A Note on the Treatment of Gender and Race in My Statistics Notes

Richard Williams, University of Notre Dame, https://www3.nd.edu/~rwilliam/

Last revised January 10, 2022

My personal web pages at https://www3.nd.edu/~rwilliam/ contain dozens, perhaps hundreds of

handouts on statistics. Early versions of some of these first saw life more than 30 years ago.
Many of these are from classes I no longer offer and may never get revised again, while others
will continue to be tweaked as I teach courses using them.

There have been many changes in statistical practice and preferred wording over those 30 years.
Traditionally most studies have (and perhaps still do) treat gender as binary and fixed at birth.
These practices are unfortunate, not only because they are inaccurate, but because they can
perpetuate harm experienced by transgender and non-binary communities (Cameron & Stinson,
2019). I became acutely aware of such problems when one of my own students said that my
terminology and examples sometimes made them feel “unwelcome” and “uncomfortable” in
class. Further, as Mize (2015) points out, the concepts of sex (biological differences between
males and females) and gender (socially constructed notions of masculinity and femininity) have
often been confounded and treated as synonymous.

It has also been common to treat race as an objective trait fixed at birth, and to oversimplify and
misuse the concept of race. Preferred wording has also changed over time. While it used to be
common to use the terms black or blacks (note the lower-case b’s) today it is often seen as better
to use phrases like Black Individuals or Black People or to at least capitalize Black. (See the
references for historical and ongoing debates about racial terminology.)

My notes (and for that matter probably most of my publications) reflect common past practices.
As I update handouts, I am making changes to reflect current, more inclusive terminology and
methods. When I revise my notes, I am not going to engage in revisionist history and pretend
studies were conducted differently than they were. Terms like female and black will continue to
appear as they did in survey questions and in value and variable labels, which are often very
short anyway to keep output easily readable. But, when interpreting results, I am going to try to
use more contemporary phrasings, e.g. refer to Black Individuals rather than blacks. I’ll also
work on this verbally and try to avoid slipping into old habits.

I’d also like to incorporate examples where gender is not treated as binary and unchanging. I
have and will continue to seek out appropriate datasets. Also, if anyone knows of good empirical
(and preferably replicable) examples where this is done I'd love to hear about them. I’ll also
periodically remind people that if studies I use in examples were being conducted today, they
might handle race and/or gender a bit differently.

Cautionary Note: Small Ns may greatly limit possible alternative gender identity analyses
though. As explained below, the General Social Survey made major changes in its questions on
gender identity in 2018. But, out of 1409 respondents, only 9 cases were identified as
transgender or another alternative gender designation. As a practical matter, I suspect most
quantitative analyses will be limited to treating sex as binary, unless the sampling strategy has
explicitly targeted collecting data from respondents with less common gender identities. Some
promising-sounding data sets that have done so are listed in the appendix.

A Note on the Treatment of Gender and Race in My Statistics Notes Page 1

 Appendix: Additional Resources
My thanks to those who have made suggestions on how my handouts can be improved. I
appreciate everyone’s patience as I work on them further. Here are some additional materials on
relevant publications and data sets that may be useful, especially for those who would like to do
research involving less common alternative gender identities.

I. References on the measurement of gender identity

Cameron, JJ, Stinson, DA. Gender (mis)measurement: Guidelines for respecting gender diversity
in psychological research. Soc Personal Psychol Compass. 2019; 13:e12506.
https://doi.org/10.1111/spc3.12506. This article also offers good suggestions for what
researchers should to do instead.

Carian, Emily K. 2019. More Inclusive Gender Questions Added to the General Social Survey.
https://gender.stanford.edu/news-publications/gender-news/more-inclusive-gender-questions-
added-general-social-survey. Retrieved January 4, 2021.

Mize, T. D. (2015) What Social Psychology Can Contribute to the Study of Sex, Gender, and
Sexual Orientation. Sociology Compass, 9: 1066– 1081. https://doi.org/10.1111/soc4.12331

Smith, Tom W., and Jaesok Son. 2019. Transgender and Alternative Gender Measurement on the
2018 General Social Survey. GSS Methodology Report No. 129.
http://www.gss.norc.org/Documents/reports/methodological-
reports/MR129%20transgender%202018.pdf. Retrieved January 4, 2022.

II. References on Racial Terminology

Racial terminology is hotly debated and has been for several decades. For some good
discussions, see

Palmer, Brian. 2010. When Did the Word Negro Become Taboo? https://slate.com/news-and-
politics/2010/01/how-old-was-harry-reid-when-the-word-negro-became-taboo.html

Eligon, John. 2020. A Debate Over Identity and Race Asks, Are African-Americans ‘Black’ or
‘black’? https://www.nytimes.com/2020/06/26/us/black-african-american-style-debate.html

Appiah, Kwame Anthony. 2020. The Case for Capitalizing the B in Black.
https://www.theatlantic.com/ideas/archive/2020/06/time-to-capitalize-blackand-white/613159/

Wong, Brittany. 2020. Here's Why It’s A Big Deal To Capitalize The Word ‘Black’.
https://www.huffpost.com/entry/why-capitalize-word-black_l_5f342ca1c5b6960c066faea5

A Note on the Treatment of Gender and Race in My Statistics Notes Page 2

 III. Possible Data Sets that include gender-identity questions

Many data sets have good measures of sexual orientation, but I suspect good data related to sex
and gender identity (e.g. transgender, intersex, non-binary) may be harder to find. Questions
about gender identity are often not asked, and when they are the Ns for people who list
themselves as transgender/intersex/non-binary may be too small to provide useful information.

The General Social Survey. https://gss.norc.org/. The General Social Survey (GSS) is a
nationally representative survey of adults in the United States conducted since 1972. Based on
the recommendations of several social scientists, the GSS modified its questions on gender
identity in 2018. For the first time in GSS history, a two-step gender question asked respondents
to separately identify the sex they were assigned at birth and their current gender (Carian 2019).
The first question reads, “What sex were you assigned at birth? (For example, on your birth
certificate)” and allows respondents to select “Female,” “Male,” “Intersex,” or “No answer.” The
second question asks, “What is your current gender?” Respondents were able to select
“Woman,” “Man,” “Transgender,” “A gender not listed here,” and “No answer.” But, out of 1409
respondents, only 9 cases were identified as transgender or another alternative gender
designation. However, a person at the GSS told me that the Ns may get higher, more questions
are coming, and that gender scholars are excited for the new content. It will be exciting to see
what the GSS produces in the future.

Social Justice Sexuality Project: 2010 National Survey.

https://www.icpsr.umich.edu/web/RCMD/studies/34363. The Social Justice Sexuality Project
(SJS) is one of the largest national surveys of Black, Latina/o, Asian and Pacific Islander, and
multiracial lesbian, gay, bisexual, and transgender (LGBT) people. The purpose of the SJS
Project is to document and celebrate the experiences of lesbian, gay, bisexual and transgender
(LGBT) people of color. It has many questions I haven’t seen anywhere else. For Gender
Identity, 2279 listed themselves as male, 2162 were female, 111 were trans M-F, 62 were trans
F-M, 132 were other, 28 claimed multiple identities, and 179 were missing. That is over 300
cases with alternative gender identities, which is far higher than I have seen in most data sets.

LGBTQData.com. http://www.lgbtdata.com/. LGBTData.com serves as a no-cost, open-access

clearinghouse for the collection of sexual orientation & gender identity data and measures. The
website encourages the collection of sexual orientation data and the analysis of data sources that
have already collected such data. While a few of these data sources were designed to collect data
primarily from LGBs, most of them are large surveys that decided to include a sexual orientation
variable. I haven’t checked these sources out.

LGBTQ datasets available on data.world. https://data.world/datasets/lgbtq. The site currently

lists three LGBTQ datasets that it has available. I haven’t checked these sources out.

Searchable LGBTQ Resource Database. https://lgbtq-ta-center.org/resource/. The California

Reducing Disparities Project (CRDP) aims to implement and evaluate community-based
programs that support mental health among California’s Lesbian, Gay, Bisexual, Transgender,
and Queer/Questioning populations. I’m not sure if you can actually find data on the site, but it
does provide links to 100s of resources related to LGBTQ-related topics.

A Note on the Treatment of Gender and Race in My Statistics Notes Page 3

2015 U.S. Transgender Survey (USTS) (ICPSR 37229).
https://www.icpsr.umich.edu/web/ICPSR/studies/37229. The 2015 U.S. Transgender Survey
(USTS) is the largest survey ever conducted (27,715 respondents) to examine the experiences of
transgender people in the United States. The survey included questions related to the following
topics (in alphabetical order): accessing restrooms; airport security; civic participation;
counseling; education; employment; family and peer support; health and health insurance; HIV;
housing and homelessness; identity documents; immigration; income; intimate partner violence;
military service; police and incarceration; policy priorities; public accommodations; faith; sex
work; sexual assault; substance use; suicidal thoughts and behaviors; unequal treatment,
harassment, and physical attack; and voting. There are no publicly available data files for this
study, but researchers can apply for access. You can download documentation for the study.

National Transgender Discrimination Survey, [United States], 2008-2009 (ICPSR 37888).

https://www.icpsr.umich.edu/web/ICPSR/studies/37888. The National Gay and Lesbian Task
Force and the National Center for Transgender Equality are grateful to each of the 6,450
transgender and gender non-conforming study participants who took the time and energy to
answer questions about the depth and breadth of injustice in their lives. These data provide
information on transgender discrimination in every major area of life, including housing,
employment, health and health care, education, public accommodation, family life, criminal
justice and government identity documents, and demographic information such as citizenship,
race, ethnicity, employment, and income. In virtually every setting, the data underscores the
urgent need for policymakers and community leaders to change their business-as-usual approach
and confront the devastating consequences of anti-transgender bias. Data can be downloaded in
Stata format. It appears to have many interesting variables you usually don’t see in other
studies.

Resources for Research on Gender Identity, Sexual Behavior, and Sexual Orientation.
https://www.icpsr.umich.edu/web/pages/RCMD/lgbtq-resources.html. (These resources are
provided by the Resource Center for Minority Data, described below.) Here you'll find data
resources for those interested in studying gender identity, sexual behavior, and sexual
orientation, or in finding datasets in which those populations are well-represented. It includes
data resources available at ICPSR and elsewhere. For instance, you'll find data from studies that
focus on these populations and data from broader studies that contain questions that enable
researchers to identify participants in these populations. Currently about 40 datasets are listed in
a downloadable spreadsheet.

You might also check out Celebrating Pride Month with Resources From ICPSR.
https://www.icpsr.umich.edu/web/pages/RCMD/news.html?node=3290. Besides highlighting
various data sets, it has interesting video and audio presentations. For example, “Quantifying
Gender Identities and Behaviors” (https://www.youtube.com/watch?v=1j1U6b436QA) reviews
challenges in creating gender inclusive categories in surveys using real-world examples (e.g.
what if your sample has 1 trans person and everyone else is cis male/female, for example?) It
also discusses how researchers, particularly grad students and undergrads, are using social media
to target distribution of surveys to transgender populations.

A Note on the Treatment of Gender and Race in My Statistics Notes Page 4

 IV. Data Sets that focus on racial and ethnic minorities

Many, perhaps most, data sets have questions on race and ethnicity. But, some data sets focus on
racial and ethnic minorities much more than do others.

Resource Center for Minority Data (RCMD; Also referred to on their web pages as Really
Cool Minority Data). https://www.icpsr.umich.edu/web/pages/RCMD/about.html,
https://www.icpsr.umich.edu/web/pages/RCMD/mission.html
RCMD is a recent initiative of ICPSR (Inter-university Consortium of Political and Social
Research). The mission of the Resource Center for Minority Data (RCMD) is to provide
educators, researchers, and students with data resources to analyze issues affecting racial and
ethnic minority populations in the United States. RCMD provides access and analytic tools for
use with the vast array of minority data available from ICPSR. RCMD enhancements assist
researchers, educators, and students in their quest to effectively utilize data on underrepresented
populations. RCMD provides links to over 400 downloadable data sets, sortable by topic area
(e.g. education, discrimination, immigration, health & well-being), at least some of which are
already in Stata format.

Coursework offered by RCMD. To raise awareness of data resources about racial and ethnic
minority populations, the ICPSR Summer Program offers (or at least has offered) a four-week
course titled “Methodological Issues in Quantitative Research on Race and Ethnicity.” This
course examines research design, measurement, and analysis issues involved in research on
minority populations in the United States. Topics include
• Reliable and valid measurement of racial and ethnic variables in major survey studies
• Innovative sampling strategies for diverse populations
• Race-related measurement errors in data collection
• Analysis traditions and trends within various social/behavioral science field

For more detail see https://www.icpsr.umich.edu/files/RCMD/brochure.pdf

Note: My thanks to Mara Kraemer for drawing my attention to the RCMD. I haven’t used it yet
myself but it looks like it has a lot of great resources that are freely available.

A Note on the Treatment of Gender and Race in My Statistics Notes Page 5