Doing sociology is hard and physics is, in some ways, absurdly easy. As a physicist, you can generate an endless supply of particles. They are all identical, and so behave in the same way. They do whatever they do, regardless of the experimenters' expectations, social pressures, or the like. The results can be slotted into a formal mathematical framework, which also helps you design better experiments. The work of these experiments is often difficult (e.g., holding a detector just about 0K), but the concepts are often relatively clear.
None of this is true in 'softer' sciences. Every person (or animal) is different and you often can't even test the same subject twice because the very act of doing so changes them. People often don't know what they want or why they do something. When they do know, they sometimes lie about it, for myriad reasons. This makes designing, analyzing, and interpreting experiments conceptually hard, even if the actual work is pretty easy.