Good input all, these are the kinds of holes/skepticism I want to hear before I head too far down my research path.

RA, understand your point about democracy change, but the dataset allows for regime change during the insurgency (Indonesia is a key case - autocracy to democracy)

To Bob's World, I would definitely opine that the focus of my paper would not be to justify forceable democratic change, but to determine why, out of 89 insurgencies (as classified by RAND) observed since 1945, there are 25 insurgent "wins" against autocracies, anocracies, monarchial, and colonial governments, but none against democracies.

Like the "democratic peace" question - I am curious as to the reason - is it a facet of democracy that prevents insurgent victory, or some other reason?

Again, my initial findings indicate that not only are democracies resistant to insurgent victory, they actually directly politcially settle LESS than other forms of goverment.