Connecting content to student experience
The examples all come from one country and one kind of question
Every worked example is US data on a disease none of your students study, and the ones who came here for something else quietly conclude the method is not for them.
What it is
Choosing examples and readings so that students can see their own questions in the method, and being explicit about why the reading list is what it is.
The fellowship framed this under recognizing the whole student, alongside contextual teaching and diverse representation. Two moves sit inside it: entering a topic through something students already have, and widening whose work appears on the list, including saying what you left off and why.
Run it
- Survey their interests in week one. One question about what they came here to study, and use the answers when you pick examples.
- Enter each topic through prior knowledge. Their own thesis data, their prior field, the disease they came here because of.
- Widen the examples past one country and one literature, and say out loud why the list is what it is.
- Name the history where it is methodological, not only ethical. Study designs that harmed communities usually also produced bad estimates, and that is a methods point as much as a moral one.
Say this
Week one, one question
“Before the second session, send me two sentences: what question brought you into this programme, and what data you expect to work with. I use these to pick the worked examples, so the more specific you are the more of your own work you will see on the slides.”
In my courses
Planned for EPI 601, odd weeks.
I will use real-world epidemiologic examples from diverse populations and global contexts (not only U.S.-centric studies) to illustrate methodological concepts. When discussing study design, I will include examples from students’ own research areas when possible, drawing on a brief survey of their interests in Week 1. I will acknowledge that epidemiologic methods have historically been applied in ways that have harmed certain communities, and discuss how improved methods can address these harms.
The 601 reading list carries this deliberately: work from Rwanda, neighbourhood-health research, and trigger studies alongside the standard sources. The note attached to this practice is that a course on causality and validity has natural openings for it, since historical designs such as Tuskegee and early environmental-justice research raise methodological questions and ethical ones at the same time. New for 601; not used before the fellowship.
Evidence
No study supports this as a discrete practice, and the fellowship materials offer none.
The adjacent evidence is about who feels able to participate rather than about examples. A cross-sectional survey in high-enrollment biology found 78 percent of students believed the instructor knew their name while instructors actually knew 53 percent (Cooper et al., 2017). Work across 56 instructors and more than 4,900 students found relationship-building talk predicted peer connectedness, with effects varying in direction across intersectional identities (Jackson et al., 2026; Seidel et al., 2015). Instructor immediacy correlates with affective learning and much more weakly with cognitive learning (Witt et al., 2004).
None of that is a randomized test of choosing different examples, and none of it should be presented as one.