You may have heard the natural sciences (chemistry, physics, biology) called the hard sciences and the social sciences (history, political science) referred to as the soft sciences. Certain sciences, such as psychology and sociology, used to be considered soft science and now are considered hard science. So, you may be wondering about the difference between hard and soft science.It isn't that one type of science is more rigorous or challenging than the other. For example, while you may consider chemistry harder to understand than anthropology, that does not have anything to do with chemistry being considered a hard science and anthropology being a soft science. Rather, it has to do with experimental design and the scientific method. Hard science involves experiments where it is relatively easy to set up controlled variables and make objective measurements. Particularly in sciences dealing with people, it may be difficult to isolate all the variables that may influence an outcome. In some cases, controlling the variable may even alter the results! Simply put, it is harder to devise an experiment in a soft science.
The terms "hard science" and "soft science" are used less often than they used to be, in part because the terminology is misunderstood and therefore offensive. People perceive "harder" to imply "more difficult" when it may be much more challenging to devise and interpret an experiment in a so-called soft science than a hard science. The distinction between the two types of science is a matter of the how strongly you can state, test and then accept or reject the hypothesis. In the modern world, the degree of difficulty is less related to discipline than it is to the specific question, so one might say the terms "hard science" and "soft science" are outdated. What do you think?
What Is an Experiment? | What Is the Difference Between a Control Group and Experimental Group?

Comments
Is there any reference point, other than the scientific method, for sciences which seem softer than chemistry and physics? If a discipline cannot fulfill the requirements of the scientific method, can it be science at all? Maybe such disciplines need to find another term.