PT152 · S1 · Q19 Researcher: In an experiment, 500 families

Yes, the author's causal story does indeed assume a difference between the book-group and the non-book group. If the non-book group also had medical self-help books, then the author would need some other way to explain the 20% difference. Ultimately, this answer is just less direct of a criticism than the correct answer is. The illegal backwards move and the failure to consider alternate explanations for why the book-group had 20% fewer visits are the two big reasoning flaws. This objection is kind of just nitpicking at a premise, making it seem like there wasn't a difference between the two groups in the first place. That doesn't provide a different way to explain the 20% difference (if both groups had self-help books, then we have no idea why one group is visiting docs 20% less than the other). If we had an argument that said, "Jumping off a tall building will lead to a broken bone. Jim has a broken bone. Thus he must have jumped off a tall building", our primary objections would be the backwards logic or the failure to consider alternate explanations, not "it is possible that a tall building has a lot of bushes at ground level to cushion one's fall".

This is similar to (A), but it doesn't do a great job telling us which families we're even talking about. In order for this to weaken, we'd need to know that the families in the book-group got medical self-help info from some other source, and that info is why they saw the doc 20% less. That would undermine the argument that medical self-help books improve family health by making it seem like the medical self-help book didn't have anything to do with the improved family health. This answer doesn't discriminate between the two groups, so it doesn't give us an alternative way to explain the data. And the author didn't need to assume that people's only access to medical self-help info is books.

In order for saying, "Hey, author, X could contribute to Y and Z" to be an objection, our author would have needed to be thinking that a certain causal factor only caused one thing. Our author does think that having a self-help book caused improved family health, and that improved family health caused 20% fewer trips to the doctor. So our author actually already accepts that a state of affair (having a self-help book) could causally contribute (directly or indirectly) to two or more effects.

If this argument hadn't been in the context of an experiment, and instead had just been a correlation like "People who own medical self-help books are less likely than those who don't to visit the doctor", then this answer could have pointed out the reverse causality possibility that FIRST came the healthy underlying factors, THEN came buying the medical self-help book. But for this argument, we know that people are given the book as part of the experiment, so it removes this line of objection.