Most of the negative examples are med/bio-related (including psychology here), whereas the single positive example is in CS/Engineering. Systematically med is d

Most of the negative examples are med/bio-related (including psychology here), whereas the single positive example is in CS/Engineering. Systematically med is dealing with a harder replication problem. Moreover, "light involvement from the academic system " is a significant stretch. Not all but most of the people doing the work have been trained within academia instead of within industry and the work has been from an academia-industry collaboration. Thinking about the proportion of people that go into academia vs industry makes the ratio more meaningful as well. The replication crisis is clearly a big issue, but the conclusions I would draw from the examples presented are more difference between fields than a condemnation of academia. To just say "move it to private instituions" is shortsighted.