The Open Notebook strives to serve journalists who cover science by fostering an inclusive and welcoming global community. To do that well, we recognize the importance of embodying the values of diversity, equity, and inclusion (DEI) in all our work, including the makeup of our editorial team and board of directors, the training and mentoring programs we offer, the topics we cover in our vast library of articles on the craft of science journalism, and the policies and practices that underlie our work. One important part of our effort to support and expand DEI across all aspects of TON is ensuring that people from a diverse range of backgrounds and experiences are represented as sources in the articles we publish.
We began systematically tracking the diversity of sources in our stories in January 2022 and have been doing so contemporaneously since then. That same year, we also began the process of conducting an historical audit of the sources who appeared in our stories over the three previous years. This report includes an analysis of the data we have collected for stories published at The Open Notebook from 2019 through 2023.
The data below reflect the initial results of this ongoing process. By sharing this information publicly, we aim to be transparent about our efforts to expand the diversity of sources represented in our stories and our goals for continued improvement. Our intention is to share updated source-tracking data on a regular basis (normally every two years).
Data Collection and Analysis
We began testing a method for collecting source demographic data in 2021, with the help of journalist Carolyn Wilke. (The system we settled on is described in the sample script and survey that we published in 2022, alongside The Open Notebook’s public guide to tracking source diversity.)
We decided to collect data on the following dimensions of diversity:
- Geographic location (country)
- Race
- Gender
- LGBTQ+ identity
- Disability
Of course, these are only a few of many different aspects of diversity that we could choose to track; in choosing these dimensions, we sought to balance completeness and practicality. It is possible to alter or add to the data we collect in the future.
We began tracking source demographics for all stories in January 2022. That process includes the following elements:
- Story authors invite sources to consider providing demographic information, either by email or phone.
- The boilerplate language we use in this invitation (see the sample script mentioned above) includes an explanation of our reasons for wanting these data. It also emphasizes that doing so is optional and that data will be kept confidential and stored securely.
- Story authors enter whatever information sources choose to provide into our secure form, along with their own name, the story title and type (e.g., reported feature, interview, etc.), and the year of publication. (For stories that include only one source, such as interviews and “A Day in the Life” profiles, authors do not provide the story title, since doing so would mean including the name of that source in the database.)
- Privacy and security considerations:
- We do not collect sources’ names.
- Story authors do not have access to the back end of the source-tracking database; this means that they cannot see any data other than those they enter.
- The only person who has access to the database is the executive director and editor-in-chief of The Open Notebook, Siri Carpenter, and the account is protected by two-factor authentication. As a further privacy measure, Carpenter does not have access to information about which individual sources correspond to which records in the database.
In addition to our contemporaneous source-tracking, Wilke, along with Allison Whitten, also assisted in launching an historical source-diversity audit of sources in TON stories published in 2019 through 2021. The purpose of conducting this three-year historical audit was to give us a more meaningful baseline from which we could set goals.
The historical data-collection process included contacting (and sending up to two follow-up emails, as needed) to about 450 individuals in our 2019-2021 story cohort, inviting them to participate in our audit. Replies went to a secure email address, and only Carpenter had access to this email address; after Carpenter entered data into the database, the original response emails were deleted so that they would not be obtainable if the account were to be hacked. The same data-protection measures described above were also used for the historical audit.
A redacted version of the data file (with story authors and titles omitted, to further protect sources’ privacy) was provided to TON’s program assistant Sarah Luft, who analyzed the following kinds of data in aggregate (i.e., across all five years); by year (2019-2023); and by story type (e.g., reported features, interviews, etc.).
The data below comprise 688 individuals included in stories published at The Open Notebook between January 1, 2019 and December 31, 2023. Collectively, these individuals were quoted or paraphrased in 103 reported feature articles, 49 “story-behind-the-story” interviews, 6 “Storygram” annotations, and 29 “A Day in the Life” stories. Below, you’ll find breakdowns of data by demographic category and year. Though we also analyzed data by the story types listed here, there were no significant differences in the patterns observed across story types.
One caveat in interpreting the data: Our response rate for the historical audit, 2019-2021, is about 64%, whereas our response rate for source demographics collected contemporaneously during 2022 and 2023 is about 93%. Considering this disparity, we should interpret the year-over-year trends, as well as the data from 2019-2021, with caution.
Geographic Location
By Global Region
To date, people based in the Global North have made up the majority of TON’s sources. From 2019 through 2023, people based in the Global South made up about one fifth of total sources. For the purposes of this reporting, Global North is defined as Europe, parts of Oceania (Australia and New Zealand), parts of Asia (for example, Israel and Japan), and parts of North America (Canada and the U.S.), while Global South is defined as Africa, parts of Asia, South America, and parts of North America (Mexico and Central America).
Over the last few years, the percentage of TON sources based in the Global South has trended upwards, from a low of roughly 6% in 2020 to about 30% in 2023.
By Continent
Since 2019, people based in North America have made up most of TON’s sources, just shy of 80%. People based in Africa, Asia, Europe, Oceania, and South American have collectively made up the additional 20% of sources. People based in Oceania have typically made up the lowest percentage of TON’s sources.
The percentage of sources from North America has been trending downwards since 2020. Meanwhile, the percentage of sources from Asia has generally trended upwards, jumping from about 1% in 2019 to 12% in 2023. The percentage of sources from South America was also at its highest percentage to date in 2023 at about 7%.
By Country
From 2019 to 2023, more than 70% of TON sources came from the United States. During this same period, between 1% and 4% of sources came from the United Kingdom, India, Canada, Mexico, Nigeria, South Africa and Kenya. The remainder of sources came from 31 countries, as shown below, with the percentage of sources from each of these countries below 1%.
Race
People who identified themselves as white (and no other racial identity) make up about 49% of sources. People of the global majority (that is, those who identified themselves as American Indian or Alaskan Native; Asian; Black or African American; Hispanic, Latino or Spanish Origin; Middle Eastern or North African; and/or Other Indigenous/Aboriginal) collectively make up about 49% of sources. The remaining 2% of sources did not select one of the above racial identities and selected “Other or Prefer to Self-Describe” or “Prefer Not to Disclose.”
The sources who selected “Other or Prefer to Self-Describe” noted ethnic and racial identities including Armenian American, Ashkenazi Jewish, Austronesian-Southeast Asian, European, Italian American, Jewish, Spanish Italian, and Tamil.
* Please note: These numbers do not add up to 100%; about 9% of sources identified as more than one race.
People who identified as Asian consistently make up the second-largest percentage of sources, following those who identified as white. Most recently, in 2023, people who identified as white made up just over half of all sources; people who identified as Asian comprised just over a quarter of all sources.
Of the remaining quarter of sources in 2023, people who identified as Black or African American and Hispanic, Latino or Spanish origin made up the third largest percentages at about 9% and 12% respectively. In the same year, people who identified as Middle Eastern or North African, Other Indigenous/Aboriginal, American Indian or Alaskan Native, or who selected “Other” or “Prefer to Not Disclose” each made up about 1% to 3% of sources.
Gender
Between 2019 and 2023, people who identify themselves as female made up the majority of TON’s sources, about two thirds in total. People who identify themselves as male made up the second largest gender group at about 26%. People who identify themselves as non-binary or genderqueer comprised another 6% of sources during this period. Note: Several sources selected more than one gender category.
Since 2019, the percentage of sources who identify themselves as male has generally trended upwards, while the percentage who identify themselves as female has generally trended downwards. In 2019, about 19% of sources identified themselves as male; in 2023, about 27% of sources identified themselves as male. In 2019, about 76% of sources identified themselves as female; in 2023, about 63% of sources identified themselves as female. Meanwhile, the percentage of sources who identify themselves as non-binary or genderqueer has remained between 3% and 6% with the exception of 2020 (when it landed at about 10%).
LGBTQ+ Identity
To date, sources who identify as LGBTQ+ have made up about one fifth of TON’s sources. Most recently, in 2023, LGBTQ+ people comprised about one quarter of TON’s total sources. As a reference point: In the U.S., about 7.6% of the population identifies as LGBTQ+ (Source: Gallup).
Disability
Sources who say they live with a disability and/or accessibility needs make up about 12% of total sources. As a reference point: It’s estimated that about 16% of people worldwide experience significant disability (Source: World Health Organization) and that 1 in 4, or 26%, of U.S. adults is disabled (Source: Centers for Disease Control).
2022 was the year with the lowest percentage of disabled sources (about 8%). In every other year, the percentage of disabled sources has landed between 11% and 15%.
Conclusions and Next Steps
These data are a baseline summary of where The Open Notebook currently stands in the diversity of sources represented in our stories. Our findings indicate areas where we have already seen improvement over time, such as a notable increase from 2019 through 2023 in representation of sources from the Global South and from LGBTQ+ communities. The findings also call attention to specific areas where we can continue to improve. For example:
-
-
- The Global North, and specifically North America, still makes up an overwhelming majority of our sources across all story types.
- Black, Indigenous, and Middle Eastern or North African sources are particularly underrepresented in our sourcing.
- Our representation of sources who identify themselves as disabled does not yet mirror the roughly one quarter of U.S. adults who are disabled.
-
This report sets the stage for our future efforts to expand the diversity of our sources, allowing us to set more ambitious goals for growth in the categories discussed above and for considering tracking additional dimensions of diversity. As creating a culture of DEI at The Open Notebook is very much an ongoing process rather than a one-time activity, we’re committed to continuing to learn and grow. Our editorial team and board are currently discussing the results of our initial five-year source-tracking analysis and their implications for our future work, including goals for the coming year or two and possible changes in the way we track source diversity.
(Editors’ Note: Special thanks to Carolyn Wilke and Allison Whitten for their help with project planning and data collection, and to Sarah Luft for her substantial work on data analysis and preparation of the report. Thanks also to members of the TON board of directors and editorial team for providing feedback and ideas.)