14 - Chapter 14: Tom W’s Specialty

Chapter 14: Tom W’s Specialty

• an easy q., no specific info abt Tom, so just consult [proportion of students in each area (base rates)

• but everything changes w.  the following personality sketch of Tom

• first given a sketch of Tom’s personality, we think of stereotypes, probably comp. sci. for Tom — S1 was activated by various hints to invoke stereotype — desc. deliberately aims at minor fields of study (comp. sci, librarian, engineer), poor fit for more popular fields —  i.e. an anti-base-rate” description — notice that source of desc. is said to be not v. trustworthy

Predicting by Representativeness

• representativeness = similarity (S1) to stereotypes — focus on fitness of sketch w. stereotypes, ignore base rates — happens even w. grad. psych. students or stat’cians who know relevant base rates, know sketch is not v. reliable
• substit’n of similarity (easy) for probability (difficult)
• representativeness vs. base rates
• if only judging similarity, OK to ignore base rates, accuracy of desc.
• but ignoring base rates & quality of evidence in probability assessments à mistakes
• for the public probability is a vague notion (cf. scientist’s precise idea), evokes S1’s mental shotgun, answers to easier q’s
• S1 assesses representativeness easily — e.g. He looks like a leader
• cf. book/movie Moneyball, rep. vs stats

The Sins of Representativeness

• judging probability by the rep’ness heuristic (stereotypes) often works — e.g. people who look friendly usually are, stereotypes hv some truth — but stero. sometimes false,  result in neglect of base rates
• sin of representativeness #1 — too willing to predict occurrence of unlikely (low base-rate) events — e.g. a person reading NY Times on NY subway. Which more likely? Has a PhD or no college degree — base rate info: more of the latter ride subway than former — usually ignore base rates when hv info abt indiv. case, improves when S2 is activated — ignoring base rates ß ignorance or laziness
• sin of representativeness #2 — neglect quality of evidence  —psych’l sketch still infl. judg’t even knowing it is unreliable (WYSIATI)
• Note: when quality of evidence is in doubt, stay close to base rate (stat’s)

How To Discipline Intuition

• Bayesian statistics — math’l rules that gov. how we shd alter assessments (prior beliefs, base rates) in light of evidence
• Don’t believe everythg that comes to mind — base rates are impt, even w. add’l evidence abt current case
• keys  to Bayesian reasoning:
• 1. anchor judg’t of probabilty on plausible base rate
• 2. question value of  evidence — not easy to do

Speaking of Representativeness

• “The lawn is well trimmed, the receptionist looks competent, and the furniture is attractive, but this doesn’t mean it is a well-managed company. I hope the board does not go by representativeness.”
• “This start-up looks as if it could not fail, but the base rate of success in the industry is extremely low. How do we know this case is different?”
• “They keep making the same mistake: predicting rare events from weak evidence. When the evidence is weak, one should stick with the base rates.”
• “I know this report is absolutely damning, and it may be based on solid evidence, but how sure are we? We must allow for that uncertainty in our thinking.”