I think there is some room in between those two.

At a minimum you'd expect a few more companies, more sources than just code review and code productivity metrics (this alone disqualifies the study because it centers on just one task: software development) etc.