The idea that raw data is good is rather pervasive. Obviously "raw" data has not been tampered with, massaged, excluded, modified, or diminished. Raw data contains all possible information and none of the potential prejudices of those who recorded the data. This is true to a certain extent, but often times the raw data is raw because it requires expert analysis, review and understanding to have the meaningful bits extracted. In science the amount of raw data generated is phenomenal. In the absence of careful analysis there is no point to it. It is not publishable, it is not relevant, it is, by itself, nothing. When someone with expertise and understanding analyzes the data, however, and extracts the meaningful aspects, then shows and explains those to others, there is a crystallization of the data. A form appears from nothing and others can build more off of it.
This is what scientists do (as opposed to technicians who may or may not be scientists), they construct form from raw data. Journalists are supposed to do something similar. If reporting with no analysis, or insight is the same a as journalism, there is no expertise associated with being a journalist. Parroting and repeating things reduces journalists to base reporters, automated mimeographs. This is not always bad, mind you. I would love for there to be real reporting on Iraq, for example. But when it comes to some things, like politics, and (ostensibly) scientific work, journalists are a must.