Research Insights: Using the [email protected] Instagram Archive to Improve Schools
Nicole Brittingham Furlonge
and Kenny Graves
This article appeared as “The Art of Listening” in the Spring 2022 issue of Independent School.
“I remember one day I went to an administrator, yet again, about something racist a student comfortably said to me in front of my whole advisory. The administrator sat me down and told me over and over how I had ‘thin skin.’ ” —A post from one of the many [email protected] Instagram accounts
This anecdote is one of thousands documenting Black and intersectional experiences at independent schools that have been shared on Instagram over the past two years. This anecdote stands out not because it is unique, but because it is indicative of a larger narrative that has emerged in these spaces over time. The [email protected] archive, however, effectively shifts the conversation concerning these experiences, providing a public social media space through which to share these matters beyond one’s individual or school context. These posts resist the idea that these incidents happen at other people’s schools or in other regions of the country. Such incidents not only happen there. They happen here.
Shortly after the emergence of the first [email protected] Instagram accounts in the summer of 2020, in which Black, Indigenous, People of Color (BIPOC) students, alumni, and faculty/staff shared stories of their experiences in predominantly white institutions and laid bare the racism, exclusion, and trauma that they had suffered in independent schools, we saw the need to help schools understand the magnitude of this information and learn how to really listen to it. So in July of 2020, the Klingenstein Center, Teachers College, Columbia University convened a team of research faculty, graduate students, and alumni, and launched a research initiative to collect and analyze the [email protected] public archive.
This initiative aligns with the center’s commitment to design, conduct, translate, and share research that leads to equitable and sustainable structures, practices, and cultures in independent and international schools. At the core of our research is accessible collaboration with schools and the people in them. Such generative collaboration aims to build capacity in schools as they work in evidence-informed ways to create environments in which all students can learn and thrive and are embraced in their fullest humanity.
With this initiative in particular, we wanted to shift the question on so many school leaders’ minds from “What do we do with this information?” or “How do we respond to this information?” to “How do we listen for opportunities for intervention and growth toward diversity, equity, inclusion, justice, and belonging (DEIJB)?” We wondered: How might the catalog of stories shared in these posts—across schools, regions, and the country—allow educators and school leaders to make more informed decisions about how we approach DEIJB work in schools?
Post by Post
Much like a classic qualitative investigation that might use interviews or focus groups to gather stories, we sought to use social media to listen to the collective voices in the [email protected] archive. Nicole Brittingham Furlonge describes this as a precise kind of “listening in print,” a term she coined in her 2018 book Race Sounds: The Art of Listening in African American Literature, to describe the ways in which some African American literature reminds us that print is not mute. It exists on the page both to scan and to tune into.
In the case of the [email protected] archive, listening in print—reading these posts to give audience to the people speaking in them—is one way educators and leaders might attend to the racism, exclusion, and trauma expressed in these posts and lived in their institutions’ histories, structures, cultures, and practices. How might we understand our schools differently, particularly in terms of DEIJB work, if we begin by listening?
In our study, we use automated text mining, an emerging form of analysis in educational research. Text mining allows researchers to synthesize a large amount of qualitative data in more efficient and comprehensive ways. First, we gathered all data from [email protected] Instagram accounts and created a database of entries and text. We then used the programming language Python to build a program that automatically groups common words and phrases into latent themes. The resulting output is a list of keywords and common phrases found throughout the larger corpus of data.
Over two years, we've solicited 12 graduate students in our full-year and two-summer master's programs to help us identify [email protected] accounts, collect and code posts, and research some background information. We have collected data from 253 [email protected] accounts involving schools or local or regional school collectives from more than 27 states. By identifying the most common keywords and associated words that occur in the posts, we can empirically uncover topics across different social media accounts. Some topics that emerged from our initial analysis include emotion, school (as an institution), relationships, words (from a teacher or peer), interpersonal exchanges, and classes (as in curriculum).
Pilot: NYC Schools
After collecting the data from 253 accounts, we focused our analysis on data by location. Our initial analysis focuses on New York City, home to the Klingenstein Center and to a robust and diverse micro-ecosystem of independent schools. This analysis examines more than 900 posts involving 11 New York City-area independent schools. Using the text mining tools and analysis, we grouped the experiences shared in these posts into the following categories: (1) racially charged incidents in the classroom, often involving curriculum and pedagogy; (2) interpersonal incidents outside of the classroom; and (3) students of color trying to share racist incidents only to be ignored or told by faculty, staff, and/or fellow students that they were being too sensitive or blowing things out of proportion.
Some of the posts share concerns about how institutional power is conferred, particularly in relation to DEIJB directorships. As one post asserts, “we would also discourage making the new DEI director read through the experiences and try to fix it. This is a community effort. These issues existed long before [named leader’s] entrance. Take ownership in the culture the community has created and participated in.”
Another post insists that the DEI director role be positioned in the organization at the same level as principals and vice principals and that the director report directly to the head of school: “If the school is committed to social justice and then elevates this position on par to the other positions of power in the school, it will show [school] is serious.”
The [email protected] archive and our research compels school leaders to develop institutional ways of thinking more expansively about what we count as data and how we use it to practice evidence-informed leadership in all facets of our work. Through this particular initiative, we continue to explore what it can sound like to maintain steady momentum when working to effect such change. Ultimately, our goal is to create a framework that describes steps school leaders can take to help disrupt racist and otherwise oppressive behavior, practices, and systems.
The Klingenstein Center is collaborating with schools to help them learn how to conduct their own local inquiry into their [email protected] accounts to uncover systemic issues extending throughout the ecosystem of their schools and schooling. Schools can use text mining and listening strategies not only to understand their individual school’s [email protected] and future social media posts but also to continue social and other forms of institutional listening beyond Instagram. Data use that also leverages collaborative inquiry among educators before, during, and after analysis can help facilitate additional meaningful discovery and action. Consider these questions when examining data in your school:
- What biases do I bring to the table?
- In what ways do we take responsibility for examining the data rather than focusing on our assumptions?
- What do I notice? (descriptions of the data)
- What do I wonder? (inferences about the data)
- What trends or patterns of behavior do I notice? (interpreting the data)
DEIJB is adaptive change work that takes time and perpetual community and institutional commitment. Based on what we’ve discovered, we recommend schools consider implementing the following adaptive changes to address systemic issues: establishing anonymous reporting systems for bias incidents, similar to Title IX for universities; reporting employment and enrollment diversity numbers more accurately; investing in multicultural psychological services for students and adults; and creating systems and structures that directly position the work of DEIJB in a collaborative relationship with the school’s teaching, student learning, and professional learning efforts. We are already working with schools to implement these changes.
- What does the data say about our ability to see and hear our school?
One more focused yet effective way is to review documents such as the narrative feedback comments teachers write for students and families at various times during the school year. What do we notice about the language we use to discuss students? What language patterns—and therefore ways of thinking or assumptions—emerge? Conducting a textual analysis of narrative reports could provide a baseline for conversations about how we talk about student progress and learning. It also can help reveal what patterns we fall into when speaking about certain students.
We are in the initial stages of this research initiative and sharing our emerging findings with schools. We are also introducing the text mining tools we are using. This is an opportunity for school leaders to understand how to take swift, resolute action in response to racism while simultaneously engaging in a more complex, sustained process of making DEIJB principles central and enduring in their schools, for their students. This is how we can shift how schools imagine, teach, and lead now and for the future.
The authors of this article facilitated the 2022 NAIS Annual Conference workshop, “Listening for Data: [email protected] Schools Instagram Text Mining and Evidence-Informed DEI Work in Independent Schools,” in which they shared the text mining code and other resources to help schools expand their capacity for evidence-informed leadership. For more information, go to klingensteincenter.org/research.