I was once a political science major, and while it wasn't a huge waste of my time, I wouldn't exactly describe it as a positive experience. For one thing I quickly grew tired of, as a fellow science major put it, "classes where my opinion counts more than facts" secondly I quickly fell in love with geology after my first class, and changed from another directionless college student, to a dedicated geology student because it is simply more fun.
Well recently I came across a post at Cosmic Variance about this New York Times editorial. If you have seen either, the editorial basically said that poly sci researchers should not bother themselves with testing their hypothesis as unsubstantiated assertions work just as well. Well as you can expect every scientist who saw that article was understandably upset about it, and my main problem goes right back to the reason I dropped that major, Political Science isn't a science no matter how much they tell themselves otherwise. At best it should be called Political Studies, as you do study political systems but because you can't come up with any universal theories or even agreement within the discipline about important phenomena, like why do voters vote the way they do.
However, I can be somewhat sympathetic, as when political studies researches look at the success that physics has had in the past centuries it can seem daunting to try to replicate that work. It would be hard for just about any field to have analogous successes to Newtonian Mechanics or General Relativity, but it is important to remember the work it took to get there, and that the physicists' work hasn't finished. Although I would assert that political science (if it followed the scientific method in any sense or fashion) would resemble Geology more than anything.
You might wonder why I would say that. Well first off I don't mean to disparage geology at all by comparing it to political science; however, if political scientists acted like scientists, their research would probably be very similar to geology. First off the whole idea that the present is the key to the past would be a necessary part to this hypothetical political science as it is to geology. Presumably political scientists would have to look through much of history and retrace events that led to important political events, then they would have to hypothesize a cause for this change, find evidence to support their assertion WHILE PAYING SPECIAL ATTENTION TO EVIDENCE THAT CHALLENGES OR DISPROVES THEIR HYPOTHESIS. That was in all-caps because that is where political studies abandons all pretext of following the scientific method, and when they do provide evidence they cherry pick it, ignoring inconvenient facts. I know that it is human nature to do so, which is why political studies needs true peer reviewed journals if they want a legitimate claim to the "science" part of the field.
Political Scientists will say that the systems they study are too complicated to follow the scientific method, thus they should be exempt. However, geology is incredibly more complicated, it covers almost 4.5 billion years of Earth history, looking at evidence which has been changed into new and hard to read forms or in some case destroyed and now requires inference from analogous areas elsewhere in the world. Geologists try to understand how mountains formed, the history of life, and they retrace the movement of continents backwards in time; in addition, I would bet that nearly every geologist has seen a geologic cross-section that resembles a Jackson Pollack painting more than any stratigraphic layers deposited under Steno's laws; however, dedicated work by many real scientists can eventually clarify these confusing maps and turn them into coherent stories.
In the end, though if those involved in political studies want to be considered scientists they would have to start applying statistical methods and probably create mathematical and computer models to try and explain their research areas. In addition, they will have to learn to look at all of their previously hypothesis with a skeptical eye and evaluate them based off of the evidence not whether or not they align with their personal beliefs. This would require a tremendous amount of work to so completely change the discipline, and I doubt that those involved would be willing to do so. This is a shame as they could presumably discover some fascinating aspects of human nature, but until it does so I will refuse to call it a science.