Where Is the Diversity in Publishing? The 2015 Diversity Baseline Survey Results

diversity102-logoBy now it’s no secret that publishing suffers from a DBS_caption1major lack of diversity problem. Thanks to years of research by the Cooperative Children’s Book Center, we have ample data to confirm what many readers have always suspected: the number of diverse books published each year over the past twenty years has been stuck in neutral, never exceeding, on average, 10 percent.

Countless panels, articles, and even conferences have been dedicated to exploring the causes and effects of this lack of diversity. Yet one key piece of the puzzle remained a question mark: diversity among publishing staff. While the lack of diversity among publishing staff was often spoken about, there was very little hard data about who exactly works in publishing.

At the beginning of 2015 we decided to conduct a survey to establish a baseline that would measure the amount of diversity among publishing staff. We believed in the power of hard numbers to illuminate a problem that can otherwise be dismissed or swept under the rug. We felt that having hard numbers released publicly would help publishers take ownership of the problem and increase accountability. We also felt that a baseline was needed to measure whether or not initiatives to increase diversity among publishing staff were actually working.

Our Diversity Baseline Survey took a year to complete. The results include responses from 8 review journals and 34 publishers of all sizes from across North America. Here are the results:

2015 Diversity Baseline Survey (DBS) Click for larger image

View a slideshow of the DBS survey results

Methodology and Response Rate
The Diversity Baseline Survey (DBS) was sent to 1,524 reviewer employees and 11,713 publishing employees for a total of 13,237 surveys deployed. The response rate was 25.8 percent. This is on par with the average for online surveys and actually a bit higher than the norm, given the sensitive nature of the questions.

In 2015, Publishers Weekly included some staff diversity questions in their annual Salary and Compensation Survey. They deployed their survey to 5,800 subscribers and had a response rate of 7.3 percent. Therefore, the DBS should yield a much more comprehensive picture of diversity in the publishing community.

The DBS was deployed directly from each publisher or review journal. A link was sent to all staff from a member of each publisher’s or reviewer journal’s human resources or executive team, often with an introduction explaining why the company was participating. Some companies even wanted to add additional questions to their surveys. The results provided here are only for questions that appeared in every survey.

The surveys were completely anonymous, and companies did not have direct access to the results. All data was analyzed and aggregated by Dr. Sarah Park Dahlen and Nicole Catlin of St. Catherine University in St. Paul, Minnesota, to ensure anonymity for individual employees.

Although our response rate was good, we still wonder: who didn’t take the survey, and how might that influence the results? With a survey of this kind, there is most likely some degree of selection bias. In other words, people who self-identify as diverse may have been more likely to take the survey. If that was the case, it would mean that our results portray publishing as more diverse than it actually is.

No voluntary survey can ever be 100 percent accurate, and no survey that asks questions about personal identity can ever be anything but voluntary. Even so, the results of the DBS offer a strong snapshot of the makeup of the publishing industry.

Notes and Analysis: What the Numbers Tell Us

According to the survey, just under 80 percent of publishing staff and review journal staff are white. The rest are comprised of Asians/Native Hawaiians/Pacific Islanders (7.2 percent), Hispanics/Latinos/Mexicans (5.5 percent), Black/African Americans (3.5 percent), and biracial/multiracial people (2.7 percent). Native Americans (0.5 percent), and Middle Easterners (0.8 percent) of publishing staff.

While all racial/ethnic minorities are underrepresented when compared to the general US population, the numbers show that some groups, such as Black/African Americans, are more severely underrepresented. This mirrors trends among children’s book authors. In 2014, just 2 percent of the books tracked by the Cooperative Children’s Book Center were by black authors. Latinos were similarly underrepresented in both places.

DBS_caption2Creating the list of ethnicities for a survey such as this was a real challenge. The racial breakdown we offered was based on the US census, with a few adjustments. For our first survey, we felt that this was the best way to break things down because it presented familiar categories that respondents had seen before.

But no list can accurately depict the complexity of this question. Within each category, there are so many different groups, and people self-identify in a wide variety of ways. The census groups White Americans, European Americans, and Middle Eastern Americans together. The census is not quite sure what to do with Latino and Hispanic people, who may or may not identify as white. And it certainly does not know how to handle the differences among Asians, Pacific Islanders, and South Asians.

We received more than 50 write-in comments for this question from people who did not feel that any of the options offered adequately represented them. Some identified as Jewish or European. Many specified that they were South Asian and didn’t feel that the overall Asian category was specific enough. And several simply called themselves “Human” and wondered why we cared so much about this. One block of data was compromised when the survey link was shared with outside spammers, which made a portion of the surveys ineligible for inclusion. These incidents and answers are all telling because they allude to the wide scope of attitudes toward this issue and how deeply the question of race resonates with people, in both positive and negative ways.

The survey reveals that publishing is about 78.2 percent women or cis-women and 20.6 percent men or cis-men. These numbers may help explain why some feel that children’s book publishing skews toward female readers. Among executive and board member positions this disparity evened out a bit, with approximately 40 percent of executives and board members identifying as men or cis-men. This reflects the reality that males still ascend to positions of power more often, even in female-dominated industries.

The gender question also reveals that about 98.7 percent of publishing staff identify as cis men or women. This means that they identify with the genders they were assigned at birth. How does this compare with the general population? We don’t really know. For many reasons, we don’t have a good count of the percentage of the general population that is transgender. That being said, the small number of transgender, gender-nonconforming, intersex, and other gender-fluid people in publishing points to the need for publishers to make sure that books on these topics are being examined for cultural and scientific accuracy by experts before they are published.

Sexual Orientation:
According to the survey, about 88.2 percent of publishing staff identify as straight or heterosexual. This may be the category in which publishing is most on par with the general population, though we can’t know for sure.

Beyond the labels we offered, many respondents added their own labels that they felt better represented them. Quite a few identified as “queer.” Others wanted to know why we were asking for such personal information at all. Overall, this question got one of the lowest response rates of the survey, an indication, perhaps, that many people did not feel comfortable sharing this information. We decided to include this question because we wanted to acknowledge this aspect of diversity, and if we didn’t include it, this segment of the workforce would remain uncounted and invisible.

The survey reveals that about 7.6 percent of publishing staff identify as having a disability. We defined disability broadly in the survey, so this does not give us an indication of the types of disabilities that are represented.

One interesting result: when broken down by department, design had a significantly higher average rate of disability (18 percent), followed by book reviewers (12 percent). Perhaps this is becauseDBS_caption3 there are more freelance design and reviewer jobs that can be done from home even when mobility is limited. Providing opportunities to people with disabilities may be an underappreciated benefit of creating more freelance positions in publishing.

The DBS results offer the opportunity to filter responses by department, giving a better picture of how diversity breaks out throughout an organization. More than one hundred thirty people wrote in comments for this question, listing departments or sub-departments beyond those listed in the survey. Because the survey was administered to companies ranging from just a few employees to several hundred or more, some departments or roles were left out. The next version of the survey will have an expanded list that is more inclusive to account for some of the staff who had to write in departments this time around.

An interesting result was the high response rate from editorial staff, who made up nearly 20 percent of survey respondents. This compares to less than 10 percent of respondents from marketing/publicity and 13.5 percent from sales. Since these ratios do not seem to match the overall breakdown by departments in publishing, we wonder if staff in some departments, such as editorial, were more likely than others to respond. If so, why? Are editorial staffs more on board with diversity initiatives than staff in other departments?

Here are the numbers:

Board Members and Executive Positions
Without a doubt, board members and those in executive positions make up the highest level of decision makers on the corporate ladder. Board members and executive positions are: 86 percent white, 59 percent cis-women, 89 percent heterosexual, and 96 percent able bodied/without a disability.

Editorial is the next most important department when it comes to the in-house staff closest to generating actual books. Editorial staff is: 82 percent white, 84 percent cis-women, 86 percent heterosexual, and 92 percent able bodied/without a disability.

Marketing and Publicity
These are the departments that promote the books. Staff members in marketing and publicity are: 77 percent white, 84 percent cis-women, 87 percent heterosexual, and 94 percent able bodied/without a disability.

Members of the sales team are the ones out there pounding the pavement and knocking on doors to sell front list and back list titles. Sales people are: 83 percent white, 77 percent cis-women, 90 percent heterosexual, and 94 percent able bodied/without a disability.

Reviewers often have a direct influence on what readers buy. Reviewers are: 89 percent white, 87 percent cis-women, 91 percent heterosexual, and 88 percent able bodied/without a disability.

What’s next?
Does the lack of diverse books closely correlate to the lack of diverse staff? The percentages, while not exact, are proportional to how the majority of books look nowadays—predominately white. Cultural fit would seem to be relevant here. Or at least in DBS_caption4publishing’s case, what is at work is the tendency—conscious or unconscious—for executives, editors, marketers, sales people, and reviewers to work with, develop, and recommend books by and about people who are like them.

So, we have our baseline numbers. What are the next moves? In future posts we will discuss initiatives already in place that will hopefully move the needle toward more diversity. We will also look at a similar publishing diversity survey that was conducted in 2014 in the United Kingdom. And we will be working on designing DBS version 2.0, which we hope will include the publishers who either didn’t hear about the survey or opted out the first time.

We also hope that the DBS will lead to more “Diversity 102” conversations about what publishers can do, including improving retention and staff training. How can company cultures be more welcoming for diverse staff? Do diverse staff members feel comfortable voicing their opinions? Are systems in place to make sure all staff are trained and well versed in diversity issues?

Publishing is not alone when it comes to having a lack of diversity problem. All media, including film, television, and theater, are having similar conversations about diversity. It is plain to see that our society as a whole has a problem. We believe we are at a crucial time right now. We all have to decide if the country in which we live is better off if we conduct our lives separately or together. The diversity problem is not the responsibility of diverse people to solve. It is a problem for everyone to solve. Now that the Diversity Baseline Survey is completed, the real work toward changing the status quo begins. It is not going to be easy. Knowing where we stand and establishing a baseline was the first step. Knowing the baseline numbers gives us a way to measure progress going forward, but only our actions can change things for the better.

Read also: Behind the Scenes Of Publishing’s First Diversity Baseline Survey and The Diversity Baseline Survey: What Happens Next?

For press inquiries or permission to reprint, please contact hehrlich[at]leeandlow[dot]com.

168 thoughts on “Where Is the Diversity in Publishing? The 2015 Diversity Baseline Survey Results”

  1. Not such a shocker. Applicant socioeconomic status is a big issue for a job that pays poorly but is highly competitive. If a person is good with words and of color, and of sufficient academic achievement that a Big Six-plus publisher is willing to consider him or her, there are many other far better paying options that will be wide open. I think if Hachette paid its editors even half of what Morgan pays its investment bankers, you’d see whole lot more publishing houses with a whole lot more non-white staff. It’s the economics and the sociology. I bet Scholastic isn’t getting all that many applicants from high-achieving rural kids who are white with laborer parents that don’t own their homes. If those kids are strong enough to qualify to be a Scholastic editor, they’re going to be wooed by American Express corporate communications, too. Who could blame them?

    1. Maybe not many, but we exist in publishing, at about the same rate. (I grew up on a small farm, in poverty, myself.)

      But not a shocker or not, we have work to do.

    2. I have to agree with this. It’s extremely difficult to break into publishing and the pay is fairly poor. There are socioeconomic bars. The people who do get in typically (with exception) come from wealth, have high academic achievement/support, and are connected. The extra programs that entrants are encouraged to participate in (Columbia, NYU, etc.) are prohibitive to those who can’t afford it.

      1. Not always. Comparing publishing with banking is apples and oranges. We’re talking about the industry as it now. I’ve seen white college drop outs get jobs at a Big Six. A person of “color” wouldn’t be able to do that. But that’s also another part of the industry. Who you know.

    3. I would add that class is an intersection with race. Not all POC are poor. There are plenty of POC who are not coming from poor backgrounds who might be interested in publishing who are still up against a gauntlet, whether they have to worry as much about money or not.

      As for those whose SES is a factor: I had to do the whole unpaid internship thing, and combined with my master’s in children’s literature program and having to take time off from school to work just to make ends meet long enough to get through another semester, combined with all sorts of other factors that stem from the poverty I grew up in (not least of which being a complete inability to manage money when I started out) I’ll be in student loan debt literally the rest of my life. Publishing *shouldn’t* be structured this way. It is not a badge of honor to make kids deal with the kinds of economic issues I dealt with–and in some ways still deal with–just because they come from a poor background.

      We can do better.

      1. I urge everyone to look at these findings sociologically. In all cases of commentary on social science research, there will be compelling anecdotes like Stacy’s. My hat’s off to her. We need more editors like her. The important thing to remember is that survey data like this points to broad trends, not individual cases. That one person can buck a trend, or that publishing is not like investment banking, corporate communications, or Federal government service is not the point. We get diverse books that buck the trends, too. But not enough, some think.

        The broader point remains: A young man or woman of color who has the right stuff to get to the final cut of a much sought-after editorial job comes with exceptional credentials no matter what his or her color, or other sociological identification factors, and also the right identifiers to meet the diversity missions of so many companies, public sector employers, not-for-profits, and the like. That person will be much sought-after by so many places committed to diversity in hiring in fact as well as in theory. Some of these young men and women may take the decision to work at the lowest paying of those options, particularly if family SES makes them on a class basis the equal of the competition. If the SES is lower, it is hard to ignore better paying work, and especially far better paying work.

        Stacy’s other point bears a close look and exploration. For example, would a high SES African-American (fe)male with identical striking academic and social capital to an high SES white (fe)male, both applying to a Big Six-plus publisher, have to run a harder gauntlet to get hired? My gut instinct is no. I suspect the path would be easier, in fact. My hypothesis would be that the closer model would be applications to highly selective colleges and universities. However, this is a subject for further research. Perhaps this is a ripe situation for a modified version of the kind of research done by Pager et al. (2009), who sent matched pairs of job applicants out for minimum wage level work and found structural bias against African-American applicants.

        I am grateful to all involved in this project for generating some decent data in an area where the conversation has been all anecdotal. Good work.

        1. A little further time with the figures yields interesting sociological insight. There are three groups that are heavily over-represented in comparison to the population as a whole. First, whites in general, but that can be explained by the SES factors elucidated above. It’s not that there aren’t qualified POC, it’s just that the historical financial set and setting for those POC candidates would have them gravitate toward far higher paying positions in the private and even non-profit educational sectors, leaving less qualified POCs to complete with higher qualified whites. Change can be expected in the next generation of these highly qualified POC, whose parents will have stronger financial stability. I would expect numbers commensurate with percentages of population around 2040. The second group is women. A social critic like Susanna Hoff Sommers might well say that the numbers here are an extension of the phenomena in elementary-through secondary education, and an extension of the so-called War Against Boys that has girls doing so much better in school than their male counterparts. There is another fact that also points to more female editors: in general, females read more and know more about books. I would bet one would see the same breakdown among librarians. The final category is those who identify as non-normative in their sexuality, who express in this survey twice their percentage of the population according to the best survey data (which has then below 5% at most, and probably closer to 3.5%). This is in many ways the least unexpected of the factors, considering the demography of major urban areas like New York and Boston, and the historic gravitation of those of non-dominant sexuality the arts (and to books), compared to e.g. electrical engineering or construction. While there are always exceptions in all fields of employment, different occupations have historically had different percentages of gays and lesbians.

          1. Did I write Susanna Hoff Sommers? I’m channeling my sociological inner Bangles. I meant of course Christina Hoff Sommers. By the way, I think her argument doesn’t hold up here. Females read way more than males. Therefore, it makes sense to have a disproportionate number of female editors and business people.

        2. I thought people went into publishing because they love books and want to work with them, and they want to work with other people who love books, and they do not want to work in a bank, even if they could get a banking job.

          The pay is worse in publishing, but working in banking can mean having no life, being under horrible stress, and having to focus on things that a book lover usually does not care about.

          1. Indeed…people who “stay” in publishing teachers, writers, social workers, librarians, etc. Have a true passion for their work. Although many times than not, the salary does not come close to the time & effort put into it. I have worked for years on the publishing side of books, and now on the retail side of books. My educational background is in Early Childhood Education, and yes, I am an African American Woman.

            It is not because African Americans don’t have the right stuff. I truly believe that in order to have change… we have to force change by writing books, self publishing, and creating our own buzz in the media and throughout the world. Change is slow, but always comes.

    4. Your argument doesn’t make any sense especially when Wall Street’s record of hiring minorities is even more abysmal with most new hires being white and male. The low pay in the publishing industry simply keeps them from going after jobs in that sector.

  2. Boy, what an eye-opener. So few African Americans involved in publishing. Is it any wonder why so few books by black authors are published? How can we rectify this situation.
    Thanks for giving us this important information.

  3. A bit meta, perhaps, but I think that a 25% response rate could be problematic. That’s a huge margin for error. If you were to try to track down pictures / profiles of all of the respondents who didn’t respond and then fill out as much as you could, you might plug that potential weakness in your statistics. It wouldn’t be a complete fix – you can’t tell for certain someone’s ethnicity / gender / disability, in all cases. But in many you should be able to give a fairly decent guess.

    1. I don’t think you can respond FOR other people when it comes to a survey like this. Also, you can’t tell someone’s ethnicity, sexual orientation, gender, or disability from a profile photo at all. That’s incredibly presumptuous, and not the point of what Lee & Low was trying to do.

      25% is a telling start. I’ve been in the industry for 10 years and I can count the non-white people on two hands. The other 75% isn’t going to be a landslide of diversity.

      1. I do not disagree with the meat of these findings – I simply don’t think that it’s methodologically sound. It sounds like a quantitative study; as such, it should stand up to rigorous examination, and this fails even a cursory examination. If it’s an industry study (against whom they compare their response rates), then sure, whatever – use it to sell cars, or make business decisions. But it’s not an industry study: it’s a sociological study, and therefore needs to be statistically valid if it’s a quantitative study, which it purports to be.

        If this were brought to me by a PhD candidate under my supervision, I’d have some serious questions, was my point. If this were in a published journal somewhere, I’d probably have to question the journal. 25% is completely inadequate to make generalizations to the population, which is what they’re attempting to do. Have a read through https://kkbiersdorff.wordpress.com/2009/09/16/how-many-is-enough/ for a good explanation of statistical significance.

  4. So do the results indicate that white, female heterosexuals in the publishing industry should be concerned about their jobs?

  5. “Does the lack of diverse books closely correlate to the lack of diverse staff? The percentages, while not exact, are proportional to how the majority of books look nowadays—predominately white. Cultural fit would seem to be relevant here. Or at least in publishing’s case, what is at work is the tendency—conscious or unconscious—for executives, editors, marketers, sales people, and reviewers to work with, develop, and recommend books by and about people who are like them.”

    I’m sure that personal preferences do have some affect on the books a person reviews and recommends. But I would hope that personal professionalism (is this a good book?) and economics (will people buy/be interested in this book?) would have a greater affect on the work product of the publishing industry.

    Also, generally accepted statistical analysis principals would check the data to see if there is a cause and effect relationship (high R value) between a person’s demographic and the demographic of the books they review and recommend. Maybe that is data that could be gathered in the next survey.

  6. I am a white person who has always felt the other ethnic groups were shunned.
    I am a children’s book author and took diversity into my decision for my first and second published children’s book. As a self-published author, I was thrilled to have my books picked up by Barnes & Noble to be shelved in their stores.
    My first book “Marshmallows Galore” portrays children of all nationalities, my second “The Wooly Adventures of Purl” is a story of a little girl of color.
    Barnes & Noble will be distributing a few hundred copies in the southern part of the country,
    I am so pleased by this. The writer’s I see are not the problem. As much attention is bringing this into the light, something is going to change.

  7. Next time around, it would be interesting to see the statistics on literary agents, who perform a key role as gatekeepers and filters for the publishing companies.

  8. This is such an important study that reminds us of the importance of diversity among people who are gatekeepers! Today I read the following article that had a telling statement about reading in particular. See: https://www.washingtonpost.com/news/education/wp/2016/01/21/teacher-race-affects-black-students-odds-of-being-labeled-gifted/

    “Black students taught by black teachers are assigned to gifted programs at almost the same rate as white students, and three times more often than black students with similar academic ability and family background who are taught by teachers of other races… The race of a black child’s teacher made a particular difference in whether he or she was identified as gifted in reading. The results raise “serious concerns,” the authors wrote, in a nation where 80 percent of black elementary school students are taught by teachers of other races.”

  9. FYI: We received some negative feedback regarding the term, differently abled. We were informed that this term was unacceptable, so we updated the infographic to reflect the more universally acceptable term of nondisabled.

  10. It would be interested to see breakdown between different genres. This survey seems to be tilted toward children’s literature. Several of the journals and publishers in the study are clearly identified as specializing in kids books. The listing of journals and publishers seems to lack adult genres of romance and science fiction. I don’t see Baen, Tor, Asimov magazine, or Locus having taken part in the survey. I’d like to know the breakdown of this part of the industry and how gender might affect the relationships between genre. I suspect that science fiction has a higher percentage of men making marketing decisions and that we may have gender polarization between genres, equivalent to the pink and blue toy aisle. This survey may have left out the blue aisle.

    1. SF/fantasy editors/publicists are still predominately women, at least at the big publishers. I cannot speak for the indies.

    2. To expand, you are actually far more likely to find a man working for one of the literary imprints or with narrative nonfiction than traditional paperback genre imprints.

  11. It seems that the headlines related to publishing companies should specifically note “Youth literature publishing industry”. I observed that several publishing companies did not participate (e.g., Pearson, Scott Forseman, Houghton Mifflin, etc.)–all companies that focus on curricular products. I am interested in seeing the results of a similar survey for this sector of the publishing industry–those who publish P-12 curriculum that often incorporates published youth literature, particularly for grades P-5. I think this addition to the current data would be powerful and even eye-opening when juxtaposed side-by-side. It would tell the story with data of who holds (or not) the broad range of power and gate keeping over the P-12 knowledge base of U.S. students. Thank you for your good work!

    1. It is true that there are a number of publishers who did not participate. Hopefully they will choose to do so in the next survey. As far as who the survey was sent to, the survey did include staff from adult publishing as well as children’s.

  12. I have only two concerns: #1) Racial variety (skin tone and prominent features) is not a guarantee or promise of true diversity nor is racial homogeneity an accurate way to declare something ‘lacks diversity’. CULTURE and LIFESTYLE are the landmarks of diversity. #2) If you’re truly wondering about diversity and whether there’s a lack of it – or not – then look to authors and examine more closely their reasons for embracing or not embracing cultural or lifestyle diversity. “Because Publishers” only goes so far. In order to write a more dynamic and diverse cast of characters you must understand the subtle and obvious differences between culture and lifestyle – two very difficult things to achieve without offending someone or being told you’re doing it wrong.

    1. Word, Judy. There is another kind of diversity that this survey does not analyze, which is political diversity. Does it matter if someday there is a broad representation of colors, ethnicities, and sexualities represented in publishing if all those voices are Leftist, or if all those voices are Conservative?

      My guess would be that publishing tilts heavily to the Left, just like being a librarian or a university professor in anything but business school tilts heavily to the Left. America is balanced politically. It would be good to see a range of editors who reflect that balance, fostering books with characters that not only look like America, but think like America.

  13. I am Hispanic/Native American, retired teacher/counselor K-12 age 80 years old, educator of Indian students over 50 years. I believe I have something to contribute, but seems age is not a diversity considered in your program. At least I have not seen any reference to it since I joined (only most recently), please correct me if I am mistaken. My issue is that in so many old cultures of the world, elders are a source of passing down many words and examples of dealing with life and people. Continue this most valuable service, I applaud you and offer my hellp in any way possible. Thank You.

    1. Hi Pat,
      Thanks for commenting! You’re right, for this study we did not include age diversity, but it would definitely be an interesting thing to look at in the future. I do believe that the wisdom of our elders is very undervalued in our society!

  14. It would be interesting to see a line of pie graphs representing the population of children broken down by the same categories–race, gender, orientation and disability. How do the statistics of people currently creating books for children compare to the actual population of children reading them? Have you made an infographic with these statistics that we can share in social media?

    1. Hi Michelle,

      We don’t have an infographic depicting this but if you search for the most recent census data it is fairly straightforward to find. We will try to include something like this for reference next time around.

Leave a Reply

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