Skip to content

About our DATA

Why was this Data Collected?

In late 2016—in the midst of Donald Trump battling for the presidential office—reports documenting incidents of hate and bias in the United States began to proliferate. As reported by sources such as the Southern Poverty Law Center (SPLC), the New York Times, The Washington Post, and the FBI, increased threats related to antisemitism and white supremacy were on rise, and many organizations noted that such proliferation had much to do with Donald Trump and his incendiary rhetoric. Due to such increased violence and the lack of reliable governmental data about it, the SPLC, ProPublica, Think Progress, and New American Media began to document incidents of hate and bias in order to help account for this intensifying phenomenon. As ProPublica argued, we need “new, more creative approaches” to document and understand both high-level and low-level incidents of hate—“from hate-inspired murders to anti-Semitic graffiti to racist online trolling.” 

The Swastika Counter Project, which began in early 2017, is a direct response to this call to action.

Around the time of Donald Trump’s inauguration, Dr. Laurie Gries, a visual rhetoric scholar, began noticing that a troubling wave of swastikas was sweeping across the United States and drawing intense concern as news reports emerged almost daily to report various sightings. Early data collection for the Swastika Counter Project confirmed that such swastika incidents were targeting a wide range of minoritized peoples and communities, often going so far as to make explicit, violent threats when delivered with accompanying texts. Simultaneously, swastikas were surfacing in relation to Donald Trump, sometimes to show support, and other times to critique his widely-acknowledged appeals to white nationalism. Considering such diverse functions, the Swastika Counter Project began with four interrelated research questions: 

  • Where are swastikas surfacing on the streets of the United States as Donald Trump gains presidential power? 
  • How are swastikas surfacing in terms of media, and with what other signs and texts?
  • Who are swastikas targeting and who is spreading them?
  • How are swastika incidents being reported and how are people and communities responding to swastika incidents?

How was this Data Collected?

To address these questions, the Swastika Counter Project used a digital research method called iconographic tracking (Gries 2015) to search for swastika incidents that occurred between January 1, 2016 to January 20, 2021, the time roughly between Donald Trump’s entrance into the 2016 Republican presidential primary and his last day in the oval office. Using the search term “swastika,” data was collected one month at a time via Google’s image and text-based search engines. In sum, researchers discovered over 1300 swastikas on the streets of the United States during this time period.

In order to learn, among other information, where swastikas were reported, in what forms they surfaced, who swastikas were targeting, and how communities were responding, researchers spent five years collecting, processing, and analyzing the reported 1300+ swastika incidents. To decipher this information, we read through thousands of reports published in national and local newspapers, magazines, religious publications, student newspapers, organization websites, social media sites, etc. To ensure reliability, all incidents were cross checked across multiple reporting sources and all data was double coded.

A Note about the Ethics of our Data Collection Process

Google Search engines were chosen for their ability to quickly crawl through millions of pages on the World Wide Web and to cull information from various media types (video, photographs, print texts) and source types (newspapers, magazines, social media, institutional websites, etc.). This broad access was necessary as many swastika incidents are only reported in community or hyper-local online news sites, student run online magazines, and/or social media sites. Still, it is well known that filter bubbles limit the kind of data one is exposed to via Google Search. To account for restricted information, researchers thus searched through 10-15 pages of Google findings for each swastika incident. Researchers also used strategies such as burning cookies, erasing web history, using incognito functions, and diversifying the computers on which we did research. 

To address both bias and limited information in sources, we crossed checked information about each incident across multiple kinds of sources. We also kept track of how source types referred to swastika incidents, noting a range of descriptive terms—from Nazi symbols to racist graffiti, to vandalism. In addition, because we wanted to avoid skewing data toward one particular targeted community, locale, or institution, we chose not to narrow searches by a particular publication outlet or organization. Due to this decision, our database of information does not document every single swastika incident that occurred in the U.S. between January 1, 2016 and January 21, 2021, but we do believe our database offers a reliable set of data via which we could identify patterns and trends relevant to this case study’s governing research questions identified above.

How was Data Analyzed?

Working in a Google spreadsheet, data analysis occurred through a rigorous practice of tagging and coding data according to a set of distinct categories we developed to inform our research. For the purposes of this project, tagging refers to labeling information by a matter of fact whereas coding refers to labeling information that requires some level of interpretation. A list of the exact data categories that emerged through our research is below. When made explicit, most of this data could be easily identified in the reporting sources. Coding, on the other hand, came into play when some interpretative gesture was needed such as identifying targeted communities. Each swastika incident was double tagged and coded to help ensure reliability of data. 

Data categories:

  • geographical location (city, state),
  • date (day, month, year),
  • reported source (national newspaper, local news, student newspaper, etc.),
  • reporting frames (vandalism, graffiti, symbol of hate, racist message, etc.),
  • accompanying text (“Trump,” “Heil Hitler,” etc.),
  • accompanying pictorial elements (iron cross, Star of David, etc.),
  • media (spray paint, marker, pinecones, etc.),
  • site of swastika incident (public space, local business, college, etc.),
  • specific location (dorm building, library, restaurant, etc.),
  • structure (bathroom stall, park bench, urban wall, automobile, etc.),
  • targeted individual/community (race, ethnicity, religious affiliation, etc.),
  • perpetrator (identified by gender and race/ethnicity),
  • community response (identified by actor (police, school administrator, parents, etc.) and rhetorical move (arrest, statement, vigil, etc.))
A Note about Tagging/Coding Difficulties

Coding especially came into play when we attempted to identify the targeted individual/community of each swastika incident. In most cases, some level of interpretation was necessary, as we rarely had access to a perpetrator’s intentions, if we were able to discover the perpetrator at all. We thus developed a coding scheme to identify targeted individuals/communities based on the local place in which the swastikas surfaced, the demographics of the neighborhood, and/or the accompanying text. In some instances, targeted communities of swastika incidents felt more easily identifiable, such as when a swastika suddenly appeared on the dorm door of a Jewish student or, as in the example above, on the home of a Muslim family. Other times, a swastika might surface on an urban wall with a racial slur such as the N-word. In such latter cases, we identified the target as Black Americans. Still in other cases, the phrase “White Power” surfaced alongside a swastika, an act we interpreted as a threat to unidentifiable non-white community members. In addition, we did not have access to targeted individual’s intersectional identities. Due to such coding difficulties, we urge those who take up this data to be aware of how this particular data point may be unverifiable. 

Another difficulty we faced is accounting for swastikas that surfaced during the time period of our study versus ones that may have existed in a particular place for a while and were suddenly noticed by a community member or reported for the first time by a local news source. Many swastikas, for instance, may have been worked into the design of certain architectural structure fifty years ago, but are just now gaining critical attention. This temporality issue could also be the case with swastikas emplaced in public spaces (such as tunnels) that might have gone unnoticed or commented upon until recently. Due to such reality, for the purposes of this project, swastika incidents should be understood as occasions in which swastikas are discovered in a particular community. 

Still yet another point to note about our data analysis deals with location. Undoubtedly, many swastikas are spreading across social media, often in hidden corners of the World Wide Web. Future research might work hard to track swastikas circulating across social media in the form of tweets, GIFs, memes, posts, and a range of other digital productions in a variety of contexts. For the purposes of this particular project, we wanted to specifically know in what geographical spaces and on-the-ground locales swastikas were circulating. We thus chose to only include swastikas either documented by social media (such as when someone takes a photo of a swastika on a playground and shared the photo on Twitter) or when a swastika surfaced in a zoom bomb at a community center or religious institution. Identification of where of swastika incidents took place should thus always be understood as a specific on-the-ground location with the recognition that during this time period swastikas were also circulating widely across the unfixed borders of the digital sphere.

Lastly, but also Important to note is that the maps, visualizations, and findings reported in this project are based on information present in the news sources listed in our database as well as, in some cases, sources sought out for missing information and cross-checking purposes. In some instances, these reporting sources may have been updated after we accessed them, and our findings may not reflected this updated information. In addition, not all information for categories listed above could be found for all incidents.

How can Data be Accessed?

Select information from our research is accessible in open-access CSV files stored in The Swastika Counter Project Archive hosted by OSF, the Open Science Framework. OSF is an open source project management tool maintained by the Center for Open Science that enables scholars to share and build upon each other’s archives.

In The Swastika Counter Project Archive, visitors can access two data sets. The first–titled “Swastika Counter Project Data Set_July 2023″–provides data about the location and dates of swastika incidents as well as data about with what words and images the swastika circulated, in what media they surfaced, and in what local sites they appeared, and on what structures. The second data set–titled “Source Data for Swastika Counter Project”–provides access to over 1300 web sources that documented swastika incidents that took place on the streets of the United States between January 1, 2016 and January 20, 2021. Note: We have done our best to provide active links to the sources, but for various reasons, some of these links may no longer be active. 

We invite scholars and other interested parties to use and extend this data set for their own research purposes.

How can you Cite this Project?

Authors: Laurie E. Gries and Kelly L. Wheeler

Title of Project: The Swastika Counter Project

Date of Publication: 2024

URL of The Swastika Counter Project Archive: https://osf.io/4fkpw/

URL of Data Advocacy Website: https://theswastikacounter.org

Cite this Website: Gries, Laurie E. and Kelly L. Wheeler. The Swastika Counter Project. 1 August 2023. https://theswastikacounter.org.

Who Contributed to this Project?

Research, Design, and Content Contributors

Laurie E. Gries: Project Design and Management; Data Collection and Analysis; Data Visualization Design; Web Design and Content

Kelly L. Wheeler: Data Analysis and Clean Up; Management of Data Visualizations and Educational Resources; Lesson Plan Development; Web Design and Content

Jacob Richman: Data Clean Up; Data Visualizations and Analysis

Jason Miller: Map Design and Implementation

Morteza Karimzadeh: Map Design

monét cooper, Mara Johnson, Heather Bruce, and Kelly L. Wheeler: Educational Lesson Plans

Rolf Norgaard and Technical Communication and Design Student Team, User Centered Study

Christopher Benson, Web Design

Advisory Board Members

Carolin Aronis, Assistant Professor, Ethnic Studies, Colorado State University. Co-Chair of Advisory Council on Jewish Inclusion at CSU.

Josh Berkenwald, Rabbi, Congregation Sinai in San Jose, California.

Melissa Borja, Assistant Professor, Asian/Pacific Islander American Studies Program, University of Michigan. Lead of The Virulent Hate Project.

Jay Dolmage, Professor of English, Waterloo University. Founding Editor of Canadian Journal of Disability Studies.

Ilana Cone Kennedy, Chief Operating Officer, Holocaust Center for Humanity, Seattle, Washington.

Keon Pettiway, Independent Researcher of Communication, Media, and Critical Design Studies. Product Design Manager at Celigo, Chapel Hill, NC.

Phaedra Pezzullo, Associate Professor of Communication, Rhetoric and Culture. University of Colorado Boulder.

Roopika Risam, Digital Humanist, Associate professor of Secondary and Higher Education and English, and the Faculty Fellow of Digital Library Initiatives at Salem State University. Member of xpmethod and team member of the Torn Apart/Separados Project.

Susan A. Youngblood, Associate Professor of Technical and Professional Communication, English Department, Auburn University. Past president of Congregation Beth Shalom, the Reform synagogue in Auburn, Alabama.

What are some Limitations and Affordances of this Project?

As with all data-driven projects, the data generated for the Swastika Counter Project is somewhat limited and skewed. As already made evident above, the target individuals/communities of swastika incidents are not completely verifiable, even as we attempted to ground our coding in material evidence. This data about perpetrators of swastika incidents is also limited as identities of perpetrator are most often not available. Also important to note is that this research does not include data from before or after Trump’s rise to presidential power between 2016 and 2021, so in and of itself, the database cannot demonstrate that incidents of antisemitism escalated during the Trump’s administration. Lastly, because this data is not inclusive of all swastikas incidents that occurred between January 2016 and January 2021, any generalizable claims about swastika incidents during this time period in the United States should be made with care.

Despite such limitations, we hope this research and project will prove useful to various stakeholders. From the open access database to the critical map and data visualizations to the general report and education resources, we especially hope that city officials, politicians, school administrators, educators, scholars, and activist organizations will use this data for their own unique purposes. We are happy to hear from any stakeholders and encourage contact via our email Theswastikacounter@admin.com.

Bibliographic Notes

To learn more about the method of iconographic tracking, see Gries, L. (2015). Still life with rhetoric: A new  materialist approach for visual rhetorics. Utah State UP.

Also, please note that some content on this website has been taken directly and/or adapted from an early publication about this project. See Gries, L. (2018). Swastika monitoring: Using digital research to track visual rhetorics of hate. In R. Wysocki and M. Sheridan (Eds.). Future matters. Logan, UT: Computers and Composition Digital Press-An Imprint of Utah State University Press.

Simplified Summary

About Our Data

Why Was This Data Collected?

In late 2016, during Donald Trump’s campaign for the presidency, reports of hate and bias incidents in the United States increased. Many organizations, including the Southern Poverty Law Center and ProPublica, began documenting these incidents due to a rise in threats related to antisemitism and white supremacy. The Swastika Counter Project started in early 2017 in response to this growing concern.

Dr. Laurie Gries, a visual rhetoric scholar, noticed a surge in swastika sightings across the U.S. after Trump’s inauguration. These incidents targeted various minoritized groups and often included violent threats in accompanying texts. With questions about where, how, and why swastikas appeared, the project aims to understand their impact on communities.

How Was This Data Collected?

Using a method called iconographic tracking, researchers searched for swastika incidents from January 2016 to January 2021. They collected data using Google’s search engines, finding over 1300 incidents. Researchers then spent five years analyzing reports from newspapers, magazines, social media, and other sources to understand the context of each incident.

A Note about the Ethics of Our Data Collection Process

How Was Data Analyzed?

Data analysis involved tagging and coding incidents in a Google Spreadsheet. Tags labeled factual information, while codes interpreted data. Each incident was tagged and coded twice to ensure accuracy. Categories included location, date, reported source, accompanying text, media used, targeted individual/community, perpetrator details, and community response.

How Can Data be Accessed?

Two datasets are available in The Swastika Counter Project Archive. One provides details about swastika incidents, while the other contains links to over 1300 sources documenting these incidents. The project encourages scholars and interested parties to use this data for research purposes.

How Can You Cite This Project?

Authors: Laurie E. Gries and Kelly L. Wheeler

Title: The Swastika Counter Project

Date of Publication: 2024

URL: https://theswastikacounter.org

Who Contributed to This Project?

Research and web content contributors included Laurie E. Gries, Kelly L. Wheeler, Jacob Richman, and others. Advisory board members provided guidance and expertise.

What are Some Limitations and Affordances of This Project?

The data is limited and may not fully represent all swastika incidents. It does not cover incidents before Trump’s presidency, and perpetrator identities are often unknown. Despite these limitations, the project aims to be useful for various stakeholders, including city officials, educators, and activists.

Bibliographic Notes

For more information about iconographic tracking, refer to Gries, Laurie E. (2015). Still life with Rhetoric: A New Materialist approach for Visual Rhetorics. Utah State UP. Some content on the website is adapted from an earlier publication by Gries (2018).