In the interest of remaining accountable to the values statement for this book, the authors tasked me with performing an audit of all the individuals, projects, and organizations referenced in Data Feminism. Quantifying these references provided important information about which perspectives were being included in the work and to what extent. At the same time, this process presented difficult-to-answer questions about identification and classification. Therefore, this methods statement will serve to explain how we approached those questions and what our answers were.
First, we had to decide what would constitute a “reference” that needed to be recorded in the audit. Every individual mentioned by name was counted as a single reference, as was every project (i.e., Kiln’s Ship Map). Corporations were mostly excluded from the audit unless they played an active role in an example and were mentioned more than once. For instance, Target is included because its pregnancy-targeted marketing strategy is used as a major example in chapter 1. On the other hand, Instagram is not included despite being mentioned multiple times in the Serena Williams example because in that case the social medium is merely the site where Williams and her fans exchanged stories of their birth experiences.
Each categorization then required its own investigation. For individuals, we attempted to verify their race, gender, country of origin (which fed into our Global South/Global North distinction), and indigeneity. Additional categorizations included whether or not references were from within the academy and if they represented an example of good data practices or “what not to do.” References were logged as “communal” if they were community-driven (e.g., Data for Black Lives), and a separate category recorded whether the reference provided a “nonvisual example” of data work or not. Each record was further classified by importance. One reference constituted “passing” importance; two to four references, “more than once;” and beyond that, “central.”
Although categories like “importance” and “nonvisual example” were straightforward to assign, others like race and gender were not. We tried to confirm these categorizations through online research, but without the self-identification of the referenced individuals, these categories are obviously subject to further inquiry. It should be noted that only individuals were counted by race or gender, unless an example hinged on a racialized or gendered assertion. For example, General Motors is discussed in relation to its discriminatory treatment of Emma DeGraffenreid, which became the impetus for legal scholar Kimberlé Crenshaw’s study of “intersectionality.” Discriminatory hiring on the basis of race and gender is the reason for General Motors’s inclusion in the book; therefore the company is listed as “white” and “man” in the audit. Also, if upon further research an organization was found to have an all-white board or staff, like Clarksons Research UK, it was categorized as “white.”
Future attempts to replicate this audit should take seriously the difficulty of clearly establishing these identity categories without formally consulting with those who are being referenced and therefore classified. Some classifications—such as gender—may not be possible to define, as individuals may choose not to publicly disclose their status as trans or nonbinary to avoid discrimination or simply because they consider it private information. Also, we would like to acknowledge that categorizing references by their social identities and structural affiliations does not inherently guarantee a representative discourse. However, we consider this an important effort to remain accountable to the values that brought forth this entire project, which include intersectional and antihierarchical thought, the honoring of a multiplicity of viewpoints, and a commitment to acknowledging our own positions and limitations.
Isabel Carter is a multimedia reporter who received their master of arts in journalism at Emerson College in 2019.