When Parnas refers to overly large groups, he explains “Academics with large groups who spend little time with each student, but put their name on all of their students’ papers, will rank above those who work intensively with a few students.”

On the other hand, there are many situations in which multi-author papers are counted less than single-author ones. This has been cited as an impediment for interdisciplinary collaboration.

Perhaps this evaluation of research is like Consumer Reports evaluating chocolate (which they’ve done): it’s relatively straightforward to recognize crap, but after that it really depends on the consumer. This is excepting the Nobel, Fields, and Turing level stuff, which it’s likely won’t be recognized until after the publication in question is no longer relevant for evaluation purposes (at least, that’s what I keep telling myself).