CAN OUTSIDERS ANALYZE COMPLEX DATA MEDICAL DATASETS? Ornstein writes about an interview with the editor-in-chief of the New England Journal of Medicine: “The worry he was initially trying to articulate, Drazen said in an interview, is that scientists not involved with original research will swoop in, conduct additional analyses (perhaps without understanding the data) and then take credit from those who spent months or years working on the underlying research.”
‘The datasets are very, very, very complex,’ [Drazen] said. ‘You don’t want someone to analyze the dataset not fully understanding it.'”
I find this sentence even more troubling than open and scornful opposition to replication studies. It implies that not all the parameters of some original studies are published. The methods of analyzing the data of some studies are apparently critical to the results but are not made public. It is presumed that somebody on the outside would not be able to understand how the data is analyzed—or “manipulated” (to use a perhaps excessively harsh term.)