Epistemic violence in STEM textbooks: A Feminist critique of biostatistics

Kulyash Zhumadilova
03/27/2023 | Reflections

One theory concerning the etiology of breast cancer states that women in a given age group who give birth to their first child relatively late in life (after age 30) are at greater risk for eventually developing breast cancer over some time period t than are women who give birth to their first child early in life (before age 20). Because women in upper social classes tend to have children later, this theory has been used to explain why these women have a higher risk of developing breast cancer than women in lower social classes. To test this hypothesis, we might identify … . (Rosner 2015, p. 42)

The passage above is the first paragraph of Chapter 3 “Probability” in Bernard Rosner’s “Fundamentals of Biostatistics” textbook. Now in its eighth edition, this textbook is very common in introductory biostatistics classes. In fact, it was the one I used during junior year at college. By sharing my personal experience, I hope to raise awareness in the STS community towards critical analysis of textbooks and epistemic violence that is committed by some of them.

My junior year was stressful: grad school applications, study burn-out, and all-consuming thoughts on what I should do after graduation. I remember working hard on problems in biostatistics class, constantly reading and rereading the textbook to get a better understanding. But each time I opened the book, I noticed a feeling of irritation.

The textbook had a disproportionate number of problems that utilize women specific cancer datasets. Although as a future professional scientist I tried to maintain a distance, it inevitably stirred worries and doubt. I learned then that the author of the text, Harvard professor Bernard Rosner, specializes in breast cancer modeling and his book uses examples and data from his research. The table below shows word search count on the latest edition of ebook. It is evident that breast cancer is mentioned disproportionately more frequently than other types of the most common cancers and comprises nearly half of all cancer examples.

An analysis of the frequency of certain words and word combination in Bernard Rosner’s “Fundamentals of Biostatistics.”
Word/word combination	Total count
“cancer”	831
"breast cancer"	368
"lung cancer"	65
"prostate cancer"	6
"skin cancer"	9
"bladder cancer"	16

The intellectual challenges of statistics are strenuous enough without the constant reminder (from your textbook no less) that in the act of exercising reproductive rights by choosing to focus on one's career, could have a negative impact on your health. STEM careers, in particular, require a significant time investment with the majority of professionals spending their twenties in graduate and postgraduate training. Reorganization of the whole STEM education-workplace pipeline is needed to make it more family friendly, but until then the “ticking clock” argument will only lead to more attrition of women from STEM. According to the textbook, educated, high income women get breast cancer more often. This vision of the future of professional women is in sharp contrast with aspirations of a female student in an upper level biostatistic class, who works hard to be in STEM. It challenges her ambitions and plants a seed of doubt.

There are a few interesting theoretical points that this skewed presentation raises which I want to survey here by analyzing my emotional response from a critical distance and tools of STS. I was inspired by the approach of feminist scholars, such as Emily Martin and Patricia A. Gowaty who examined scientific narratives and experiments for implicit biases.

Breast cancer is a serious and, unfortunately, a prevalent issue. According to the American Cancer Society, breast cancer accounts for a third of the new cancer diagnoses in women each year. The diagnosis usually happens later in life (the median age is 62). Although no definite causes have been discovered, research has linked it to genetics, lifestyle, and reproductive history. Thus, it is important to educate women on risk factors of breast cancer. However, unsolicited and intrusive rhetorical forms that show up in problems in a biostatistics textbook do not inform but rather induce normative prescriptions of behavior and pathologize reproductive choices.

Although the example from above acknowledges class difference, it ascribes it to values concerning reproductive choices by emphasizing difference in average age of first birth. Such idealism, however, disregards substantial material differences between high and low income lifestyles. Since 1996 studies sponsored by the Silent Spring Institute, a scientific research organization dedicated to uncovering the environmental causes of breast cancer, are looking into endocrine disruptors as a cause. A 1999 study surveyed women from low and high income standing and with high and low breast cancer incidence. Although no definite results were produced, the study suggested that chemical exposure (pesticide use, professional lawn service, dry cleaning, etc.) rather than solely breastfeeding is responsible for higher breast cancer rates. It is interesting that even though the current edition of the textbook was published long after these studies, no updates have been made to suggest this complexity in breast cancer rates and lifestyle.

I anticipate how some will object by insisting that I should have treated the textbook examples from a “scholarly distance” without inferring them to my personal life. But is it possible to maintain a distance when what you study relates to your life directly (or perhaps pretends to)? With the authority that we as a society have granted to science, it has the power to influence how people think and see the world. This is especially true for those who aspire to be scientists and adhere to its realism. Demanding courses like biostatistics required considerable practice, so I would encounter an example on breast cancer almost every week for four months. You don’t have to be a psychoanalyst to see that it would “stick”.

The fundamental issue here is the interconnection between knowledge and experience. Why did no one – editors, reviewers, focus groups, professors – question how women who would read this textbook could feel? Women in STEM need support and encouragement, not didactic discouragement or moralizing lessons about the harms of pursuing professional self-realization. The biostatistics class was challenging and filling it with breast cancer examples only made it worse.

References:
Rosner, Bernard. Fundamentals of Biostatistics. 5th ed. Duxbury, 2000.
Rosner, Bernard. Fundamentals of Biostatistics. 8th ed. Boston: Cengage Learning, 2015.

Kulyash Zhumadilova is a doctoral candidate in Science, Technology and Society (STS) at Virginia Tech. She received her MSc in Life Sciences from Skoltech and BSc in Biology from Nazarbayev University. Kulyash’s primary research interest is history and philosophy of molecular biology, but she also engages with broader issues in STS, political economy and academia.

Published: 03/27/2023