in the summertime, waiting until there was green on the trees and an abundance of natural lighting. They emphasized warm colors, natural light, and an overall aesthetic sense of vibrancy in their creative. Story subjects were often portrayed in imagery with natural light beaming directly onto their faces, using aesthetic decisions to lift them up as the heroes in their own narratives.

&nbsp;

![DatametArt5.jpeg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art5_7e859700c7.jpeg "A still from the documentary short about the Hinton Community.")

## Who Are Your Protagonists?
As The Hub prepped to film their mini-documentaries, it became clear: stories are so much more than visuals. Armed with the Tough Cookies audience profile, The Hub began working with a local creative agency (led by a founder who identifies as a “Tough Cookie” audience member) to produce short-form documentaries. Cultural fit, messaging fit, and motivation fit became the core components of how to build the “characters” at the center of these stories — in other words, West Virginian viewers had to see themselves in and be seen by the documentaries to engage with the mission of The Hub.

Lori McKinney of The RiffRaff Arts Collective is part of a community in Princeton, WV that has partnered with fellow residents to use creativity, art, food, and performance to economically and culturally revitalize The Mercer Street Grassroots District. That story is at the heart of the  [documentary short](https://wvhub.org/redefinedwv-princeton/)  The Hub produced, centering the video around the people of Princeton and their embodiment of the Tough Cookies ethos, focusing on  **a devoted, tireless, creative dedication** to their community’s growth. McKinney herself said affirmatively, “The ways that we shape our world create the world that we want to live in. It’s about what we do together.”

&nbsp;

![DatametArt6.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art6_238e9986a5.jpg "A still from the documentary short about the Monticello neighborhood of Clarksburg, WV.")

## Storytelling Leads to Growth
Being able to pick an aesthetic, pair it with a protagonist (or a group of them), and build a clear framework around individualized experiences is the heavy lifting of narrative change storytelling. Having the research is only the beginning — being able to responsively build a story that feels authentic requires accepting (and rejecting) discovered elements about target audiences to create messaging and strategies that will aid in building overwhelmingly powerful narratives of change.

And, indeed, a specifically developed and chosen protagonist-centered ethos at the heart of these stories from The Hub has paved the way for effective engagement —  [the video about the Kelly Miller Community Center](https://wvhub.org/redefinedwv-monticello/)  in the Monticello neighborhood of Clarksburg, WV was a wild success in the Center’s fundraising efforts, resulting in an unexpected $140,000 grant after swaying the heart of a major local funder.

&nbsp;

![DatametArt7.jpg](https://harmony-strapi-storage.s3.amazonaws.com/Datamet_Art7_a2b63f96a8.jpg "A still from the documentary short about the Monticello neighborhood of Clarksburg, WV.")

## What Comes Next?
Results like this with The West Virginia Community Development Hub provide a creative model for high-impact storytelling that can be sustained — but requires effective distribution of the created media to scale the project and ensure that narrative change leads to the systemic changes, such as changes to policy, that are necessary to break the cycle of intergenerational poverty present in West Virginia and create much-needed economic momentum. It is certainly one thing for the research to influence growth results and lead to small wins (like a powerful video incentivizing a grant), but another entirely to be able to maximize these impacts to drive lasting, transformative community change. Scaled distribution will allow for programs to be advanced and grown, and each new produced piece of carefully constructed media will pull efforts forward one story at a time.

If you are interested in leveraging audience and culture work like this in your own community outreach and narrative intervention efforts,  [get in touch](https://harmonylabs.org/contact)  to start a conversation.

&nbsp;

---

&nbsp;

_This blog is based on research funded in part by the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Bill & Melinda Gates Foundation._

DataMetArt.jpeg

How The West Virginia Community Development Hub Used the Narrative Observatory @Harmony Labs to Build Transformative Visual Media

When Data Met Art: Redefining Narratives of Poverty Through Research-Driven Storytelling

The West Virginia Community Development Hub

wvhub.org

Two years ago, Harmony Labs and the  [Bill & Melinda Gates Foundation](https://www.gatesfoundation.org/)  set out to build the Narrative Observatory, a first-of-its-kind tool to support researchers and practitioners working to change narratives about poverty and economic mobility.

The work, as we envisioned it in our original  [white paper](https://docsend.com/view/j3xbxce6r6edack6), was complex, highly collaborative, experimental, hard. Success was not assured. To be sure, the trail we ended up blazing through this work was dense with occasional setbacks and failures, and big, difficult learnings — which, in the end, only strengthened the Narrative Observatory.

In September 2021, the  [Narrative Observatory](https://narrativeobservatory.org/)  launched as a website and prototype infrastructure for data-driven cultural strategy, with a strong circle of partners and charter users committed to helping us build something that would improve outcomes in their own work, across a range of social issues, from economic mobility to racial justice, immigration, and denuclearization.

&nbsp;

![HLabsDataNarChg2.gif](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Data_Nar_Chg2_9bc1d01dc2.gif)

## Initial Findings From Charter Users
Over two years, we worked with ten charter users to discover their needs with respect to cultural strategy and media making, and to experiment with the outputs best suited to satisfy those needs. Charter users included: the  [Bill & Melinda Gates Foundation](http://gatesfoundation.org/)  and their challenge grantees, the  [Robert Wood Johnson Foundation](https://www.rwjf.org/),  [Mozilla Foundation](https://foundation.mozilla.org/en/),  [Equis Labs](https://www.equislabs.us/),  [A/B Partners](https://www.abpartners.co/),  [Race Forward’s Butterfly Labs](https://www.raceforward.org/),  [NARAL](https://www.prochoiceamerica.org/),  [Caring Across Generations](https://caringacross.org/),  [Nuclear Threat Initiative](https://www.nti.org/),  [Jim Joseph Foundation](https://jimjosephfoundation.org/), and  [United We Dream](https://unitedwedream.org/). The work we’ve done for and with these partners includes:

## Audience Segmentation
We started by doing bespoke audience clustering, like  [this work on Black audiences online for A/B Partners](https://docsend.com/view/xafiesp86f9ju4yk)  or  [this work on Latinx audiences for Equis Research](https://medium.com/harmony-labs/what-media-matters-for-latinx-americans-6b9752c32b5d), before we developed a single values-based audience segmentation, via the Bill & Melinda Gates Foundation’s poverty and economic mobility portfolio. We are now using this segmentation for most Narrative Observatory work, which provides a useful framework for ongoing learning across projects and continuous refinement.

&nbsp;

![HLabsDataNarChg3.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Data_Nar_Chg3_610994627d.jpg "The 4 values-distinct segments around which the majority of audience research is mapped.")

## Audience Profiles
Audience profiles bring to life audience segments by describing and embodying each segment’s values, beliefs, attitudes, knowledge, and/or media behaviors. In our audience profiles, we most often use the four-quadrant values-based audience segmentation, applied explicitly in  [the Narrative Observatory website](https://narrativeobservatory.org/audiences/). We can also generate audience profiles from other audience segmentation, as in  [this work for NARAL](https://www.abortiondisinfo.org/audiences)  using the audience segmentation from  [Story at Scale](https://www.storyatscale.org/). We have developed many ways to bring audience profiles to life, including the  [search-based case studies](https://www.abortiondisinfo.org/case-studies)  created for NARAL and content ecosystems generated for a  [recent exploration](https://static1.squarespace.com/static/5c993096f4a30b0001b73418/t/61e89eb9d4ab9913bc309804/1642634940820/elnuevonoticiero.pdf)  of polarizing YouTube channels for EquisLabs.

&nbsp;

![HLabsDataNarChg4.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Data_Nar_Chg4_c256ef6ae1.jpg "Segmented content ecosystems within Latinx YouTube consumption in work with EquisLabs.")

## Narrative Profiles
Narratives emerge from audiences, built and reproduced and spread from the ways people interact with stories in media. Narratives serve as atmosphere, moods that color people’s perception and influences how people create and reflect culture. Our narrative profiles are always built “bottom up,” which means we start by having annotators code media content for relevance to a particular social issue. We use annotator judgments to train a relevance model to help us determine on which media content to focus additional narrative analysis, then, through machine clustering and human decision making, are able to produce digestible sets of narratives to characterize the terrain for one or more audiences. The audience-specific narrative profile can be seen as an extension of the audience profile, as seen in  [this work](https://docsend.com/view/zdn5h5r8z3cez8ud)  with Mozilla about the narrative landscape surrounding AI.

&nbsp;

![HLabsDataNarChg5.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Data_Nar_Chg5_f2cc711331.jpg "A topical and persistent narrative of “The Grind” was showcased on The Narrative Observatory website.")

## Story Opportunities & Threats
These outputs are prescriptive in nature and primarily suggests  _change-making_ intentions, identifying possible “north stars” in audiences’ current media diets toward which action and response can be directed, work that often takes the form of content briefs. Content briefs make explicit recommendations about the stories, story beats, endings, characters, worlds, tones, and/or aesthetics that are likely to produce desired responses in order to bring about some evolution in reality. The work of generating content briefs is best done in close collaboration with content creators, wherein the Harmony Labs team helps content creators form story interests and intuitions into hypotheses, which creators quickly embody in low-fidelity content, and which Harmony Labs then tests. We have done this work with cohorts of content creators including  [Race Forward’s Butterfly Lab](https://www.raceforward.org/practice/arts-narrative-culture/butterfly-lab-immigrant-narrative-strategy),  [Mozilla Foundation’s Creative Media Awards program](https://medium.com/harmony-labs/ai-art-advocacy-fb58989d920f), and for Bill & Melinda Gates Foundation challenge grantees in partnership with  [Purpose](https://www.purpose.com/)— and also directly with creative agencies, as in our work with  [Nuclear Threat Initiative](https://medium.com/harmony-labs/dispatches-from-data-science-selling-the-ending-d9660f299fd3?source=user_profile---------4-------------------------------).

&nbsp;

![HLabsDataNarChg6.jpg](https://harmony-strapi-storage.s3.amazonaws.com/H_Labs_Data_Nar_Chg6_2fbebe849a.jpg "1 of 9 creative projects from Black artists as part of Mozilla’s Creative Media Awards program.")

## What’s Next for the Observatory
Over the course of the project, the audience and narrative vocabulary that has been developed within the Narrative Observatory has become progressively adopted by more than a dozen organizations in their narrative and cultural work. Some examples of how we’ve seen Narrative Observatory outputs put into practice include:

-   Persuading audiences that collective solutions to poverty exist by helping storytellers blend stories focused on individual characters with insights about the systems that contribute to economic disadvantages
-   Moving denuclearization advocacy away from fear-based mobilization strategies and toward empowering language of positive futures proven — in other words, selling an aspirational future that we can achieve together
-   Aiming partners engaged in immigration, reproductive rights, racial justice, and civic engagement advocacy away from an exclusive focus on “the base” and toward adjacent or peripheral audiences, resulting in content that is equally effective with the base and generally less polarizing for adjacent audiences

Building off these early “wins,” moving forward we’re focused on continuing to innovate in our technical approach to understanding audiences relative to their place in culture; to identifying, measuring, and tracking narratives within audiences over long time scales; and to surfacing audience-specific story opportunities and threats. We’re also working closely with our partners and  [advisors](https://medium.com/harmony-labs/introducing-our-new-advisory-board-785e6948a86f)  to identify ways in which the outputs and insights generated from the Narrative Observatory might best be integrated into their overall strategy and their media making practices.

Finally, we’re always on the lookout for ways to extend this work to new issues areas and cultural domains, with new partners. If you’re interested in leveraging the Narrative Observatory for your organization,  [get in touch](https://harmonylabs.org/contact)  to start a conversation.

As we reflect on what we’ve accomplished and where we intend to go next, we’d also like to extend a note of appreciation and gratitude to the community that has provided critical support to this work along the way. Our work is collaborative by necessity and design and would not be possible without our funders, partners, and community. Amidst society redefining a “new normal” over the last few years, together our work evolved in scope and ambition, seeking to answer questions about our ever-evolving world and we look forward to continuing this work together in the years to come.

&nbsp;

---

&nbsp;

_This blog is based on research funded in part by the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Bill & Melinda Gates Foundation._

HLabsDataNarChg1.jpg

What’s Next for the Narrative Observatory @Harmony Labs

Trailblazing Data-Driven Narrative Change

![mike.jpg](https://harmony-strapi-storage.s3.amazonaws.com/mike_7a93191c91.jpg)

_Over the last year, Harmony Labs has been developing an audience-narrative architecture to help the Robert Wood Johnson Foundation (RWJF) and its partners find “narrative strategies” and actual storytelling that would activate people in the U.S., especially healthcare professionals, to dismantle structural racism in America’s healthcare and public health systems. The foundation of this work was research on audiences and the health equity narrative landscapes they inhabit. This initial research helped surface the most promising opportunities for content testing and validation, from which our project partner [Story Strategy Group](https://www.storystrategy.co/) convened groups of professional storytellers, anti-racism experts, community leaders, and content strategists to develop new stories that might transport audiences toward a target narrative of health equity. This piece is one in a three-part series from some of the subject matter experts and storytellers who worked on this project. You can explore the full findings [here](https://harmonylabs.org/project/health-equity)._

_The post was originally published on [Medium](https://medium.com/harmony-labs/whats-the-metastory-d9e18ea5d232)._

&nbsp;

---

&nbsp;

I am a [story consultant](http://storysaveslives.com/) who works with social justice organizations and public defenders to use storytelling to get audiences to empathize and act. I’ve consulted on hundreds of criminal and civil cases and have an 80% success rate, including several acquittals in jury trials where our client was facing life in prison or the death penalty. I attribute this success to providing attorneys with actual storytelling techniques that trigger empathy.

My social justice storytelling work involves interviewing people and identifying the stories they’re living: their story purpose, allies, obstacles, and goals. The information I glean is then used by social justice organizations to shape stories that align with their audience’s stories. By doing so, we can find common themes that unite disparate groups and encourage them to work together. I believe that of the hundreds of people I’ve spoken with, almost every person could empathize with an issue if it related to their story goal.

The Health Equity Narratives Project, sponsored by the Robert Wood Johnson Foundation, was designed to provide advocates for health justice with narrative strategies to support their goals, including the bold and urgent objective of dismantling structural racism in healthcare systems.

My role on the project was to take the anecdotes or issues identified by healthcare experts, turn them into stories, then ensure the stories we told achieved the project goals, within the [constraints of Facebook image ads](https://www.facebook.com/business/ads-guide/update/image), and a testing-friendly format selected by the research team.

Our goal with these testable stories was to get audiences to recognize the problem of racial inequities in the healthcare system and agree that action must be taken to address the imbalances suffered by people of color. Achieving this meant challenging people’s biases.

## Storytelling Empathy, and Epiphany
Stories are universal, and make us human. We use stories to justify our life choices. Almost all of us can name a story that changed our lives. When asked how we chose our partner, or a career, we will tell a story. This is the power of storytelling in our lives. When the right story is told in the right way, an audience’s views on the world — and their attitudes, beliefs and behavior — can change.

In my work, I use the terms “story” and “storytelling” literally. A story is a mechanism that must be constructed for an audience. It must have certain key features — a protagonist, a story goal,and obstacles. This is the DNA of western storytelling, and it’s what people expect in a movie, novel or TV show, as well as in a criminal trial, ad campaign, or social justice initiative.

The goal of the storyteller is to create empathy. Action can be achieved with empathy.

In my trial work, the jury is the audience, who usually are biased against the client, who is often part of a marginalized group, such as the poor, mentally ill, or unhoused. Bias against such groups makes it easy for juries to convict the client, and law enforcement and prosecutors often rely on such bias to make their jobs easier.

I help the defense attorney tell the jury true stories about the incident, client, and case that prove the jury’s assumptions false. In the moment the jury realizes their assumptions were wrong, they have an epiphany.

Epiphanies are outcomes _determined by the audience themselves_ based on the events of the story. Having an epiphany is more powerful and feels more true than simply being told what to believe.

The importance of an epiphany is that it opens the door to empathy. For example, in one murder trial I consulted on, the client was silent and avoided eye contact. The prosecution depicted him as a dangerous “loner” living in a shack and roaming public parks at night. We represented the client’s actions to show that he was an introvert — an avid reader who lived in a room behind his father’s house out of deference to his stepmother, who resented his presence. Instead of spending his free time at his father’s house, he went to the park to avoid a confrontation with his stepmother. The same actions given proper context allowed the jury to have the epiphany of seeing him differently.

It’s important to note that we did not call the client an introvert — we let the jury come to this conclusion based on the actions of his story. This re-framing of the client’s personality and actions caused the jury to have an epiphany that their assumptions were false and allowed them to review the entirety of the case with empathy. They ultimately found the client not guilty. Post-verdict interviews proved the validity of using storytelling to allow the jury to empathize and act impartially.

This is not an isolated example. My experience working on hundreds of cases bears out that stories with reversals of expectations can cause the audience to reconsider their opinion of the client and the entire case.

The successful implementation of storytelling in trial has yielded extremely good results — and there’s no reason that it can’t be applied to social justice issues, too, provided we tell the right stories the right way. In the Health Equity Narratives Project, I wanted to use storytelling in the same manner, to create the same effect.

## Telling the Story of Healthcare Inequities
To create the first round of test ads, I led a creative team that took ideas about racism in healthcare and healthcare systems, as well as stories about equity and access, and put them into simple design concepts with one image and very few words.

The ads featured declarative statements, such as, “The healthcare system is biased against Black mothers.” The results were as expected — the ads resonated with those that we expected to agree with them and had little effect on those groups that tended not to believe or care about such issues.

For the second and third rounds of testing, we presented the same issues, but as stories — which we know is an onramp to creating empathy. Using information gathered from experts in racism in healthcare, I crafted specific stories. Creative teams then wrote copy and applied images, which included the four key story elements of:

1. A protagonist;
2. A goal the protagonist is trying to achieve;
3. Obstacles challenging the protagonist from attaining their goal, which could be an antagonist, a situation, or the protagonist themself; and
4. An outcome.

With structural racism and multiple inequities prevalent throughout our healthcare system, we were sometimes able to combine multiple problems into one story. For example, here are three general problems faced by people of color in healthcare, according to experts:

- Bias against the patient’s race or culture by healthcare practitioner
- An indifferent and illogical bureaucratic healthcare system
- A shortage of healthcare providers

We created two versions of one story that addressed all three of these issues:

>![ex11.png](https://harmony-strapi-storage.s3.amazonaws.com/ex11_60872b160a.png)
>![ex22.png](https://harmony-strapi-storage.s3.amazonaws.com/ex22_a023eb348a.png "Version 1 & Version 2")

The images and copy in Version 1 tell the story of a doctor named Jean-Pierre who immigrated to the U.S. but is unable to practice medicine due to bureaucracy.

Version 2 is a general version of the same story. No one is named, but the images depict and imply the story of specific protagonists.

Versions 1 and 2 both worked with certain audiences, supporting our theory that presenting the same issue in a story format invites viewers in and creates the empathy people need to call for changes to our current healthcare system.

## The Importance of Addressing the Metastory
A metastory is a general pre-existing story or belief held by the audience. A storyteller must know their audience’s metastory and have a relevant story that can address it in the right manner — perhaps reinforcing a beneficial metastory or countering a harmful one.

For example, in a jury trial, the jury’s pre-existing metastory might be that a homeless Black loner is most likely a murderer. By providing context and detail to his actions, we can prove the jury’s metastory to be incorrect, which can compel them to acquit.

For many social justice issues, audiences are rooted in feelings of scarcity, believing the metastory that they will have to sacrifice their goals if the needs of others are addressed. We specifically rebutted that metastory in this project by showing that everyone would benefit if the needs of a specific group were addressed. [Specifically, we’d have more doctors if we eliminated the bureaucracy that prevented immigrant healthcare workers from doing the jobs they were trained for](https://www.migrationpolicy.org/news/us-health-care-system-immigrant-health-professionals-are-talent-pool).

![ex33.jpg](https://harmony-strapi-storage.s3.amazonaws.com/ex33_a90eb072ca.jpg)

This story tested well with a range of audiences because they could see how the action would benefit everyone. The key overlapping story points were:

- Almost everyone has suffered bureaucratic frustration in the healthcare system
- The healthcare system feels indifferent to human need
- Helping even a subset of people would make the system more efficient
- Having more healthcare providers with language skills and cultural sensitivity to people of color is good

## Key Story Questions
As part of the Health Equity Narrative Project, our team ultimately created 105 pieces of content that sought to counter harmful metastories related to healthcare access and structural racism in our healthcare system and encourage people to take action to rectify these injustices. Of those 105 “mini stories,” many were highly successful in engaging audiences and/or moving them towards a better and more equitable narrative about healthcare and healthcare systems.

The lessons we learned are not specific to just this project. Others can apply our learnings to their own work provided they leverage the power of stories and storytelling.

When creating stories for social justice purposes, I recommend keeping these key questions in mind:

- What do you want your audience to do?
- What are the audience’s existing metastories? What stories are they currently living
- For the stories you intend to tell your audience, what are the obstacles/antagonists blocking the protagonist and how does your audience relate to them?
- Does the call to action help or harm your audience?

Creating stories that address these points is the best way to get all audiences to empathize and take action — which we need to do if we’re going to be successful at getting more people to join us in the fight for a better, more just, and more equitable future.

News

Sort

Narrative Strategy for Racial Equity

by Melinda Weekes-Laidlow Jan 20, 2023

VR: Going Beyond the Empathy Machine

by Harmony Labs Jun 17, 2016

Viral Political Ads May Not Be As Persuasive As You Think

Aug 7, 2018

Media Consumption in the American Midwest

by Harmony Labs Oct 20, 2022

archive

In Search of Healthy Narratives

Equity in Research… For Equity in Health

Audience-Centered Advocacy : Toward a New Model?

A New Year for Harmony Labs

ICYMI: Audiences 101

Introducing Our New Advisory Board

About

01

02

03

04

Approach

01

02

03

04

Work

01

02

News

01

02

03

Other

01

02

03

04

Email us at hello@harmonylabs.org

©2023 HARMONYLABS

by Melinda Weekes-Laidlow
Jan 20, 2023

by Harmony Labs
Jun 17, 2016

by Harmony Labs
Oct 20, 2022