Devising good tests to measure scientific knowledge is not simple. Questions about values and attitudes can be asked again and again over the years because they will be understood the same way by everyone who hears them; for example, Dr. Miller's surveys regularly ask people whether they agree that science and technology make life change too fast (for years, about half of Americans have answered yes) or whether Americans depend too much on science and not enough on faith (ditto).
You do not need to know a dam bit about science to know that the industy can not even get insurance still. Good business sence tells you nukes are a bad idea.