There are many who do not think the social sciences—including economics—should be considered to be sciences. They do so for a variety of reasons.
One objection I keep seeing is that because you cannot make accurate predictions within the social sciences, they cannot be sciences. By this logic, biology isn’t a science, either, because you cannot make predictions in biology. Some of the earliest work in genetics fooled many people into believing you could, but it turns out that 1:1 gene:expression is extremely rare. Mendel lucked out with his peas. Most traits are expressed in rather complex ways, within the context of the other genes, epigenetic effects, and the environment. This generally renders prediction impossible.
Another similar objection is that the social sciences are too complex to understand. But the presence of complexity doesn’t mean understanding is impossible. It is possible to discover patterns and emergent laws in a complex system. In economics, we have discovered the law of supply and demand. It’s always and everywhere true—all other things held constant. The fact that different contexts will affect the degree of elasticity or overwhelm the effect or require major digging down to see how it applies in no way means supply and demand isn’t a law of economics. Complexity actually requires the emergence of laws for there to be complexity. Otherwise, you don’t have complexity, you simply have randomness.
Too many others simply reject the social sciences as sciences because the discoveries don’t fit their ideologies. In this they are like those creationists who reject evolution as science (declaring as “merely a theory”). Just because you want to increase the minimum wage because you think it’s “fair” to do so, that doesn’t mean that the economic law of supply and demand is bunk. If you increase the minimum wage, you will increase unemployment (again, everything else being held constant).
Just because the social sciences aren’t simple like physic and chemistry, that doesn’t mean they’re not sciences. They are sciences. They are complexity sciences. And in many ways, chemistry and physics are only recently starting to catch up and describe many of their processes in similar ways. Does the unpredictablility of the weather over the long term mean meteorology isn’t a science? It would be foolish to claim so. Meteorology is a complexity science. Because it deals with a relatively small number of variables and known laws, we have a fairly high level of predictability. But even so, the weather person can be off even the next day.
We have to get our minds around the fact that there are complexity sciences out there. When we study them, we are trying to work out their emergent laws. The fact that the processes we are studying are complex doesn’t mean they don’t have laws. That’s an important point to understand. Those laws do exist, and those laws interact with each other to create emergent patterns of behavior. In other words, precise predictions are impossible in complex systems, but pattern predictions most certainly are not. We have to get past our demands for absolute, clear, black-or-white answers and embrace instead strange attractor, edge-of-order-and-chaos, ambiguous answers. Because those answers are also true.