Evaluating Pedagogical Choices with an Eye Toward LGBTQ+ Students

by Zoe Rehnberg and Allison Theobold

In this post, we apply the three-question design decision framework outlined by Dr. Nicole Dalzell (in the previous blog post) to evaluate curricular and pedagogical choices that specifically impact lesbian, gay, bisexual, transgender, and queer (LGBTQ+) individuals in the classroom.

This post builds on conversations started by statistics and data science educators over the last five years. In their exposé of teaching randomization, Miller and Hardin (2019) pressed statistics educators to consider how discussions of “randomizing gender” (i.e., balancing gender through randomization) might affect transgender and non-binary students. Thornton, Green, and Benn (2019) continued this conversation, outlining “recommendations for making the statistics and data science community more inclusive and supportive of gender non-conforming and LGBTQ persons.” Drawing on the perspectives of non-cisgender identifying students, Rao, Mader, and Friedlander (2023) outlined gender inclusivity as “fundamentally about humanizing data,” a sentiment reiterated by Thornton et al. (2022). While these recommendations have been generally well received, it is important to note that some recommendations have been met with resistance.

In this post, we continue these conversations by presenting two concrete examples of pedagogical decisions that particularly affect how LGBTQ+ students learn in our classrooms. 

For each example, we will apply the three key questions (KQ) outlined by Dr. Dalzell:

• KQ1: What student needs are and are not supported by this design decision?
• KQ2: How can we adapt to reflect student needs not supported by the initial design?
• KQ3: Is applying this design decision sustainable for both me and the students?

Example 1: Using Data with Binary Gender

As statistics instructors, we can spend hours searching for datasets that perfectly illustrate the concepts we teach. Suppose we are working on creating a multiple linear regression lesson, and find a rich dataset that is well-suited to this topic. The dataset has a moderate sample size, an interesting application area, and a wide range of covariates to investigate. But then we dig a little deeper into the data and find a variable labeled “gender,” coded as female or male. Digging even deeper into how this variable was collected, we find that gender was “identified” by the researchers based on each subject’s picture (Hamermesh & Parker, 2005).

With this information in hand, we’re now faced with a decision about the effectiveness of using this dataset in class (KQ1). For some students, this will just be another dataset that is used to illustrate a concept. However, for transgender, non-binary, and intersex students, the inaccurate exchange of gender and sex and the incorrect representation of binary gender could impede their learning, making them feel alienated and unable to focus on the statistical concepts being taught.

Having identified students who are not supported by these data, we consider two possible adaptations (KQ2).

Adaptation 1: Pick a different dataset

We initially chose these data as a tool for teaching multiple linear regression. It is now clear, however, that this is not an effective tool for some of the students in our classrooms. Therefore, a clear adaptation to our original design decision is to find a different dataset to illustrate this concept.

Let’s now consider the sustainability of this adaptation (KQ3). First, it could be time-intensive to find a new dataset. This would particularly be true for educators teaching many courses, those teaching coordinated courses over which they have little autonomy, and those teaching more advanced topics for which illustrative datasets might be scarce. Additionally, most (if not all) statistics textbooks contain examples using binary gender. Thus, avoiding examples using binary gender would preclude the use of such a textbook in class.

Adaptation 2: Facilitate a class discussion about flaws in the data collection methods

This might be a better adaptation if we are short on time. If we inherited course material from a colleague and need to teach this material tomorrow, we might not have time to swap out the dataset. This could also be a preferred adaptation if you want to explicitly bring attention to the differences between gender and sex, and offer guidance on how these variables should be collected and used in an analysis.

One sustainability concern (KQ3) for this adaptation is fairly clear – this discussion takes classroom time. Many of our classes already contain too much content, so it might be tricky to find the time to devote to such a conversation. Additionally, we, as instructors, need the tools and confidence to facilitate this type of classroom discussion successfully. So, not everyone will feel prepared to take on this role.

Example 2: Asking for Student Pronouns

Sharing pronouns has become a common feature of today’s classrooms. Indeed, many teachers consider allowing students to share their pronouns as a way to make their classroom more “inclusive.” While many students are comfortable being open about their gender identity, not everyone is (KQ1). In fact, for certain members of the LGBTQ+ community, sharing pronouns is inherently risky. 

If you have even glanced at the news, you should be aware that members of the transgender community face prejudice on a daily basis—murder, ostracism, homelessness, denial of medical services. So, a transgender student coming into your classroom is already poised not to trust strangers, and a stranger asking for your pronouns is not always welcome. Even if you are a supportive ally of LGBTQ+ students, think about the faculty at your university. Can you say with certainty that every faculty member would make LGBTQ+ students feel welcome in their classroom? 

For these reasons, we believe asking students for their pronouns¹ should be done with care. Asking for a student’s pronouns has the advantage of helping students who aren’t cisgender feel seen. On the flip side, asking for a student’s pronouns has the disadvantage of making students who aren’t cisgender feel seen. Not every transgender and non-binary student is “out,” and those who are out to their friends may not be out to everyone. This adds another layer to why students might not feel comfortable sharing their pronouns with you. Finally, pronouns may change during your class. Students may see themselves as cisgender in August, but may have a different self-perception in December.

Therefore, our design decision—asking students to share their pronouns—might not support many of the students it aims to put at ease. So, we ask (KQ2), what adaptations could we make?

---

¹Notice we are referring to “pronouns” not “preferred pronouns.” A pronoun is like someone’s name, we don’t “prefer” people to refer to us as Zoe and Allison, those are our names.

Adaptation 1: Make it optional as part of an icebreaker

Icebreaker exercises at the beginning of a course work to establish trust between you (the instructor) and the students, and provide you with valuable information about each student. Thus, you could choose to ask students for their pronouns as part of an icebreaker activity at the beginning of the course, but this approach requires additional considerations. First, students should never feel pressured to share their pronouns. Pressuring non-binary and transgender students to share their pronouns puts students in a vulnerable position, effectively breaking any sense of trust between you and the student. So, the icebreaker activity should be written, not spoken, and stating your pronouns should be optional.  

This adaptation does, however, have some sustainability concerns (KQ3). First, as stated earlier, there is the possibility that a student’s pronouns could change during your class. Repeatedly asking for students’ pronouns is not sustainable from the instructor’s perspective, but only asking for pronouns once puts the burden on students to tell us if their pronouns have changed. Second, for instructors of large courses, reading (and remembering) student pronouns is not feasible, which leads us to our second adaptation.

Adaptation 2: Don’t use pronouns in your classroom

An alternative approach is removing pronouns from your classroom. This could be done in a few different ways. One option would be to exclusively use “they / them” pronouns for every student. This adaptation has the advantage of normalizing the use of they / them pronouns, but this also challenges student’s identities who prefer she / he / other pronouns. In addition, for many instructors it may take a great deal of mental energy (KQ3) to break the habit of using gendered pronouns.

Another approach would be to remove pronouns by instead using terms like “colleague” or “peer” when referring to members of the classroom. For example, suppose Student A responds to a question posed by you (the instructor) and Student B would like to ask a follow-up question to what Student A said. Rather than using a gendered pronoun, Student B could instead refer to Student A as “my colleague.” This adaptation has two distinct advantages; first it does not challenge student identities, and second it cultivates language surrounding a classroom community. Our colleague (🙂) Dr. Dalzell has used this approach, and takes it one step further by deciding as a classroom community how to refer to one another.

Conclusion

We have chosen to focus this post on design decisions that mainly affect LGBTQ+ students in our classrooms. This framework, however, is an invaluable tool for assessing all choices we make in our classrooms and how they affect students, and should be applied broadly for a variety of student identities.

---

Collaborators:

• Nicole Dalzell, PhD, Associate Teaching Professor at Wake Forest University

---

References

Miller, J. and Hardin, J. (2019). The evolution of variables and the existence of trans people, AMSTATNEWS. 

Thornton, S., Green, B., and Benn, E. (2019), Friends and allies: LGBT+ inclusion in statistics and data science. Significance, 16: 39-41. https://doi.org/10.1111/j.1740-9713.2019.01280.x

Thornton, S., Roy, D., Parry, S., Lalonde, D., Martinez, W., Ellis, R., and Corliss, D. (2022). Towards statistical best practices for gender and sex data. Significance, 19(1): 40-45. https://doi.org/10.1111/1740-9713.01614 

Rao, V., Mader, J., and Friedlander, E. (2023). Gender inclusive activities in an introductory statistics class [Poster]. USCOTS, State College, PA. https://www.causeweb.org/cause/uscots/uscots23/f02-gender-inclusive-activities-introductory-statistics-class

Hamermesh, D. S. and Parker, A. (2005). Beauty in the classroom: Instructors’ pulchritude and putative pedagogical productivity. Economics of Education Review, 24(4). https://doi.org/10.1016/j.econedurev.2004.07.013