That study turned out to be quite flawed, because the cases were ordered [1].
In particular, cases were grouped by prison and within each prison, prisoners without an attorney went last. As the well-known saying goes, "the man who represents himself has a fool for a client": the pro se prisoners fare far worse than those with a lawyer. Judges tried to complete an entire prison before taking a meal break, and therefore ended one session with the statistically weakest cases and began the next with stronger set, thereby guaranteeing the result.
If you look at the original data, some other oddities pop out. A physiological process, like becoming hungry, should depend more on the wallclock time than the number of cases heard. However "However, note that in an analysis that included both the cumulative minutes variable and the ordinal position counter, only the latter was significant."
Ah, thank you! The more I look into a lot of these (in)famous behavioral psychology studies, many don’t hold up or aren’t replicable. As a layman who is simply interested in the topic, I appreciate the correction and insight!
In particular, cases were grouped by prison and within each prison, prisoners without an attorney went last. As the well-known saying goes, "the man who represents himself has a fool for a client": the pro se prisoners fare far worse than those with a lawyer. Judges tried to complete an entire prison before taking a meal break, and therefore ended one session with the statistically weakest cases and began the next with stronger set, thereby guaranteeing the result.
If you look at the original data, some other oddities pop out. A physiological process, like becoming hungry, should depend more on the wallclock time than the number of cases heard. However "However, note that in an analysis that included both the cumulative minutes variable and the ordinal position counter, only the latter was significant."
[1] https://www.pnas.org/content/108/42/E833?ijkey=21207497c684f...