When other researchers took a closer look, it turned out that in each session, prisoners without attorneys (who are less likely to be granted parole) were scheduled last, just before the breaks. After accounting for this and several smaller factors, the original conclusion could no longer be supported. [1]
https://www.goodreads.com/book/show/35412097-when
And direct from the author some insights from the book:
'Afternoons are the Bermuda Triangles of our days. Across many domains, the trough represents a danger zone for productivity, ethics, and health. Anesthesia is one example. Researchers at Duke Medical Center reviewed about 90,000 surgeries at the hospital and identified what they called “anesthetic adverse events”— either mistakes anesthesiologists made, harm they caused to patients, or both. Adverse events were significantly “more frequent for cases starting during the 3 p.m. and 4 p.m. hours.” The probability of a problem at 9 a.m. was about 1 percent. At 4 p.m., 4.2 percent. In other words, the chance of something going awry was four times greater during the trough than during the peak. On actual harm (not only a slip up but also something that hurts the patient), the probability at 8 a.m. was 0.3 percent — three-tenths of one percent. But at 3 p.m., the probability was 1 percent — one in every one hundred cases, a threefold increase. Afternoon circadian lows, the researchers concluded, impair physician vigilance and “affect human performance of complex tasks such as those required in anesthesia care.”'
https://www.linkedin.com/pulse/youre-more-likely-screw-up-af...
Ps: I, on the other hand, am really more productive and attentive to details after lunch and specially after 2pm. Mornings for me are just good for physical activities..
[1] https://journals.sagepub.com/doi/10.1177/1745691616652873
2. For income vs. IQ, there will be high variance. This is obvious.
3. He intentionally omits R-values for many of his plots to hide their predictive power. Even smaller R-values, given a large sample size and good p-test, show the power of a correlation.
> I am still attached to every study that I cited, and have not unbelieved them, to use Daniel Gilbert’s phrase. I would be happy to see each of them replicated in a large sample. The lesson I have learned, however, is that authors who review a field should be wary of using memorable results of underpowered studies as evidence for their claims.
So it sounds like stands by the claims in the book. If that's the case, then I'm not sure what would be in a second edition.
For example patients with more severe illness could be more eager to see the doctor asap, and therefore be more likely to take the morning appointments, skipping work?
Patients will probably also worry about doctors getting tired, and try to schedule accordingly. I know I do worry about that kind of thing.
No doubt doctors also do get tired, but perhaps, like the judges, they also arrange their schedules accordingly?