| Chapter 3 People are generally overconfident by 10-20% People get tired, have other things on their mind, etc => are not consistent Statistical models operate relentlessly and consistently => better predictors Experts are overconfident about their abilities to predict Most people look only for confirming evidence The only way to falsify an "if X then Y" statement is to find "X and not Y" In science we always have to look for disconfirming evidence Self-fulling prophecies: People receive confirming feedback for their misconceptions because they only look for confirming evidence We should try to get rid of biases by sensitizing people to the possibility that prejudice may be present We can frame questions in a way that discourages confirming answers Self-negating prophecies Holding on to irrational beliefs prevents people from behaving in rational ways We should ask experts what variables we should take into account when making a judgment but we should let a mechanical model combine the variables in a consistent manner. |
 |