Instructor Notes — Chapter 42
Suggested allocation: one 90-minute session, or two 50-minute sessions split at §42.5. The material is not technically difficult; the difficulty is entirely about maintaining a distinction that students find unstable.
The one thing to protect
This chapter fails in a specific, predictable way: students convert it into permission to dismiss.
The failure looks like competence. A student learns to spot an affiliate link, spots one, and announces the claim is false. They have not misunderstood the chapter — they have compressed it, in exactly the way §42.4 describes content being compressed, and for the same reason: the compressed version is more satisfying and easier to deploy.
Everything in this session should be organized around holding two propositions at once:
- Incentive predicts which claims get made. (True, important, the chapter's contribution.)
- Incentive does not establish that any given claim is false. (Also true, and the first thing students drop.)
Rating 4 in §42.9 exists to make this examinable. If your students can state why "an affiliate relationship predicts a creator's claims are inaccurate" earns ❌ in the same chapter that spends 9,000 words on affiliate relationships, they have the chapter.
Common misconceptions, and where they surface
"So telehealth is bad." Surfaces within ten minutes, every time. §42.2 is written specifically to resist it and students still arrive there, because "both halves at once" is cognitively expensive and "pick a side" is free. Counter by making them argue the access half first, on the record, before any critique is allowed.
"So disclosure should be mandatory." It largely already is. The point of §42.3 is not that disclosure is missing but that it is the wrong instrument for the dominant problem. Students who reach for stronger disclosure have not yet absorbed the selection effect.
"Platforms should just rank by accuracy." The best teachable moment in the chapter. Ask them to specify the input. They cannot, because it does not exist as a property of the file. This converts a policy preference into an engineering impossibility in about ninety seconds, and it lands hard.
"The compounding thing was just fraud." §42.8 and Case Study 1 §7 push back. Much of it was legal at the time, the demand was real, and the patients were responding to a pricing and coverage failure. Students who cannot say this back have absorbed the chapter as a morality tale.
"Uncompensated sources are trustworthy." The symmetric error, and it appears later than the first one — usually in the dossier work. Unpaid people are wrong constantly. Independence removes one distorting pressure and supplies no expertise.
Sequencing that works
- Open with the funnel diagram on the board, unlabeled, and have the class fill in who is paid at each step before reading §42.1's version. Ten minutes. They will get steps 1–4 and miss step 5, and discovering the omission is worth more than being told.
- Then the friction inversion. Ask for examples of friction in care that are protective. Do not accept the first three; push until someone names follow-up.
- §42.2 with the access half first, argued by the students, not by you.
- Break, if splitting the session.
- §42.4–42.5 as a pair. Run the compression ladder live on a sentence from an earlier chapter.
- Close on §42.9, and make them run the five questions on something congenial. The calibration test is the assessment that matters.
Assessment guidance
- Exercises AG (read applied to congenial content) and AI (interest map) are the two that distinguish understanding from recitation. Weight accordingly.
- Quiz items 7, 15, and 21 are the discriminating items. A student who misses all three has the compressed version of the chapter.
- Any submission concluding "therefore they are lying" from a financial interest alone should be
marked down regardless of quality, per the global rule in
answers/ch42.md.
Sensitivity notes
- Some students in the room are patients. Anyone who obtained treatment through the structures described may hear the chapter as an accusation about their own judgment. Name this at the start. Case Study 1 §7 is the text to point at: those patients were behaving rationally under constraints somebody else set.
- Some students are creators. In any large class, several make sponsored content. The chapter does not require anyone to conclude they have behaved badly, and the discussion should not either.
- Some are clinicians or clinicians-in-training, and §42.6 concerns them directly. The chapter's position is explicitly not "clinicians should not make content." Say so out loud.
- Do not let names into the room. Students will offer them. Redirect to the role: "so, a fitness creator with an affiliate arrangement — what does that predict?" This is a rule about the classroom, not just the text, and enforcing it is itself instructive.