Deep Dive: Listening to the Narrative Space

As the second step in the CHL methodology, listening to the narrative space means taking a broad look at the shape of the narrative(s) we chose to engage with in Step 1 of the CHL methodology (Module 4), where we created our point of view statement. Listening explores the relationship of the interlocked parts of a narrative, such as the spaces, actors, actions, momentum, and sentiment of a conversation. Listening is essentially a data collection and coding process where we aim to find out how the narratives (at whatever scale) we are engaging with come alive in the world. We construct a listening model to set the parameters for our inquiry which is an entry point for the narrative we want to analyze. 

Once we identify the conversation by collecting the data outlined in the parameters of the listening model, we codify the data to analyze the relationship between the actors and the narratives. 

There are two listening methods available to us, depending on your capacity and the nature of the conversations tracked:

Both approaches allow us to see and analyze the landscape of a narrative including the main actors involved, the spaces (physical and digital) where it takes place, and the shared messages that are being exchanged. As social movements become increasingly plugged into technologies, both for organizing and communication, there is a need to understand the ways in which this involvement shapes social movements and in turn how they are hacked and reinterpreted for the purposes of mobilization.

A listening model is a theoretical and practical tool developed to research and make sense of narratives. A listening model should be unique to an inquiry and include specific parameters which will be described below. This tool should be oriented to the analysis of complex systems, which is defined as the analysis of “how relationships between parts give rise to collective behaviors of a system and how said system interacts and forms relationships with its environment”.  

What this essentially means is that the tool aims to help understand the relations of narratives, actors, dynamics, themes, and actions that work within a narrative frame. When we do social listening to analyze narratives for narrative change and social movements, we are able to locate

• Nodes, which are the basic units of narrative communities, which can be people, publications, groups or businesses
• Narrative communities, which are social networks that are involved in an active discussion about a specific topic or issue.
• Echo chambers, which describe narrative communities which primarily talk amongst themselves and are insular.
• The power dynamics and structures between/within narrative communities, that show which narrative forms are dominant and which ones are emergent.

Listening to Narratives

Now that you know what listening means for Culture Hack, we will define what kind of data we need and how we are going to collect it. In mapping a narrative there is a need to listen to the current conversations relating to it. Listening to narratives mean you would pay attention to the following (among other things):

1. Demographics: Given your point of view, what demographics can you use to refine your listening that will help you discover the significant narrative communities that will emerge from your data (e.g. political affiliation, approximate age, gender, etc)

2. Themes: Given your Point of View, what are the themes you want to analyze in your narrative. (Think about what topics can be derived from the narrative you want to know in depth, perhaps a narrative about the climate crisis may be related to the defense of the territory or indigenous communities; or a narrative about reproductive rights has to do with access to safe abortion, all narratives have topics that are related, make sure you think about what are the different vectors that make up the narrative.)

3. Platforms: In which places (physical or digital) will you carry out your search for information on this narrative? (Specify all the platforms on which you will carry out your search. If there are digital spaces you can consider  Twitter, Instagram or Facebook. You can also search forums and news portals. 

4. Date range: What time period does this span? (Choose a specific moment – preferably in the present – about the narrative you are researching. The dates you choose can revolve around a significant event such as BLM or Fridays For Futures, an important political summit or social movement.)

5. Geography: In what countries, languages, nations or territories are you performing your search? (Generally once you choose the momentum of a conversation you can easily define in which specific territories you want to conduct your search. For example, if we want to explore a narrative related to the decriminalization of abortion in Latin America we could choose some countries like Argentina, Mexico and El Salvador and so on with each conversation you want to listen to.)

Based on your considerations on the 5 points above, you should now be able to indicate how you collect data, in other words, create your strategy:

6. Strategy: How will you collect your data? (This is the most important step to build your listening model, as you will have to define a strategy to collect your data. You could choose to use small listening methods like surveys, in-depth interviews, and manual collection on websites or look for broader patterns using big listening. You can also choose a hybrid strategy.)

If you decide to pursue big listening, and use a search engine for this (outlined in the resources section below), remember to develop your key search criteria by listing a serious of key words corresponding to the narratives you want to listen to (see an example of this in the Indigenous Futures listening model below).

If you are still having trouble here are some questions for you to think about individually or collectively:

• What events have shaped or mobilized public opinion?
• Who has been influential in bringing this narrative to the general public?
• Is your subject related to any mainstream topics?
• If there is no actual conversation about the topic you are working on, what is a broader/related space you can listen and hack into? 

Activity: Defining your listening data

This table is useful to help you define your listening activities; in other words define the data you will listen to. Create as many of these tables as you think necessary for your data collection process.

Listening Data	Description
Demographics
Themes
Platforms
Date range
Geography
Strategy

Listening Data	Description
Demographics
Themes
Platforms
Date range
Geography
Strategy

Collecting and coding your data

Step 1 – Prepare for coding

Coding data is simply looking for themes within the data you collected. Once you identify, collect and organize your data, it is necessary to code it before doing further analysis. The coding process allows you to make your data meaningful in relation to your POV and therefore is always an interpretive process. Code is a word or short phrase that labels your data based on the patterns or themes you find.

There are two types of coding, one is an approach that assigns the codes to the data before retrieving it (deductive). The other type is creating codes once the data is collected and the patterns start to emerge (inductive).  All data can be coded, however, it would be an endless and impractical task to try to code absolutely everything. Therefore, after reviewing all the data and annotations you have made at the time of collection, you can begin to identify themes that emerge. How it is really important that you know what it is you are looking for hence it is highly recommended that you keep a copy of your Point of View, concerns on the narrative research and goals of the intervention to guide your coding decisions. 

Given your point of view and the listening activities you have outlined, what are the key questions you want to ask the data?

Q1.

Q2.

Q3.

Step 2 – Put together your listening model

Now that we have reviewed what data is important for our intervention, outlined what narratives we want to find and how to code our data, we can put together a listening model that will be our complete guide for data collection, coding and analysis. The listening model should follow the structure outlined should cover the following main aspects:

• Point of View
• Small Listening activities
• Big Listening activities
• Coding: What are we searching for (hypothesis) & Questions for the data

Step 3 – Collect your data!

See the resources section below for a list of tools you can use for your small or big listening.

Step 4 – Coding

Now that you have collected your data, you can code it given the key questions you have articulated previously. Write a short description of what your data is telling you. For this exercise, we will use data from a poll about ‘climate change’ as an example. This coding sometimes is about copy-pasting, but sometimes will require you to interpret the answers.

Example: “The UN, Scientists, Greta Thumberg and the youth movement always talk about climate change”

• Nodes: Greta Thumberg, UN

 Example: “The UN, Scientists, Greta Thumberg and the youth movement always talk about climate change”

• Communities: Scientists, Youth Movement for Climate Justice

 Example: “Yes, climate change affects us all and our future on this planet. I have been in several climate strikes and am planning to organize with my friends at school.”

• Keywords: Future, Planet, Activist, Youth

Step 5 – Questioning your data

Write down three questions that you want to answer with the data set you just created, some of them can be qualitative and some of them could be quantitative. These questions should be aligned to your point of view.

Example Questions:

• From my data, what would be the profile of people interested in my narrative?
• Was there any Node or Community mentioned several times that I didn’t know about?
• Was the overall poll useful and what can I learn to make it better next time?

Example of a Listening Model – Indigenous Futures

This listening model pursued big listening. They developed their keywords used to search for relevant narratives to listen to, based on their consideration of the demographics, themes, platforms, date range, geography, of the spaces they wanted to listen to (outlined in the section “Listening to Narratives” of this module).

Overview: 

Indigenous Peoples are 6.2% of the world’s population but their territories contain almost 80% of the planet’s remaining biodiversity. In the heart of those diverse ways of knowing & being, beats resilience: for more than 500 years, indigenous nations around the world have dealt with systems of oppression that threaten bodies, territories and cultures. This project focuses on co-creating replicable and scalable narratives, tools and practices with indigenous peoples as an amplifier of movements in defense of cultural and ecological diversity in times of climate crisis, pandemics and war.

Objectives: 

1. To track communities (allies) that are already communicating these core ideas. The aim here will be to identify their frames and their impact in the narrative space.
2. To track the  ‘broad’ narrative landscape, that is defining the key narratives and themes that we have identified as important, or potentially ‘hackable’.

For our Indigenous futures case study, data collection began in August 2021 until October 2021. Our data collection captured a conversation dating back to May of 2021. We researched both English speaking and Spanish speaking in narrative spaces to ensure that our data did not only reflect a Global North perspective. In our choice of keywords, we sought to find whether the current public conversation connects Indigenous activism and struggles with climate change – both in Spanish and English speaking spaces. For this, we looked at two types of conversations:

1. Climate Change (General): We tracked conversations around key events of the summer of 2021 like the release of the IPCC report. We also wanted to track the perspective of people who are not necessarily activists or involved in the climate change conversation but who are living in its effects. So we looked for conversations related to recent climate disasters such as hurricanes, fires, floods, fires, droughts, and we cross referenced them with mentions of “climate change”, “global warming”, etc. 

2. Indigenous Climate Spaces: We also focused on Indigenous narrative spaces by looking at specific movements or current issues through keywords #Landback, #StopLine3, Indigenous people against carbon report, etc. We wanted to understand how Indigenous communities addressed climate change and/or connect it to their current problems or resistance. 

The data led us to spend some time on social media, reading posts, looking at comments to get a better understanding of the conversation.

The table below identifies our key search criteria, the date range of collection & related objectives:

	Narrative	Search Criteria (English)	Date Range	Objective
1	Cop 26	“COP26” + “COP 26” + “COP-26” + #COP26	Aug-Oct	2
2	Indigenous	“Indigenous” + “Indigenous Peoples”	Aug-Oct	1
3	Climate	“Climate change” + “climate Collapse” + “climate emergency” + “climate crisis” + #climatechange + #climatecrisis + #climateemergency + #climatecollapse	Aug-Oct	2
4	Indigenous + CC	“Indigenous” + ( ”climate change” OR “climate crisis” OR “climate emergency” OR “climate collapse”)	Aug-Oct	1
5	Land Defenders	“land defender” + “land defense” + “defenders of the land”	Aug-Oct	1
6	Indigenous day	#IndigenousDay #IndigenousPeoplesDay #IndigenousPeoples #IndigenousRights #WeAreIndigenous	Aug-Oct	1
7	IPCC	“IPCC” + #IPCC + #IPCCFindings + #ClimateReport	Aug-Oct	2

Step 2: Your Listening Model

Overview:

________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Objectives:

__________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

In the table below identify your key search criteria, the date range of data collection & the related objectives:

	Narrative	Search Criteria	Date Range	Objective
1
2
3
4
5
6

Develop insights from your data

Once you’ve defined your listening model, collected data, and coded it, you can start to develop insights.

These questions will help you create a list of initial insights, from which you will then start to identify narrative communities (in Module 6):

 What are the overall trends and patterns? Who/What are the most influential voices? What are they saying? Do they interact? What are we not seeing in the data?

Resources

A) Listening Tools for Small Listening:

1. Surveys, interviews and polls.

Both methods can help you collect qualitative data from participants. Choose the method according to the depth and needs of the subject.  For example, if you prefer to collect more complex data, then choose an in-depth interview with few questions; on the other hand, if you need simple data quickly, then go for a survey. See our annex on how to develop and conduct polls. 

2. Manual collection

Manual collection is reading through websites, social media posts, documents and articles to identify language and shared narrative patterns.

3. Memex

Memex is a useful tool for manual collection as it is an open-source software that can help you save and organize your bookmarks, underline, annotate, and share what you find online. 

4. Zotero

Zotero is a free and open-source reference management software to manage bibliographic data and related research materials. It includes features such as web browser integration, online syncing, generation of in-text citations, footnotes, and bibliographies, as well as integration with the word processors Microsoft Word, LibreOffice Writer, and Google Docs.

5. Data Scraper (Chrome Extension)

Any data scraper tool can help you scrape any HTML web page. You can extract text, tables and lists from any page and turn them into CSV files, APIs or spreadsheets. It’s useful to browse academic articles, mine the text and build a corpus that could be helpful for linguistic analysis. 

B) Listening tools for Big Listening:

1. Media Cloud

is an open-source platform for studying media ecosystems. The tools of this software are designed to analyze, visualize and deliver information to answer quantitative and qualitative questions about the content of online media, it collects most of its content through the RSS feeds of media sources they follow, they only have data from the media sources from the time they started scraping its RSS feeds.

2. Socioviz

Is a free social network mining and network analysis tool for Twitter, with this platform you can analyze any topic or hashtag, discover global trends on Twitter, identify influencers in a conversation and export data for analysis in any other data visualization software. 

3. Culture Hack Platform

An integrated set of tools and techniques to track, research and intervene in cultural narratives. We capture and map large volumes of social data then analyze the networks, language, and deep logics of the narrative to develop actionable insights for strategic narrative interventions.

4. Gephi

Gephi is a free open-source software for visualization of networks and graphs. It can handle big-data sets, reading files in .csv, .gexf and .gephi. Gephi includes a number of useful metrics such as centrality, community detection and random layouts. It can do real-time visualization as well as over time comparisons of data

5. NodeXL

NodeXL Basic is a free, open-source template for Microsoft Excel 2007, 2010, 2013 and 2016 that makes it easy to explore network graphs.  With NodeXL, you can enter a network edge list in a worksheet, click a button and see your graph, all in the environment of Excel.

Public APIs: Some websites have public APIs to gather data, such as Twitter, Reddit or Instagram, many others also have Python packages that make it easier to access the data you might need: Tweepy for Twitter or PRAW for Reddit.