Instructor Notes — Chapter 41
Part VIII opener. Difficulty: beginner. Prerequisites: Chapters 6, 8, 12. Estimated contact time: one 90-minute session, or two 50-minute sessions split at §41.5.
What this chapter is for
Chapter 41 is the pivot from evaluating a drug to evaluating an argument about a drug. Students arrive from Chapter 40 with a complete method and a completed dossier. This chapter shows them the information environment that method has to survive in.
Its intellectual payload is small and sharp: cultural claims have populations, endpoints, and falsifiers, and they get rated the same way pharmacological claims do — except for the ones that aren't empirical claims at all, and recognizing which is which is the skill.
Everything else in the chapter serves that.
The three things students must leave with
- One word covering multiple non-interchangeable products is a safety problem, not a style problem. If they leave saying "semaglutide" and "which product?", the chapter worked.
- Speculation about individuals is unfalsifiable, which is a stronger objection than "it's mean." The epistemic framing is deliberate; it reaches students who reject the moral framing.
- A joke transmits a social fact, not information. And the format prevents evaluation.
Teaching sequence that works
Open with the labels, not the culture. Put the §41.2 table on the board with the brand column blank and have students fill it in. Most classes cannot. Then reveal it. The genuine surprise — "wait, those two are the same drug?" — does more than any lecture, and it inoculates against the common student reaction that this chapter is a media-criticism digression.
Do the base-rate arithmetic on the board. Slowly, with the numbers visible. The 80%/80%/10% example in §41.3 lands hard because students expect "80% accurate" to mean "usually right." Watching 26 flags produce 8 correct answers is the moment the ❌ rating becomes obviously correct rather than merely asserted. Emphasize that the numbers are invented for the demonstration.
Run §41.4 as a structured disagreement, never as a debate to be won. Assign the for case to students who instinctively oppose it and vice versa. Then reveal that the chapter refuses to decide, and ask whether that refusal is a cop-out. It is a good argument and it should be had.
Save §41.5 for last if you split the session. It is the chapter's strongest passage and it needs room.
Where students reliably go wrong
They collapse the two facial-volume claims. Expect "but it IS a real thing, my aunt's face changed." Correct gently and structurally: yes, and that is claim one, which is well supported. The rating is on claim two, the backward inference. Rule 6 — one phenomenon, two claims, two ratings — is the fix, and this is the best example of it in the whole book.
They hear §41.7 as a concession rather than a finding. Some students will treat the "what the culture gets right" section as the author being fair-minded. Push back: it is a substantive claim that culture accomplished a public-understanding transfer that decades of deliberate education did not. That is uncomfortable for a science course and should be sat with.
They convert the no-names rule into a discussion ban. It is not. Students can discuss types, patterns, mechanisms, industries, and platforms in full detail. What they cannot do is identify an individual. If a discussion stalls, it is usually because a student had only an example and no argument — which is diagnostic.
They reach for "everyone knows." Have the §41.5 hedges list visible during discussion. Point at it rather than arguing.
They over-apply the epistemic argument. A few students will conclude that no inference about anyone is ever permissible. Clarify: the objection is to confident conclusions where no observation could disconfirm. Clinicians infer about individuals constantly — with examination, history, and a reference standard. The difference is the evidence available, not the act of inferring.
Assessment guidance
- Exercises §C ("Evaluate this claim") are the graded core. Award marks for stating the claim precisely enough to be rateable — population, endpoint, timeframe — before any rating is assigned. A correct rating with a vague claim should score below an arguable rating with a precise one.
- Exercise 41.22 and quiz item 11 both test NOT RATED. Watch for students who treat NOT RATED as a weaker ❌. That is the single most common conceptual failure in this chapter.
- Exercise 41.26 is the best short discriminator between students who have absorbed the method and students who have memorized the ratings.
- Case Study 41.1 question 5 (what if a good diagnostic study existed?) separates students who understand the argument's structure from those who have adopted its conclusion. Good answers notice that stages 3 and 4 would be partially rehabilitated and stage 4's concealment allegation would not be.
- Do not grade Exercise 41.32 for content. It asks for an honest self-report about a hard habit; grade for completion only, and do not require students to share it.
Timing and cuts
If you have only 50 minutes: §41.2, §41.3, §41.5. Those three carry the chapter. §41.6 can be assigned as reading with quiz item 16 as the check. §41.7 must not be cut — a version of this chapter without it is exactly the unfair chapter the text warns against, and students will notice.
A note on classroom safety
Some students in the room are taking these drugs. Some are taking insulin. Some have been the subject of exactly the speculation this chapter analyzes. Set the no-names rule at the top of the session, extend it explicitly to people in the room and their families, and do not invite personal disclosure in either direction. If a student volunteers, thank them and move on without building the session around it. §41.8's justification tax is not an abstraction for everyone present.