Social science - Topics

Gödel showed that any consistent formal system contains true statements it cannot prove — certainty has a ceiling and we found it from inside A proof settles a question in a way no experiment ever has to — mathematics is the only field where certainty comes standard What is the difference between social science and humanities? The Tacoma Narrows collapse taught engineers more about resonance than any wind tunnel study before it Statistics is math that makes peace with uncertainty — the branch where being approximately right beats being precisely wrong Goodhart's law: when a measure becomes a target it ceases to be a good measure — the deepest fact social science learned about institutions Safety factors and rules of thumb encode a century of failure — the know-how inside engineering codes was bought with collapsed bridges The Riemann hypothesis: the most famous unsolved problem in math — worth a million dollars and far more in borrowed credibility You cannot randomize a revolution — causal claims in social science are built on natural experiments and creative proxy designs instead Engineering is science with a deadline — a bridge must stand whether or not the theory behind it is elegant Quantitative vs qualitative methods is a fake war — numbers tell you how much while interviews tell you why — most real questions need both The replication crisis started in psychology and spread across the social sciences — half the famous findings did not survive being rerun Language shapes cognition more than grammar books admit and less than Sapir-Whorf fans hope Eyewitness memory is the most confident and least reliable evidence in court — confidence tracks certainty of belief and not accuracy of recall Hawthorne effect: people change behavior when they know they are observed — the observer is part of the experiment whether you like it or not Economics has math envy and sociology has physics envy — both borrow methods from fields whose objects misbehave less The WEIRD problem: most psychology findings come from Western educated industrialized rich democratic samples and get generalized to everyone Self-report is the social scientist's microscope and its broken lens — people cannot accurately report why they did something Algorithmic feeds replace editors with engagement metrics — what you see is now an optimization output rather than a judgment Sociology and economics look at the same problem and refuse each other's evidence Applied math borrows certainty from pure math and pays a tax at the modeling step — the map is where the doubt enters Wikipedia is more reliable than most encyclopedias because its errors get fixed in hours and not in editions T-shaped skills: depth in one thing and breadth in many — the shape consultancies drew and everyone else copied Why survey weighting matters more than sample size How many hours of contact it takes to turn a stranger into a friend