Judgements Flashcards
Misperceptions of Randomness
Randomness does not imply uniformity
Heuristics 1: Anchoring & Adjustment
A wheel of fortune is rigged to stop at either 10 or 65
People are asked to read the result, decide whether the proportion of nations in the UN that are African is larger or smaller than the number, and then to make a numerical estimate
10 -> 25%
65 -> 45%
Currently, the correct figure is just under 30%, and was similar when the experiment was carried out.
Adjustment
Adjustment can occur as a deliberate process (system 2)
But also subject to unconscious influences
Epley and Gilovich – interference from other processes; shaking head leads to moving further from anchor, nodding head to less
There is also evidence of a priming (system 1) component to anchoring via evoked images
Heuristic 2: Availability
Examples in which availability misleads:
Do more English words have “R” as the first or as the third letter?
Famous name demonstration. Subjects believe that gender-balanced lists contain more women (men) if the women (men) are more famous.
Ease of recall is often good clue to probability - but not always.
Heuristic 3: Representativeness
Tom W experiments
Three tasks:
rank base rates for graduate students in 9 areas of study
Rank each area for how well a description (old, and based on poorly validated tests) of Tom W fits typical graduate student in that area
Rank areas for how likely Tom W is now a graduate student in that area
Results for Tasks 2 and 3 are very similar, suggesting substitution of question
The two problems with using Representativeness
Ignoring base rates
Computer science graduate students were relatively rare when the experiments were first carried out
Using poor (or even useless) information
The information in the description of Tom W is described as old, and based on “psychological tests of uncertain validity”
Regression to the mean
We have very poor intuitions about this phenomenon, and don’t readily recognise it in the real world
Outcome bias
Events (and the choices that led to them happening) are judged by their actual outcomes, not by whether the decisions were good ones when they were made
The Illusion of Validity
This feedback did not affect the confidence the assessors felt in making later judgements.
Confidence reflects coherence and ease of making a judgement
The Illusion of Skill Some “Experts” are not expert
Barber and Odean (2002) – analysis of about 163,000 stock trades by brokers
On average the bought stock did WORSE than the sold stock by 3.2%
Brokers who made least trades tended to do better
Year on year correlations in success were low to zero
