Factors Affecting Statistical Reasoning
The application of statistical heuristics can be hindered by biases such as base rate neglect and sample size neglect.
THE IDEA
FACTORS THAT AFFECT STATISTICAL REASONING
What factors make it difficult to apply statistical heuristics when these are required, and what factors can make it easier?
When to use it
Evaluating the accuracy of medical tests and interpreting results
Making investment decisions based on historical market data
Assessing the effectiveness of a new policy or intervention from small pilot studies
Interpreting survey results or polling data with appropriate margins of error
FROM THEORY TO EVERYDAY LIFE
In practice
Base Rate Neglect in Medical Diagnosis
A doctor receives a positive test result for a rare disease. Despite the disease's low base rate (0.1%), and even though the test is 99% accurate, the doctor concludes the patient likely has the disease, ignoring the high probability of a false positive due to the base rate.
Sample Size Neglect
A manager observes that a small team of 5 employees has a higher productivity rate than the company average, and attributes this to a new management style, without considering that small samples are more variable and likely to regress to the mean.
A MOMENT TO REFLECT
Put it to work.
Think of a recent decision you made that involved probabilities. Did you consider the base rate? Why or why not?
How might presenting statistical information as natural frequencies (e.g., '10 out of 1000') instead of percentages affect your understanding?
In what situations do you find it hardest to rely on statistical reasoning instead of intuition?
CONTINUE THE CONVERSATION
Source texts
- Heuristics and Biases The Psychology of Intuitive Judgment (Thomas Gilovich, Dale Griffin, Daniel Kahneman)