I think the NSF's numbers are pretty good, and I'd bet that they do track your field (unless you are a non-scientist). The problem is likely that the NSF decided to get out of the business of analyzing their own data apart from doing Science and Engineering Indicators. Unfortunately, there isn't anyone really doing analysis who caters to a student / postdoc audience. The big hole in NSF's data on science careers is that they don't track anybody who didn't get their PhD in the US, which means that they don't know anything about the majority of postdocs.
As for US numbers vs foreign, I know that estimates for China and India's degree production have been questionable and that Vivek Wadhwa has been doing some work on getting them right. I hadn't heard about the Gathering Storm screwup, though - do you have a pointer anywhere?
I think the NSF's numbers are pretty good, and I'd bet that they do track your field (unless you are a non-scientist). The problem is likely that the NSF decided to get out of the business of analyzing their own data apart from doing Science and Engineering Indicators. Unfortunately, there isn't anyone really doing analysis who caters to a student / postdoc audience. The big hole in NSF's data on science careers is that they don't track anybody who didn't get their PhD in the US, which means that they don't know anything about the majority of postdocs.
As for US numbers vs foreign, I know that estimates for China and India's degree production have been questionable and that Vivek Wadhwa has been doing some work on getting them right. I hadn't heard about the Gathering Storm screwup, though - do you have a pointer anywhere?