News databases: Diversity without equity or inclusion

The Problem

Back in September, 2020, I sent out a call for help across AISL and other school librarian-oriented lists in hopes of finding databases that provide “diverse, inclusive, and equitable access to perspectives mirroring the composition of our country in magazines, historical newspapers, and contemporary news.” Generally, database companies sell “core” collections that are positioned as “high quality sources,” comprised almost entirely of white-perspective news outlets. Then they up-sell from a menu of discrete “ethnic” packages to provide “alternate perspectives.” Students deserve better.

Thank you to the many folks who responded hoping to hear of a good database in which to invest. Sadly, the answer is…so far I’ve found no way to buy this unicorn of the database world. Ultimately, I started doing my own diversity audit of our databases and others on the market to try to better articulate the nature of the problem.

I am currently only part way through this process. First semester ended up (happily) being much more crowded with instruction than I had anticipated. Even the terminology I use to think about this set of issues is still in crude form. Here is an update on what I have learned so far, however. To date I have focused on US news, historical and contemporary, and have only been able to compare offerings from two companies. This work has served — at the very least — as evidence that the problem is real and pressing.

Inclusion

In the fall, I had not yet fully realized the insidious nature of the juxtaposition we often attribute to databases: quality sources vs. alternative perspectives. I’ve been sitting with this formulation increasingly in the intervening months, and contemplating how our professional narrative around databases is driven by the marketing efforts of the database companies themselves. Consider the act of marketing a database as “providing researchers access to essential, often overlooked perspectives” that exists because the perspectives have been intentionally overlooked and isolated to sell us another database. So how much does the title list of an intentionally curated “ethnic” database (which mysteriously includes the LGBTQ+ collection, by the way) overlap with a product intended for high school?

ProQuest: Compared title lists for Research Library Prep and Ethnic Newswatch databases.

Please note that the “Overlap (%)” column conveys how many of the “specialized” Ethnic Newswatch titles also appear in “general” Research Library Prep. It does not convey the percentage of Research Library Prep that are/are not white perspective — those numbers would apparently be vanishingly small. 

An issue that struck me immediately as I got started was that scholarly journals comprise, by far, the largest mass of content in Ethnic Newswatch that is also available in Research Library Prep. These sources differ distinctly from newspapers or popular magazines; academic discourse may well be quite removed from the community it studies. That is, a large percentage of the authorial and editorial work is carried out within a realm of authority modeled on European institutions and constructed in our academic halls of privilege. To put it plainly: the perspectives appearing in the University of Pennsylvania Press’ Hispanic Review may not reflect community voices in the same way that those appearing in La Prensa Texas newspaper do. Both source types provide important points of view; their creation does not serve the same purpose.

Important as it is to have a diversity of voices in our scholarly works, they provide fundamentally different types of evidence from newspapers. Not to mention, they are not accessible to most K-12 students.

Gale: General OneFile, In Context: High School, OneFile: High School Edition, OneFile: News, In Context: US History

It has been challenging to figure out how to do a diversity audit, but since many database companies seem to start monetizing diversity with Black American newspapers, I decided to work from lists of existing and historical Black papers, including: National Newspaper Publishers Association Current Members and Princeton University Library – African-American Newspapers (1829-present). I used titles from these lists to search within Gale’s title lists, and found:

Checking Black American newspaper titles against Gale title lists yielded vanishingly few overlaps.

In the process of looking at the titles that are listed, the Atlanta Daily World and the Chicago Defender — historically both very important publications in the 20th Century United States — only had coverage from 2014-present, with exceptions from 2015/16 to the present. Meaning, in fact, they only have a handful of issues of each paper.

Once again, these databases provide news sources that almost entirely reside within historically white readerships.

Equity

In another sense, it does not functionally matter if a database includes sources from diverse sets of communities. When the algorithm privileges white perspective publications, searchers may never encounter other points of view.

Spot checks of ProQuest’s ranking of newspaper results in Research Library Prep confirmed that their methods for ranking heavily favored specific titles. Specifically, the New York Times dominated results, with a smattering of hits from the Chicago Tribune, Washington Post, US Fed News Service and Targeted News Service

I ran a series of searches, noting how many unique titles were returned for each search, as well as those titles’ spreads across the top 100 results. In essence, how many pages would I need to scroll through to access more than a few titles? I searched for [ the ] — as a word that appears universally in English-language newspapers — and also for words like [ miami ], [ skagit ], and [governor] — each of which strongly suggest local news. In every case, the results looked something like the results for [ the ]:

  • Returned 81 unique titles
  • Top 100 results:
    • 95 results from the NYT
    • Other titles ranked: #37, 41, 71, 91, 95
    • Other publications in the top 100 results: Chicago Tribune, Washington Post, Targeted News Service 

Progress!

However, there is good news. A Gale sales rep who is on one or more school library lists began wondering about this issue themselves, and carried out an independent audit that they then presented to their acquisitions department. As a result, when I last checked in this past November, Gale publisher relations personnel have identified:

  • Licensed periodicals where the issues aren’t current
    • Updates are in various states of progress 
  • Important periodicals with lapsed agreements 
    • Updates are in various states of progress  
  • Over 140 new periodicals from the following communities:  “African Americans, Arab Americans, Hispanic Americans/ Latinx, Native Americans, Ability Diverse, LGBTQ+, Women, & more”
    • Requests have been sent to publisher relations to pursue license agreements

Though not within the scope of my current work, Gale has also taken a look at their reference overviews and biographies and have made efforts to offer more coverage, as well.

The solution?

Does this issue interest you? Would you like to join me in fighting for single databases that are diverse, equitable, and inclusive? Whether you would like to audit a database you have, suggest a consistent method for auditing, share findings at your state conference, or talk to your database companies once we have a clear report — kindly reach out. If the idea is that we are better together, let’s unite and make a difference!

I am deeply grateful to my director, Jole Seroff, for being so invested in and supportive of this exploration, and colleague Sara Kelley-Mudie for helping me focus my thinking.

3 thoughts on “News databases: Diversity without equity or inclusion

  1. Your research is inspiring and impressive. I surely hope AISL can rally and write to some of these companies asking for more equity and inclusion.

  2. This is remarkably insightful – thank you for identifying this and calling it what it is, and I am especially heartened to know that there may soon be at least some kind of solution. Here’s to a better tomorrow, Tasha!

  3. Gale is starting to audit their collection and address the disparity that is a great beginning. It would be good to have a consistent method for auditing databases. A method the databases could apply to their own content. Thanks for increasing our awareness of the need for Equity, Diversity, and Inclusion in databases, and for beginning the process of auditing these research tools.

Leave a Reply to Alyssa Mandel Cancel reply

Your email address will not be published. Required fields are marked *