Critical Thinking for the Information Age презентация

validity
If you’ve had one course spend a few minutes looking at material to see if it’s helpful
If you’ve had several courses in statistics it makes sense to go straight to Lesson 2
No math
No math

a function of their size
E.g., your judgment about Bill’s honesty or the quality of food in the new restaurant becomes more accurate the more evidence you have
This is especially true when there’s a lot of error in your sample
The big problem to be overcome: recognizing that there is error

two variables
E.g., mothers’ height and daughters’ height, IQ and income
Difficult to detect some correlations
Worse: we detect lots of correlations that aren’t really there
You need protection from the media – because many reporters don’t understand these principles

correlational evidence
The concept of natural experiments:
E.g., the town without toothache
How to do experiments on yourself
E.g., does coffee make you more or less efficient?
The terrific costs society pays for the experiments it doesn’t do

rare, the next value you encounter is probably going to be less extreme
E.g., Joan is probably not going to be as extraordinarily generous the next time you see her
Concept of base rate
Predictions about a case should take into account what other similar cases are like

lack some important concepts
Illusion of Objectivity
Fundamental Attribution Error
Heuristics: rules of thumb that can lead us astray when assessing probability and causality
Confirmation bias: when testing hypotheses we tend to look only for evidence that would be supportive, not for equally valuable evidence that might be contradictory

Costs
How to avoid taking actions that make potentially more valuable actions impossible
Sunk Costs
How to avoid carrying out an action for no better reason than that you paid to do it

Resolving Contradictory Propositions

York: Farrar, Straus & Giroux
Available in paperback, Kindle and audio