To call a discipline a science, it needs to use the scientific method, not just sometimes, but always. It needs testable hypotheses. Psychology sometimes has this, but often it does not. There is certainly reason for non-science fields to use scientific methods at times, but that doesn't necessarily mean those fields should be called sciences.
An illustrative example is the difference between "medicine" and "medical science". Your doctor has studied medicine. He or she, in addition to studying some medical science, has learned interview techniques, psychology, and various other aspects of a craft that are, in no way, scientific. If you talk to a doctor in a non-medical setting (many are specifically trained not to reveal ignorance to patients in order to maintain patient confidence), you'll find they're remarkably ignorant about the how's or why's of the human body, except when it comes to something they've been trained to spot and fix. A huge portion of their training centers on knowing when to do nothing at all, because medical intervention almost always carries it's own risks. In his or her daily job, your doctor does not employ the scientific method. At least, you should probably hope you are not being experimented on by your family physician! Some doctors do research in the field of "medical science", but this really is an entirely different job from being a family physician, surgeon, etc.. Medicine is a highly skilled craft that sometimes uses science, but it is not itself a science.
Also, your assertion that to call a discipline a science, it needs to use the scientific method, is well... demonstrably false, simply by virtue of the fact that we are having this debate.
The scientific method produces very reliable knowledge, but it is not the only way to produce knowledge that is reliable. There is also knowledge that we rely on that is not as rock solid as knowledge obtained from the scientific method, but which is still valuable and still falls under the realm of science, because it is still part of the systematic pursuit of progressively more reliable knowledge.
Yeah, as evidenced by
http://www.nature.com/news/over-half-of-psychology-studies-f...
Indeed! Just look at the puny and half-wrong progress made in physics and compare it to the enormous progress made in astrology and psychology which transformed our lives and understanding of everything!
I mean, they've been at it for hundreds of years, and they still haven't realized that "it is not the only way to produce knowledge that is reliable" (never mind what "reliable" means, who cares about such details anyway). Wish they were also blessed with magic ball.
And hey! They've been sucking our tax money like vampires for so many years!
But what do you mean "used"?
In politics, the term "science" is extremely potent. Claiming that science is on your side is commonly done to dramatic effect in order to sway opinions. This effect is very useful in those cases where science tells us something irrefutably important, such as the future of the temperature of Earth's atmosphere or the effectiveness of medical interventions. But it's also very abusable when the standard of proof is relaxed, since it becomes easy to use defeasible reasoning to produce "science" which supports someone's ideology with the borrowed charisma of the physics department.
And now, my favorite example:
http://thelastpsychiatrist.com/2010/06/are_certain_behaviors...
So no science as a field should remain well defined in scope and meaning, in the same way crafting shouldn't be blurred with engineering.
Oh, so when science is defined in the way Popper defines it, it is "arbitrary confines", but when it is defined in a way to legitimize cargo cult science, it is "well defined".
And no, it's not just that majority of psychology results are "not easily reproducible".
Furthermore, again, science isn't just data fitting and extrapolation. A real theory gives you a lot more than what you put into it, including novel phenomena. Social "sciences" have been playing science for a long time, and still they don't have any universal laws or any deep understanding of what they're studying. I'm not saying data collection, interpolation and extrapolation (or as Rutherford put it, "stamp collecting") isn't a bad or useless thing. It just isn't science.
The key point here is that science is not limited to the scientific method. The scientific method just produces more reliable knowledge, but is not the be all end all of science, it's just a limited subset, just like the "hard" sciences are a limited subset of science
This sound a lot like trolling, but I'll bite for this once.
You can't have less/more reliable knowledge. Science can't be half-wrong and half-right. You're either right or you're wrong. Your theory can't be vague and non-falsifiable. And you have to get a lot more than what you put into your theory. Regardless of how much you want to call it science, that is just pseudoscience.
I personally really don't want to be associated with such "scientists". Which is why I think twice using the word science/scientist and tend to use the word physics/physicists. All thanks to scientists who are able to think out-of-the-box, unlike the narrow-minded stupid physicists who don't know anything but scientific method!
