Diversity of Authors in Notices and Beyond

Introduction

I was invited by the Editor in Chief of the Notices of the American Mathematical Society to survey, analyze, and discuss findings on the diversity of authorship in the Notices from 2019–2021. To conduct this study, the 760 authors from this period were emailed a diversity, equity and inclusion (DEI) survey. There were 229 (30.1%) respondents. Here we examine nine questions and statements for comment from a Likert scale survey, analyze observational data on diversity of authorship, and compare these results to national studies from the National Center of Education Statistics (NCES ) 2020 report on US professors by race and the National Science Foundation (NSF) 2021 report for mathematics and statistics doctoral recipients.

The nine survey questions/statements are: 1. What is your race/ethnicity? 2. What is your gender? 3. What are your preferred pronouns? 4. Please share your age group. 5. The Notices magazine of the AMS promotes an environment that allows writers to feel confident that their work matters. 6. You were confident that your article would be fairly considered for publication in Notices. 7. In which Notices section was your work published? (Check all that apply). 8. Diversity, Equity, and Inclusion: DEI programs are useful in the mathematics profession. 9. Are you an AMS member? This DEI survey was not designed by the AMS.

Diversity appears in various forms and in overlapping ways. Actions that promote fair treatment of all in organizations are complex. The University of Iowa offers a well-rounded sociological definition of diversity as, “…all aspects of human difference, social identities, and social group differences, including but not limited to race, ethnicity, creed, color, sex, gender, gender identity, sexual identity, socio-economic status, language, culture, national origin, religion/spirituality, age, (dis)ability, and military/veteran status, political perspective, and associated preferences (https://diversity.uiowa.edu/resources/dei-definitions)”. The demographic aspects receiving the most research presently include race and ethnicity, gender, gender identity, sexual orientation, and age (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6701939/), all of which are represented and analyzed within this study.

The PEN America report and research studies shared that there is and always has been a “persistent lack of racial and ethnic diversity among employees and authors” in the publishing industry (https://pen.org/report/race-equity-and-book-publishing/). This article focuses on writing and publishing, revealing a hypothesis that due to the persistent lack of diversity, equity, and inclusion and long-standing history within the AMS and the mathematics profession in general, the Notices authorship would be as lacking in diversity as other comparable reports about the mathematics profession and studies of publication and diversity of authorship in journals more generally (https://www.nature.com/articles/d41586-023-01457-4).

A formula doesn’t care who solves it. —Janet Barber

Diversity and the Notices of the American Mathematical Society

The Notices is a membership magazine that publishes articles on the applications and theory of mathematics, the history of mathematics, mathematics education, current issues facing mathematicians, as well as other math-related topics. Having published with this magazine previously about discrimination and being cited in their AMS Diversity Report, “Towards a Fully Inclusive Mathematics Profession” 1, it was encouraging to see a DEI statement from the AMS that reads,

The AMS supports equality of opportunity and treatment of all participants regardless of gender, gender identity or expression, race, color, national or ethnic origin, religion or religious belief, age, marital status, sexual orientation, disabilities, veteran status, or immigration status.

The first issue of the Notices appeared in 1953 during turbulent racial disparities in America, including within the AMS. For example, according to the AMS diversity report, there were instances of African American mathematicians not being allowed to reserve lodging in the same hotel as white mathematicians at most conferences throughout these unfortunate times 13. Though by 1951 the AMS had established a nondiscrimination policy, the report stated that the AMS turned a blind eye, and African Americans had to secure accommodations outside of the conference venue while white (and other) mathematicians were allowed to stay at the conference hotel. Besides the hotel issue, racism and research snubs continued even as late as the 1970s. Longstanding negative histories can fester and persist, as the AMS acknowledged the existence of recent discriminatory practices and disrespectful behaviors 1. Nonetheless, according to the DEI report, the current American Mathematical Society is trying to be more inclusive in employment, research, and publication. One way that the Notices is taking responsibility is by challenging itself to report on its DEI issues via this study.

Overall Findings

Question #1: What is your race/ethnicity?

The question of race/ethnicity in DEI research is always informative; therefore in order to analyze the diversity of Notices authors (2019–2021), the respondents were asked to self-identify with culturally responsive and extended categories for personal identification. The results are in the Race and Ethnicity table below.

Table 1.

Race and ethnicity.

Answer Choices	Responses	#Respondents
African	0.89%	2
Asian or Asian American	7.11%	16
Arab American, Middle Eastern, or North African	1.33%	3
Alaskan/Aboriginal or Indigenous Canadian	0%	0
Bi/Multiracial	1.78%	4
Black or African American	3.56%	8
East Asian	0.89%	2
Hispanic, Latino/a/x, or Spanish origin	5.33%	12
Mexican or Mexican American	0.89%	2
Middle Eastern (non-African)	0.89%	2
Native American or Indigenous	0%	0
Native Hawaiian or other Pacific Islander	0.44%	1
South Asian	0.44%	1
Southeast Asian	0.89%	2
White/European American, European	75.56%	170
TOTAL (4 skipped)		225

Additional ethnicity data

The DEI survey for this project was conducted during the COVID-19 pandemic and lockdowns, and although it was important to study and note respondents’ replies to the nine questions, it was also useful to identify demographic and diversity information obtained by direct observation of Notices authorship. This enabled us to have some idea about those who were unable to or chose not to respond to the survey’s study. Using this data offered another representative metric for comparative measure and narrative. For example, this observational data indicated that only 68% of Notices’ authors were white, while according to the survey, 76% of Notices’ authors were white (see Table 2 below).

The observational data used Notices writers’ photos, social media accounts (such as LinkedIn), biographies (when available), and one on one conversations with authors or people who knew them to carefully determine their race and/or ethnicity. Though gender and sexual orientation were not part of this DEI data, it was learned that there were nine (0.01%) LGBTQIA+ authors (seven white and two Black) in 2019–2021.

Table 2.

Demographic observational data. (Not based on survey data in Table 1.)

Race/Ethnicity	# of Authors	% out of 760
African American/African/Black	56	7.3%
Asians	100	13%
Hispanic, Latino/a/x, Mexican, Pac. Isl.	76	10%
MENA (Middle Eastern, North African)	12	1.6%
TOTAL	244 (overall out of 760)	32%
Comparative Information:
White/European American, European	516	68%

Mathematicians: National reports

Data from the NCES 2020 report as well as data from the NSF 2021 report for mathematics and statistics doctoral recipients were used as comparative measures (see Table 5). Though these data may not be perfect for comparisons, they have proven useful for this study. Also, for further comparison, note the 2017–2018 AMS finding that 80% of new doctorates in mathematics who were US citizens are white 1; https://www.ams.org/profession/data/annual-survey/2018Survey-NewDoctorates-Report.pdf.

The National Center for Education Statistics 2020 report (see Table 3) showed overall US full-time university faculty were white at 73% (https://nces.ed.gov/fastfacts/display.asp?id=61), and though not all authors for this study were professors, most were, and the percentage of white authors who responded to this survey (76%) was similar to NCES data. These statistics demonstrated even less diversity than the National Science Foundation (NSF) survey (Table 4), though all showed a lack of diversity in higher education venues, intentional or not.

Table 3.

Findings from the National Center for Education Statistics (NCES) https://nces.ed.gov/fastfacts/display.asp?id=61.

US Professors by Race (all levels)	Percentages
White	73%
Asian	12%
Black or African American	6%
Hispanic	6%
Unknown	1%
American Native & Alaska Native	Less than 0.5%
Pacific Islander	Less than 0.5%

(From Digest of Education Statistics 2022, table 315-20).

Table 4.

National Science Foundation (NSF).

US Survey of Mathematics Doctorates	Percentages
White	61%
Asian	30%
Black or African American	2.4%
Hispanic or Latino	4.1%
Other	1.6%
American Indian & Alaska Native	0.9%

Source: NSF, 2021 National Center for Science and Engineering Statistics Survey of Doctorate Recipients. Adapted from Table 8. US residing employed doctoral scientists and engineers, by field [Mathematics and statistics] of doctorate, ethnicity, race, and sex: 2021 https://ncses.nsf.gov/surveys/doctorate-recipients/2021#data

Table 5.

Comparative demographic data.

Research Statistics, Studies, Reports >>>>	Observational data (Notices)	Survey data	NCES data, professors	NSF data, PhDs/math
Race/Ethnicity	% of 760	% of 229	% NCES	% NSF
African American/African/Black/Biracial	7.3%	6.23%	6%	2.4%
Asians (combined)	13%	9.33%	12%	30%
Hispanic, Latino/a/x, Mexican, Pacific Islanders	10%	5.33%	6%	4.1%
MENA (Middle Eastern, North African)	1.6%	2.22%	N/A	N/A
Other races and ethnicities	0.1%	0.89%	3.0%	2.5%
White/European American, European	68%	76.0%	73.0%	61.0%
BIPOC Percentages	32%	24%	27%	39%

Note that the Black authorship, mostly PhDs, of the Notices was more diverse than the NSF findings for the Black mathematics doctoral profession as a whole (see Table 5).

Question #2: What is your gender?

Two-thirds of the writers for the Notices identified as male (Table 6). The National Center for Education Statistics (NCES) fall 2021 findings for all levels of professors across all fields in the United States by gender were: Females 47% and Males 51%.

Table 6.

Gender identity from survey data.

Answer Choices	Responses	Respondents
Female	34.80%	79
Male	64.76%	147
Nonbinary	0.44%	1
Other	0%	0
TOTAL (2 skipped)		227

Analyzing NCES findings, this project’s data showed that within the realm of mathematics professors, the Notices (or the mathematics field itself) is below the norm in the participation of women and above the norm for participation of males across all fields. Note that the NCES report did not account for other gender categories, nor did their report detail sum to 100% because of their data rounding (https://nces.ed.gov/fastfacts/display.asp?id=61).

Question #3: What are your preferred pronouns?

Out of 229 participants, six did not answer. For the remaining 223, see Table 7.

Table 7.

Personal gender pronouns.

Answer Choices	Responses	Respondents
She/her/hers	33.63%	75
He/him/his	60.99%	136
They/them/theirs	1.35%	3
Other preferences	4.04%	9
TOTAL		223

Question #4: Please share your age group

Authors of articles in the Notices who are 40 to 60 years of age are typically already secured in their careers. A person’s age can have a major impact on their knowledge, skills, experience, professional position, and even discipline of study. This data (Table 8) provided background information to better understand generational responses to Questions/Statements #5 and #6 and Table 11.

Table 8.

Age group.

Answer Choices	Responses	Respondents
60+	35.96%	82
50–59	25.44%	58
40–49	22.81%	52
30–39	14.04%	32
Under 30	1.75%	4
TOTAL		228

Question #5: The Notices magazine of the AMS promotes an environment that allows writers to feel confident that their work matters

Table 9.

Confident that their work matters.

Answer Choices	Responses	Respondents
Strongly disagree	5.70%	13
Disagree	1.75%	4
Neither agree or disagree	15.79%	36
Agree	51.19%	119
Strongly agree	24.56%	36
TOTAL	228	(1 skipped)

Most participants (76%) felt confident that their work matters (see Table 9). However, a higher percentage of males in the 60+ age category disagreed with the sentiment that the Notices fosters an environment where their work matters. On a more positive note, see Table 10 (broken down by gender).

A total of 226 participants responded to this question with 174 agreeing that the Notices promoted an environment that allows writers to feel confident that their work matters.

Table 10.

Gender breakdown for confidence that their work matters.

	Disagree or Strongly Disagree	Neutral	Agree or Strongly Agree
Gender (N=226)	N %	N %	N %
Female	8 10.1%	17 21.5%	54 68.4%
Male	9 6.2%	17 11.6%	120 82.2%
Nonbinary		1 100.0%
TOTAL	17 7.5%	35 15.5%	174 77.0%

Question #6: You were confident that your article would be fairly considered for publication in Notices

With 228 participants responding to this question, 90% strongly agree or agree that their article would be fairly considered for publication, 51% and 39% respectively (see Table 11). While about 6% were neutral, this left 4% for disagree/strongly disagree. The gender breakdown is 93% of males and 87% of females who feel that their articles will be fairly considered for publication.

Other than the “Under 30” age group with just four (4) respondents, most age groups responded that they strongly agree that their article would be fairly considered. The breakdown by age demonstrated a higher percentage 6.1% of the 60+ age category that disagreed with this sentiment compared to younger respondents. From a DEI and social science standpoint, this is worth mentioning. One elder, a male AMS member, suggested that the Notices DEI survey itself was an attempt to weed out white men. What’s driving these elder writers’ concerns? Nevertheless, when broken down by age group for Question/Statement #6, we find a strong “agreement” for all race-ethnicities and age groups.

Table 11.

Confident that article would be fairly considered for publication.

	Disagree or Strongly Disagree	Neutral	Agree or Strongly Agree
Age Group (N=228)	N %	N %	N %
Under 30		2 50.0%	2 50.0%
30–39		3 9.4%	29 90.6%
40–49	1 1.9%	2 3.8%	49 94.2%
50–59	2 3.4%	4 6.9%	52 89.7%
60+	5 6.1%	2 2.4%	75 91.5%
TOTAL	8 3.5%	13 5.7%	207 90.8%

Question #7: In which Notices section was your work published? (Check all that apply)

There were 220 respondents. Most of the writers’ articles were published in the Features section at 29%. The next largest was the Early Career section at 23.6%, followed by the Memorial section (which is extensive in almost all issues) at 17.7%.

Table 12.

Notices section.

Answer Choices	Responses	Respondents
Features	29.09%	64
Education	6.82%	15
Early Career	23.64%	52
Sectional Sampler	2.27%	5
Memorial Tribute	17.73%	39
History	2.27%	5
Book Review	7.27%	16
Bookshelf	1.36%	3
Communication	12.27%	27
Other	11.36%	25
TOTAL		251

Those published in the features section were predominately white and male at 61.9% and 70%, respectively. See more demographic findings about features in Tables 13, 15, and 16.

Table 13.

Features authors by race/ethnicity.

Race/Ethnicity	Responses N	Percentages
Asian or Asian American	6	9.5%
Arab, American, MENA	1	1.6%
Bi/Multiracial	2	3.2%
Black or African American	3	4.8%
East Asian	1	1.6%
Hispanic, Latino/a/x, Spanish	10	15.9%
Mexican or Mexican American	1	1.6%
White/European American, European	39	61.9%
TOTAL	63	100%
		38.1% = nonwhite

Table 14.

DEI programs are useful.

DEI PROGRAMS INQUIRY	Disagree or Strongly Disagree	Neutral	Agree or Strongly Agree
Race/Ethnicity (N=220)	N %	N %	N %
African			2 100.0%
Asian or Asian American		1 6.3%	15 93.8%
Arab American, ME, NA		2 66.7%	1 33.3%
Alaskan/Aboriginal or Indigenous Canadian
Bi-racial	1 25.0%		3 75.0%
Black or African American			8 100.0%
East Asian		1 50.0%	1 50.0%
Hispanic, Latino/a/x, or Spanish origin	1 8.3%	2 16.7%	9 75.0%
Mexican or Mexican American	1 50.0%		1 50.0%
Middle Eastern (non-African	1 50.0%		1 50.0%
Native American or Indigenous		1 100.0%
South Asian			1 100.0%
Southeast Asian			2 100.0%
White/European American, European	6 3.6%	22 13.3%	137 83.0%
TOTAL	10 4.5%	29 13.2%	181 82.3%

Table 15.

Features authors by gender.

Gender	N	Percentages
Women	19	30%
Men	44	70%
TOTAL	63	100%

Table 16.

Features authors by age.

Age Group	N	Percentages
Under 30	1	2%
30–39	10	16%
40–49	16	25%
50–59	21	33%
60+	16	25%
TOTAL	64	100%

NSF’s study showed that mathematics doctorates were 61% white. Table 13 shows that 61.9% of feature articles for the Notices were written by white authors. This is less than the NCES data, which showed 73% of professors across all fields were white, and less than the survey data which had 76% white respondents. This percentage is also less than the observation data which showed 68% of Notices authors were white. Therefore, 61.9% demonstrated the success of the Notices’ efforts to include feature authors who did not identify as white, though special issues in honor of Black History and Hispanic Heritage likely contributed to this effort.

Another important discovery was that Hispanic, Latino/a/x, or Spanish ethnicities were found to be published at the second highest rate after white authors. People of more senior ages may be more published in the features area because they have had longer research careers.

Question #8: Diversity, equity, and inclusion: DEI programs are useful in the mathematics profession

Data gathered for Question #8 demonstrated overwhelming agreement with the importance and usefulness of DEI programs in the mathematics profession and community. Out of 229 respondents, 224 answered this question. While 81% agreed with this statement, the breakdown was 55.8% strongly agreed and 25% agreed, 13.39% were neutral, and the remaining respondents, 5.8% either disagreed or strongly disagreed. There was 100% agreement from African/Black/African American respondents, 100% agreement from South Asians and Southeastern Asians, and 83% agreement from white respondents. Note: N=220, as four (4) skipped Question #1 (Excel breakdown by race) and five (5) skipped Question #8 (see Table 14).

Question #9: Are you an AMS (American Mathematical Society) member?

This question was asked because AMS members may have a vested interest in publishing in their membership magazine. Responses showed that 67% of the 228 respondents were members of the AMS (see Table 17).

Table 17.

AMS membership.

Answer Choices	Responses	Number of Respondents
Yes (member)	66.67%	152
No (nonmember)	33.33%	76
TOTAL (1 skipped)		228

One interesting note, apart from the fact that the more senior-aged group was more published in features, was that AMS members, not surprisingly, tended to be more published in the features area of the Notices than nonmembers.

Table 18.

AMS membership and confidence article would be fairly considered for publication in the Notices (Deduced from Questions 6 and 9).

AMS MEMBERSHIP RESPONSES (N = 228)	Disagree or Strongly Disagree	Neutral	Agree or Strongly Agree
Yes (member)	5%	7%	88%
No (nonmember)		4%	96%

However, after combining Questions/Statements #6 and #9, it was found that nonmembers of the AMS agreed or strongly agreed at a higher rate than members to feeling confident that their article would be fairly considered for publication (Question #6). See Table 18.

When combining data from Questions 5, 6, and 9, 133 members and 73 nonmembers, a total of 206 agree or strongly agree both that their work matters and that their article would be fairly considered for publication in the Notices.

Conclusion

Supporting the Editor in Chief’s interest and request for this study, 33 issues of the Notices of the American Mathematical Society from 2019–2021 were examined for diversity of authorship, and found low but typical percentages for writers who did not identify as white, as three-fourths of survey respondents were white. Though there were some slight category constraints in design, it was found that the diversity in authorship for this project resembled demographic findings of racial inequalities in not only academic journals (https://www.nature.com/articles/d41586-023-01457-4) but also that of mathematics scholarship, faculty, and doctoral recipients. These similar findings included the Notices’ own AMS leadership Diversity Report information as well as NCES 2020 research.

The study revealed a secondary prevalence of Asian/Asian American authors in the Notices overall. However, Hispanic, Latino/a/x, and Spanish ethnicities were shown to be published at a higher rate (15.9%) in the prominent features section of the Notices (2019–2021) than Asians (11.1%) and other races and ethnicities (11.2%), except for whites (61.9%). Also, note that Black authorship (6.23%) in the Notices overall was greater than data from the NSF 2021 report of Black representation among mathematics PhDs (2.4%). This and the data for nonwhite authors in the features section demonstrated the success of the Notices’ efforts to include authors who did not identify as white.

However, reading the literature and studying the data and results for this project, there is a need to consider investigating the complexity of how and why white writers and those who identify (or are perceived) as white live in a place of “privilege” with more access to resources and therefore more publishing opportunities than those who do not identify as white, visually or otherwise. Discrimination, consciously or unconsciously, oftentimes plays a part in publishing oversights. This problem grows out of appalling systemic and intentional racism and discrimination in the sciences and other fields. Faulty perceptions, which foster false narratives, are often why certain marginalized and underserved populations in mathematics and other disciplines are not fairly published 1234.

Though becoming more controversial and politicized during 2022–2023, DEI issues are still relevant and needed. At the time of this survey, results showed overwhelming agreement from Notices writers that diversity, equity, and inclusion programs are useful in the mathematics community. This mindset could lead to promising and more robust DEI efforts for publishing in the Notices and other academic journals in the future.