Correlation vs Causation: What's the Difference?
Correlation means two things move together; causation means one makes the other happen.
How we think: biases, mental models, fallacies, numbers that mislead, and how to check what you read before you believe or share it.
HACKS VITAE ARTWORK (AI-ASSISTED)
Mindset matters because a settled way of seeing a problem can steer what you notice and what you try.
Read it →Correlation means two things move together; causation means one makes the other happen.
A randomised trial lets chance decide who gets which treatment; an observational study watches without assigning; a meta-analysis combines the results of existing studies.
Factfulness argues that most of us carry an out-of-date, too-gloomy picture of the world, and names ten instincts that keep it that way.
Nicholas Carr argues that the internet is changing how we think, training attention to scatter.
Emotional triggers such as fear, scarcity, urgency and social proof make you feel something so that you act.
Our reading: give most weight to sources that lose something when they are wrong, or have a track record you can check.
Carl Sagan's 1995 book argues that science is a way of thinking, not just a body of knowledge, and that a public without it is easy prey for pseudoscience.
Critical thinking is reasonable, reflective thinking about what to believe or do, and AI raises the stakes because people tend to over-rely on confident machines.
Influence argues that people often say yes on a single cue, and that persuaders lean on six such cues; the 2021 edition adds a seventh, unity.
Mindset argues that people differ in whether they see ability as fixed or as something that grows, and that this belief shapes how they meet setbacks.
Nudge argues that no choice is ever presented in a neutral way, so whoever designs it can steer people towards better options without banning anything.
Adam Grant argues that the ability to rethink may matter more than raw intelligence, and that we would do better to treat our views as hypotheses, as scientists do.
Thinking, Fast and Slow describes two ways of thinking, fast and automatic (System 1) and slow and effortful (System 2), which Kahneman presents as a metaphor.
We traced seven tools to where they started; the Kelly criterion began as Kelly's 1956 paper about information, not advice.
Each of the 52 thinking tools gets a verdict and a source: 11 true, 30 partly true, 7 disputed, 2 unverified and 2 myths.
Several famous stories are not in the sources the way they are told: Cortés wrote that he had his ships beached, not burned.
"Mental models" means two different things: in psychology it goes back to Kenneth Craik (1943), and the self-help sense is Charlie Munger's, from a 1994 talk.
"Scarcity mindset" mixes two unrelated ideas: a self-help attitude from The 7 Habits of Highly Effective People and research on how shortage takes up attention.
Of the fourteen findings, two held up cleanly, seven are real but smaller or narrower than the famous version, four are genuinely disputed, and one has not been confirmed.
Use four moves, called SIFT: Stop, Investigate the source, Find better coverage, and Trace claims to their original context.
Ask nine questions, starting with the methods, not the abstract: study type, peer review, sample size, p-values, effect size, replication, funding and retractions.
Go back to the original: for a quote, search digitised book archives for the earliest dated printing; for a source, check that it exists and says what it is claimed to say.
A fallacy is a flaw in reasoning, not a false conclusion, so naming one never proves the other side wrong.
Misleading numbers are usually true figures missing the comparison, base or sample that would change their meaning.
Of fourteen popular claims, nine are myths, three partly true, one cannot be settled and one holds up.
Of twelve claims, four animal lessons are myths, the boiling frog is disputed, one is a Game of Thrones line, and one shop trick, the 9-ending price, holds up.
Dark psychology is a popular label for the ways people manipulate and control others, not a research field.
Survivorship bias is drawing conclusions from the cases that made it through, while the failures have dropped out of view.