Counting the Collection: Conducting a Diversity Audit of Adult Biographies

Diversity audits assess representation in library materials, but most examples focus on children’s books. One librarian shares her methods and challenges in tackling adult biographies.

Diversity audits assess representation in library materials, but most examples focus on children’s books. One librarian shares her methods and challenges in tackling adult biographies.

Creating an adult nonfiction collection that includes representation of people from all communities ensures that libraries are being conscientious and inclusive, in providing materials and resources that value the distinctive characteristics, experiences, and perspectives of all people. To recognize and embrace diversity, libraries need to offer information resources that benefit and reflect the lived experiences, ideas, and cultures of every community member, including all marginalized populations.

Maintaining a multicultural collection, rich in Own Voices titles, is important whether or not the groups represented are present in the local community. Corinne Duyvis coined the term “Own Voices” to describe an author from a underrepresented group writing from their own perspective or about their own experiences, as opposed to an author from an outside perspective writing about a marginalized people.

A key benefit to having Own Voices titles in a collection is that such representation can mirror the lived experiences of marginalized readers. They include cultural authenticity and character identity that capture nuances of underrepresented traditions, which allows members of the group portrayed to see the rich variation of their experience valued, and the achievements of members of their community celebrated—achievements which have too often been elided or erased. Offering books about the lives and histories of people from marginalized groups is also important in communities where the particular identity depicted might not be present. Such works are crucial not only for allowing community members of color, for instance, to see themselves accurately represented, but for building awareness about the experiences of marginalized people among readers from other backgrounds. Learning about BIPOC, neurodivergent, unhoused, LGBTQIA+, disabled, or other marginalized populations by reading their stories gives non-marginalized readers accurate representations, helping them to break down biases and become more culturally competent.

For the sake of all patrons, then, it’s crucial to evaluate holdings to discover if there are collection gaps in which marginalized populations are misrepresented, underrepresented, or not represented.

Conducting a diversity audit of library collections is not a new concept. It requires evaluation of the authors’ authority to tell the story, as well as the authenticity of the story being told. Karen Jensen, in her School Library Journal article “Diversity Auditing 101: How To Evaluate Your Collection,” provides useful resources to get started with a diversity audit. Following her guidance, we decided to start by auditing BIPOC representation in the adult biography section (which our library calls the “Lives” collection). Our adult nonfiction is arranged in eight broad categories called “glades” to encourage browsing within interest groups; “Lives” contains around 4,500 biography titles. Books are organized in Dewey order within each glade. Biographies seemed to be the best section to identify books reflecting the diverse experiences, perspectives, backgrounds, or beliefs of all people. Tackling this project was a bit daunting at first, but exciting at the same time. We offer our experience as an example because adult nonfiction diversity audits are not well documented.


Mapping the process included looking at what percentage of the current collection represents marginalized populations, as well as determining a minimum target percentage of the diversity we hope to reflect. Using U.S. Census, CDC, and local data, we set the minimum target percentage for each marginalized population group to be midway between the local and national population statistics—though even these numbers may represent an undercount of some BIPOC communities, particularly those who are undocumented. The mapping also included running reports for each glade in our collection that would enable us to identify books in the catalog by and/or about BIPOC people, history, or experiences.

After collating the reports by broad subject headings, we looked at the print books identified to assess the level and quality of information about BIPOC subjects and whether they provide a positive or negative representation. During this step, we needed to physically look at the books in each glade and identify those on the shelf that were not listed in our reports. Analyzing the reports and our findings showed us which marginalized populations and experiences were not sufficiently represented. Adding tags to catalog records to improve discovery for patrons and staff was the next step. Finally, we evaluated and measured the impact and value of auditing the collection.



The controlled vocabulary of the Library of Congress Subject Headings (LCSH), as well as limited author characteristics and information in bibliographic records of books, make it more challenging determine whether a library collection reflects voices and experiences from a diversity of backgrounds. Depending on how granularly a library plans to assess the representation in their collection, a range of subject headings can be selected.

Cornell University Libraries have identified some subject headings useful for auditing diversity: “Feminism”; “Gender”; “Sexuality”; “Race”; “Class”; “Disability/Ability.” Using all or some of these headings will help identify what books are in the library, where gaps exist, and opportunities to make the collection more inclusive and representative. To keep our task manageable, we focused on one topic at a time, with race as the first subject of our audit.

Identifying race in existing titles sounds fairly simple, but deciding how to evaluate a nonfiction collection has its challenges: while nonfiction is definitionally based on fact, the author’s subjective interpretation cannot help but be present, so bias is highly possible. Therefore, looking at biographies and accounts of real people seemed like a sensible place to begin. The next challenge was recognizing that the LCSH terms are not consistent enough to capture all relevant subject areas through catalog searches. Some searchable terms might be missed in an audit, while other books that are about race might not be identified, perhaps due to inconsistent labeling in the subject heading field. For example, My Korean Deli: Risking It All for a Convenience Store, by Ben Ryder Howe, is one of three books in our entire adult nonfiction collection that has “Korean Americans” in the subject heading but not “Asian Americans.” To aid in findability, both subject headings should ideally be included. Compare that to Minor Feelings, by Cathy Park Hong, which has “Asian Americans” listed in the subject heading, but does not include the specific subject heading “Korean Americans” despite the fact that the work is centered in Hong’s perspective as a Korean American person. Moreover, some of the LCSH terms can themselves be problematic or outdated. For instance, the Asian continent comprises nearly 50 countries, with more than 60 percent of the world’s population and a wide range of ethnic identities; using the LCSH term “Asian Americans” as a racial category can be overgeneral. Similarly, the LCSH term “Minorities,” mentioned below, not only centers whiteness but is increasingly inaccurate in many locales.

There are also titles that might not be identified in a catalog search because of they don’t have race-specific subject heading terms—for instance, Hana Ali’s At Home with Muhammad Ali. This book has “Boxers” as the subject heading field in line 650 of the MARC record, and no mention of race, even though other books about Ali do have “African Americans” in the subject headings.

Elizabeth Hobart, in her College and Research Libraries News article “Antiracism in the Catalog: An Analysis of Records,” writes that catalogs are often unable to support discoverability of the wealth of backgrounds, cultures, or abilities represented in book titles, due to the information that is captured (or isn’t captured) in bibliographic records and subject headings. Nor do LCSH terms identify the author’s race or their other identities.

We looked at existing subject heading terms in our bibliographic records and identified the most commonly used, then chose the below terms to audit (all these terms are from the Library of Congress database of subject headings):

  • “African Americans”; “Blacks”
  • “Asian Americans”; “Chinese Americans”; “Japanese Americans”; “East Indian Americans”
  • “Indians”; “Indians of North America”; “American Indians”; “Native Americans”
  • “Hispanic Americans”; “Latin Americans”; “Spanish Americans (Latin America)”
  • “Racially mixed people”; “Racially mixed families”
  • “Minorities” [LOC’s catch-all term for non-white or marginalized racial groups]


Once we’d compiled reports on each of the above subject heading terms, the full list was used to compare with the physical books on the shelf. In evaluating the credibility and authenticity of the collection, our goal was to consider the following criteria for each book identified on the list:

  • Is it an Own Voices title, written by a person from the same marginalized identity? We determined this by reading the biography of the author on the cover of the book, checking Goodreads, or doing a Google search on the author. Authors tend to reveal more about themselves in biography books than in other nonfiction sections, which makes this easier. If it’s not Own Voices, does the author have some credibility to write about the population? (For instance, a civil rights historian who specializes in the subject of their book.)
  • Is it a “single story”? Chimamanda Ngozi Adichie warned in her 2009 TEDGlobal talk that a single story “shows people as one thing, as one thing only.” If it is a single story, are there other books in this section of the collection that are written from a different perspective?
  • Does the author show prejudice? Does the book reflect partiality or preference for or against a person or an idea? Does the author use so-called “benevolent” stereotyping (where ostensibly positive traits are attributed to all or most members of a group)?
  • Is this a positive or a negative representation? Do we want it on our shelf to reflect a diverse population? Not all representation is necessarily good representation.

We needed to evaluate all the books in the collection that met our criteria, even if they weren’t listed in the reports. To accomplish this, books with race-related topics that were found on the shelf, but not listed in the reports, were captured with a simple Diversity Audit Shelf Check form that we made in Office 365. Using a mobile device, staff members could capture the details of any book that should have been included on the report but was not identified because of inconsistent subject headings in the catalog. The details recorded include the title, author, barcode, call number, and category of the book.

After collaborating with local and state librarians to determine the most effective way make these books discoverable, we decided to add a set of internal subject headings (listed below) to bibliographic records for any new book added to our adult nonfiction collection. We also added these subject headings to those books that had slipped through the parameters we picked to generate our reports. This update of our bibliographic records will enable us to run future diversity audits more accurately and ensure we are meeting our objectives in marketing and promoting new titles added to the collection. More importantly, the subject headings will increase discovery of these books through our catalog and encourage their circulation in the community, thereby increasing the audit’s impact. The categories are broad and will be used only internally, for auditing, until LCSH terms are consistent and accurate enough to capture all works by authors from marginalized populations and/or about relevant subjects. These are the categories we’ve added to the subject headings in line 650 of the MARC records: “Diversity Race”; “Diversity Gender”; “Diversity Sexual Orientation”; “Diversity Ability”; “Diversity Religion”; and “Diversity Ethnicity.” These categories will be allocated in the NOTES field for every relevant book purchased through our print book vendor, which will transfer the NOTES information to the line 650 subject heading in the book’s bibliographic record. The headings are added when we order the book, and library staff use a book’s description and reviews to determine what subject headings to add to each record. We will only add this category for new books going forward; we won’t change the records of books currently in our collection. That could be another project, if needed.

It is important to note that our decision to use these six categories is not without drawbacks. Rachel Ivy Clarke and Sayward Schoonmaker, in their Advances in Classification Research Online article “The Critical Catalog: Giving Voice to Diverse Library Materials through Provocative Design,” cite several good arguments that prove it can be problematic to add more bibliographic data about the identities of authors: the limits of controlled vocabulary can affect the creator’s agency, and it also risks outing an author who might be discriminated against or attacked for their identity. Clarke and Schoonmaker suggest instead to tag all books by or about white, male, cis people—an idea to seriously consider in the future for any libraries engaging in this work. This isn’t currently feasible for us, as it would incur a considerable change to our bibliographic records. For now, therefore, we are using these categories to facilitate discovery within our collection and enable us to measure the impact of our audit.



Once the reports were compared to the books on the shelves, it was time to analyze the results. We initially found that we met our lowest target percentage goal for representation of some population groups. In some cases, we exceeded those percentage goals—for instance, in titles with African American subject headings in our “Classics” glade: these were 12.4 percent of the total “Classics” collection (see graph). However, on closer inspection, the representation was skewed. Our collection has large clusters of books by or about, for example, some particularly widely known Black Americans, like Maya Angelou, Toni Morrison, or Barack Obama, but it lacks diversity in lived experiences or lesser-known authors. There were many “single stories” in this section, and we became aware of the need to find more related books to share shelf space with these titles. We also identified books that we needed to weed out—books that included biased or negative representations of marginalized populations, or did not add value to our community.


The next step to ensure this process is effective, and adds value to the collection and the library, is to gather data to measure and determine the impact of the audit. By adding the new subject headings, we will be able to run reports and count item circulation statistics. By increasing discoverability, we will be more able to promote these books in online book lists, in social media posts, in our displays, or in staff reading lists. Our vendors have the capacity to send us pre-populated carts with highly recommended books about underrespresented groups. We continue to consult resources like NovelistPlus and GoodReads, and to look at book lists, book awards, and publisher reviews to identify new authors from marginalized populations, as well as new titles reflecting perspectives or experiences that may be underrepresented in the collection.

Keeping staff abreast of equity, diversity, and inclusion issues ensures that they will be able to promote books about underrepresented populations through recommendations and book talks. We can match our collection’s diversity with our programming, like Dr. Simran Jeet Singh’s lecture, delivered at Darien Library, “Representation Matters: What Being Seen Means to the Marginalized”; this helps us be an institution that promotes “interest, information, and enlightenment” to all people in our community, as set out in the American Library Association’s Library Bill of Rights. Finally, our Collection Development Policy must be revisited and updated to ensure that we maintain robust and integrous representation of all people.



We predict that our audit will take a year to complete. We have found this work to be time-consuming and challenging, but we need to focus on its desired outcomes: continually working towards a more inclusive adult nonfiction collection that includes, reflects, and represents all populations. The Library of Congress is gradually changing its subject heading terms to be more inclusive and to reflect equity, diversity, and inclusion initiatives, but libraries need not wait on those efforts to finish before they make all titles discoverable. Libraries can manipulate their own metadata and bibliographic records to improve access and measure increased representation right now. Taking time to update the catalog will improve discoverability, especially for titles that do not often appear in popular reviews or publications.

As Hobart aptly wrote in “Antiracism in the Catalog,” updating the catalog, to consider race, ability, sexual orientation, ethnicity, religion, gender, and other characteristics, will help patrons find the resources they want and will make “the catalog stronger”: an example of how improved accessibility truly benefits everyone! It would be easy to find excuses to avoid this call to action, or to promote diversity only in small displays at certain times of the year, but striving for anything less than a fully inclusive collection wouldn’t be conducive to the role of libraries and librarians: we are information providers, dedicated to welcoming everyone and providing equitable access to resources for the communities we serve.

Colleen Wood has been a Knowledge and Learning Services Librarian at Darien Library, CT, for three years, and serves on its equity, diversity, and inclusion committee. She is a facilitator for LJ’s Equity in Action workshops and will be an instructor for the MLIS program at the University of Maryland this fall. She is co-chair of the American Library Association’s International Relations Round Table Sister Libraries committee.

Comment Policy:
  • Be respectful, and do not attack the author, people mentioned in the article, or other commenters. Take on the idea, not the messenger.
  • Don't use obscene, profane, or vulgar language.
  • Stay on point. Comments that stray from the topic at hand may be deleted.
  • Comments may be republished in print, online, or other forms of media.
  • If you see something objectionable, please let us know. Once a comment has been flagged, a staff member will investigate.



We are currently offering this content for free. Sign up now to activate your personal profile, where you can save articles for future viewing