Political science with scientific rigor? Hah!

Fred Schwarz writes in NRO's The Corner that The American Political Science Association (APSA) is pretending that what it is doing is "science applied to politics". This sort of thing always ruffles my feathers because it's clearly just going through the scientific motions without actually bothering to establish the required base of knowledge that any scientific examination would require.  Here's the comment I left there. 

Statistical rigor? These are pretensions. We are missing some of the very basic prerequisites to achieve any sort of scientific examination of the american political scene. 
Try to find:
1. A comprehensive list of all governments in the US (according to the contradictory definitions of our collective 51 sovereign entities that can, and do, define them).
2. The jurisdictional bounds of all governments
3. A list of what each government does
4. A comprehensive list of the officeholders for all governments
You cannot write, scientifically, about who is doing government, much less politics, what they are doing, and where they are doing it with any sort of numerical precision without these prerequisites, and a number of other basics besides. For instance any comprehensive scientific fiscal analysis would require a meta chart of accounts so you could do apples to apples fiscal comparisons across governments. We can't do that either. 
There are practical consequences for this lack of scientific rigor prerequisites. People get arrested for accidentally driving into a gun hostile jurisdiction. There's no app for that but with items 1, 2, and some simple survey work, one could be made. A police militarization app could be similarly constructed. The list of practical improvements possible is rather long.