Not “the soft option” — a set of disciplines with their own empirical methodology for studying something genuinely harder to measure than physical matter: human behaviour at scale.

The popular STEM-versus-social-sciences framing treats the second half as less rigorous, mostly because its subject — human behaviour, meaning, and social structure — resists the clean, repeatable measurement physical sciences take for granted. That’s a claim about difficulty of subject matter, not absence of method: the social sciences have their own developed empirical and statistical apparatus, refined specifically for the messier, harder-to-isolate variables human behaviour actually presents.

Source: Bhattacherjee, A. et al. Social Science Research: Principles, Methods and Practices, open textbook, 2019.