Addressing Racism in Research Methods and Statistics Courses

Teaching Tips: Addressing Racism in Research Methods and Statistics Courses
(Published in January, 2021, in the ASC DWC #SayHerName Special Edition Newsletter)

“Data do not tell us a story. We use data to craft a story that comports with our understanding of the world. If we begin with a racially biased view of the world, then we will end with a racially biased view of what the data have to say.” –Bonilla-Silva & Zuberi, 2008, p.7

As I stood in a crowded atrium listening to students express their anger and frustrations about racism on our campus to university administrators, a colleague from another department whispered to me: “I’m so glad I don’t have to address these issues in my classes.” My eyes widened and I replied, “You may not have to, but you should.”

Although I predominately teach undergraduate and graduate-level research methods and statistics classes, I intentionally incorporate critical conversations about racism into these classes. Discussions of racism should not be reserved for courses with race in the title. Racism, especially anti-Blackness, is persistent and pervasive (Bell, 2008; Muhammad, 2010). As professors and mentors, we are complicit if we are silent about racism.

Quantitative Data are Not Objective

Students often tell me they want to take statistics because “statistics are objective” and they want to be able to learn “the truth” about their topic of interest. The problem is, as Joel Best (2001) puts in Damned Lies and Statistics: “all statistics are created through people’s actions: people have to decide what to count and how to count it, people have to do the counting and the other calculations, and people have to interpret the resulting statistics, to decide what the numbers mean” (p.27). As long as humans are involved, there is nothing objective about statistics.

In the late 1800s, Francis Galton, a cousin of Charles Darwin, was the first to use statistics to attempt to support racist eugenic practices (Roberts, 2011; Zuberi, 2001). Galton believed that intelligence is inherited. He is considered the “father of social statistics” as he was the first to apply evolutionary biology to statistical logic to develop his racial theory of eugenics (Zuberi, 2001, p.33). He thought that if he could demonstrate correlations between race and intelligence, he could advance his racist ideas that “human evolution could be accelerated by a self-conscious policy of selective mating practices” (Zuberi, 2001, p.34), (e.g., forced sterilization of “inferior” people) (Roberts, 2011).

Galton failed to confirm his hypothesis of the intellectual inferiority of people of color, yet in 1905 Alfred Binet and Theodore Simon built on Galton’s work to create an IQ test (Kendi, 2019). Then, a decade later, the IQ test was used widely in the U.S. by eugenicist Lewis Terman to attempt to demonstrate “enormously significant racial differences in general intelligence” (as cited in Kendi, 2019, p.102). Building on Terman’s work, Carl Brigham created the Scholastic Aptitude Test (SAT) to “reveal the natural intellectual ability of White people” (as cited in Kendi, 2019, p.102).

More than 100 years later, researchers and professors continue to point to IQ tests, grades, and standardized tests to demonstrate the “achievement gap” and institutions of higher education continue to use them to decide whether a student is worthy of admission. Yet, these quantitative “indicators” of intelligence or achievement are inherently racist assessments – and the “achievement gap” is indicative of a more appropriately named “education debt,” or the cumulative impact of disparities in access to opportunities and resources (Ladson-Billings, 2006).

When teaching students, it is important to address this history of statistics. It is also important to reflect on the construct validity of measures from an anti-racist lens. I know I’ve been guilty of using test scores as a variable when I need to demonstrate a concept with a continuous variable. However, from an anti-racist perspective, when you use test scores or GPA, you are measuring something other than intelligence – you are measuring opportunity, privilege, and access to resources.

Race is Not a Control Variable

“Race is not a biological category that is politically charged. It is a political category that has been disguised as a biological one.” (Roberts, 2011, p.4)

When talking to students about multivariate models and including race as a control variable, you should ask, what is it, exactly, that they are trying to measure? Perhaps they attempting to measure differences among groups of people with lived experiences of racism? Or the effects of racism? Or racism? Race is a socially constructed and not fixed (James, 2008), or homogeneous (Holland, 2008). As James (2008) writes in the edited volume titled White Logic, White Methods:

Often race is presented as a ‘demographic’ or ‘control’ variable, implying a theoretical neutrality not supported by the substance of the arguments or techniques used in the research. In this way, race has become, to use a bad pun, a ‘black hole’ of social scientific research… The use of race as a control variable flattens out the meanings of racial differences and replaces it with a generic notion of difference. This technique represents a seemingly theoretical and presumed neutral usage of race. However, using race as an independent variable without any contextualization or explanation implies that the causal mechanism for social differences lies in the categories themselves. (p.43).

When a student asks, what is the effect of race on, say, incarceration rates? They are (likely) seeking to measure the impact of racism or inequality. It is not race that affects incarceration or police shootings. It is racism. Therefore, in our research methods and statistics courses, it is our job to work with our students to determine what should be in the model, other than race – or as an interaction with race (Holland, 2008; Stewart & Sewell, 2011). Examining interaction effects can help us understand the complexity of the experiences of the people in our sample – and their experiences with political and structural inequality – from an intersectional lens (Bowleg, 2008).

When Parsimony Introduces More Complexity

I still cringe when I remember a paper I co-presented at American Evaluation Association early in my career. The aim of our paper was to analyze whether there were differential impacts of a bystander intervention program for students from various racial or ethnic identities. We were excited to share our results, given that our sample was majority students of color (which was rare for bystander intervention research).

The panelists before us presented on all the reasons why collapsing racial categories is problematic in evaluation research, with an emphasis on the diverse cultural context of Asian Americans, Native Hawaiians, and Pacific Islanders. Then we presented our paper – an evaluation – with 11 racial/ethnic categories collapsed into five: Asian, Black, Latino, White, and Other. We did this so that we would have enough power to run the analyses – but, to be honest, we did it blindly without necessarily considering how collapsing might affect the interpretations of our findings. During the Q&A portion of our session, presenters and attendees engaged in a powerful conversation about the dilution of racial and ethnic identities in social research.

The lack of homogeneity among racial and ethnic identities may make any conclusions we attempt to make meaningless. For example, Asia has 4.5 billion people in 48 countries with more than 2300 languages spoken and several religious identities. In addition, racial and ethnic categories are inherently dynamic, as social constructions. Over the last 200+ years, the US Census’s racial categories have changed at least 24 times (Roberts, 2011).

We knew from the data for our AEA that there were immigrant and US-born students in each category, and that the “other” category included multiracial, not specified, and other races/ethnicities. People who share a racial or ethnic identity may have common experiences with discrimination, but from an intersectional lens (Crenshaw, 1989), it is not very meaningful to only include a dichotomous (0/1) race variable without interacting it with or considering the identity in the context of other variables (e.g., gender identity, sexual orientation, socioeconomic class, religious identity, nationality, etc.). As Bowleg (2008) writes:

the key interpretative task is to derive meaning from the observed data on the one hand, and to on the other, interpret this individual level data within a larger sociohistorical context of structural inequality that may not be explicit or directly observable in the data. (p.320)

Anti-Racist Practice in the Research Methods / Statistics Classroom 

As a white woman, I seek to engage in anti-racist practice and pedagogy (see Kishimoto, 2016) in my role as a faculty member. As examples, I decenter whiteness in my syllabus, challenge binary thinking in class discussions, interrogate positivist and problematic epistemic superiority, and help students understand how their positionality will affect the research questions they ask and how they interpret data. I facilitate a culture of respect and empathy with how I treat the students and my expectations of how they will treat one another. I emphasize student well-being, avoid spotlighting students of color in conversations about racism, and emphasize ‘the danger of a single story’ (Ngozi, 2009). I also engage in my own ongoing self-reflection and self-education because there is always more to learn and you do not know what you do not know.

As early as the 1890s, “white social scientists presented [data about ‘black criminality’] as objective, color-blind and incontrovertible,” (Muhammad, 2010, p.4) without attention to the structural conditions that mean low-income Black communities are underesourced and overpoliced (patterns that remain today). It is easy to talk about racism in research as a historical artifact (e.g., discussions of the Tuskegee Syphillis Study or Henrietta Lacks’ immortal cells) or an anomaly (e.g., the relatively recent misuse of Havasupai Tribe’s DNA samples). However, given that many focal points of criminological research involve marginalized or vulnerable populations who may not have the autonomy we think they have to consent to research, it is important to consider that IRB approval does not mean a study is ethical or beneficial to the community (Green, 2020). We continually face ethical issues related to vulnerabilities among the participants in our research studies, limits to confidentiality when someone may be harmed, witnessing or learning about illegal activity, and fulfilling our commitment to “do no harm.”

As research methods and statistics faculty, it is essential that racism and other forms of oppression are addressed throughout the semester as our biases affect what we choose to study (research design), how we operationalize concepts (measurement), how we gather information (data collection) and from whom (sampling), how we interpret information (data analysis), and how we share what we learned (dissemination).


In a casual conversation with a white male Chair of a criminology department, I asked, “What’s it like to be a Chair in the era of Black Lives Matter?” He responded, “I hadn’t thought about that.” Our field will continue to fail BIPOC people and communities if we are not actively thinking about – and intentionally engaging in – how to be anti-racist in our classrooms, our departments, our universities, our research, our families, and our communities. Ignoring racism – because you don’t think it’s relevant or it makes you too uncomfortable – harms our BIPOC students, our white students, our BIPOC colleagues, and the profession. If only I had a dime for every time I’ve heard someone say that “2020 is a reckoning.” I hope it is – as hope is what has kept me engaged in racial and social justice work for more than two decades. As Kendi (2020) writes, “Once we lose hope, we are guaranteed to lose. But if we ignore the odds and fight to create an antiracist world, then we give humanity a chance to one day survive, a chance to live in communion, a chance to be forever free.” (p.238)

For further reading, I recommend:

  • Applying Indigenous Research Methods by Sweeney Windchief (ISBN-13: 978-1138049062)
  • Black Feminism in Qualitative Inquiryby Venus E. Evans-Winters (ISBN-13: 978-1138486225) ​​
  • The Condemnation of Blackness: Race, Crime, and the Making of Modern Urban America by Khalil Gibran Muhammad (ISBN-13: 978-0674062115)
  • Data Feminism by Catherine Dignazio and Lauren Klein (ISBN-13: 978-0262044004)
  • Pedagogy of the Oppressed by Paulo Friere (ISBN-13: 978-0826412768)
  • Research is Ceremony by Shawn Wilson (ISBN-13: 978-1552662816)
  • Teaching to Transgress by bell hooks (ISBN-13: 978-0415908085)
  • Thicker than Blood: How racial statistics lie by Tukufu Zuberi (ISBN-13: 978-0816639090)
  • White logic, White methods: Racism and methodology by Tukufu Zuberi & Eduardo Bonilla-Silva (ISBN-13: 978-0742542815)


Bell, D. (1987). And we are not saved: The elusive quest for racial justice. Basic Books.

Best, J. (2001). Damned Lies and Statistics. Berkeley, CA: University of California Press.

Bonilla-Silva, E. & Zuberi, T. (2008). Toward a definition of white logic and white methods. In T. Zuberi & E. Bonilla-Silva (Eds.), White logic, White methods: Racism and methodology (pp.3 – 30). Lanham, MD: Rowman & Littlefield.

Bowleg, L. (2008). When Black + lesbian + woman ≠ Black lesbian woman: The methodological challenges of qualitative and quantitative intersectionality research. Sex Roles, 59, 312-325.

Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory, and antiracist politics. University of Chicago Legal Forum, 139, 139–167.

Green, A. (2020 Sept 8). A new research experiment in Kenya raises questions about ethics. Devex. Retrieved from

Holland, P. (2008). Causation and race. In T. Zuberi & E. Bonilla-Silva (Eds.), White logic, White methods: Racism and methodology (pp. 93 – 110). Lanham, MD: Rowman & Littlefield.

James, A. (2008). Making sense of race and racial classification. In T. Zuberi & E. Bonilla-Silva (Eds.), White logic, White methods: Racism and methodology (pp. 31 – 46). Lanham, MD: Rowman & Littlefield.

Kendi, I.X. How to be an antiracist. New York, NY: Penguin Random House.

Kishimoto, K. (2016). Anti-racist pedagogy: From faculty’s self-reflection to organizing within and beyond the classroom. Race, Ethnicity, and Education, 21(4), 540-554.

Ladson-Billings, G. (2006). From the achievement gap to the education debt: Understanding achievement in US schools. Educational Researcher, 35(7), 3-12.

Muhammad, K.G. (2010). The condemnation of Blackness: Race, crime, and the making of modern urban America. Cambridge, MA: Harvard University Press.

Ngozi, C. (2009). The danger of a single story. TEDGlobal. Retrieved from

Roberts, D. (2011). Fatal invention: How science, politics, and big business re-create race in the 21st century. New York, NY: The New Press.

Stewart, Q.T. & Sewell, A. (2011). Quantifying race: On methods for analyzing social inequality. In Stanfield, J. (Ed.), Rethinking race and ethnicity in research methods. Walnut Creek, CA: Left Coast Press.

Wallerstein, N. & Duran, B. (2006). Using community-based participatory research to address health disparities. Health Promotion Practice, 7(3), 312-323.

Zuberi, T. (2001). Thicker than blood: How racial statistics lie. Minneapolis, MN: University of Minnesota Press.

Author Biography:

Jane E. Palmer, Ph.D., M.S.W, is a professorial lecturer in the department of justice, law & criminology and the director of the Community-Based Research Scholars program at American University (AU) in Washington, DC. She is also a faculty affiliate at AU’s Antiracist Research & Policy Center, a faculty fellow with the Metropolitan Policy Center, and a non-resident fellow in community-engaged methods at the Urban Institute. At AU, she has been involved in a variety of anti-racist initiatives including helping to design the curriculum for a mandatory anti-oppression course for all first-year students, the development of the parameters for required upper-level “diversity and equity” courses, and an active participant in a faculty learning community on inclusive pedagogy.


Leave a Reply

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